The European Nitrogen Assessment: Sources, Effects and Policy Perspectives

This page intentionally left blank The European Nitrogen Assessment Sources, Effects and Policy Perspectives A centur...

Author: Mark A. Sutton | Clare M. Howard | Jan Willem Erisman | Gilles Billen | Albert Bleeker | Peringe Grennfelt | Hans van Grinsven | Bruna Grizzetti

18 downloads 1365 Views 29MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

This page intentionally left blank

The European Nitrogen Assessment Sources, Effects and Policy Perspectives A century ago, when the world depended on fossil nitrogen and manure recycling, there was insufficient reactive nitrogen to feed the growing human population. With the invention of the Haber–Bosch process, humans found a way to make cheap reactive nitrogen from the almost inexhaustable supply of atmospheric di-nitrogen. What humans did not anticipate was that the massive increase in reactive nitrogen supply, exacerbated by fossil fuel burning, would lead to a web of new environmental problems cutting across all global-change challenges. The European Nitrogen Assessment presents the first full, continental-scale assessment of reactive nitrogen in the environment and sets the problem in context by providing a multidisciplinary introduction to the key processes in the nitrogen cycle. Issues of up-scaling from field, farm and city to national and continental scales are addressed in detail with emphasis on opportunities for better management at local to global levels. A comprehensive series of maps showing nitrogen pools and fluxes across Europe also highlight the location of the major threats and allow a comparison of national budgets for the first time. Five key societal threats posed by reactive nitrogen are assessed, providing a framework for a set of policies that can be used for joined-up management of the nitrogen cycle in Europe. This includes the first cost–benefit analysis for different reactive nitrogen forms and consideration of future scenarios. Incorporating a handy technical synopsis and summary for policy makers, this land-mark volume is an essential reference for academic researchers across a wide range of disciplines, asÂ€well as for stakeholders and policy makers in Europe and beyond. It is also a valuable tool in helping communicate the key environmental issues and future challenges to the wider public. Mark Sutton is an environmental physicist investigating human alteration of the nitrogen cycle, with specific attention to ammonia. He is coordinator of the major integrated project ‘NitroEurope’, a 5-year effort, bringing together 64 research institutes to ask how nitrogen is affecting the European greenhouse gas balance. Dr Sutton is vice-chair of the ‘Nitrogen in Europe’ (NinE) programme of the European Science Foundation, the Director of the European Centre of the International Nitrogen Initiative (INI) and co-chair of the Task Force on Reactive Nitrogen of the UN-ECE Convention on Long-range Transboundary Air Pollution. Clare Howard is currently engaged in a postdoctoral fellowship in knowledge transfer, with an emphasis on research networks which focus on nitrogen. Dr Howard is project coordinator for the European Nitrogen Assessment and for the Task Force on Reactive Nitrogen, which sits beneath the Working Group on Strategies and Review of the Convention on Long Range Transboundary Air Pollution. Her research interests involve the modelling of biogeochemical cycles of nitrogen and carbon and assessing uncertainty in model systems. Jan Willem Erisman heads the Biomass, Coal and Environmental Research Unit of the Energy Research Centre of the Netherlands (ECN) and is a professor in Integrated Nitrogen studies at Vrije Universiteit, Amsterdam. His research focuses on atmosphere–biosphere exchange of gases and aerosols related to acidification and eutrophication and climate change. He was instrumental in establishing the International Nitrogen Initiative, the Nanjing Declaration on Nitrogen Management, the EU 6th Framework research program NitroEurope and for chairing the European Science Foundation project NinE and the EU COST Action 729.

Gilles Billen is research director of the Centre National de la Recherche Scientifique (CNRS) at the University Pierre and Marie Curie (Paris) where his research covers many aspects of biogeochemistry, with an emphasis on the nitrogen, phosphorus and silica cycles. His main expertise is on the assessment and modelling of the ecological functioning of hydrosystems, including marine, estuarine and freshwater environments. From 1997 to 2007, he was the Director of the PIREN-Seine programme, a large interdisciplinary research programme on the Seine river watershed. Albert Bleeker works as a senior scientist at the Energy Research Centre of the Netherlands, in the department of Air Quality and Climate Change. He has almost 20 years of experience in the field of nitrogen, where his main expertise is on the atmospheric emission, transport and deposition of nitrogen at various spatial scales, as well as studies on the effect of nitrogen in the natural environment. Currently, he is the Nitrogen in Europe (NinE) Programme Co-ordinator and a member of the COST 729 Management Committee. Peringe Grennfelt has a background in atmospheric chemistry. His research includes regional air pollution problems in Europe, in particular acidification, nitrogen deposition and tropospheric ozone. He has coordinated several national and international research programmes including the EU project Network for the support of European Policies on Air Pollution (NEPAP). He is presently leading the Mistra Climate Policy Research Programme (Clipore) and the Swedish Clean Air Research Programme (SCARP). Hans van Grinsven works at the Netherlands Environmental Assessment Agency where he conducts research and coordinates projects related to agriculture and environment, focusing on nitrogen and phosphorus, and sustainable food production. Dr van Grinsven was responsible for evaluations of national implementation of the EU Nitrates Directive and was also closely involved in the evaluations of the implementation of the EU Water framework Directive and EU NEC directive. Bruna Grizzetti is a researcher in the field of large scale modelling of nutrient and water transfer. She works on modelling nutrient pressures on water at European scale in support to the implementation of environmental European policies, such as the Water Framework Directive, Nitrates Directive and the Marine Strategy. Since 2007, Dr Grizzetti has been a member of the Coordination Team of the European Nitrogen Assessment process, supported through the European Science Foundation.

TFRN

The European Nitrogen Assessment has been prepared through coordinated action led by the Nitrogen in Europe (NinE) Research Networking Programme of the European Science Foundation, the NitroEurope Integrated Project supported by European Commission’s 6th Framework Programme and the COST Action 729. The Assessment is a contribution to the work of the Task Force on Reactive Nitrogen (TFRN), led by the UK and the Netherlands, in support of the long-term goals of the UN-ECE Convention on Long-range Transboundary Air Pollution (CLRTAP). In parallel, the Assessment represents a European contribution to the work of the International Nitrogen Initiative (INI), a joint project of the International Geosphere Biosphere Programme (IGBP) and the Scientific Committee on Problems of the Environment (SCOPE), providing evidence to underpin many United Nations and other multi-lateral agreements. The actual assessment work has been carried out by 200 experts from 21 countries and 89 organizations which kindly provided support for this work. The ENA has been conducted as a scientifically independent process. The views and conclusions expressed are those of the authors, and do not necessarily reflect policies of the contributing organizations.

Acknowledgements

The European Nitrogen Assessment was prepared by the list of contributors given on page ix, with the support of the NinE Programme of the European Science Foundation, theÂ€ NitroEurope IP (funded by the European Commission 6th Framework Programme), the COST Action 729, the Task Force on Reactive Nitrogen and the International Nitrogen Initiative. The editors gratefully acknowledge the wider support which the assessment received, in the form of all those attending and hosting the ENA workshops, internal and

external reviewers of chapters and summaries. We particularly thank Agnieszka Eljasz of CEH, Susan Francis and Laura Clark of Cambridge University Press, Ellen Degottv.W.Rekowski and Paola Campus of the European Science Foundation, Peter Coleman of Defra, Anastasios Kentarchos of the European Commission, Matti Johannsson and Tea Aulavuo of the Secretariat to the UN-ECE Convention on Long-range Transboundary Air Pollution for their support through this process.

The European Nitrogen Assessment Sources, Effects and Policy Perspectives Edited by

Mark A. Sutton

NERC Centre for Ecology and Hydrology

Clare M. Howard

NERC Centre for Ecology and Hydrology and University of Edinburgh

Jan Willem Erisman Energy Research Centre of the Netherlands

Gilles Billen

CNRS and University of Paris VI

Albert Bleeker

Energy Research Centre of the Netherlands

Peringe Grennfelt

Swedish Environmental Research Institute (IVL)

Hans van Grinsven

PBL Netherlands Environmental Assessment Agency

Bruna Grizzetti

European Commission Joint Research Centre

C A M B R I D G E U N I V E R SI T Y P R E S S Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Tokyo, Mexico City Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title:Â€www.cambridge.org/9781107006126 © Cambridge University Press 2011 © Editorial contributions by Bruna Grizzetti, European Union 2011 © Chapter 17, European Union 2011 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2011 Printed in the United Kingdom at the University Press, Cambridge A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication Data The European nitrogen assessmentâ•›:â•›sources, effects, and policy perspectivesâ•›/â•›[edited by] Mark A. Sutton ... [et al.]. â•…â•… p.â•… cm. Includes bibliographical references and index. ISBN 978-1-107-00612-6 (hardback) 1.â•‡ Nitrogen compounds–Environmental aspects–Europe. 2.â•‡ Nitrogen cycle–Europe.â•… 3.â•‡ Nitrogen fertilizers–Government policy–Europe.â•… I.â•‡ Sutton, Mark A. TD196.N55E96 2011 363.738–dc22â•…â•…â•… 2010051120 ISBN 978-1-107-00612-6 Hardback Additional resources for this publication at www.cambridge.org/ena Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents List of contributorsâ•¯â•¯â•¯â•¯â•¯â•¯page xi Forewordâ•¯â•¯â•¯â•¯â•¯â•¯xxiii Summary for policy makersâ•¯â•¯â•¯â•¯â•¯â•¯xxiv Technical summaryâ•¯â•¯â•¯â•¯â•¯â•¯xxxv

1 Assessing our nitrogen inheritanceâ•¯â•¯â•¯â•¯â•¯â•¯1 Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti

Part Iâ•‡ Nitrogen in Europe:Â€the present positionâ•¯â•¯â•¯â•¯â•¯â•¯ 2 The European nitrogen problem in a global perspectiveâ•¯â•¯â•¯â•¯â•¯â•¯9 Jan Willem Erisman, Hans van Grinsven, Bruna Grizzetti, Fayçal Bouraoui, David Powlson, Mark A. Sutton, Albert Bleeker and Stefan Reis 3 Benefits of nitrogen for food, fibre and industrial productionâ•¯â•¯â•¯â•¯â•¯â•¯32 Lars Stoumann Jensen, Jan K. Schjoerring, Klaas W. van der Hoek, Hanne Damgaard Poulsen, John F. Zevenbergen, Christian Pallière, Joachim Lammel, Frank Brentrup, Age W. Jongbloed, Jaap Willems and Hans van Grinsven 4 Nitrogen in current European policiesâ•¯â•¯â•¯â•¯â•¯â•¯62 Oene Oenema, Albert Bleeker, Nils Axel Braathen, Michaela Budňáková, Keith Bull, Pavel Čermák, Markus Geupel, Kevin Hicks, Robert Hoft, Natalia Kozlova, Adrian Leip, Till Spranger, Laura Valli, Gerard Velthof and Wilfried Winiwarter 5 The challenge to integrate nitrogen science and policies:Â€the European Nitrogen Assessment approachâ•¯â•¯â•¯â•¯â•¯â•¯82 Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, William J. Bealey, Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti

Part IIâ•‡ Nitrogen processing in the biosphere 6 Nitrogen processes in terrestrial ecosystemsâ•¯â•¯â•¯â•¯â•¯â•¯99 Klaus Butterbach-Bahl, Per Gundersen, Per Ambus, Jürgen Augustin, Claus Beier, Pascal Boeckx, Michael

Dannenmann, Benjamin Sanchez Gimeno, Ralf Kiese, Barbara Kitzler, Andreas Ibrom, Robert M. Rees, Keith A. Smith, Carly Stevens, Timo Vesala and Sophie Zechmeister-Boltenstern 7 Nitrogen processes in aquatic ecosystemsâ•¯â•¯â•¯â•¯â•¯â•¯126 Patrick Durand, Lutz Breuer, Penny J. Johnes, Gilles Billen, Andrea Butturini, Gilles Pinay, Hans van Grinsven, Josette Garnier, Michael Rivett, David S. Reay, Chris Curtis, Jan Siemens, Stephen Maberly, Øyvind Kaste, Christoph Humborg, Roos Loeb, Jeroen de Klein, Josef Hejzlar, Nikos Skoulikidis, Pirkko Kortelainen, Ahti Lepistö and Richard Wright 8 Nitrogen processes in coastal and marine ecosystemsâ•¯â•¯â•¯â•¯â•¯â•¯147 Maren Voss, Alex Baker, Hermann W. Bange, Daniel Conley, Sarah Cornell, Barbara Deutsch, Anja Engel, Raja Ganeshram, Josette Garnier, Ana-Stiina Heiskanen, Tim Jickells, Christiane Lancelot, Abigail McQuatters-Gollop, Jack Middelburg, Doris Schiedek, Caroline P. Slomp and Daniel P. Conley 9 Nitrogen processes in the atmosphereâ•¯â•¯â•¯â•¯â•¯â•¯177 Ole Hertel, Stefan Reis, Carsten Ambelas Skjøth, Albert Bleeker, Roy Harrison, John Neil Cape, David Fowler, Ute Skiba, David Simpson, Tim Jickells, Alex Baker, Markku Kulmala, Steen Gyldenkærne, Lise Lotte Sørensen and Jan Willem Erisman

Part IIIâ•‡ Nitrogen flows and fate at multiple spatial scales 10 Nitrogen flows in farming systems across Europeâ•¯â•¯â•¯â•¯â•¯â•¯211 Steve Jarvis, Nick Hutchings, Frank Brentrup, Jørgen Eivind Olesen and Klaas W. van der Hoek 11 Nitrogen flows and fate in rural landscapesâ•¯â•¯â•¯â•¯â•¯â•¯229 Pierre Cellier, Patrick Durand, Nick Hutchings, Ulli Dragosits, Mark Theobald, Jean-Louis Drouet,

vii

Contents

OeneÂ€Oenema, Albert Bleeker, Lutz Breuer, Tommy Dalgaard, Sylvia Duretz, Johannes Kros, Benjamin Loubet, Joergen Eivind Olesen, Philippe Mérot, Valérie Viaud, Wim de Vries and Mark A. Sutton

18 Nitrogen as a threat to European air qualityâ•¯â•¯â•¯â•¯â•¯â•¯405 Jana Moldanová, Peringe Grennfelt, Åsa Jonsson, David Simpson, Till Spranger, Wenche Aas, John Munthe and AriÂ€Rabl

12 Nitrogen flows and fate in urban landscapesâ•¯â•¯â•¯â•¯â•¯â•¯249 Anastasia Svirejeva-Hopkins, Stefan Reis, Jakob Magid, Gabriela B. Nardoto, Sabine Barles, Alexander F. Bouwman, Ipek Erzi, Marina Kousoulidou, Clare M. Howard and Mark A. Sutton

19 Nitrogen as a threat to the European greenhouse balanceâ•¯â•¯â•¯â•¯â•¯â•¯434 Klaus Butterbach-Bahl, Eiko Nemitz, Sönke Zaehle, Gilles Billen, Pascal Boeckx, Jan Willem Erisman, JosetteÂ€Garnier, Rob Upstill-Goddard, Michael Kreuzer, Oene Oenema, Stefan Reis, Martijn Schaap, David Simpson, Wim de Vries, Wilfried Winiwarter and Mark A. Sutton

13 Nitrogen flows from European regional watersheds to coastal marine watersâ•¯â•¯â•¯â•¯â•¯â•¯271 Gilles Billen, Marie Silvestre, Bruna Grizzetti, Adrian Leip, Josette Garnier, Maren Voss, Robert Howarth, Fayçal Bouraoui, Ahti Lepistö, Pirkko Kortelainen, Penny Johnes, Chris Curtis, Christoph Humborg, Erik Smedberg, Øyvind Kaste, Raja Ganeshram, Arthur Beusen and Christiane Lancelot 14 Atmospheric transport and deposition of reactive nitrogen in Europeâ•¯â•¯â•¯â•¯â•¯â•¯298 David Simpson, Wenche Aas, Jerzy Bartnicki, Haldis Berge, Albert Bleeker, Kees Cuvelier, Frank Dentener, Tony Dore, Jan Willem Erisman, Hilde Fagerli, Chris Flechard, Ole Hertel, Hans van Jaarsveld, Mike Jenkin, Martijn Schaap, Valiyaveetil Shamsudheen Semeena, Philippe Thunis, Robert Vautard and Massimo Vieno 15 Geographical variation in terrestrial nitrogen budgets across Europeâ•¯â•¯â•¯â•¯â•¯â•¯317 Wim de Vries, Adrian Leip, Gert Jan Reinds, Johannes Kros, Jan Peter Lesschen, Alexander F. Bouwman, Bruna Grizzetti, Fayçal Bouraoui, Klaus Butterbach-Bahl, Peter Bergamaschi and Wilfried Winiwarter 16 Integrating nitrogen fluxes at the European scaleâ•¯â•¯â•¯â•¯â•¯â•¯345 Adrian Leip, Beat Achermann, Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Wim de Vries, Ulli Dragosits, Ulrike Döring, Dave Fernall, Markus Geupel, Jürg Herolstab, Penny Johnes, Anne-Christine Le Gall, Suvi Monni, Rostislav Nevečeřal, Lorenzo Orlandini, Michel Prud’homme, Hannes I. Reuter, David Simpson, Günther Seufert, Till Spranger, Mark A. Sutton, John van Aardenne, Maren Voß and Wilfried Winiwarter

Part IVâ•‡ Managing nitrogen in relation to key societal threats 17 Nitrogen as a threat to European water qualityâ•¯â•¯â•¯â•¯â•¯â•¯379 Bruna Grizzetti, Fayçal Bouraoui, Gilles Billen, Hans van Grinsven, Ana Cristina Cardoso, Vincent Thieu, Josette Garnier, Chris Curtis, Robert Howarth and Penny Johnes

viii

20 Nitrogen as a threat to European terrestrial biodiversityâ•¯â•¯â•¯â•¯â•¯â•¯463 Nancy B. Dise, Michael Ashmore, Salim Belyazid, Albert Bleeker, Roland Bobbink, Wim de Vries, Jan Willem Erisman, Till Spranger, Carly J. Stevens and Leon van den Berg 21 Nitrogen as a threat to European soil qualityâ•¯â•¯â•¯â•¯â•¯â•¯495 Gerard Velthof, Sébastien Barot, Jaap Bloem, Klaus Butterbach-Bahl, Wim de Vries, Johannes Kros, Patrick Lavelle, Jørgen Eivind Olesen and Oene Oenema

Part Vâ•‡ European nitrogen policies and future challenges 22 Costs and benefits of nitrogen in the environmentâ•¯â•¯â•¯â•¯â•¯â•¯513 Corjan Brink, Hans van Grinsven, Brian H. Jacobsen, Ari Rabl, Ing-Marie Gren, Mike Holland, Zbigniew Klimont, Kevin Hicks, Roy Brouwer, Roald Dickens, Jaap Willems, Mette Termansen, Gerard Velthof, Rob Alkemade, Mark van Oorschot and Jim Webb 23 Developing integrated approaches to nitrogen managementâ•¯â•¯â•¯â•¯â•¯â•¯541 Oene Oenema, Joost Salomez, Christina Branquinho, Michaela Budňáková, Pavel Čermák, Markus Geupel, Penny Johnes, Chris Tompkins, Till Spranger, Jan Willem Erisman, Christian Pallière, Luc Maene, Rocio Alonso, Rob Maas, Jacob Magid, Mark A. Sutton and Hans van Grinsven 24 Future scenarios of nitrogen in Europeâ•¯â•¯â•¯â•¯â•¯â•¯ 551 Wilfried Winiwarter, Jean-Paul Hettelingh, Alex F. Bouwman, Wim de Vries, Jan Willem Erisman, James Galloway, Zbigniew Klimont, Allison Leach, Adrian Leip, Christian Pallière, Uwe A. Schneider, Till Spranger, Mark A. Sutton, Anastasia Svirejeva-Hopkins, Klaas W. van der Hoek and Peter Witzke

Contents

25 Coordinating European nitrogen policies between international conventions and intergovernmental organizationsâ•¯â•¯â•¯â•¯â•¯â•¯570 Keith Bull, Robert Hoft and Mark A. Sutton 26 Societal choice and communicating the European nitrogen challengeâ•¯â•¯â•¯â•¯â•¯â•¯585 David S. Reay, Clare M. Howard, Albert Bleeker, Pete Higgins, Keith Smith, Henk Westhoek, Trudy Rood,

Mark R. Theobald, Alberto Sanz-Cobeña, Robert M. Rees, Dominic Moran, Kate Ravilious and Stefan Reis

Glossaryâ•¯â•¯â•¯â•¯â•¯â•¯602 Indexâ•¯â•¯â•¯â•¯â•¯â•¯607

ix

This page intentionally left blank

Contributors

John van Aardenne European Commission Joint Research Center Institute for Environment and Sustainability via Enrico Fermi 2749 21027 Ispra (VA) Italy Wenche Aas NILU, Norwegian Institute for Air Research PB 100 2027 Kjeller Norway Beat Achermann Federal Office for the Environment Air Pollution Control and NIR Division Air Quality Management Section CH-3003 Bern Switzerland Rob Alkemade Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH The Netherlands Per Ambus Risø DTU National Laboratory for Sustainable Energy, Technical University of Denmark Biosystems Division Frederiksborgvej 399 4000 Roskilde Denmark Michael Ashmore University of York Environment Department Heslingon YO10 5DD United Kingdom Juergen Augustin Leibniz-Centre for Agricultural Landscape Research (ZALF) Eberswalder Strasse 84 Muencheberg

D-15374 Germany Alex Baker School of Environmental Sciences University of East Anglia Norwich NR4 7TJ United Kingdom Hermann W. Bange IFM-GEOMAR, Leibniz-Institut für Meereswissenschaften Düsternbrooker Weg 20 Kiel D-24226 Germany Sabine Barles Université Paris EstÂ€– LATTS, Institut Français d’Urbanisme 4 rue Alfred NobelÂ€– Cité Descartes Champs-sur-Marne 77420 France Sébastien Barot IRD-Bioemco, Bioemco ENS 46 rue d’Ulm 75230 Paris Cedex 05 France Jerzy Bartnicki Norwegian Meteorological Institute P.O. Box 43 Oslo NO-0313 Norway Claus Beier RisØ DTU, National Laboratory for Sustainable Energy Ecosystems Research Programme P.O. Box 358 4000 Roskilde Denmark

xi

List of Contributors

Salim Belyazid Belyazid Consulting and Communication AB, Österportsgatan 5a 21128 Malmö Sweden Leon J. L. van den Berg Radboud University Nijmegen Heyendaalseweg 135 6525 AJ Nijmegen The Netherlands Peter Bergamaschi European Commission Joint Research Centre Institute for Environment via Enrico Fermi 2749 and Sustainability 21027 290 Ispra (VA) Italy Haldis Berge Norwegian Meteorological Institute PO 43 Blindern 0313 Oslo Norway Arthur Beusen Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Gilles Billen University Pierre & Marie Curie 4 place Jussieu 75005 Paris France Albert Bleeker Energy Research Centre of the Netherlands P.O. Box 1 1755 ZG Petten The Netherlands Jaap Bloem Alterra Wageningen University and Research Centre Soil Science Centre P.O. Box 47 6700 Wageningen The Netherlands Roland Bobbink B-Ware Research Centre Radboud University P.O. Box 9010 9500 GL Nijmegen The Netherlands Pascal Boeckx Ghent University Faculty of Bioscience Engineering

xii

Coupure 653 9000 Gent Belgium Fayçal Bouraoui European Commission Joint Research Centre via Enrico Fermi 2749 21027 Ispra (VA) Italy Lex Bouwman Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Nils-Axel Braathen OECD 2 rue André-Pascal F-75775 Paris Cedex 16 France Cristina Branquinho Universidade de Lisboa, Faculdade de Ciências Centro de Biologia Ambiental, Campo Grande, Bloco C2, 5º Piso, sala 37 1749–016 Lisboa Portugal Frank Brentrup Yara International, Centre for Plant Nutrition and Environmental Research Hanninghof 35 48249 Duelmen Germany Lutz Breuer Institute for Landscape Ecology and Resources Management Research Centre for BioSystems, Land Use and Nutrition Heinrich-Buff-Ring 26 35392 Giessen Germany Corjan Brink Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Roy Brouwer VU University Amsterdam Institute for Environmental Studies De Boelelaan 1085 1081 HV Amsterdam Netherlands Michaela Budňáková Ministry of Agriculture of the Czech Republic Těšnov 17 117 05 Praha 1 Czech Republic


Keith R. Bull Centre for Ecology and Hydrology Lancaster Environment Centre Library Avenue Lancaster LA1 4AP United Kingdom Klaus Butterbach-Bahl Karlsruhe Institute of Technology Institute for Meterology and Climate Research Atmospheric Environmental Research Kreuzeckbahnstrasse 19 82467 Garmisch-Partenkirchen Germany Andrea Butturini University of Barcelona Department of Ecology Faculty of Biology avd. Diagonal 645 8028 Barcelona Spain John Neil Cape Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Ana C. Cardoso European Commission Joint Research Centre Institute for Environment and Sustainability via Enrico Fermi 2749 21027 Ispra (VA) Italy Pierre Cellier INRA, UMR EGC 78850 Thiverval-Grignon France Pavel Čermák Central Institute for Supervising and Testing in Agriculture Hroznová Street 2 656 06 Brno Czech Republic Daniel J. Conley Lund University Department of Earth and Ecosystem Sciences Sölvegatan 12 223 62 Lund Sweden Sarah E. Cornell University of Bristol QUEST, School of Earth Sciences

Queens Road Bristol BS8 1RJ United Kingdom Chris J. Curtis University College London Environmental Change Research Centre Gower Street London WC1E 6BT United Kingdom Cornelis Cuvelier European Commission Joint Research Centre P.O. Box 410 21020 Ispra (VA) Italy Tommy Dalgaard Aarhus University Department of Agroecology and Environment P.O. Box 50 8830 Tjele Denmark Michael Dannenmann University of Freiburg Institute of Forest Botany and Tree Physiology Georges Köhler Allee 53/54 79110 Freiburg Germany Frank Dentener European Commission Joint Research Centre via Enrico Fermi 2749 21027 Ispra (VA) Italy Barbara Deutsch Stockholm University Department of Applied Environmental Science Svanthe Arrheniusväg 8 11418 Stockholm Sweden Roald Dickens Department for the Environment Food and Rural Affairs 17 Smith Square London SW1P 3JR United Kingdom Nancy B. Dise Manchester Metropolitan University Department of Environmental and Geographical Sciences John Dalton East Building, Chester Street Manchester M1 5GD United Kingdom

xiii


Ulrike M. Doering European Commission Joint Research Centre Institute for Environment and Sustainability P.O. Box 290 21020 Ispra (Va) Italy Anthony Dore Centre for Ecology and Hydrology Bush Estate Penicuik EH26 9HF United Kingdom Ulrike Dragosits Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Jean-Louis Drouet INRA UMR INRA/AgroParisTech Environment and Arable Crops 78850 Thiverval-Grignon France Patrick Durand INRA UMR 1069 SAS 35000 Rennes France Sylvia Duretz INRA UMR EGC 78850 Thiverval-Grignon France Anja Engel Alfred Wegener Institute for Polar and Marine Research Am Handelshafen 12 27515 Bremerhaven Germany Jan Willem Erisman Energy Research Centre of the Netherlands P.O. Box 1 1755 ZG Petten the Netherlands Ipek Erzi TUBITAK Marmara Research Centre Environment Institute P.O. Box 21 41470 Gebze Kocaeli Turkey

xiv

Hilde Fagerli Norwegian Meteorological Institute P.O. Box 43 0313 Blindern Norway David Fernall Department for Environment, Food and Rural Affairs Kingspool, Peasholme Green York YO1 2PX United Kingdom Chris R. Flechard Soils, Agro-hydro systems and Spatialization 65 rue de St-Brieuc 35042 Rennes France David Fowler Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom James Galloway University of Virginia P.O. Box 400772 Charlottesville VA 22901 United States of America Raja S. Ganeshram University of Edinburgh School of GeoSciences Grant Institute West Mains Road Edinburgh EH16 5NW United Kingdom Josette Garnier UMR Sisyphe UPMC & CNRS, , 4 place Jussieu 75005 Paris France Markus Geupel Federal Environment Agency, Germany Wörlitzer Platz 1 6844 Dessau Germany Ing-Marie Gren Swedish University of Agricultural Sciences Department of Economics 750 07 Uppsala Sweden


Peringe Grennfelt IVL Swedish Environmental Research Institute Ltd Aschebergsgatan 44 P.O. Box 5302 400 14 Gothenburg Sweden

Jean-Paul Hettelingh National Institute for Public Health and the Environment Coordination Centre for Effects P.O. Box 1 3720 BA Bilthoven The Netherlands

Hans van Grinsven Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands

Kevin Hicks University of York Stockholm Environment Institute Grimston House Heslington YO10 5DD United Kingdom

Bruna Grizzetti European Commission Joint Research Centre via Enrico Fermi 2749 21027 Ispra (VA) Italy Per Gundersen University of Copenhagen Forest and Landscape Denmark Rolighedsvej 23 1958 Frederiksberg Denmark Steen Gyldenkærne Afdeling for Systemanalyse Danmarks Miljøundersøgelser Frederiksborgvej 399 4000 Roskilde Denmark Roy M. Harrison University of Birmingham School of Geography, Earth and Environmental Sci. Edgbaston Birmingham B15 2TT United Kingdom Anna-Stiina Heiskanen Finnish Environment Institute P.O. Box 140 251 Helsinki Finland Josef Hejzlar Institute of Hydrobiology Biology Centre AS CR Na Sadkach 7 370 05 Ceske Budejovice Czech Republic Ole Hertel University of Aarhus National Environmental Research Institute P.O. Box 358 4000 Roskilde Denmark

Peter Higgins University of Edinburgh Holyrood Road Edinburgh EH8 8AQ United Kingdom Klaas W. van Der Hoek National Institute for Public Health and the Environment P.O. Box 1 3720 BA Bilthoven The Netherlands Robert Hoft Convention on Biological Diversity 413, Saint Jacques Street, suite 800 Montreal QC H2Y 1N9 Canada Mike Holland University of Reading EMRC Whitchurch Hill Reading RG8 7PW United Kingdom Clare M. Howard Centre for Ecology and Hydrology Bush Estate Penicuik EH23 4RB United Kingdom Robert W. Howarth Cornell University Department of Ecology and Evolutionary Biology Corson Hall Ithaca NY 14853 United States of America Christoph Humborg Stockholm University

xv


Department of Applied Environmental Science Svanthe Arrheniusväg 8 10691 Stockholm Sweden Nicholas J. Hutchings University of Aarhus Research Centre Foulum 8830 Tjele Denmark Andreas Ibrom Risø National Laboratory for Sustainable Energy Frederiksborgvej 399 4000 Roskilde Denmark Hans van Jaarsveld Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Brian H. Jacobsen University of Copenhagen Institute of Food and Resource Economics Rolighedsvej 25 1958 Frederiksberg Denmark Steve Jarvis The European Journal of Soil Science Centre for Rural Policy Research University of Exeter Amory Building, Rennes Drive Exeter EX4 4RJ United Kingdom Michael E. Jenkin Atmospheric Chemistry Services Okehampton EX20 1FB United Kingdom Lars Stoumann Jensen University of Copenhagen Faculty of Life Sciences Department of Agriculture and Ecology Thorvaldsensvej 40 1871 Frederiksberg C Denmark Timothy Jickells University of East Anglia School of Environmental Sciences Norwich NR4 7TJ United Kingdom

xvi

Penny Johnes University of Reading Aquatic Environments Research Centre Whiteknights Reading RG6 6DW United Kingdom Age W. Jongbloed Wageningen UR Livestock Research Edelhertweg 15 8219 PH Lelystad The Netherlands Åsa Jonsson IVL Swedish Environmental Research Institute P.O. Box 5302 400 14 Göteborg Sweden Øyvind Kaste Norwegian Institute for Water Research Jon Lilletuns vei 3 4879 Grimstad Norway Ralf Kiese Karlsruhe Institute for Technology Institute for meteorology and Climate Research Atmospheric Environmental Research Kreuzeckbahnstrasse 19 82467 Garmisch-Partenkirchen Germany Barbara Kitzler Federal Research and Training Centre for Forests, Natural Hazardo and Landscape Seckendorff-Gudent-Weg 8 1130 Vienna Austria Jeroen de Klein Wageningen University and Research Centre Aquatic Ecology and Water Quality Management Group P.O. Box 47 6700 AA Wageningen The Netherlands Zbigniew Klimont International Institute for Applied Systems Analysis Schlossplatz 1 2361 Laxenburg Austria Pirkko Kortelainen Finnish Environment Institute (SYKE) P.O. Box 140 00251 Helsinki Finland


Marina Kousoulidou Aristotle University of Thessaloniki Department of Mechanical Engineering Laboratory of Applied Thermodynamics 54124 Thessaloniki Greece Natalia Kozlova North-West Research Institute of Agricultural Engineering and Electrification (SZNIIMESH) P.O.Tiarlevo, Filtrovskoje shosse, 3 196625 Saint-Petersburg-Pavlovsk Russian Federation Michael Kreuzer ETH Zurich Institute of Plant, Animal and Agroecosystem Science Universitätstrasse 2 8092 Zurich Switzerland Johannes Kros Alterra, Wageningen University and Research Centre P.O. Box 47 6700 AA Wageningen The Netherlands Markku Kulmala University of Helsinki Department of Physics P.O. Box 64 14 Helsinki Finland Joachim Lammel Yara International Centre for Plant Nutrition and Environmental Research Hanninghof 35 48249 Duelmen Germany Christiane Lancelot Université Libre de Bruxelles Ecologie des Systèmes Aquatiques ESA, CP 221 Boulevard du Triomphe 1050 Bruxelles Belgium Patrick Lavelle CIAT Km 17, Recta Cali-Palmira Apartado Aéreo 6713 Cali Colombia

Anne-Christine Le Gall INERIS Economics and Decision for the Environment, Chronic Risks Division Parc Technologique Alata, BP2 60550 Verneuil en Halatte France Allison Leach University of Virginia P.O. Box 400123 Charlottesville VA 22904 United States of America Adrian Leip European Commission Joint Research Centre Institute for Environment and Sustainability via E Ferminrico 2749 21027 Ispra (VA) Italy Ahti Lepistö Finnish Environment Institute (SYKE) P.O. Box 140 251 Helsinki Finland Jan Peter Lesschen Wageningen University and Research Centre Alterra P.O. Box 47 6700 AA Wageningen The Netherlands Roos Loeb B-ware Research Centre P.O. Box 6558 6503 GB Nijmegen The Netherlands Benjamin Loubet INRA, INA PG UMR Environm & Grandes Cultures 78850 Thiverval-Grignon France Rob Maas Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Stephen C. Maberly Centre for Ecology and Hydrology Lancaster Environment Centre Library Avenue Lancaster LA1 4AP United Kingdom

xvii


Luc Maene International Fertilizer Industry Association 28 rue Marbeuf 75008 Paris France Jakob Magid Copenhagen University Department of Agriculture and Ecology Thorvaldsensvej 40 1873 Copenhagen Denmark Abigail McQuatters-Gollop Sir Alister Hardy Foundation for Ocean Science Citadel Hill Plymouth PL1 2PB United Kingdom Philippe Merot INRA 65 rue de Saint-Brieuc, CS84215, 35042 Rennes France Jack J. Middelburg Utrecht University Faculty of Geosciences Budapestlaan 4 3584 CD Utrecht The Netherlands Jana Moldanová IVL Swedish Environmental Research Institute Ltd Box 5303 400 14 Göteborg Sweden Suvi Monni European Commission Joint Research Centre Institute for Environment and Sustainability via Enrico Fermi 2749 21027 Ispra (VA) Italy Dominic Moran Scottish Agricultural College King’s Buildings Edinburgh EH9 6GU United Kingdom John Munthe IVL Swedish Environmental Research Institute P.O. Box 5302 400 14 Gothenburg Sweden

xviii

Gabriela B. Nardoto Universidade de Brasília Faculdade UnB Planaltina Área Universitária 1 Vila Nossa Senhora de Fátima, Planaltina 73.340–710 Brasília Brazil Eiko Nemitz Centre for Ecology an d Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Rostislav Neveceral Czech Hydrometeorological Institute Na Sabatce 17 14000 Praha Czech Republic Nikolaos P. Nikolaidis Technical University of Crete Department of Environmental Engineering University Campus 73100 Chania Greece Oene Oenema Wageningen University and Research Centre Alterra P.O. Box 47 6700 AA Wageningen The Netherlands Jorgen E. Olesen Aarhus University Department of Agroecology and Environment Blichers Alle 20 8830 Tjele Denmark Mark van Oorschot Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Lorenzo Orlandini European Commission – DG AGRI Rue de la Loi 130–05/20 1000 Brussels Belgium Christian Pallière Fertilizers Europe Avenue E. Van Nieuwenhuyse 6 1160 Brussels Belgium


Gilles Pinay University of Birmingham School of Geography Birmingham B15 2TT United Kingdom Hanne Damgaard Poulsen Aarhus University Department of Animal Health and Bioscience P.O. Box 50 8830 Tjele Denmark David Powlson Rothamsted Research Harpenden AL5 2JQ United Kingdom Michel Prud’homme International Fertilizer Industry Association 28 rue Marbeuf 75008 Paris France Ari Rabl ARMINES/Ecoles des Mines de Paris 6 av. Faidherbe 91440 Bures sur Yvette France David S. Reay University of Edinburgh School of Geosciences CECS, High School Yards Edinburgh EH8 9XP United Kingdom Robert M. Rees Scottish Agricultural College West Mains Road Edinburgh EH9 3JG United Kingdom Gert Jan Reinds Wageningen University and Research Centre Alterra P.O. Box 47 6700 AA Wageningen The Netherlands Stefan Reis Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom

Hannes Isaak Reuter Gisxperts gbr Eichenweg 42 06849 Dessau Germany Michael O. Rivett University of Birmingham Water Sciences Group Birmingham B15 2TT United Kingdom Trudy G. A. Rood Netherlands Environmental Assessment Agency P.O. Box 303 3721 AH Bilthoven The Netherlands Joost Salomez Flemish Government K. Albert II-laan 20 1000 Brussels Belgium Benjamin Sanchez Gimeno CIEMAT Avda. Complutense 22 28040 Madrid Spain Alberto Sanz-Cobena Technical University of Madrid Av/ Complutense s/n, Ciudad Universitaria 28040 Madrid Spain Martijn Schaap TNO Built Environment and Geosciences P.O. Box 80015 3508 TA Utrecht The Netherlands Doris Schiedek National Environmental Research Institute Frederiksborgvej 399 4000 Roskilde Denmark Jan K. Schjoerring University of Copenhagen Department of Agriculture and Ecology Thorvaldsensvej 40 1871 Frederiksberg C Denmark Uwe A. Schneider KlimaCampus, Hamburg University Research Unit Sustainability and Global Change Bundesstrasse 55

xix


20146 Hamburg Germany Valiyaveetil Shamsudheen Semeena Norwegian Meteorological Institute P. O. Box 43 0313 Blindern Norway Günther Seufert European Commission Joint Research Centre Institute for Environment and Sustainability P.O. Box 050 21027 Ispra (VA) Italy Jan Siemens University of Bonn Institute of Crop Science and Resource Conservation – Soil Sciences Nussallee 13 53115 Bonn Germany Marie Silvestre CNRSÂ€– FR3020 FIRE 4 place Jussieu 75005 Paris France David Simpson Norwegian Meteorological Institute EMEP MSC-W P.O. Box 43 0313 Blindern Norway Ute Skiba Centre fro Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Carsten Ambelas Skjøth Aarhus University P.O. Box 358 4000 Roskilde Denmark Caroline Slomp Utrecht University Department of Earth Sciences Budapestlaan 4, 3584 CD Utrecht The Netherlands Erik Smedberg Stockholm University Baltic Nest Institute

xx

Stockholm Resilience Centre 10691 Stockholm Sweden Keith A. Smith University of Edinburgh Institute of Atmospheric and Environmental Science West Mains Road Edinburgh EH9 3JN United Kingdom Lise Lotte Sørensen Risø National Laboratory for Sustainable Energy P.O. Box 49 4000 Roskilde Denmark Till Spranger Federal Ministry for the Environment, Nature Conservation and Nuclear Safety Stresemannstrasse 128–130 10117 Berlin Germany Carly J. Stevens The Open University Department of Life Sciences Walton Hall Milton Keynes MK7 6AA United Kingdom Mark A. Sutton Centre for Ecology and Hydrology Bush Estate Penicuik EH26 0QB United Kingdom Anastasia Svirejeva-Hopkins Potsdam Institute for Climate Impact Research Telegrafenberg A31 14473 Potsdam Germany Mette Termansen University of Aarhus Department of Policy Analysis Frederiksborgvej 399 4000 Roskilde Denmark Mark Theobald Technical University of Madrid/Centre for Ecology and Hydrology Department of Agricultural Chemistry and Analysis Ciudad Universitaria, s/n 28040 Madrid Spain


Vincent Thieu UMR 7619 Sisyphe CNRS/UPMC 4 place Jussieu 75005 Paris France Philippe Thunis European Commission Joint Research Centre Institute for Environment and Sustainability via Enrico Fermi 2749 21020 Ispra (VA) Italy Chris Tompkins Independent consultant United Kingdom Robert Upstill-Goddard Newcastle University School of Marine Science and Technology Ridley Building Newcastle-upon-Tyne NE47 9BL United Kingdom Laura Valli CRPA Corso Garibaldi 42 42100 Reggio Emilia Italy Robert Vautard LSCE/IPSL laboratoire CEA/CNRS/VSQ Orme des Merisiers 91191 Gif/Yvette Cedex France Gerard L. Velthof Wageningen University and Research Centre Alterra P.O. Box 47 6700 AA Wageningen The Netherlands Timo Vesala University of Helsinki Department of Physics P.O. Box 48 14 Helsinki Finland Valérie Viaud INRA, UMR 1069 SAS 65 rue de Saint-Brieuc 35000 Rennes France

Massimo Vieno University of Edinburgh School of Geosciences The King’s Buildings Edinburgh EH9 3JN United Kingdom Maren Voss Leibniz-Institute of Baltic Sea Research Warnemuende Seestrasse 15 18119 Rostock Germany Wim de Vries Alterra, Wageningen University and Research Centre Centre Soil, Droevendaalsesteeg 4, Wageningen 6708 PB The Netherlands Jim Webb AEA Energy and Environment Gemini Building, Harwell Business Centre Didcot OX11 0QR United Kingdom Henk J. Westhoek Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Jaap Willems Netherlands Environmental Assessment Agency P.O. Box 303 3720 AH Bilthoven The Netherlands Wilfried Winiwarter International Institute for Applied Systems Analysis Schlossplatz 1 2361 Laxenburg Austria Peter Witzke EuroCARE GmbH Nussallee 21 53115 Bonn Germany Richard F. Wright Norwegian Institute for Water Research Gaustadalleen 21 349 Oslo Norway Sönke Zaehle Max Planck Institute for Biogeochemistry

xxi


Biogeochemical Systems Department Hans-Knöll-Strasse 10 07745 Jena Germany Sophie Zechmeister-Boltenstern Federal Research and Training Centre for Forests Natural Hazards and Landscape Seckendorff Gudent Weg 8

xxii

1131 Vienna Austria John F. Zevenbergen TNO Defence, Security and Safety Lange Kleiweg 137 2288 GJ Rijswijk The Netherlands

Foreword

Addressing the grand challenges of society depends fundamentally on firm scientific evidence. Today, Europe faces several of these challenges, as outlined in the Europe 2020 strategy adopted by the Commission on 3 March 2010, including climate change, energy and food security, health and an ageing population. Research and innovation are crucial to address these challenges effectively. For that reason, the Commission launched the ‘Innovation Union’ flagship initiative, with the aim to re-focus research and development as well as innovation policy on these grand societal challenges. In this framework we very much welcome the European Nitrogen Assessment. It is fair to say that nitrogen will be a new story for many people. Yet we can here clearly identify a case of science at its best: innovative thinking that enables the development of connections from evidence-based policies to evidence-tested decisions. The Assessment highlights how human production of reactive nitrogen has literally changed the world. Since the invention of the Haber-Bosch process a century ago, humans have been able to double the world’s circulation of nitrogen compounds, resulting in nitrogen fertilizers sustaining around 3 billion people, almost half of the world population. It is therefore obvious that nitrogen is essential, not only to meeting the challenge for food security, but, with the increasing importance of biofuels, also for energy security. Yet with this achievement, originating from European innovation a century ago, has also come an inheritance of environmental effects that cuts across all global ecosystems. As the Assessment reveals, excess reactive nitrogen contributes to climate change; it adversely affects water, air and soil quality, and is putting unsustainable pressure on ecosystems and biodiversity in Europe. Moreover, the surplus of nitrogen compounds leaking into air and water may lead to a substantial health risk for vulnerable human populations.

The Assessment highlights how nitrogen is related to each of the great challenges that European society faces, and the need to develop joined up approaches to address them. In this respect the European Nitrogen Assessment is an important step, building scientific and institutional bridges and sharing different perspectives. It is rewarding to see different environmental disciplines being brought together, and scientists proactively seeking to engage European industry, policy makers and the public. These significant commitments also emphasize the importance of critical mass in the European Research Area. The Assessment is a key output from a large amount of ongoing research in Europe and elsewhere, but in particular from the NitroEurope Integrated Project supported by the European Commission’s 6th Framework Programme and the Nitrogen in Europe (NinE) Research Networking Programme of the European Science Foundation. With the involvement of Action 729 of the COST Programme, the necessary expertise has been gathered to drive the Assessment. The message of 200 leading European experts from different disciplines and perspectives is surely that we need to take steps forward. Only by joining forces to face the societal challenges will European research provide the scientific basis and the evidence needed for solutions. If European innovation has handed us down a nitrogen inheritance, threatening the environment as a price for a solution to nourish the growing world population, it is only right that European science should lead the way in responding to the challenge. Robert-Jan Smits Director General for Research, European Commission Professor Marja Makarow Chief Executive, European Science Foundation

xxiii

Summary for policy makers Lead authors:Â€Mark A. Sutton and Hans van Grinsven Contributing authors:Â€Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Keith Bull, Jan Willem Erisman, Peringe Grennfelt, Bruna Grizzetti, Clare M. Howard, Oene Oenema, Till Spranger and Wilfried Winiwarter

Main messages Too much nitrogen harms the environment and the economy • Over the past century humans have caused unprecedented changes to the global nitrogen cycle, converting atmospheric di-nitrogen (N2) into many reactive nitrogen (Nr) forms, doubling the total fixation of Nr globally and more than tripling it in Europe. • The increased use of Nr as fertilizer allows a growing world population, but has considerable adverse effects on the environment and human health. Five key societal threats of Nr can be identified:Â€to water quality, air quality, greenhouse balance, ecosystems and biodiversity, and soil quality. • Cost–benefit analysis highlights how the overall environÂ� mental costs of all Nr losses in Europe (estimated at €70–€320 billion per year at current rates) outweigh the direct economic benefits of Nr in agriculture. The highest societal costs are associated with loss of air quality and water quality, linked to impacts on ecosystems and especially on human health.

Nitrogen cascade and budgets • The different forms of Nr inter-convert through the environment, so that one atom of Nr may take part in many environmental effects, until it is immobilized or eventually denitrified back to N2. The fate of anthropogenic Nr can therefore be seen as a cascade of Nr forms and effects. The cascade highlights how policy responses to different Nr forms and issues are inter-related, and that a holistic approach is needed, maximizing the abatement synergies and minimizing the trade-offs. • Nitrogen budgets form the basis for the development and selection of measures to reduce emissions and their effects in all environmental compartments. For instance, the

xxiv

European nitrogen budget highlights the role of livestock in driving the European nitrogen cycle.

Policies and management • Existing policies related to Nr have been largely established in a fragmented way, separating Nr forms, media and sectors. Despite the efforts made over many years to reduce Nr inputs into the environment, most of the Nr-related environmental quality objectives and environmental action targets have not been achieved to date. • The five societal threats and N budgets are starting points for a more-holistic management of Nr. The Assessment identifies a package of 7 key actions for overall management of the European nitrogen cycle. These key actions relate to:Â€Agriculture (3 actions), Transport and Industry (1 action), Waste water treatment (1 action) and Societal consumption patterns (2 actions). • The key actions provide an integrated package to develop and apply policy instruments. The need for such a package is emphasized by cost–benefit analysis that highlights the role of several Nr forms especially nitrogen oxides (NOx), ammonia (NH3) and Nr loss to water, in addition to nitrous oxide (N2O), in the long term.

International cooperation and communication • Tackling Nr necessitates international cooperation. There are various options to implement multi-lateral environmental agreements; a possible inter-convention agreement on nitrogen needs to be further explored. • Communication tools for behavioural change should be extended to nitrogen, such as calculating nitrogen ‘food-prints’. Messages should emphasize the potential health co-benefits of reducing the consumption of animal products to avoid excess above recommended dietary guidelines.

Summary for policy makers

1.â•‡ Why nitrogen? Concerns and the need for new solutions 1.â•‡ Nitrogen is an abundant element on earth, making up nearly 80% of the earth’s atmosphere. However, as atmospheric di-Â� nitrogen (N2), it is unreactive and cannot be assimilated by most organisms. By contrast there are many reactive nitrogen (Nr) forms that are essential for life, but are naturally in very short supply. These include ammonia, nitrates, amino acids, proteins and many other forms. Until the mid nineteenth century, limited availability of these Nr compounds in Europe severely constrained both agricultural and industrial productivity [1.1, 2.1].1 2.â•‡ With increasing population in the late nineteenth century, rates of biological nitrogen fixation were not sufficient for crop needs and Europe became increasingly dependent on limited sources of mined Nr (guano, saltpetre, coal). At the start of the twentieth century, several industrial processes were developed to fix N2 into Nr, the most successful beingÂ€the Haber–Bosch process to produce ammonia (NH3) [1.1, 2.1]. 3.â•‡ Since the 1950s, Nr production has greatly increased, representing perhaps the greatest single experiment in global geoengineering [1.1]. Europe’s fertilizer needs have been met, as well as its military and industrial needs for Nr [3.2, 3.5]. In addition, high temperature combustion processes have substantially increased the formation and release of nitrogen oxides (NOx) [2.4]. While the Nr shortage of the past has been solved, Europe has stored up a nitrogen inheritance of unexpected environmental effects [1.1]. 4.â•‡ Europe remains a major source region for Nr production, with many of the environmental impacts being clearly visible and well studied. There is a wealth of evidence on sources, fate and impacts of Nr. However, the complexity and extent of the interactions mean that scientific understanding has become scattered and focused on individual sectors. A parallel fragmentation can be seen in environmental policies related to nitrogen, which are typically separated by media (air, land, water, etc.), by issue (climate, biodiversity, waste etc) and by Nr form [4.4, 5.3]. 5.â•‡ While this specialization has advanced understanding, European science and policies related to nitrogen have to a significant degree lost sight of the bigger picture. The occurrence of Nr in many different Nr forms and media, means that each component should not be considered in isolation. A more comprehensive understanding of the nitrogen cycle is therefore needed to minimize the adverse effects of Nr in the environment, while optimizing food production and energy use [5.3].

2.â•‡ Role and approach of the European Nitrogen Assessment 6.â•‡ A key challenge is to synthesize the science and understanding of nitrogen into a form that is useful to governments and society. This involves bringing the different Nr forms, disciplines and stakeholders together. 1

References in this summary (e.g., [1.1, 11.1]) refer to chapter and section numbers of the European Nitrogen Assessment.

7.â•‡ The European Nitrogen Assessment (ENA) was established in response to these needs. It was coordinated by the Nitrogen in Europe (NinE) programme of the European Science Foundation, drawing on underpinning research from across Europe, but especially the NitroEurope Integrated Project co-funded by the European Commission, with input from the COST Action 729. The Assessment provides a European contribution to the International Nitrogen Initiative (INI) [1.3]. 8.â•‡ The lead policy audience for the Assessment is the Geneva Convention on Long-range Transboundary Air Pollution (CLRTAP), established under the auspices of the United Nations Economic Commission for Europe (UNECE). Through its Task Force on Reactive Nitrogen, the Convention has formally adopted the Assessment as a contributing activity to its work [1.3]. 9.â•‡ In addition to supporting CLRTAP, the Assessment is targeted to provide scientific and policy support to the European Union and its Member States, as well as other multi-lateral environmental agreements, including the Global Partnership on Nutrient Management facilitated by UNEP [1.5]. 10.â•‡ Recognizing these needs, the goal of the European Nitrogen Assessment was established:Â€to review current scientific understanding of nitrogen sources, impacts and interactions across Europe, taking account of current policies and the economic costs and benefits, as a basis to inform the development of future policies at local to global scales [1.4]. 11.â•‡ The Assessment process was conducted through a series of five open scientific workshops between 2007 and 2009. Draft chapters were submitted to internal and external peer review [1.3].

3.â•‡ Disruption of the European nitrogen cycle Fertilizers, energy and transport:Â€drivers for increased nitrogen inputs 12.â•‡ Production of Nr is a key input for agriculture and industry, and a persistent side-effect of combustion for energy and transport. Industrial production in Europe of Nr in 2008 was about 34 Tg per year (where 1 Tg = 1 million tonnes) of which 75% is for fertilizer and 25% for chemical industry (production of rubbers, plastics, and use in electronic, metals and oil industry) [3.5]. The trend in mineral fertilizer represents the largest change in overall Nr inputs to Europe over the past century (Figure SPM.1). 13.â•‡ The combustion of fossil fuels has allowed a substantial increase in industrial production and transportation, reflected in the greatly increased emission of nitrogen oxides, which only over the last 20 years have partly been controlled. By contrast, the total contribution of crop biological nitrogen fixation has decreased significantly. 14.â•‡ The provision of Nr from the Haber–Bosch process removed a major limiting factor on society, permitting substantial population growth and improving human welfare.

xxv


Figure SPM.1 Estimated trend of anthropogenic reactive nitrogen inputs to the European Union (EU-27) [5.1] (1 Tg equals 1 million tonnes).

Figure SPM.2 Simplified view of the N-cascade, highlighting the capture of atmospheric di-nitrogen (N2) to form reactive nitrogen (Nr) by the Haber–Bosch processÂ€– the largest source of Nr in Europe. The main pollutant forms of Nr (orange boxes) and five environmental concerns (blue boxes) are summarized. Blue arrows represent intended anthropogenic Nr flows; all the other arrows are unintended flows [1.2]. For fuller description including other Nr sources, see [5.2].

However, accounting for natural sources, humans have more than doubled the supply of Nr into the environment globally [1.1], and more than tripled this supply in Europe (Figure SPM.3) [16, supplementary material]. 15.â•‡ As of the year 2000, Europe creates about 19 Tg per year of Nr, of which 11 Tg per year is from chemical fertilizers, 3.4 Tg per year from combustion sources, 3.5 Tg per year from food and feed import and 1 Tg per year by crop biological N-fixation (BNF) (Figure SPM.3).

The nitrogen cascade 16.â•‡ Human production of Nr from N2 causes a cascade of intended and unintended consequences. The intended cascade is that each molecule of Nr contributes to soil fertility and increased yields of crops, subsequently feeding livestock and humans, allowing the formation of amino acids, proteins and

xxvi

DNA. In a well managed system, the intention is for the Nr in manures and sewage to be fully recycled back through the agricultural system (blue arrows in Figure SPM.2). 17.â•‡ Reactive nitrogen, is however, extremely mobile, with emissions from agriculture, combustion and industry leading to an unintended cascade of Nr losses into the natural environment (Figure SPM.2). Once released, Nr cascades through the different media, exchanging between different Nr forms and contributing to a range of environmental effects, until it is finally denitrified back into N2. An important consequence of the cascade is that the environmental impacts of Nr eventually become independent of the sources, so that nitrogen management requires a holistic approach. This is important, both to minimize ‘pollution swapping’ between different Nr forms and threats, and to maximize the potential for synergies in mitigation and adaptation strategies [2.6, 5.2].

Summary for policy makers Europe (EU27), around 1900. N fluxes in TgN/yr

Europe (EU27), around 2000. N fluxes in TgN/yr atmospheric N2 pool

atmospheric N2 pool

2.1

3.5

9.7 crop N2 fix

atmosph. NH3NOxN2O N2 fix industry & traffic

0.6 1.9

atmosph. deposition Fertilizers

1.9

4

9.6 N2 fix industry & traffic

crop production

gaseous losses

3.4

3.8

0.2 2.1 4

crop N2 fix

atmosph. NH3NOxN2O

Livestock & human nutrition

Net atmospheric export

2.4

atmosph. deposition

3.8

1 Fertilizers

Net import of food & feed

17.6 crop production

Livestock & human nutrition

11.2 Soils

Soils Losses to water

2.3

Export by rivers to the sea

gaseous losses

13.5

Losses to 4.5 water


Figure SPM.3 Simplified comparison of the European nitrogen cycle (EU-27) between 1900 and 2000. Blue arrows show intended anthropogenic nitrogen flows; orange arrows show unintended nitrogen flows; green arrows represent the nearly closed nitrogen cycle of natural terrestrial systems [16.4 and 16 supplementary material].

A new nitrogen budget for Europe 18.â•‡ One of the tasks addressed in the European Nitrogen Assessment has been to construct a comprehensive nitrogen budget for Europe (EU-27 for the year 2000), considering each of the major flows in the nitrogen cascade [16.4]. In parallel, the estimates have also been compared with 1900 [16, supplementary material]. By combining all the nitrogen flows, such budgets provide improved perspective on the major drivers and the most effective control options. 19.â•‡ Figure SPM.3 summarizes the European nitrogen budget in its simplest form [derived from 16.4]. The budget for 2000 shows that overall human perturbation of the nitrogen cycle is driven primarily by agricultural activities. Although the atmospheric emissions of NOx from traffic and industry contribute to many environmental effects, these emissions are dwarfed by the agricultural Nr flows. 20.â•‡ It is important to note the magnitude of the European Nr flow in crop production, which is mainly supported by Nr fertilizers. The primary use of the Nr in crops, however, is not directly to feed people:Â€80% of the Nr harvest in European crops provides feeds to support livestock (8.7 Tg per year plus 3.1 Tg per year in imported feeds, giving a total of 11.8â•›Tg per year). By comparison, human consumption of Nr is much smaller, amounting to only 2 Tg per year in crops and 2.3 Tg per year in animal products. Human use of livestock in Europe, and the consequent need for large amounts of animal feed, is therefore the dominant human driver altering the nitrogen cycle in Europe [16.4]. 21.â•‡ These major intended alterations in Nr flows cause many additional unintended Nr flows (Figure SPM.3). Overall, NH3 from agriculture (3.2 Tg per year) contributes a similar

amount to emissions of Nr to the atmosphere as NOx (3.4 Tg per year). Agriculture also accounts for 70% of nitrous oxide (N2O) emissions in Europe, with total N2O emissions of 1 Tg per year. The food chain also dominates Nr losses to ground and surface waters, mainly as nitrates (NO3), with a gross load of 9.7 Tg resulting mainly from losses due to agriculture (60%) and discharges from sewage and water treatment systems (40%) [16.4]. 22.â•‡ The comparison between 1900 and 2000 shows how each of these flows have increased, including denitrification back to N2. Denitrification is the largest and most uncertain loss, as it occurs at many different stages during the continuum from soils to freshwaters and coastal seas. Although emissions of N2 are environmentally benign, they represent a waste of the substantial amounts of energy put into human production of Nr, thereby contributing indirectly to climate change and air pollution. This is in addition to the impact on climate change of N2O formed especially as a byproduct of denitrification.

Achievements and limitations of current policies 23.â•‡ Peak production of Nr in Europe occurred in the 1980s, which was linked to agricultural over-production and lack of emissions regulations. Since that time, the introduction of policies and other changes affecting agriculture (including the Common Agricultural Policy, Nitrates Directive and the restructuring of Eastern Europe after 1989), as well as stringent emission controls, e.g., for large combustion plants (EC Large Combustion Plants Directive, UNECE Sofia Protocol and Gothenburg Protocol, etc.) and the EURO standards for

xxvii


Figure SPM.4 Estimated trends in European reactive nitrogen emissions between 1900 and 2000 (EU-27) [5.1].

road transport vehicles, have led to decreases in the emissions (FigureÂ€SPM.4) [4.4]. 24.â•‡ Overall, emissions of combustion NOx have reduced by ~30% since 1990, but much greater NOx reductions per unit output have been achieved. These have been offset by an increase in traffic and energy consumption. The net emission reduction is therefore a clear example of decoupling, as emissions would have increased by over 30% if no measures had been implemented. The extent of success of the technical measures can be in part attributed to the involvement of a small number of players (e.g., electricity supply industry, vehicle manufacturers) and the fact that the costs of these measures could be easily transferred to consumers [4.5]. 25.â•‡ Agricultural measures have resulted in only a modest reduction in total agricultural Nr inputs for the EU-27 of ~15% (Figure SPM.1). This small overall reduction is reflected in the trends in NH3 emissions (FigureÂ€ SPM.4). Most of the reductions that have been achieved to date can be attributed to reductions in fertilizer use and livestock numbers, especially in EasternÂ€ Europe after 1989. Although management improvements will have contributed to reduced emissions (e.g., nitrate leaching and loss to marine areas), there has as yet been little quantitative achievement of measures to reduce N2O and NH3 emissions from agriculture on a European scale. The fact that currentÂ€ Nr emission reduction policies in agriculture (e.g., Nitrates Directive, Oslo and Paris Commission for the protectionÂ€of the North East Atlantic, UNECE Gothenburg Protocol and National Emissions Ceilings Directive) have only made limited progress can be linked in part to the large number of diverse actors (including many small farms), the diffuse nature of the Nr emission sources, and the challenge of passing any perceived costs onto consumers [4.5]. As a consequence, agriculture is the sector with the largest remaining emission reduction potential. 26.â•‡ Several instances of pollution swapping in Nr control have been observed. These include the introduction of three way catalysts in vehicles, which increased NH3 and N2O emissions (although overall Nr emissions were still greatly reduced), and the implementation of the Nitrates Directive, prohibiting wintertime manure spreading, which has led to a new peak in springtime NH3 emisssionsÂ€[9.2].

xxviii

4.â•‡ The benefits and efficiency of nitrogen in agriculture Nitrogen fertilizers feed Europe 27.â•‡ There is no doubt that human production of Nr has greatly contributed to the increase in productivity of agricultural land. Without anthropogenic Nr, a hectare of good agricultural land in Europe, with no other growth limitations, can produce about 2 tonne per ha of cereal annually. With typical additional inputs from biological nitrogen fixation (BNF), it can produce about 4–6 tonne per ha, and with addition of chemical fertilizer about 8–10 tonne per ha. Synthetic Nr fertilizer has been estimated to sustain nearly 50% of the world’s population, and is essential for the EU to be largely self-sufficient in cereals. For pork, poultry and egg production, Europe strongly depends on soybean imports from America [3.1]. 28.â•‡ Agronomic efficiency provides an indicator of the Nr-benefit to the farmer (kg crop production per kg applied N). Typically, fertilizer rates in the eastern EU Member States are up to four times lower than in the 15 ‘old’ Member States, but agronomic efficiencies are comparable (Figure SPM.5). The use of Nr is profitable as there is a robust financial return of €2–5 on every euro invested in Nr fertilizer, depending on the market price of cereals and fertilizer [3.6].

Grain and meat production considerably differ in their Nr losses to the environment

29.â•‡ The nitrogen recovery (kg N taken up by a crop per kg applied N) provides a measure of environmental N-loss in crop production. For cereals it varies 30%–60% across Europe, indicating that 40%–70% of the fertilizer Nr applied is lost to the atmosphere or the hydrosphere [3.2]. 30.â•‡ The nitrogen recovery in animal farming is inherently lower than in crops, with only 10–50% of Nr in feed being retained in liveweight and 5%–40% in the edible weight (Figure SPM.6). Accounting for the additional Nr losses in feed production, the overall efficiency of Nr use for meat production is around half these values. For this reason, the full chain of animal protein production


Cereal yield (kg / ha)

10000 8000

NL

DE

FR SI

IE AT DK HU

6000 RO

4000

SK SE

BG

PL EE

LV

2000 0

BE

CY

05

0

FI

IT

CZ

UK

GR LT ES

PT

100

150

200

Nitrogen application (kg N / ha)

Figure SPM.5 Variation of nitrogen fertilizer use on winter wheat across the European Union (EU 15: blue, EU 12: red) around the year 2000. The variation indicates that there is substantial scope to increase performance and reduce environmental effects [3.2].

generates much more losses to the environment than plant protein production. 31.â•‡ About one third (7.1 Tg per year in 2000) of the total farm input of Nr to soil comes from animal manures. This represents about two thirds of the Nr from animal feeds, while the fraction of Nr in animal manures that is lost to the environment is typically double that of mineral Nr fertilizer, highlighting the importance of proper measures to maximize the effectiveness of manure reuse [3.2].

Variation in nitrogen use efficiency highlights the potential for solutions 32.â•‡ The overall efficiency of European agriculture (ratio of N in food produced to the sum of synthetic N fertilizer used plus food and feed imports) is about 30% since 2000 [derived from 16.4, see Figure SPM.3]. The wide variety in N application rates and nitrogen use efficiency across Europe indicates that there is a huge scope to improve resource efficiency and reduce environmental effects (Figure SPM.5). 33.â•‡ In the EU, protein consumption exceeds recommended intake by 70% [26.3] and the share of animal proteins in this total is increasing. Even a minor change in human diet, with less animal protein consumption (or protein from more efficient animals), would significantly affect the European nitrogen cycle.

5.â•‡ The key societal threats of excess nitrogen 34.â•‡ From a longer list of around 20 concerns, the Assessment identifies five key societal threats associated with excess Nr in the environment:Â€ Water quality, Air quality, Greenhouse balance, Ecosystems and biodiversity, and Soil quality. Together, these threats can be easily remembered by an acronym as the ‘WAGES’ of excess nitrogen, and visualized by analogy to the four ‘elements’ (water, air, fire, earth) and quintessence of classical Greek cosmology (Figure SPM.7). These five threats provide a framework that incorporates almost all issues related to the longer list of concerns associated with excess Nr [5.4].

Figure SPM.6 Range of Nr recovery efficiencies in farm animal production in Europe (kg N in edible weight per kg N in animal feed) [3.4, 10.4, 26.3], see also supplementary material for Chapter 3. A higher recovery efficiency is indicative of a smaller nitrogen footprint. Accounting for the full chain from fertilizer application to Nr in edible produce, overall nitrogen use efficiency in animal production for the EU-27 is around 15%–17% [3, 10, supplementary material]. While intensive systems tend to have a higher Nr recovery, they also tend to have larger Nr losses per ha unless efforts are taken to reduce emissions [10.4].

Nitrogen as a threat to European water quality 35.â•‡ Water pollution by Nr causes eutrophication and acidification in fresh waters [7.4, 8.8]. Estuaries, their adjacent coastlines and (near) inland seas are also affected by eutrophication from Nr with inputs to the coastal zone being four times the natural background [13.7]. Biodiversity loss, toxic algal blooms and dead zones (fish kill) are examples of effects [8.8]. Nitrate levels in freshwaters across most of Europe greatly exceed a threshold of 1.5 to 2 mg Nr per litre, above which waterbodies may suffer biodiversity loss [7.5, 17.3]. 36.â•‡ High nitrate concentrations in drinking water are considered dangerous for human health, as they might cause cancers and (albeit rarely) infant methaemoglobinaemia. About 3% of the population in EU-15 is potentially exposed to levels exceeding the standard for drinking water of 50 mg NO3 per litre (11.2 mg Nr per litre) and 6% exceeding 25 mg NO3 per litre [17.3]. This may cause 3% increase of incidence of colon cancer, but nitrate is also considered to be beneficial to cardiovascular health [22.3]. 37.â•‡ Although aquatic eutrophication has decreased to some extent since the 1980s, agreed international policies have not been fully implemented. In addition, increasing nitrate in groundwaters threatens the long-term quality of the resource, due to long residence times in aquifers [7.5,Â€17.2]. Achieving substantial progress at the European scale requires integration of sectoral policies, reducing overall inputs of Nr to watersheds [4.5, 13.7, 17.5].

Nitrogen as a threat to European air quality 38.â•‡ Air pollution by nitrogen oxides (NOx) and ammonia (NH3) causes formation of secondary particulate matter (PM), while emissions of NOx also increase levels of nitrogen dioxide

xxix


42.â•‡ Overall, European Nr emissions are estimated to have a net cooling effect on climate of −16 mW per m2, with the uncertainty bounds ranging from substantial cooling to a small net warming (−47 to +15 mW per m2). The largest uncertainties concern the aerosol and Nr fertilization effects, and the estimation of the European contributions within the global context [19.6]. The estimate of the Intergovernmental Panel on Climate Change (IPCC) for indirect N2O emissions from Nr deposition is considered to be an underestimate by at least a factor of 2 [6.6, 19.6]. 43.â•‡ There are many opportunities for ‘smart management’, increasing the net cooling effect of Nr by reducing warming effects at the same time as other threats, e.g., by linking N and C cycles to mitigate greenhouse gas emissions through improved nitrogen use efficiency [19.6].

Nitrogen as a threat to European terrestrial ecosystems and biodiversity Figure SPM.7 Summary of the five key societal threats of excess reactive nitrogen, drawn in analogy to the ‘elements’ of classical Greek cosmology. The main chemical forms associated with each threat are shown [5.4]. Photo sources: Shutterstock.com and garysmithphotography.co.uk.

(NO2) and tropospheric ozone (O3). All of these are causes for respiratory problems and cancers for humans, while ozone causes damage to crops and other vegetation, as well as to buildings and other cultural heritage [18.2, 18.5]. 39.â•‡ Models estimate that PM contributes to 300–400 thousand premature deaths annually in Europe leading to a reduction in life expectancy due to PM of 6–12 months across most of central Europe. Nr contributes up to 30%–70% of the PM by mass [18.3, 18.5]. However, the individual contributions of NOx- and Nr-containing aerosol to human health effects of air pollution remain uncertain [18.2]. 40.â•‡ Although NOx emission decreases have reduced peak O3 concentrations, background tropospheric O3 concentrations continue to increase. By comparison to the limited progress in reducing NOx emissions, there has been even less success in controlling agricultural NH3 emissions, which therefore contribute to an increasing share of the European air pollution burden [4.5, 18.6].

Nitrogen as a threat to European greenhouse balance 41.â•‡ Reactive nitrogen emissions have both warming and cooling effects on climate. The main warming components are increasing concentrations of nitrous oxide (N2O) and tropospheric ozone, which are both greenhouse gases. The main cooling effects are atmospheric Nr deposition presently increasing CO2 removal from the atmosphere by forests, and the formation of Nr containing aerosol, which scatter light and encourage cloud formation [19].

xxx

44.â•‡ Atmospheric Nr deposition encourages plants favouring high Nr supply or more acidic conditions to out-compete a larger number of sensitive species, threatening biodiversity across Europe. The most vulnerable habitats are those with species adapted to low nutrient levels or poorly buffered against acidification. In addition to eutrophication, atmospheric Nr causes direct foliar damage, acidification and increased susceptibility to pathogens [20.3]. 45.â•‡ Although there are uncertainties in the relative effects of atmospheric nitrate (NO3−) versus ammonium (NH4+), gaseous ammonia (NH3) can be particularly harmful to vegetation, causing foliar damage especially to lower plants [20.3]. This emphasizes the threat to semi-natural habitats occurring in agricultural landscapes [9.6, 11.5]. While uncertain, Nr deposition is expected to act synergistically with climate change and ground-level ozone [20.2]. 46.â•‡ Thresholds for atmospheric concentrations and deposition of Nr components to semi-natural habitats are exceeded across much of Europe, and will continue to be exceeded under current projections of Nr emissions. In order to achieve ecosystem recovery, further reductions of NH3 and NOx emissions are needed [20.5]. Due to cumulative effects of Nr inputs and long time-lags, rates of ecosystem recovery are expected to be slow, and in some cases may require active management intervention in the affected habitats [20.5].

Nitrogen as a threat to European soil quality 47.â•‡ Soil integrates many of the other Nr effects, highlighting their interlinked nature. The major Nr threats on soil quality are soil acidification, changes in soil organic matter content and loss of soil biodiversity. Soil acidification can occur from the deposition of both oxidized and reduced Nr, resulting from NOx and NH3 emissions, reducing forest growth and leading to leaching of heavy metals [21.3]. High levels of Nr deposition to natural peatlands risk losing carbon stocks through interactions with plant species changes, although this effect is poorly quantified [6.6, 19.4].


48.â•‡ Addition of Nr typically has a beneficial effect in agricultural soils, enhancing fertility and soil organic matter [6.4 , 21.3]. However, Nr losses increase, while some soil fungi and N-fixing bacteria are reduced by high N availability. The interactions between Nr and soil biodiversity, soil fertility and Nr emissions are not well understood [21.3]. 49.â•‡ European forest soils are projected to become less acidic within a few decades, mainly as a result of reduced SO2 and NOx emissions. Ammonia emissions have only decreased slightly and NHx is increasingly dominating soil acidification effects over large parts of Europe [20.3, 21.4].

6.â•‡ The economics of nitrogen in the environment Estimated loss of welfare due to nitrogen emissions in Europe 50.â•‡ The social costs of the adverse impacts of Nr in the European environment are estimated. Expressed as € per kg of Nr emission, the highest values are associated with air pollution effects of NOx on human health (€10–€30 per kg), followed by the effects of Nr loss to water on aquatic ecosystems (€5–€20 per kg) and the effects of NH3 on human health through particulate matter (€2–€20 per kg). The smallest values are estimated for the effects of nitrates in drinking water on human health (€0–€4 per kg) and the effect of N2O on human health by depleting stratospheric ozone (€1–€3 per kg) [22.6]. 51.â•‡ Combining these costs with the total amount of emissions for each main Nr form, provides a first estimate of the annual Nr-related damage in EU-27 (Figure SPM.8). The overall costs are estimated at €70–€320 billion per year, of which 75% is related to air pollution effects and 60% to human health. The total damage cost equates to €150–€750 per person, or 1–4% of the average European income [22.6] and is about twice as high as the present ‘Willingness to Pay’ to control global warming by carbon emissions trading [22.6]. 52.â•‡ Environmental damage related to Nr effects from agriculture in the EU-27 was estimated at €20–€150 billion per year. This can be compared with a benefit of N-fertilizer for farmers of €10–€100 billion per year, with considerable uncertainty about long-term N-benefits for crop yield [22.6]. 53.â•‡ Apart from the uncertainties inherent in valuing the environment, including the use of ‘willingness to pay’ approaches for ecosystem services, the main uncertainties in these estimates concern the relative share of Nr in PM to human health effects and of Nr to freshwater eutrophication effects [22.6].

Future European nitrogen mitigation and scenarios 54.â•‡ Internalizing the environmental costs for N-intensive agriculture in North Western Europe provides economically optimal annual Nr application rates that are about 50 kg per ha (30%) lower than the private economic optimum rate for the

Figure SPM.8 Estimated environmental costs due to reactive nitrogen emissions to air and to water in the EU-27 [22.6].

farmer. This highlights the importance of increasing nitrogen use efficiency and accounting for external effects on the environment in providing N-recommendations to farmers [22.6]. 55.â•‡ The results also highlight the small overall cost due to N2O emissions compared with NOx, NH3 emissions and Nr losses to water (Figure SPM.8). Although unit costs of N2O, at €6–€18 per kg Nr emitted, are similar to the other issues, N2O emissions are much smaller (para. 21), so that total European damage costs due to N2O are much less than from the other Nr forms. Based on the ‘willingness to pay’ approach and current values, this indicates that the highest policy priority be put on controlling European NOx and NH3 emissions to air and Nr losses to water, as compared with the control of N2O emissions. It is important to target measures that have maximum synergy, reducing emissions of all Nr forms and impacts simultaneously. However, where some measures involve limited trade-offs between Nr (‘pollutant swapping’), Figure SPM.8 indicates that further control of NOx, NH3 and Nr to water would be justified economically even if a proportionate percentage increase in N2O emission were to occur. 56.â•‡ Estimated costs of technical measures to reduce emissions of NOx, NH3 and N2O are available in the IIASA GAINS model. Based on these estimates, future scenarios up to 2030 compare current reduction plans with maximum feasible reduction and a cost optimization approach. This comparison indicates substantial scope for further reductions in NOx and NH3 emissions, supporting the case for revision of the Gothenburg Protocol [24.6]. Although not assessed here, preliminary indications suggest that costs of NH3 abatement measures (€ per kgÂ€Nr) are cheaper than previously estimated, being the subject of ongoing review.2 United Nations Economic Commission for Europe (2010), Options for Revising the 1999 Gothenburg Protocol to Abate Acidification, Eutrophication and Ground-level Ozone:Â€Reactive Nitrogen (ECE/ EB.AIR/WG.5/2010/13).

2

xxxi

Summary for policy makers 6000 5000

NH3_R26 NH3_R45

Gg N/yr

4000

Figure SPM.9 Nitrogen emission scenarios for the EU-27, following the Representative Concentration Pathways (RCP) for three different storylines on radiative forcing. The storyline names indicate the radiative forcing exerted in 2100, between 2.6 (R26), 4.5 (R45) and 8.5 (R85) W per m2 [24.6].

NH3_R85 NOX_R26

3000

NOX_R45 NOX_R85

2000

N2O_R26 1000 0 2000

2020

2040

2060

2080

57.â•‡ Future long-term scenarios emphasize the possibility for major reductions in NOx emissions (by 75% or more for 2000 to 2100), due to improved technologies combined with projected decreases in energy use for some scenarios (Figure SPM.9). By contrast, the anticipated trends for NH3 and N2O are much less clear. A high CO2 scenario representing unrestricted development (+8.5 W/m2 radiative forcing) indicates an increase in NH3 emissions, which does not occur with the more optimistic climate scenarios (+2.6 and +4.5 W/m2 radiative forcing). But even these scenarios highlight a long-term outlook where NH3 quickly becomes the dominant form of Nr emission to the atmosphere, and a key challenge for control policies [24.6]. 58.â•‡ The long term outlook for scenarios of Nr use and emissions must also consider the possible extent of future renewable energy production. There is potential for substantial synergy in increased forest cover, where the main Nr input is atmospheric deposition, allowing increased scavenging of air pollutants and a contribution to carbon sequestration [9.4, 19.4]. By contrast, the increased use of fertilizer Nr to support intensively managed bioenergy and biofuel crops can involve significant tradeoffs, requiring that additional N2O, other Nr and N2 losses be balanced against the carbon benefits (para. 22) [2.4, 24.5].

7.â•‡ The potential for integrated approaches to manage nitrogen A holistic view to managing the nitrogen cascade 59.â•‡ Given the range of adverse environmental effects in the Nr cascade, the most attractive mitigation options are those that offer simultaneous reductions of all N pollutants from all emitting sectors and in all environmental compartments. 60.â•‡ An integrated approach to Nr management holds the promise of decreasing the risks of inconsistency, inefficiency and pollution swapping. Efforts at integration should recognize the varying level of success in Nr policies (para. 23–26) aiming to ensure balance in mitigation efforts between sectors.

xxxii

2100

Integration puts higher demands on interdisciplinarity and consensus building between science, policy and stakeholders [4.6, 23.4]. 61.â•‡ Integrated policies are also justified within sectors, such as agriculture, because of the large number of actors and the connection between sources, sectors and effects [23.4]. The Common Agricultural Policy of the EU provides a potentially powerful incentive to improve sustainability of agricultural production.

Seven key actions for better management of the nitrogen cascade 62.â•‡ Seven key actions in four sectors provide a basis for further developing integrated approaches to N management [23.5].

Agriculture (1)â•‡ Improving nitrogen use efficiency in crop production This includes improving field management practices, genetic potential and yields per Nr input, with the potential to reduce losses per unit of produce, thereby minimizing the risk of pollution swapping [3.3, 22.6, 23.5]. (2)â•‡ Improving nitrogen use efficiency in animal production As with crops, this includes management practices and genetic potential, with an emphasis on improving feed conversion efficiency and decreasing maintenance costs, so reducing losses per unit of produce and the extent of pollution swapping [3.4, 10.3, 23.5]. (3)â•‡ Increasing the fertilizer N equivalence value of animal manure Increasing fertilizer equivalence values requires conserving the Nr in manure during storage and land application (especially reducing NH3 emissions where much Nr is lost), while optimizing the rate and time of application to crop demand [3.4, 10.3, 23.5].

Transport and Industry (4)â•‡ Low-emission combustion and energy-efficient systems These include improved technologies for both stationary


combustion sources and vehicles, increasing energyefficiency and use of alternative energy sources with less emission, building on current approaches [4.5, 23.5, 24.6].

Waste water treatment (5)â•‡ Recycling nitrogen (and phosphorus) from waste water systemsâ•‡ Current efforts at water treatment for Nr in Europe focus on denitrification back to N2. While policies have been relatively successful [4.6], this approach represents a waste of the energy used to produce Nr (para. 22). An ambitious long-term goal should be to recycle Nr from waste waters, utilizing new sewage management technologies [12.3, 23.5].

Societal consumption patterns (6)â•‡ Energy and transport savingâ•‡ Against the success of technical measures to reduce NOx emissions per unit consumption, both vehicle miles and energy use have increased substantially over past decades. Dissuasion of polluting cars and far-distance holidays, and stimulation of energy-saving houses and consumption patterns can greatly contribute to decreasing NOx emissions [23.5]. (7)â•‡ Lowering the human consumption of animal protein European consumption of animal protein is above the recommended per capita consumption in many parts of Europe. Lowering the fraction of animal products in diets to the recommended level (and shifting consumption to more N-efficient animal products) will decrease Nr emissions with human health co-benefits, where current consumption is over the optimum [23.5, 24.5, 26.3]. 63.â•‡ Key Action 4 involves technical measures that are already being combined with public incentives for energy saving and less polluting transport (Key Action 6), linking Nr, air pollution and climate policies (cf. Figure SPM.9). Similarly, each of the Key Actions in the food chain (1–3, 7) offers co-benefits with climate mitigation and the management of other nutrients, including phosphorus. Given the limited success so far in reducing agricultural Nr emissions, more effort is needed to link the Key Actions, both to learn from the successes and to ensure equitability between sectors.

8.â•‡ Challenges for society and policy Nitrogen in multilateral environmental agreements and future research 64.â•‡ International treaties, such as Multilateral Environmental Agreements (MEAs), have done much to protect the global environment, promoting intergovernmental action on many environmental issues, but none has targeted nitrogen management policy holistically [4.3, 25.2]. 65.â•‡ A new international treaty targeted explicitly on nitrogen could be a powerful mechanism to bring the different elements of the nitrogen problem together. While a new convention would be complex to negotiate and could compete with

existing structures, a joint protocol between existing conventions could be effective and should be explored [25.3, 25.4]. 66.â•‡ New coordinating links on nitrogen management between MEAs should be further developed, including the Global Partnership on Nutrient Management facilitated by the United Nations Environment Programme, the Task Force on Reactive Nitrogen of the UNECE Convention on Longrange Transboundary Air Pollution and the links with other UNECE Conventions. There is the opportunity for the UNECE Committee on Environmental Policy to develop nitrogen management links between UNECE Conventions, while the European Union and its Member States have important roles to play in harmonization and coordination [25.4]. 67.â•‡ Such coordination actions will require ongoing support from the scientific community, especially given the many remaining uncertainties inherent in developing the long-term vision of a holistic approach. Research programmes should put a higher priority on quantifying the nitrogen links between the traditional domains of disciplines, media and environmental issues, providing data and models that can underpin future negotiations and policies.

Societal choice, public awareness and behavioural change 68.â•‡ European society is facing major choices regarding food and energy security, and environmental threats including climate change, water, soil and air quality and biodiversity loss. These issues are intricately linked to the nitrogen cycle and have a strong global context, with the decisions of European individuals on life-style and diet having a major role to play [26.3]. 69.â•‡ In Europe, different scenarios and models suggest a strong 75% decline of NOx, while emissions of NH3 and N2O display an uncertain future outlook (Figure SPM.9) [24.6]. The constraints that have so far limited reductions in Nr emissions from agriculture include many stakeholders, an open farming system with diffuse losses, the desire to maintain high outputs for European agro-economy and food security, and possible concerns about how to transfer anticipated costs to consumers (para. 25). Changes in agricultural practices to achieve substantial reductions of European Nr emissions in the coming decades therefore require awareness and broad support from policy, industry, farmers, retailers and consumers [23.3, 26.3]. 70.â•‡ The comparison between combustion and agricultural Nr emissions highlights the need to engage the public. This should emphasize mutual responsibility along the whole food-supply chain, support the basis for transferring any mitigation costs to the consumer, and emphasize that the substantial costs of environmental impacts fully justify taking action [4.5, 23.5, 26.3]. 71.â•‡ At present, public and institutional awareness of the global nitrogen challenge is very low. The comparison with carbon and climate change highlights how the nitrogen story is multifaceted, cutting across all global-change

xxxiii


themes. This complexity is a barrier to greater public awareness, pointing to the need to distil easy messages that engage the public [5.4, 26.4]. 72.â•‡ Simple messages for nitrogen include contrasting its huge benefits for society against the environmental threats, and emphasizing the need to extend existing footprinting

xxxiv

approaches, for example to calculate ‘nitrogen foodprints’. Perhaps the strongest message to the public is that there are substantial health benefits to be gained by keeping consumption of animal products within recommended dietary limits. It is an opportunity to improve personal health and protect the environment at the same time [23.5, 24.5, 26.3].

Technical summary Lead authors:Â€Mark A. Sutton and Gilles Billen Contributing authors:Â€Albert Bleeker, Jan Willem Erisman, Peringe Grennfelt, Hans van Grinsven, Bruna Grizzetti, Clare M. Howard and Adrian Leip

Part Iâ•‡ Nitrogen in Europe:Â€the present position Nitrogen inheritance 1. Gaseous di-nitrogen (N2) constitutes 78% of the earth’s atmosphere. It is a rather inert chemical, being nearly unavailable for the biological cycle. The other nitrogen forms are much more reactive; these include nitrate (NO3−), ammonium (NH4+) and ammonia (NH3), gaseous nitrogen oxides (NOx), nitrous oxide (N2O) and many other inorganic and organic nitrogen forms. Collectively, they are termed ‘reactive nitrogen’ (Nr). They are normally scarce in natural environments, with their low availability limiting the productivity of natural ecosystems. This was also the case for agricultural production before 1900, which long remained dependent on the recycling of Nr in human waste and manure, and the capacity of legumes to fix atmospheric N2 biologically. 2. With a growing human population through the nineteenth century and the need for more Nr, Europe increasingly operated a ‘fossil nitrogen economy’, dependent on the addition of nitrogen fertilizers from mined sources, including from guano, coal and saltpetre. The ‘nitrogen problem’ of the time was that these sources were fast becoming insufficient to meet Europe’s escalating need for fertilizer Nr, and its military need for Nr in explosives [1.1].1 3. The situation changed dramatically shortly after 1900, with the invention of the Haber–Bosch process. This allowed the cheap industrial production of ammonia from di-nitrogen and hydrogen, permitting mass production of synthetic Nr fertilizers. By the 1930s, the European shortage of Nr had become a problem of the past, with Nr use in agriculture strongly increasing from the 1950s [1.1, 2.2]. 4. The deliberate production and release of Nr in the Haber– Bosch process can be considered as perhaps the greatest single experiment in global geo-engineering that humans have ever made [1.1]. In Europe, human production of Nr fully met its 1

References in this summary refer (e.g., [1.1, 2.2]) to chapter and section numbers of the European Nitrogen Assessment.

objectives to underpinning food and military security, while supplying a vital feedstock for many industrial processes [3.2, 3.5]. What was not anticipated was that this experiment would lead to a ‘nitrogen inheritance’ of unintended consequences [1.1], with Nr leaking into the environment in multiple forms, causing an even larger number of environmental effects [1.1, 2.6]. Simultaneously, the increasing extent of fossil fuel combustion for transportation and electricity production has led to a massive unintentional additional release of Nr into the atmosphere, mainly as nitrogen oxides (NOx) [2.4]. 5. At the global scale, together with crop biological nitrogen fixation, these processes have altered the nitrogen cycle to an unprecedented extent, and much more than that of carbon or phosphorus. Humans introduce more Nr into the biosphere than all natural processes together [2.5, 13.2, 16.4]. Europe (EU-27) is a hot spot in this sense, producing 10% of global anthropogenic Nr, even though its surface covers less than 3% of the total world continental area [2.5, 13.2].

Benefits, threats and current policies 6. The value of the benefits brought to the European economy by the production of fertilizers and the combustion of fuel is substantial. For example, the economic benefit of applying Nr fertilizers to wheat in the EU is estimated at around €8 billion per year. Accounting for benefits to other crops, livestock production, downstream food processing and many industrial benefits (including mining and chemical synthesis), the total benefits of Nr production will be very much larger [3.6]. By contrast, the formation of NOx through high temperature combustion processes has no economic benefit, as control efforts focus on denitrifying Nr rather than its use [5.1]. 7. Against these benefits must be listed the many effects on the environment and human health, as a result of Nr pollution. Several of these threats have been addressed by governmental policy measures related to abating atmospheric pollution or limiting nitrogen contamination of groundwater and surface water resources. These include several European directives, such as the Nitrates Directive, Water Framework Directive, Groundwater Directive, Ambient Air Quality Directive, National Emissions Ceilings Directive, Urban Waste Water Treatment Directive,

xxxv

Technical Summary

Figure TS.1 Simplified view of the nitrogen cascade, highlighting the major anthropogenic sources of reactive nitrogen (Nr) from atmospheric di-nitrogen (N2), the main pollutant forms of Nr (orange boxes) and nine main environmental concerns (blue boxes). Estimates of anthropogenic N fixation for the world (Tg /yr for 2005, in black) are compared with estimates for Europe (Tg /yr for 2000, in blue italic). Blue arrows represent intended anthropogenic Nr flows; all the other arrows are unintended flows [5.2].

Marine Strategy Framework Directive, Integrated Pollution Prevention and Control (IPPC) and the Habitats Directive [4.4]. 8. The policy responses also include European commitments to multi-lateral environmental agreements, including the United Nations Economic Commission for Europe (UNECE) Convention on Long-Range Transboundary Air Pollution, the UN Framework Convention on Climate Change, the UN Convention on Biological Diversity, and the Oslo and Paris, Helsinki and Barcelona Conventions for the protection of the North East Atlantic, the Baltic Sea and the Mediterranean Sea, respectively [4.3]. 9. Review of the current policies highlights a tendency to address individual Nr species from specific source sectors (agriculture, traffic, industry), media (air, freshwater, marine), and for specific issues (climate, urban air pollution, biodiversity, water quality, etc.). Until now, there has been little focus on developing policies that recognize the full extent and complexity of the nitrogen cycle [4.3, 4.4]. 10. Trends in environmental pollution show that significant progress has been made in reducing emissions of NOx to air. These policies have benefited from the availability of measures targeted at few stakeholders (e.g. electricity generation companies, industry, vehicle manufacturers), but have nevertheless been offset by increases in overall transport use and energy consumption. In the same way, significant progress has been

xxxvi

made in reducing water pollution due to water treatment policies that have engaged water companies [4.6]. 11. By contrast, from a European perspective, Nr pollution from agriculture has shown only modest reductions in response to policies over the past 20 years [4.6]. Particular challenges faced in controlling Nr losses from agriculture are the many (often small) stakeholders, a highly diverse open system, and the perception of difficulty in passing anticipated costs to consumers [4.6]. In addition, the low price of Nr fertilizer, combined with its clear benefits to agricultural production, does not provide a strong incentive for farmers to use less than the (private) economic optimum [3.3].

Nitrogen cascade and the need for integration 12. The lack of a holistic approach to developing nitrogen policies can partly be explained by the sectoral approach taken in government departments, and can also be linked to the tendency to scientific specialization. Nitrogen research communities have developed to become highly fragmented, providing a major challenge to integrate across all nitrogen forms, proÂ� cesses and scales [5.2]. 13. The ‘nitrogen cascade’ (Figure TS.1) provides a useful concept to demonstrate the benefits of a holistic understanding

Technical Summary

of the nitrogen cycle [2.6, 5.2]. Anthropogenic fixation of N2 to Nr raises the energy state of the nitrogen, with the energy being gradually dissipated as the Nr is converted through many different forms, until it is eventually denitrified back to N2, the thermodynamically stable form of nitrogen in most environments. Even though Nr may be emitted from different proÂ� cesses and sectors, once released into the environment, the origin gradually becomes less relevant, causing multiple environmental effects. Each molecule of Nr emitted may cause several effects before eventual denitrification [2.6, 5.2]. 14. The cascade concept highlights the potential for tradeoffs and synergies in managing Nr. For example, if losses of Nr to one pathway are reduced, this can easily increase losses of another Nr form (e.g., some meaures to reduce nitrate leaching may increase ammonia emissions and vice versa). Such tradeoffs are sometimes termed ‘pollution swapping’. By contrast, because of the cascade, some measures may have benefits in reducing multiple forms and impacts of Nr [1.2]. 15. It is concluded that a more holistic approach to managing the nitrogen cycle would have benefits to ensure more effective control of the different forms of Nr pollution and impacts. Such a development must be based on more-effective integration across environmental disciplines, providing the foundation to link traditionally separate policy domains [5.3].

Approach of the European Nitrogen Assessment 16. Developing a more joined up approach to managing the nitrogen cycle necessarily proceeds gradually, and this has been encouraged through the European Nitrogen Assessment process. The overall goal of the Assessment was established as:Â€ to review current scientific understanding of nitrogen sources, impacts and interactions across Europe, taking account of current policies and the economic costs and benefits, as a basis to inform the development of future policies at local to global scales [1.4]. 17. In developing this vision of gradual integration, based on analysis of the present position (Part I), the Assessment first examines the nitrogen turn-over processes in the biosphere (Part II), and then addresses nitrogen flows at different spatial scales (Part III). 18. One of the key conclusions of the first part of the Assessment is that the complexity of the nitrogen cycle needs to be distilled to highlight the priority concerns. This is important to limit the number of interactions when developing integrated approaches [5.4]. 19. Recognizing these issues, the Assessment process established a comprehensive list of around 20 problems related to nitrogen. The list was first distilled down to nine ‘main environmental concerns’, setting the agenda for the Nitrogen in Europe (NinE) programme, as shown in Figure TS.1 [5.4]. 20. In a second stage, the list was reduced to five ‘key societal threats’ of excess nitrogen, identified as:Â€Water quality, Air quality, Greenhouse balance, Ecosystems and biodiversity, and Soil quality. These five threats provide a framework that automatically includes many of the other issues, balancing the complexity of the nitrogen cycle with the need for simplification [5.4].

21. The short-listing of the five key threats also provides a useful tool to communicate the nitrogen challenge to society. Together the five threats make an acronym as the ‘WAGES’ of excess nitrogen, while they can be also envisaged in direct analogy to the ‘elements’ of classical Greek cosmology (Figure TS.2) [5.4]. 22. The Assessment applies the framework of five key societal threats to summarize the scale of the nitrogen challenge facing Europe (Part IV). Finally, the threats are brought together to examine the future perspective for European nitrogen policies (Part V). The Assessment used a network approach, where expert teams were formed for each chapter based on open invitations, including discussions of outlines during workshops held for each of the five parts (I–IV). The chapters take a variety of approaches across the Assessment, reflective of variation in data availability (e.g., limited data for some parts of Europe) and the nature of the issues being assessed. Draft chapters were subjected to internal and external peer review before being finalized.

Part IIâ•‡ Nitrogen processing in the biosphere 23. In recent decades substantial advances have been achieved in our understanding of the processes that govern nitrogen cycling in terrestrial environments (including natural and agricultural ecosystems), in aquatic environments (including freshwater, estuarine and marine ecosystems) and in the atmosphere. Each of these environments has been considered, integrating the processes of all relevant nitrogen forms.

Reactive nitrogen turnover in terrestrial ecosystems 24. The understanding of N cycling in terrestrial ecosystems has undergone a paradigm shift since 1990. Until then, the perception was that:Â€(1) Nr mineralization is the limiting step in N cycling; (2) plants only take up inorganic Nr; and (3) plants compete poorly for Nr against microbes and use only the Nr which is ‘left over’ by microbes. Since then studies have shown that plants compete effectively for Nr with microÂ� organisms and take up organic N in a broad range of ecosystems [6.4]. 25. On the ecosystem scale, soils are the main reservoir for Nr. This is more pronounced for agricultural systems than for forest systems, with more than 90%–95% of Nr being stored in the soil. Nitrogen stocks of managed systems are typically depleted and with retention processes negatively affected [6.2, 6.4]. 26. In cereal farming, the use of only mineral Nr fertilizers, instead of animal manures or composts, as well as the simplification of the crop rotation scheme that this has made possible, has in some cases resulted in a decline of soil organic matter. In the long-term this practice of using only mineral fertilizers has decreased the buffer capacity of the soil towards inorganic N inputs, thus increasing its propensity to Nr leaching [6.4]. 27. Nitrogen fixation in non-agricultural legumes or in other N-fixing organisms remains difficult to quantify, hampering a

xxxvii

Technical Summary

Figure TS.2 Summary of the five key societal threats of excess reactive nitrogen, drawn in analogy to the ‘elements’ of classical Greek cosmology. The main chemical forms associated with each threat are shown [5.4]. Photo sources: Shutterstock.com and garysmithphotography.co.uk.

better understanding of the importance of biological N2 fixation for most terrestrial ecosystems [6.3]. 28. Nitrogen-enriched terrestrial ecosystems lose significant amounts of N via nitrate leaching and gaseous emissions (N2, N2O, NO, NH3) to the environment. Estimates of denitrification to N2 remain highly uncertain, due to difficulties in measurement and a high degree of temporal and spatial variability. There remain substantial uncertainties in the average fraction of Nr applied to fields that is emitted as N2O, ranging from 1% to 3.5%–4.5% of fertilizer N applied, using bottom-up and topdown estimates, respectively. Further research is needed to better understand the relative contribution of direct and indirect N2O emissions [6.5]. 29. In forests, the C:N ratio of the forest leaf litter or top mineral soil is a good indicator of Nr status related to nitrate leaching. At C:N above 25, mineral Nr is usually retained, whereas below 25, nitrate leaching increases with increasing Nr deposition [6.5] (Table TS.1).

Reactive nitrogen turnover and transfer along the aquatic continuum 30. Major sources of Nr in the aquatic environment include households and sewage discharges together with diffuse pollution losses from agricultural practices. 31. Nitrate retention by riparian wetlands is a frequent justification for conservation and restoration policies of these systems. However, their use for mitigating NO3 contamination of river systems must be treated with caution, since their effectiveness is difficult to predict, and side effects observed include

xxxviii

increased dissolved organic matter and N2O emissions, together with loss of biodiversity [7.5]. 32. Release of dissolved organic nitrogen has often been neglected, while it can play a significant role, particularly in upland semi-natural catchments, but is not determined in most routine European water quality monitoring programmes [7.3]. 33. The effects of increased Nr loadings to aquatic environments include acidification and loss of biodiversity in seminatural environments, and eutrophication in more disturbed ecosystems. Standing waters are particularly sensitive to both acidification and eutrophication, since the longer residence time in these systems leads to greater interaction between the biota and changing water chemistry [7.4]. 34. The richest submerged plant communities in lakes have been observed to be associated with winter nitrate concentrations not exceeding 2 mg N/l, and this has been proposed as an appropriate target concentration for enriched shallow European lakes to reach ‘good ecological status’ [7.5]. 35. Although phosphorus is often the main limiting element controlling primary production in freshwater systems, Nr has been reported as a factor limiting or co-limiting biological production in some eutrophicated lakes, and control of both Nr and P loading is needed in impacted areas, if ecological quality is to be restored [7.4]. 36. The importance of storage and denitrification in aquifers is a major uncertainty in the global N cycle, and controls in part the response of catchments to land use or management changes. In some aquifers, the increase of N concentrations will continue for decades even if efficient mitigation measures are implemented now [7.5]. 37. Nitrogen inputs from human activities have led to ecological deterioration in large parts of the coastal oceans along European coastlines, including harmful algal blooms and anoxia. The riverine Nr-loads are the most pronounced Nr source to coasts and estuaries, while atmospheric Nr deposition and N2 fixation also contribute significantly [8.8]. 38. A large imbalance of Nr with respect to silica inputs causes the development of severe harmful algal blooms. Especially affected by eutrophication are the major European estuaries (e.g., Rhine, Scheldt, Danube and the coastlines receiving their outflow), North Sea, Baltic Sea, and Black Sea, as well as some parts of the Mediterranean coastline [8.10]. 39. Marine biodiversity is reduced under high nutrient loadings, affecting nutrient recycling negatively. Recovery of communities may not be possible if eutrophication and anoxia persist for long time periods of several years [8.9]. 40. The European coastal zone plays a major role in denitrification of Nr to N2. Export of Nr to the sea is estimated at 4.5 TgN per year, most of which will be denitrified to N2. Globally, coastal denitrification is estimated at 61 TgN per year, including 8 Tg per year in estuaries. Comparison with independant estimates of estuarine and sediment denitrification implies that coastal systems impert 54–197 TgN per year from the open ocean globally, compensating the Nr losses due to sediment denitrification [8.8].

Technical Summary Table TS.1 Characteristics of coniferous forest ecosystems with low, intermediate and high N status, as grouped according to total Nr input [6.5]

Nitrogen status

Low N status (N-limited)

Intermediate

High N status (N-saturated)

Input (kg N per ha per yr)

0–15

15–40

40–100

Needle N% (in spruce)

< 1.4

1.4–1.7

1.7–2.5

C:N ratio (g C per g N)

> 30

25–30

< 25

Soil N flux density proxy (litterfall + throughfall) (kg N per ha per yr)

< 60

60–80

>80

Proportion of input leached (%)

5000 autotrophic 5000 – 1000 1000 – -1000 balanced -1000 – -5000 -5000 – -10000 heterotrophic < -10000

itoring communities (terminology, methodology, units) and the exchange of information [synthesis of 13, 14].

European scale nitrogen flows 65. European nitrogen flows and budgets have been estimated for terrestrial ecosystems (agriculture, forest and other ecosystems) and for all systems combined (also including urban, transport, industrial and aquatic flows). 66. For the year 2000, a comparison of four mass balance models estimated Nr inputs to agriculture in the EU-27 at 23–26 Tg N, being mainly due to fertilizer and animal manure. For emissions from agriculture, the comparison showed similar relative estimates for NH3 (2.8–3.9 Tg N) and N2O (0.35–0.46 Tg N), but diverging results for soil NOx (0.02–0.20 Tg N). The largest absolute uncertainties were for NH3 [15.6]. 67. Inputs of Nr to soils from both fertilizers and manure increased between 1970 and 2010 by ~20% for the EU-27. Although cattle numbers decreased, this trend was more than

Figure TS.3 Difference between nitrogen autotrophic and heterotrophic contributions for territories in the EU-27. Autotrophy is defined as the amount of nitrogen in crop and grass produced by agriculture. Heterotrophy is the amount of nitrogen feed ingested by livestock and food consumed by humans. The map shows overall net autotrophy (positive values) and net heterotrophy (negative values) [synthesis from 13.2 and 16.4].

offset by increased Nr excretion rates per cow (e.g., increased milk production per cow) [15.5]. These increases are consistent with overall estimated increases in NH3 and N2 emissions and Nr leaching by ~10% between 1970 and 2000, with emissions decreasing slightly since peak values around 1985. Emissions per unit agricultural area have increased by 20%–30% as a result of intensification over the period in western Europe [15.5]. 68. The estimated distribution of overall Nr losses to the environment is shown in Figure TS.4. Emissions to the atmosphere reflect the distribution of livestock production (dominating NH3 emissions) and human population centres (dominating NOx emissions) across Europe. The distribution of Nr inputs to aquatic systems is dominated by nitrate losses, which are largest in areas with high livestock density and precipitation excess, while more localized peaks are associated with urban waste-waters [16.3]. 69. An overall budget of the present N cycle has been established (around 2000 for EU-27) and is summarized in

xli

Technical Summary

Figure TS.4 Distribution of reactive nitrogen emissions across Europe (kg N per km2 for 2000) including emissions to air as NOx, NH3 and N2O, and total losses to aquatic systems, including nitrate and other Nr leaching and wastewaters [16.3].

FigureÂ€TS.5. The budget highlights the central role of crop production and livestock farming. The annual Nr brought to agricultural soils at 27.5 Tg N (consisting of 11.2 Tg N as synthetic fertilizers, 7.1 Tg N as manure, 2.4 Tg N as atmospheric deposition and 1.0 Tg N through biological nitrogen fixation and 5.8Â€Tg N as crop residues) is in surplus over the requirements of crop production (17.6 Tg N). The annual Nr surplus of 9.9 Tg contributes to substantial Nr leaching to surface and groundwater (6 Tg N), denitrification to N2 (4.5 Tg N), volatilization as NH3 (1.6 Tg N) and emission of N2O and NO (0.5 Tg N). The overall balance for agricultural soils implies a small annual depletion of soil Nr stocks, although this term is considered to be very uncertain since it is calculated by difference of several uncertain estimates. It represents the regional average of a net loss of Nr in soils of autotrophy-dominated regions with arable farming, and a net gain of Nr in soils of heterotrophyÂ�dominated regions with intensive livestock farming [16.4]. 70. In order to provide an annual consumption of animal products by humans of 2.3 Tg N, livestock farming in Europe uses five times as much Nr from crops and imported feed (11.8 Tg), driving the overall European agricultural N cycle. By comparison, direct consumption by humans of crops grown in Europe represents only 2.0 Tg N per year. The handling of livestock excreta also leads to direct gaseous emissions of 1.5 Tg N per year. Overall, in order to produce 4.3 Tg N of food

xlii

annually for the European population (not including 0.4 Tg N in imported food and feed), three times as much Nr is emitted to the environment, which corresponds to a nitrogen use efficiency of 30% (as compared with a global average of 50%) [16.4]. 71. To these emissions should be added 3.7 Tg N per year of wastewater discharged to surface waters and 3.4 Tg N per year of NOx emitted through fossil fuel combustion by the energy, industry and transportation sectors [9.2, 16.3]. 72. As a whole, at around the year 2000, the EU-27 is a net commercial importer of animal feed and human food of 3.5 Tg N per year (mainly due to feed). Conversely, Europe is a net exporter of Nr to the environment:Â€ by atmospheric transport (2.4 Tg N per year) and by river export to marine systems (4.5 Tg N per yr). The largest single sink for Nr is denitrification to N2 in soils, river sediments and the sediments of European shelf regions, estimated at 9.3 Tg N per year, which is one of the most uncertain estimates [16.4, 8.8]. A net transfer from anthropogenic to natural systems is also implied, as estimated annual losses (0.7 Tg N) are less than half the estimated inputs (1.6 Tg N) mainly from atmospheric Nr deposition. 73. A similar budget has been reconstructed for the same territory (the present EU-27) in the beginning of the twentieth century (Figure TS.6). The estimates are necessarily less certain than for 2000, and the hypotheses leading to this

Technical Summary

Europe (EU27), around 2000. N fluxes in TgN/yr

Atmospheric N2 pool Net atmosph. export

2.4

9.7

6

3.5

3.1

N2fix indust & traffic

4

3.4 3.8 Nat N2fix

Wood exp.

Atm depos

0.2

3.8

0.3

2

crop N2fix

Atmospheric NH3, NOx, N2O

0.4

7

11.8

Fertilizers

2.3

Net import of food & feed

5

2

11.2

1.0

Crop production

Human nutrit.

Livestock farming

17.6

5

5.8

1

3.1 wwt

3.6 Semi-nat soils

0.2

1

2.4

13.6

Agricult soils

1.4 NH3,NOx & N2O emission

2.1

8

1.5 0.1

landfill

4.5 6.2 1.8

Leachg & runoff

Denitrification

3

2.7

4.6 0.6

4.5


Figure TS.5 The N cycle at the scale of EU-27 [simplified from 16.4] for the year 2000. Fluxes in green refer to ‘natural’ fluxes (to some extent altered by atmospheric Nr deposition), those in blue are intentional anthropogenic fluxes, those in orange are unintentional anthropogenic fluxes. The numbered green circles indicate a package of seven key actions for overall integrated management of the European nitrogen cycle (see para. 111) [23.5].

retrospective reconstruction are described in the ENA supplementary material [16]. 74. By comparison with present-day amounts, the use of mineral fertilizers around the year 1900 (mainly from Chilean salpetre, guano and coal) was very small. The primary source of new Nr in agriculture was biological N fixation by legume fodder crops, typically grown once every three years in triennial rotations. The nitrogen fixed by legume crops was brought to arable soils by the incorporation of crop residues and application of animal manures. Losses of Nr from agricultural soils, though already significant, were only one third of the current level (Figure TS.6). 75. Annual atmospheric deposition of Nr was around 1.9 Tg N in 1900, roughly half of which was deposited each to agricultural and semi-natural land. This was roughly half of the atmospheric deposition in 2000 at 3.8 Tg N, of which only 37% is deposited to semi-natural land, reflecting the overall reduction in area of semi-natural land (Figure TS.6). As with 2000, the overall difference between Nr removals and gains for

agricultural soils is not considered significant, while the average gains to semi-natural soils including forests (1.1 Tg N) were larger than the removals (0.6 Tg N), but less so than in 2000.

Part IVâ•‡ Managing nitrogen in relation to key societal threats 76. For each of the five key societal threats of excess reactive nitrogen, the Assessment examined the scale of the concern, including, where possible, information on temporal trends and the progress made through any existing policy measures.

Water quality 77. Anthropogenic increase of Nr in water poses direct threats to humans and aquatic ecosystems. High nitrate concentrations in drinking water are considered dangerous for human health, as they might cause cancers and (albeit rarely) infant methaemoglobinaemia. There is also evidence for benefits of nitrate

xliii

Technical Summary Europe (EU27), around 1900. N fluxes in Tg N /yr

Atmospheric N2 pool

2.1

N2fix indust & traffic

0.6

2

Crop N2fix

Atmospheric NH3, NOx, N2O

4

1.9 Nat N2fix

Wood exp.

0.2

0.2

Atm depos

1.9

Fertilizers

0.2

9.6

7.6

Crop production

1.1

semi-nat soils

0.3 0.3

1.2 0.9

Agricult soils

1 NH3,NOx & N 2O emission

1.2

6

1

0.6

2

0.8

1.2 1.2 Leachng & runoff

Denitrification

Human nutrit.

Livestock farming

0.9

1.2

1.2 2.3


Figure TS.6 Reconstruction of the N cycle at the scale of current EU-27 for the year 1900 (16.4, supplementary material). Same colour code is used as in Figure TS.5.

for cardiovascular health and protection against infections. In aquatic ecosystems the Nr enrichment produces eutrophication, which is responsible for toxic algal blooms, water anoxia, fish kills and biodiversity loss [8.8, 17.3]. 78. In addition to high Nr concentrations in European waterbodies, increasing nitrate in groundwaters threatens the long-term quality of the resource, as nitrate may have long residence time in the aquifers, and it can be expected that past fertilizer strategies will impact for many decades the quality of European groundwaters [7.5, 17.2]. 79. About 3% of the population in EU-15 using drinking water from groundwater resources is potentially exposed to concentrations exceeding the standard for drinking water of 50 mg NO3/l (11.2 mg N/l), with 5% of the population is chronically exposed to concentrations exceeding 25 mg NO3/l (5.6 mg N/l), which may double the risk of colon cancer for above median meat consumers [17.3]. 80. A value of 1.5 mg N/l has been considered as the total Nr limit above which freshwater bodies may develop loss of

xliv

biodiversity and eutrophication. Except in Scandinavia and in mountainous regions, this level is already exceeded in most European freshwater bodies (Figure TS.7) [7.5, 17.3]. 81. With Nr inputs to the coastal zone at four times the natural background (para. 61), large areas along Europe’s coastline are suffering severe eutrophication problems, with anoxia and/ or the proliferation of undesirable or toxic algae. Particularly affected are the south-eastern continental coast of the North Sea, the Baltic Sea (except the Gulf of Bothnia), the coasts of Brittany, the Adriatic Sea and the western coastal Black Sea [8.10, 13.7, 17.3]. 82. Although eutrophication is decreasing, existing international policies have not been fully implemented, and even under favourable land use scenarios, Nr export to European waters and seas is anticipated to remain a problem in the near future. Achieving substantial progress at the European scale requires integration of sectoral policies, reducing overall inputs of Nr to watersheds, e.g., through changes in agriculture and other N flows [4.5, 13, 17.5].

Technical Summary

Figure TS.7 Indication of the nitrogen threat to water quality. Potential risk of eutrophication for surface freshwater based on estimated total Nr concentrations. The three classes of risk are:Â€low, 1.5 mg/l as total Nr concentration in water [17.3.3].

Air quality 83. Emissions of NOx and NH3 contribute to several negative effects on human health and ecosystems. In addition to effects of NO2, secondary pollutants play key roles. These include ground level ozone (O3), formed photochemically in the presence of NO2 and volatile organic compounds (VOC), and inhalable particulate matter (PM), formed from oxidation of NO2 to HNO3 and reaction with NH3 to form ammonium nitrates. NOx, O3 and PM cause or aggravate asthma, reduced lung functions and bronchitis. Chronic exposure may increase the probability of respiratory or cardiovascular mortality and cancers [18.2]. 84. Direct NO2 and ozone damage to vegetation has been recognized for a long time, as well as to materials, buildings and objects of cultural heritage. There is a difficulty of ascribing health effects to NO2 per se at ambient levels rather than considering NO2 as a surrogate for a traffic-derived air pollution mixture [18.2]. 85. The role of particulate ammonium and nitrate in human health effects is still under discussion. Current approaches assume damage on a mass basis for PM with a median diameter less than 2.5 μm (PM2.5). Nr compounds contribute up to 30–70% of PM2.5 mass in Europe [18.5]. Overall, models estimate a loss of statistical life expectancy due to PM of 6–12 months across most of central Europe (Figure TS.8). There has been a low success in controlling NH3 emissions in Europe

which needs to be further assessed, in particular in connection with the development of new agricultural policies [18.6, 4.5]. 86. In the EU-27 countries, 60% of the population lives in areas (mainly urban) where the annual EU limit value of NO2 is exceeded. Levels have decreased since 1990, although the downward trends have been smaller or even disappeared after 2000. Although episodic O3 levels have decreased since 1990 due to VOC and NOx control, continental background concentrations have increased, with O3 levels remaining a threat to human health and ecosystems [18.5, 4.5].

Greenhouse balance 87. European anthropogenic Nr emissions have a complex effect on climate by altering global radiative forcing. They directly affect the greenhouse gas balance through N2O emissions and indirectly affect it by increasing tropospheric O3 levels, altering methane (CH4) fluxes, and by altering biospheric CO2 sink (including atmospheric Nr deposition and O3 effects). Aerosol formed from NOx and NH3 emissions also has a cooling effect [19]. 88. A first assessment has been made of the overall effect (between 1750 and 2005) of European Nr emissions on radiative forcing. The main warming effects of European anthropogenic Nr emissions are estimated to be from N2O (17 (15Â€– 19) mW/m2) and from the reduction in the biospheric CO2 sink by tropospheric O3 (4.4 (2.3Â€– 6.6) mW/m2). The main cooling

xlv

Technical Summary Figure TS.8 Indication of the nitrogen threat to air quality. Across Europe Nr typically accounts for up to 0.3–0.7 of total particulate matter load with a median diameter less than 2.5 μm (PM2.5) on a mass basis [18.5]. Assuming that health effects are proportionate to mass [18.2.2] the map indicates the loss in statistical human life expectancy (in months) attributable to total PM2.5 [18.6, supplementary material].

12

effects are estimated to be from increasing the biospheric CO2 sink by atmospheric Nr deposition (−19 (−30 to −8) mW/m2) and by light scattering effects of Nr containing aerosol (−16.5 (−27.5 to −5.5) mW/m2) (Figure TS.9) [19.6]. 89. Overall, European Nr emissions are estimated to have a net cooling effect, with the uncertainty bounds ranging from substantial cooling to a small net warming (−15.7 (−46.7 to +15.4) mW/m2) [19.6]. 90. The largest uncertainties concern the aerosol and Nr fertilization effects, and the estimation of the European contributions within the global context. Published estimates suggest that the default N2O emission factor of 1% used by the Intergovernmental Panel on Climate Change (IPCC) for indirect emissions from soils following Nr deposition is too low by at least a factor of two [6.6, 19.6]. 91. Industrial production of Nr can be considered as having permitted increased livestock and human populations (and associated food, feed and fuel consumption). The expected substantial net warming effect of these wider Nr interactions remains to be quantified. Although individual components of Nr emissions have cooling effects, there are many opportunities for ‘smart management’ linking N and C cycles. These can help mitigate greenhouse gas emissions, while reducing the other Nr-related environmental threats [19.7].

Terrestrial ecosystems and biodiversity 92. Atmospheric Nr deposition is a significant driver of biodiversity loss in terrestrial ecosystems. Rates of Nr deposition have substantially exceeded critical load thresholds for natural and semi-natural areas since the increase in agriculture-, energyand transport-related emissions from the 1950s, resulting in a considerable loss of biodiversity in Europe [5.1, 8.2, 20.4].

xlvi

93. Nr deposition affects vegetation diversity through direct foliar damage, eutrophication, acidification, and pathogen susceptibility. The most vulnerable habitats are those with species adapted to low nutrient levels or sensitive to acidification, and include grassland, heathland, wetlands and forests [20.3]. First estimates of the overall reduction in biodiversity due to Nr deposition across Europe have been made (Figure TS.10), while the reductions in Nr- sensitive species will be even greater [20.4]. 94. Although it is not yet clear to what extent oxidized Â�versus reduced Nr (e.g., NO3−, NH4+) have different effects on biodiversity, gaseous ammonia (NH3) can be particularly harmful to vegetation, especially lower plants, through direct foliar damage [20.3]. Changes in plant communities may also indirectly affect faunal biodiversity. It is likely that Nr deposition acts synergistically with other stressors, in particular climate change, acid deposition and ground-level ozone, but these synergies are poorly understood [20.2]. 95. Because of cumulative effects of Nr inputs, it is likely that biodiversity has been in decline for many decades due to Nr deposition. This implies that recovery after a reduction in deposition is likely to be slow, and in some cases may require active management intervention in the habitats affected [20.5].

Soil quality 96. The major Nr threats on soil quality for both agricultural and natural soils are soil acidification, changes in soil organic matter content and quality and loss of soil biodiversity linked to eutrophication. Application of N fertilizers and manure and atmospheric Nr deposition cause soil acidification, which leads to a decrease in crop and forest growth and increases leaching of components negatively affecting water quality, such as heavy metals [21.3].

Technical Summary

426 [382 to 469]

Fossil fuel & landuse change CO 2 biospheric CO2

Long-lived greenhouse gases

(incl. atmos. fertilization & O3 effect)

CH4 (decreased atmospheric lifetime & and decreased soil uptake)

N2 O

-74 [-86 - -62] -19 [-30 - -8 ] 4.4 [2.3 - 6.5] 24.5 [22-27 ] -4.6 (-6.7 - -2.4 ) 0.13 [0.03 - 0.24 ] 17.0 [14.8 - 19.1] 17.0 [14.8 - 19.1] 7.5 [4.5 - 10.5 ]

Halocarbons

1 year

93.1

73.7

6.2%

6.6%

141.7

127.6

19.4%

26.3%

Female cattle for replacement < 1 year

44.3

39.5

13.1%

14.0%

Female cattle for replacement > 1 year

95.9

76.7

6.1%

6.4%

157.0

144.2

17.8%

23.3%

Veal calves on milk

10.6

10.7

51.2%

49.8%

Veal calves on fodder maize

30.8

27.4

28.9%

28.6%

Male beef cattle < 1 year

28.9

26.0

28.0%

30.7%

Ruminant animals Dairy:Â€Regions with high fodder maize ration

Dairy cows; lactating cows Dairy:Â€Regions with low fodder maize ration

Dairy cows; lactating cows Beef cattle:

Male beef cattle > 1 year

72.6

53.8

10.9%

15.7%

110.7

84.9

10.1%

11.4%

Sheep (including lambs)

25.0

14.4

9.3%

13.5%

Goats (including kids)

19.9

16.0

15.9%

25.2%

Suckling cows

Horses

58.4

1.9%

Monogastric animals Fattening pigs

14.3

12.9

29.8%

35.6%

Sows (including piglets)

33.8

30.8

28.0%

36.8%

Broilers

0.61

0.53

41.0%

49.9%

Laying hens < 18 weeks

0.38

0.34

22.9%

26.9%

Ducks for meat

1.12

0.76

34.3%

49.3%

Turkeys

1.98

1.71

37.5%

44.8%

NUE is calculated as the ratio of nitrogen in milk and meat over nitrogen intake with roughage and concentrates. The milk production per cow for the two indicated years was 6050 and 7926 litres per year. The calculation method made use of country-wide average feed intake levels for all individual animal categories concerning concentrates, ensiled grass and fodder maize. For ruminants, the amount of consumed grass was assumed to close the energy demand of the animals.

Excretion rates can be calculated as part of an input–output balance approach on animal level, using the simple equation Excretion = feed intake – animal products. Assessments of the N excretion rates and the corresponding N use efficiencies for the main animal categories in the Netherlands show that dairy cattle N excretion rates in regions with a high share of fodder maize in the ration are lower than those in regions with low share (Table 3.7). An increasing

milk yield does not necessarily lead to increased N excretion rates, partly caused by decrease of the amount of applied fertilizer N in the same period, resulting in a lower protein content of the consumed grass. However, according to Witzke and Oenema (2007) comparing data from EU member states, there is a reasonably close positive relationship between milk yield and N excretion rates, with a standard variation across Europe of ±23%. It should be underlined that these N excretion rates are in fact the resultant of different animal rations with varying protein contents between the EU member states.

51

Benefits of nitrogen for food, fibre and industrial production

Ruminants generally have a lower NUE than pigs and poultry, as seen from Table 3.7. Finally, young animals have a higher NUE than older animals. Using the same principle for NUE, an intensive Italian survey was undertaken for estimating N excretion rates. Based on about 10 000 cows with an average milk yield of 8366 litre of milk per year, an average N excretion of 116 kg per year was found. Compared to the Netherlands, the Italian rations were mainly based on corn silage with lower feed N concentrations, thereby reducing N excretion (Xiccato et al., 2005). Feeding trials with high yielding cows on a research farm of the Swedish University of Agricultural Sciences showed that varying the crude protein content of the ration between 135 and 184 g crude protein per kg feed resulted in NUE for milk production between 18% and 40%, with the lower protein content increasing NUE (Nadeau et al., 2007). Another approach is to use a model based on feed intake according to the energy requirements for maintenance, meat and milk production; this will enable improved optimization of NUE and, hence, lowering of N excretion rates (Vérité and Delaby, 2000; Peyraud and Delaby. For an overview of N recovery efficiencies in EU-27, the USA and the Netherlands see supplementary material for ChapterÂ€3. 2004; Dämmgen et al., 2009).

3.4.5â•‡ Economic value of N in dairy farming To illustrate the economic value of N in livestock production, the case of dairy farming is used here as an example.

Grassland productivity and N response Grassland productivity is affected by climatic factors such as rainfall and temperature and depends on the specific farm management. Nitrogen is one of the key factors to improve the productivity of grasslands, assuming no other nutrients are limiting. Soil N supply (SNS) is that originating from other sources than fertilizer. Recommendations for fertilization of grassland and arable land take the SNS into account, implying that the amount of inorganic fertilizer or animal manure can be lowered accordingly. An analysis of Dutch nitrogen fertilizer experiments on grassland during the period 1934–1994 showed that increasing fertilizer applications resulted in higher N uptake by the grass, but at the same time the N use efficiency (NUE) decreased. The analysis also showed that grazing leads to a higher SNS and to a lower NUE, compared to cutting. These effects were stronger with pure grazing than with mixed grazing and cutting (Vellinga and André, 1999). For reasons of uneven distribution and high local N loadings with urine, N in the manure which is deposited directly in the meadow has a much lower mineral fertilizer equivalent (MFE) value than N in manure spread on the field provided appropriate abatement techniques of ammonia loss are implemented. Today, the optimum N fertilization rate is based on economic and environmental targets. Until about 1990, the economic criterion was a marginal N response of 7.5 kg dry matter of herbage per kg N fertilizer applied. This criterion implied economically optimal fertilizer rates up to around 400 kg N/ ha grassland (Prins, 1983) when harvested as cut sward. Much lower values (around 200 kg N/ha) are found for grazed grass swards (Deenen and Lantinga, 1993; Lantinga et al., 1999;

52

Nevens and Reheul, 2003). Nowadays environmental targets are directed to lower protein contents of herbage and to meeting the Nitrate Directive, both leading to lower fertilization levels in the Netherlands (Vellinga et al., 2004; Oenema et al., 2011, Chapter 4 this volume). Information on the geographical distribution and corresponding productivity of European grasslands has been published recently (Smit et al., 2008). The potential grass yield varies strongly between regions in Europe and reflects areas with different natural productivity levels (Peeters and Kopec, 1996; Smit et al., 2008). The potential production of herbage dry matter (DM) can be divided into three classes: • 10–15 ton/ha:Â€North-western Europe (Atlantic coastal area), • 5–10 ton/ha:Â€North and east Europe, • 0–5 ton/ha:Â€Southern Europe (semi-arid Mediterranean, not irrigated). Grass yield trials in NW Europe (cut grass only) show annual DM yields up to about 16 ton/ha at N rates of 300–500 kg/ha (Figure 3.17). For cut grass, annual yields without N fertilizer application range between 2 and 6 ton/ha and the maximum yield between 8 and 18 ton/ha.

Effect on milk production Based on datasets of 19 dairy farm groups (Bos et al., 2003; Raison et al., 2006; Bleken et al., 2005; Aarts et al., 2008) there is a fairly good correlation between the fertilizer N rate applied to fodder crops (mainly pasture) and the milk production (R2 = 0.57, same dataset as in Figure 3.16). A survey of 139 dairy farms in the Atlantic area, ranging from Ireland to Portugal, shows a ratio of milk production per unit N applied to fodder crops ranging from about 29 kg milk per kg N on extensive farms with grazing to 547 on intensive farms without grazing (Raison etÂ€al., 2006). Data suggest that intensive farms are more efficient with fertilizer N (more milk per kg N applied) but this is also an effect of a higher proportion of feed concentrates and imported feedstuffs in total N-input. The N-losses for production of these concentrates or imported fodder are not accounted for in the farm balance, and hence the true NUE for the overall production of milk may be lower (see also Section 3.4.3).

Economic return on N for milk production The economic return on N for milk production (ERoN; see also Section 3.3.2) were derived for grass yield response curves in the UK and in the Netherlands (Figure 3.18). The fertilizer N rate used in the calculations was that needed to obtain grass yields supporting a maximum annual milk production of 10.3Â€ton/â•›ha for the Dutch case and 5.9 ton/ha for the UK case. Because present-day intensive dairy farming uses considerable amounts of feed concentrates, a fixed annual value of 100 kg/ha N as feed concentrates was used, typical for intensive dairy farming. ERoN values for price levels since 2000 (fertilizer:milk price ratio increased from 2 to 3 and is still rising, see FigureÂ€3.13), range from 2–7 € per kg milk/€ per kg N when targeting maximum milk yields per ha, and exceed 15 €/€ at present N-fertilizer levels in the Netherlands and the UK. ERoN values tend to be higher than those for arable agriculture, because,

Lars Stoumann Jensen and Jan K. Schjoerring 20

Figure 3.17 Examples of grass yield response curves for cut grass in response to annual N Â�fertilizer inputs.

Grass yield (ton/ha DM)

18 16

R 2 = 0.97

14 12 10 8 6

R 2 = 0.77

4 Schils et al. (1999) Netherlands Shiel et al. (1999) UK

2 0 0

100

200

300

400

500

600

700

N-fertilizer (kg/ha) 30

Figure 3.18 Gross economic return on nitrogen fertilizer (ERoN) application to grass for milk production as function of fertilizer N:milk price ratio (€/kg N per €/kg milk) for average fertilizer N rates (indicated in legend) and use of feed concentrates in the Netherlands and the UK. N response based on trials in the Netherlands and in the UK (see Figure 3.17).

UK Fert-70 kg/ha N, Feed-100 kg/ha N

25

20 (euro/euro)

Gross economic return on N

NL Fert-170 kg/ha N, Feed-100 kg/ha N

15

10

5

0 0

1

2

3

4

5

6

7

8

9

Fertilizer-N:milk price ratio

in contrast to wheat and oilseed rape, grass yield continues to increase up to N-rates beyond 400 kg/ha. The high ERoN for fertilizer N in milk production compared to crop production indicates how large the incentive is for the dairy farmers to apply large, and likely also excessive, amounts of fertilizer N for grasslands. Although farmers in south-eastern Europe generally use lower N fertilization rates than in north-western Europe, with significant possibilities for increasing their milk production level, the low ERoN under their production conditions (climate, soils, etc.) typically provides little incentive for them to increase fertilizer N use.

3.5â•‡ Industrial uses of dinitrogen gas and reactive nitrogen based compounds Industrial uses of nitrogen cover a range of different applications. The gas dinitrogen is used to maintain for instance an inert atmosphere, while reactive nitrogen forms (especially

ammonia and nitric acid) are used as ingredients in the chemical industry (rubbers, plastics including nylon, melamine), in the electronics industry for etching and pickling (nitric acid), in production of primary metals via leaching (nitric acid) and for cleaning catalysts used in for instance petroleum refining. In the food industry, ammonia is used for refrigeration of foods, and in the medical field it is used to refrigerate medical samples. Industrial dinitrogen gas uses worldwide are summarized in Table 3.8, while the following sections summarize they ways in which major reactive N compounds are used (Maxwell, 2004). In Europe, industrial and other uses than fertilizers consume 23% of total European ammonia production (Table 3.9) but 35% of total European ammonia is exported, so industrial and other non-fertilizer uses of ammonia constitute roughly one third of the total ammonia consumption. From Table 3.9 it can also be seen that western Europe is a net importer of ammonia, the majority coming from eastern Europe and central Asia.

53

Benefits of nitrogen for food, fibre and industrial production Table 3.8â•‡ Uses of industrial dinitrogen gas worldwide (Maxwell, 2004)

Application

Market share (%)

Chemical industry

33

Oil and gas extraction

14

Electronics

13

Primary metals

11

Petroleum refining

10

Food industry

5

Glass

2

Rubber and plastics Miscellaneous

1 11

3.5.1â•‡ Ammonia Ammonia is one of the best known bulk chemicals in the world, and the major synthetic nitrogen products made from ammonia are shown in Figure 3.19. Ammonia is predominantly used as feedstock for the production of fertilizers. It is produced by the Haber–Bosch process, invented by Fritz Haber in 1908, and turned into an industrial scale process by Carl Bosch in the years after. In the Haber–Bosch process, hydrogen from natural gas is combined catalytically with free nitrogen gas in the air at high temperature and pressure, yielding ammonia (Domene and Ayres, 2001). According to Yara (2009), 83% of all ammonia produced globally was used for fertilizer production. Beside from the direct fertilizer use of ammonia, it can also be used as a reactant with respective acids to produce ammonium nitrate, ammonium phosphate and ammonium sulphate. Reacting liquid ammonia with carbon dioxide at about 190 ºC and elevated pressure according to the so-called Basaroff reactions yields the fertilizer urea (Maxwell, 2004). The remainder of the ammonia is used in various other proÂ� cesses. The most important compound made using ammonia as a reactant is nitric acid. In three steps, ammonia is converted to nitric acid based on the so-called Ostwald process, using precious metals as catalysts at elevated pressure and temperature (Buchel et al., 2000). Nitric acid in turn is predominantly used to make explosives, like ammonium nitrate, nitroglycerine, trinitrotoluene and nitrocellulose. Ammonia is further used in the production of the cyclic amide caprolactam, a feedstock for the production of nylon-6, amines, polyacrylonitrile, hydrazine, polyurethanes, resins based on phenol or melamine, formaldehyde, nitriles, sodium nitrate, sodium cyanide and many others. Nitrogen compounds are particularly used in technologies for cleaning flue gases after fossil fuel combustion. This includes the reduction of nitrogen oxides using ammonia, both catalytically as well as non-catalytically (Caton and Xia, 2004; Wojciechowska and Lomnicki, 1999; Baukal, 2003). Ammonia is also used for the removal of sulphur dioxide, although the technology is rather new and as such it is not widely used. With this technique, an electron beam passes through the flue gas

54

and ammonia reacts with sulphur dioxide to yield ammonium sulphate (Chmielewski et al., 2002). Despite its toxicity, ammonia is regaining its position as a refrigerant due to the environmental concerns associated with chlorofluorocarbons. It has favourable thermodynamic properties, especially a low boiling point and high heat of evaporation, and is widely available for low prices (Redwood, 2010; Stoecker, 1998). In the metal industry, ammonia is used for the extraction of metals such as copper, gold and tungsten from their respective ores (Wohler, 2009). In the extraction process, the metal ores are suspended in an ammonia solution and subsequently heated, thereby creating the corresponding metal-amines, which can be isolated. Ammonia can also be used for annealing/nitriding of steel (Ross, 1988) and as a corrosion inhibitor after conversion to quaternary ammonium compounds (Sastri, 1998).

3.5.2â•‡ Ammonium nitrate Ammonium nitrate is predominantly used as a fertilizer and as an ingredient for explosives and propellants. In the United States, industrial explosives (including ammonium nitrate) account for approximately 4% of total reactive nitrogen output (Domene and Ayres, 2001). The major N containing explosives apart from ammonium nitrate are TNT, PETN, Tetryl, Nitroguanidine and Nitroglycerin. The principal non-military use of explosives is in coal mining, followed by quarrying, surface mining and construction work. All of the nitrogen contained in explosives is released directly into the atmosphere the moment they are used, mainly as free dinitrogen, but in the case of ammonium nitrate the majority is released as NO. Detonation of nitroglycerin also releases a significant proportion as N2O, as much as 97 kg per ton of nitroglycerin, according to theoretical model calculations (Domene and Ayres, 2001). Ammonium nitrate mixed with a suitable fuel, mostly fuel oil and as such abbreviated as ANFO, is a well known blasting agent used in the mining industry and for construction purposes, in which it has largely replaced dynamite (Persson et al., 1993; Tatiya, 2005; Monroe and Hall, 2006). It is used in a 95:5 weight ratio of prilled ammonium nitrate to fuel oil. Another well known mixture is AMMONAL, which consists of 60 wt% ammonium nitrate, 20 wt% trinitrotoluene (TNT) and 20 wt% aluminium. Unfortunately, owing to the large scale availability of both ingredients of ANFO, the mixture has also been used for the construction of so-called improvised explosive devices (Turkington, 2009).

3.5.3â•‡ Urea Urea is predominantly used as a fertilizer, since it has the highest nitrogen content of all known solid nitrogenous fertilizers (Schepers and Raun, 2008). Combined with formaldehyde, it forms a resin that is used in adhesives and plastics in general (called urea formaldehyde resins). The production of the bulk chemical melamine, a feedstock for predominantly plastics, is based on the ring closure of three urea molecules at elevated temperatures. On a smaller scale, urea is used as an ammonia source for removal of nitrogen oxides in flue gases, as an

Lars Stoumann Jensen and Jan K. Schjoerring Table 3.9â•‡ European potential nitrogen supply and demand balances in 2008 (FAO, 2008)

Europe, total

Central Europe

Western Europe

(million ton N and % of supply) NH3 max. prod. capacity (as N)

37.5

NH3 actual prod. (as N)

33.6

100%

4.8

100%

9.7

100%

19.0

100%

N fertilizer consumption

14.4

43%

2.7

55%

8.6

88%

3.1

17%

Non-fertilizer N demand & others

7.7

23%

0.6

13%

5.2

54%

1.9

10%

+11.5

+34%

1.5

+32%

−4.1

−42%

+14.0

+74%

Balance (+:Â€export, −:Â€import)

Monoammoniumphosphate(MAP) NH4H2PO4

6.2

Diammoniumphosphate(DAP) (NH4)2HPO4

10.3

Eastern Europe + Central Asia

Hydrogen cyanide HCN

21.0

Ammoniumsulphate(MAP) (NH4)2SO4

Ammoniumnitrate NH4NO3

NH3

H3PO4

CH4

H3PO4

C2H6 H2SO4 O2+H2O

NaOCl HCOH CH3OH

ClCH2CH2Cl

Ammonia NH3

CO2

Other or organics anics

Heat eat & catal catalyst s

Fig. 3.19 Synthetic nitrogen products made from ammonia (modified from Maxwell, 2004).

55


intermediate product in the pharmaceutical industry, as well as a reactant for the production of urea nitrate.

3.6â•‡ Economic value of reactive N use to the European economy There are various ways to approach the economic value of reactive N. Taking a global perspective, Erisman et al. (2008) argued that nearly 50% of the world human population in 2008 could be fed thanks to Haber–Bosch derived Nr applied as fertilizer. The global revenue on sale of fertilizers in 2005 amounted to nearly 30 billion USD (25 billion €; Yara, 2009). For the EU-27 countries it was estimated by Yara (2009) that the increase of wheat production in 2008 due to use of mineral N fertilizer was 64 million tons. This estimation is based on a comparison with the wheat yields achieved in ecological farming without mineral N fertilizer. The fertilizer-derived increase in wheat production represents a net economic gain (grain value minus fertilizer costs) of 7.6 billion € per year for the entire EU, or 280 €/ha. However, this net gain is sensitive to the relatively volatile world market prices for grain and fertilizer, as seen during the 2008–9 food crisis and subsequent financial breakdown, and assumptions on the potential yields in absence of mineral N fertilizer. At the level of a farm or a crop, the cost of N fertilizer is just one of several production factors. As described in previous sections, the economic return on investment in N (ERoN) is a very robust measure of importance for the farm economy and, hence, for the farmer decisions. Judging from Figures 3.14 and 3.18, the following current ERoN values can be summarized. The farmer will make a profit from N inputs if ERoN is above Product

ERoN (€ product / € fertilizer N)

Winter Wheat:

2–7

Oilseed Rape:

1–5

Milk:

10–15

one and the range in ERoN depends on (i) actual N fertilization level and (ii) shape of the response curve. A lower maximum yield for oilseed rape, wheat and grasslands is commonly found in south-eastern compared to north-western Europe. This is due especially to water limitation and implies a tendency for ERoN to be relatively low in south-eastern Europe compared to northwestern Europe, where climatic conditions favour higher potential yields under economically optimal fertilizer N input. The ERoN ranges presented in this chapter mean that for most farmers there is a huge economic profit from use of Nr, especially in relation to livestock production. The high ERoN for fertilizer N in milk production compared to crop production indicates how large the incentive is for the dairy farmers to apply large, and likely also excessive, amounts of fertilizer N for grasslands. In addition to chemical fertilizer, manure and biological nitrogen fixation are other sources of N that can be affected by farm management. The economy of N at the farm level is therefore quite complex. Costs of purchasing and handling of various N sources are quite different and change in time, e.g. depending

56

on the price of energy (natural gas) and environmental policies (see Oenema et al., 2011, Chapter 4, this volume). Compiling a comprehensive, robust inventory of the economic benefits to society of reactive nitrogen is not a simple matter. As indicated above, a coarse estimate may be that about half the value of European agricultural production may be considered as dependent on Nr supply. However, in a review of yield differences between organic and conventional farming in Europe, Offermann and Nieberg (2000) found that organic cereal yields are typically 60%–70% of those under conventional management, vegetable yields are often just as high as under conventional management and pasture and grassland yields in the range of 70%–100% of conventional yields. The derived consequences for economic profit or benefit are quite complicated as Offermann and Nieberg (2000) also state that the majority of the studies evaluated report an increase of labour needs, on average in the range of 10%–20% (but higher for vegetables), the cost of which has to be accounted for. Therefore the economic benefits of Nr use in agriculture are not easy to estimate. In the case of industry, the overall economic value includes nearly all explosives (including the economic value of military security; Erisman et al., 2008), the value of coal and other products mined with explosives, and the wide diversity of other nitrogen-containing chemical compounds. For industrial uses, however, i.e. especially explosives and plastics, there are alternatives for using Nr, and therefore the real value of Nr becomes very difficult to assess. For agricultural production, there is no simple substitute to Nr at the scale of its current level of use, but also the Nr contribution to agriculture is challengeable (Bruges, 2007). Although we have estimated that between 30%–50% of the current food production, population and GDP may be derived from use of Nr, to some extent Nr has also replaced labour. Historically, human development has been driven by the big transition in which labour force for agriculture was transferred to industry and services. The continued productivity in agriculture was ensured partly with fossil energy for machinery and Nr, partly with modern pest control agents and breeding for improved crop genotypes. Another issue is that economic benefits in the modern definition include the externalities, i.e. the negative effects (or benefits) of Nr for which there is no market. This issue is discussed at length in Chapter 22 of this volume (Brink et al., 2011). The real societal price of food is that including the external costs, or alternatively formulated, is the price of food produced without any external effects. Including externalities of Nr use (and of P, pesticides, fossil fuels, etc.) in the price may then enable transfer of part of the labour back to food production to maintain food production at lower external inputs. However, this approach would not be easily applicable in a market based economy. Given the many uncertainties in the assessment of the economic value of reactive N use to the European economy, the coarse estimate at the beginning of Section 3.6 may be as valid as any estimate derived from more refined calculations. Based on this and the additional data and arguments presented in this chapter it can be concluded that the overall benefits of N use are very substantial.

Lars Stoumann Jensen and Jan K. Schjoerring

3.7â•‡ Perspectives and recommendations

3.8â•‡ Conclusions

The need to maintain food and energy security under an increasing world population poses major challenges to supply the quantity and quality of commodities (including biofuels), given the few options to increase arable land area. With its resource of relatively fertile and productive soils, Europe has a clear capability for contributing to this, and it may be argued that Europe also has a moral obligation to do so. However, increased land use changes elsewhere in the world may not exclusively be due to an eventually diminishing agricultural production in Europe if inputs of reactive nitrogen are significantly reduced, but these possible secondary effects of reducing European fertilizer N rates must be taken into account. At the same time, environmental concerns, including agricultural responses to climatic change, as well as the need to feed the growing global population, represent a major challenge for further improvement of nitrogen benefits, i.e. to increase the use efficiency of the reactive nitrogen applied. The following recommendations for policy decisions and research priorities can be made. • Initiatives, whether voluntary or legislative, to reduce the use or surplus of nitrogen in agriculture, including inorganic fertilizer N, should take account of the need to maintain the nitrogen benefits in agricultural productionÂ€– food, feed and biomass productivity should be maintained while improving N use efficiency. • Modified field management practices for N conservation, modifications to livestock diets and recycling of wastes can enhance benefits per unit Nr used, and should be strongly promoted as best available technology (BAT). • New developments, combined with stimulatory incentives for farmers, should promote innovative technological tools to improve resource-efficiency and the overall benefits of NÂ€use: (i) management strategies involving N-conserving field practices (e.g. catch crops, reduced soil tillage, better timing of N inputs, etc.), (ii) modifications to livestock diets for decreasing N excretion rates, (iii) enhanced manure N use efficiency through improved environmental technologies for management, recycling and field application of manures, (iv) improved accounting of field level N responses depending on cropping practices, soil fertility and climate.

• Although considerable uncertainty exists in the assessment of overall benefits of reactive nitrogen, particularly as regards the economic value of Nr in industrial production, it can be concluded that Nr is very much a key factor for achievement of food security and social welfare in Europe. • Maintaining food and energy security under an increasing world population poses major challenges to supply the quantity and quality of commodities (food, feed, fibre and fuels). Changing the input of reactive nitrogen significantly to European agriculture may influence conversion of natural land areas to cropped land elsewhere in the world. • Future legislative actions to reduce the use or surpluses of nitrogen in agriculture should take account of the need to maintain benefits for food security and farm economy in Europe. • There is still a large potential for increased nitrogen efficiency in European agriculture by better management strategies, improved recirculation of nitrogen in waste materials, adoption of new fertilizer technologies, crop monitoring tools and new crop cultivars, all demanding improved skills of the individual farmers and their advisoryÂ€service. • The economically optimal N application rate for crops varies significantly across field, farms and regions, depending on crop type, crop N response, farm type, soil type and climate. • Crop N use efficiency can be increased by improving prediction of the economically optimal N rate, but at the current relatively low ratio between nitrogen fertilizer costs and crop prices, farmers often have relatively little economic incentive to restrict N application, so long as environmental effects are considered as externalities. • Nitrogen use efficiency for livestock production can be greatly improved, especially with optimized feed protein and amino acid composition, but also by animal breeding. Although intensification of livestock production with external feedstuff may increase N use efficiency, it should be noted that this may lead to larger local surplus of N (and other nutrients), necessitating application of environmental technologies for waste and manure processing to avoid increased environmental load. • For dairy farming, nitrogen use efficiency can be improved by adjusting the nitrogen content of the feed to the requirements of the cattle and by minimizing the ammonia losses from animal housing and during manure application.

• New research initiatives should focus on: (i) breeding plant species and crop varieties with improved nitrogen use efficiency through increased root length density at depth, high capacity for N accumulation in the stem, high maximum N-uptake rate and N remobilization during grain-filling, (ii) improved composition of major feed crops and novel feed additives, e.g. proteins from bio-fuel production waste and other means of increasing feed N responses per unit mass Nr used, (iii) new technologies for improving fertilizer application and sensing of crop N demand, including tools for improved utilization of N in agricultural and urban waste materials to increase overall N use efficiency.

Acknowledgements The authors of this gratefully acknowledge support from the European Science Foundation for the NinE programme and the COST Action 729. In addition, the lead authors acknowlÂ� edge financial support from The Danish Research Council for Technology and Production (274–08–0439), The Strategic Research Council (09–067246), The Hofmansgave Foundation, The Danish Ministry of Food, Agriculture and Fisheries

57


(3304-FVFP-09-B-004) and the European Commission (NitroEurope Integrated Project). Christian Pallière acknowledges the support of Fertilizers Europe and Joachim Lammel and Frank Brentrup the support of Yara International.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press: www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Aarts, H. F. M, Daatselaar, C. H. G. and Holshof, G. (2008). Bemesting, Meststofbenutting en Opbrengst van Productiegrasland en Snijmaïs op melkveebedrijven. Plant Research International B.V., Wageningen. Akiyama, H., Yan, X. Y. and Yagi, K. (2010). Evaluation of effectiveness of enhanced-efficiency fertilizers as mitigation options for N2O and NO emissions from agricultural soils:Â€meta-analysis. Global Change Biology, 16, 1837–1846. Balmford, A., Green, R. E. and Scharlemann, J. P. W. (2005). Sparing land for nature:Â€exploring the potential impact of changes in agricultural yield on the area needed for crop production. Global Change Biology, 11, 1594–1605. Barlóg, P. and Grzebisz, W. (2004). Effect of timing and nitrogen fertilizer application on winter oilseed rape (Brassica napus L.) I. Growth dynamics and seed yield. Journal of Agronomy and Crop Science, 190, 305–313. Baukal, C. E. (2003). Industrial Combustion Pollution and Control. CRC Press, Boca Raton, FL. Bellarby, J., Foereid, B., Hastings, A. and Smith, P. (2008). Cool Farming:Â€Climate Impacts of Agriculture and Mitigation Potential. Greenpeace International, Amsterdam, The Netherlands. (available at http://www.greenpeace.org). Berry, P. M. and Spink, J. H. (2009). Canopy Management and Late Nitrogen Applications to Improve Yield of Oilseed Rape, HGCA Project Report No. 447. Birkmose, T. S. (2009). Nitrogen recovery from organic manures:Â€Improved slurry application techniques and treatmentÂ€– the Danish scenario. In: Proceedings 656, International Fertilizer Society, York, UK. Bleken, M. A., Steinshamn, H. and Hansen, S. (2005). High nitrogen costs of dairy production in Europe:Â€worsened by intensification. Ambio, 34, 598–606. Bos, J., Pflimlin, A., Aarts, H.F.M. and Vertès, F. (eds.) (2003). Nutrient management at farm scale. In: First Workshop of the EGF working group ‘Dairy farming systems and environment’, Quimper, France 23–25 June 2003. Plant Research International, Wageningen. Brentrup, F. and Palliere, C. (2008). GHG emissions and energy efficiency in European Nitrogen fertilizer production and use. In: Proceedings 639, International Fertilizer Society, York, UK. Brink, C., van Grinsven, H., Jacobsen, B. H. et al. (2011). Costs and benefits of nitrogen in the environment. In:Â€The European Nitrogen Assessment, ed. M.A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Bruges, J. (2007). The Big Earth Book:Â€Ideas and Solutions for a Planet in Crisis. Alastair Sawday, London. Buchel, K. H., Moretto, H. H. and Woditsch, P. (2000). Industrial Inorganic Chemistry, 2nd edn, Wiley-VCH, New York.

58

Burney, J. A., Davis, S. J. and Lobella, D. B. (2010). Greenhouse gas mitigation by agricultural intensification. Proceedings of the National Academy of Sciences of the United States of America, 107, 12052–12057. Cassman, K. G., Dobermann, A., Walters, D. T. and Yang, H. (2003). Meeting cereal demand while protecting natural resources and improving environmental quality. Annual Reviews of Environmental Resources, 28, 315–358. Caton, J. A. and Xia, Z. (2004). The selective non-catalytic removal of nitric oxides from engine exhaust streams:Â€comparison of three processes. Journal of Engineering for Gas Turbines and Power, 126, 234–240. CBS (2009). Animal Manure and Minerals 1990–2008* (Dierlijke mest en mineralen 1990–2008*). Centraal Bureau voor de Statistiek, Den Haag/Heerlen (in Dutch). Cerrato, M. E. and Blackmer, A. M. (1990) Comparison of models for describing corn yield response to nitrogen fertilizer. Agronomy Journal, 82, 138–143. Chien, S. H., Prochnow, L. I. and Cantarella, H. (2009). Recent developments of fertilizer production and use to improve nutrient efficiency and minimize environmental impacts. Advances in Agronomy 102, 267–322. Chmielewski, A. G., Tyminski, B., Pawelec, A. and Licki, J. (2002). Industrial plant for flue gas treatment with high power electron accelerators. In: AIP Conference Proceedings, 680 (1), International Conference on the application of accelerators in research and Industry, Denton, Texas, USA. Dalling, M. J. (1985). The physiological basis of nitrogen redistribution during grain filling in cereals. In: Exploitation of Physiological and Genetic Variability to Enhance Crop Productivity, American Society of Plant Physiologists, Chicago, IL, pp. 55–71. Dämmgen, U., Haenel, H.-D., Rösemann, C. et al. (2009). Improved national calculation procedures to assess energy requirements, nitrogen and VS excretions of dairy cows in the German emission model GAS-EM. LandbauforschungÂ€– vTI, Agriculture and Forestry Research, 3, 233–252. Dampney, P., Dyer, C., Goodlass, G. and Chambers, B. (2006). Crop responses to Different N Fertilizer Materials. Component report for Defra report NT 2605. Defra, London. de Fries, W., Leip, A., Reinds, G. J. et al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Deenen, P. J. A. G. and Lantinga, E. A. (1993). Herbage and animal production responses to fertilizer nitrogen in perennial ryegrass swards. I. Continuous grazing and cutting. Netherlands Journal of Agricultural Science, 41, 179–203. Domene, L. A. F. and Ayres, R. U. (2001). Nitrogen’s role in industrial systems. Journal of Industrial Ecology, 5, 77–103. Dourmad, J. Y., Seve, B., Latimier, P. et al. (1999). Nitrogen consumption, utilisation and losses in pig production in France, The Netherlands and Denmark. Livestock Production Science, 58, 261–264. EFMA (2003). Understanding nitrogen and its use in agriculture. European Fertilizer Manufacturers Association (now Fertilizers Europe), report, 64 pp. EFMA (2007). Forecast of food, farming and fertilizer use in the European Union 2007–2017. European Fertilizer Manufacturers Association (now Fertilizers Europe), report, 7 pp. EFMA (2009). Forecast of food, farming and fertilizer use in the European Union 2009–2019. European Fertilizer Manufacturers Association (now Fertilizers Europe), report, 20 pp.

Lars Stoumann Jensen and Jan K. Schjoerring Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. and Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. EU Glossary (2010). Common agricultural policy (CAP). Available at http://europa.eu/scadplus/glossary/agricultural_policy_en.htm (accessed 4 Aug 2010). Eurostat (2009). Agricultural statistics, Annual selling prices of crop products per 100 kg; wheat and rape 1995–2008. Accessed at http:// epp.eurostat.ec.europa.eu/portal/page/portal/agriculture/data/ database June 2010. Fangueiro, D., Pereira, J., Coutinho, J., Moreira, N. and Trindade, H. (2008). NPK farm-gate nutrient balances in dairy farms from Northwest Portugal. European Journal of Agronomy, 28, 625–634. FAO (2008). Current World Fertilizer Trends and Outlook to 2012. Food and Agriculture Organization of the United Nations, Rome. FAO (2009). FAOSTAT Database Collections. Food and Agriculture Organization of The United Nations. Rome. http://faostat.fao.org/. Fefac (2009). Feed and Food: Statistical Yearbook 2007. Fefac, Brussels. Feller, C., Fink, M., Maync, A. et al. (2001) Düngung im Freilandgemüsebau. Institut für Gemüse- und Zierpflanzenbau Großbeeren/Erfurt e.V. Foulkes, M. J., Hawkesford, M. J., Barraclough, P. B. et al. (2009). Identifying traits to improve the nitrogen economy of wheat:Â€Recent advances and future prospects. Field Crops Research, 114, 329–342. Galloway, J. N., Schlesinger, W. H., Levy, H. II, Michaels, A. and Schnoor, J. L. (1995). Nitrogen fixation:Â€atmospheric enhancement–environmental response. Global Biogeochemical Cycles, 9, 235–252. Godfrey, D., Hawkesford, M. J., Powers, S. J., Millar, S. and Shewry, P. R. (2010). Effects of crop nutrition on wheat grain composition and end use quality. Journal of Agricultural and Food Chemistry, 58, 3012–3021. Grant, R., Pedersen, L. E., Blicher-Mathiesen, G. et al. (2009). Landovervågningsoplande 2007, Report no. 709 from DMU/NERI (in Danish, with English summary). Gyldenkærne, S. and Mikkelsen, M. H. (2007). Projection of the Ammonia Emission from Denmark from 2005 until 2025, NERI Technical Report No. 239. National Environmental Research Institute, University of Aarhus. Henke, J., Breusted, G., Sieling, K. and Kage, H. (2007). Impact of uncertainty on the optimum nitrogen fertilisation rate and agronomic, ecological and economic factors in an oilseed rape based crop rotation. Journal of Agricultural Science, 145, 455–468. Herridge, D. F., Peoples, M. B. and Boddey, R. M. (2008). Global inputs of biological nitrogen fixation in agricultural systems. Plant and Soil, 311, 1–18. IPCC (2007). Intergovernmental Panel on Climate Change:Â€Technical Summary. In:Â€Climate Change 2007:Â€The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, I. B. Averyt, M. Tignor and HL. Miller (eds.)]. Cambridge University Press. Irigoyen, I., Lamsfus, C., Aparicio-Tejo, P. and Muro, J. (2006). The influence of 3,4-dimethylpyrazole phosphate and dicyandiamide on reducing nitrate accumulation in spinach under Mediterranean conditions. Journal of Agricultural Science, 144, 555–562. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:Â€The

European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.Â€W. Erisman et al. Cambridge University Press. Jenkinson, D.S. (2001). The impact of humans on the nitrogen cycle, with focus on temperate arable agriculture. Plant and Soil, 228, 3–15. Johnston, A. E. and Poulton, P. R. (2009). Nitrogen in agriculture:Â€an overview and definitions of nitrogen use efficiency. In: Proceedings 651, International Fertilizer Society, York, UK. Johnston, A. E., Poulton, P. R. and Coleman, K. (2009). Soil organic matter:Â€its importance in sustainable agriculture and carbon dioxide fluxes. Advances in Agronomy, 101, 1–57. Jongbloed, A. W. and Lenis, N. P. (1998). Environmental concerns about animal manure. Journal of Animal Science, 76, 2641–2648. Karp, A. and Shield, I. (2008). Bioenergy from plants and the sustainable yield challenge. New Phytologist, 179, 15–32. Kitou, E., Fernandez, R., Gugele, B., Goettlicher, S. and Ritter, M. (2009). Annual European Community Greenhouse Gas Inventory 1990–2007 and Inventory Report 2009. European Commission, DG Environment European Environment Agency. http://www.eea. europa.eu/publications/european-community-greenhouse-gasinventory-2009/. Kohn, R. A., Dou, Z., Ferguson, J. D. and Boston, R. C. (1997). A sensitivity analysis of nitrogen losses from dairy farms. Journal of Environmental Management, 50, 417–428. Lantinga, E. A., Deenen, P. J. A. G. and van Keulen, H. (1999). Herbage and animal production responses to fertilizer nitrogen in perennial ryegrass swards. II. Rotational grazing and cutting. Netherlands Journal of Agricultural Science, 47, 243–261. LEI (2009). Agricultural prices in the Netherlands. Milk- and fertilizer (CAN) prices 1995–2008. Accessed June 2010 at http://www.lei. wur.nl/NL/statistieken/Agrarische+prijzen/. Leip, A., Achermann, B. Biller, G. et al. (2011). Integrating nitrogen fluxes at the European scale. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Lewis, A. J. and Southern, L. L. (2001). Swine Nutrition. CRC Press, Boca Raton, FL. Lopez-Bellido, R. J., Lopez-Bellido, L., Benıtez-Vega, J. and LopezBellido, F. J. (2007). Tillage system, preceding crop, and nitrogen fertilizer in wheat crop:Â€II. Water utilization. Agronomy Journal, 99, 59–65. Makowski, D., Wallach, D. and Meynard, J. M. (1999). Models of yields, grain protein and residual mineral nitrogen responses to applied nitrogen for winter wheat. Agronomy Journal, 91, 377–385. Marschner, H. (1995). Mineral Nutrition of Higher Plants, Academic Press, London. Maxwell, G. R. (2004). Synthetic Nitrogen Products: A Practical Guide to the Products and Processes. Kluwer Academic/New York. Monroe, C. E. and Hall, C. (2006). Explosives in Metal Mining and Quarry Operations, Merchant Books. Mosier, N., Wyman, C., Dale, B. et al. (2005). Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresource Technology, 96, 673–686. Nadeau, E., Englund, J.-E. and Gustafsson, A. H. (2007). Nitrogen efficiency of dairy cows as affected by diet and milk yield. Livestock Science, 111, 45–56. Nevens, F. and Reheul, D. (2003). Effect of cutting or grazing grass swards on herbage yield, nitrogen uptake and residual soil nitrate at different levels of N fertilisation. Grass and Forage Science, 58, 431–449.

59

Benefits of nitrogen for food, fibre and industrial production Oenema, O., Bleeker, A., Braathen, N. A. et al. (2011). Nitrogen in current European policies. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Offermann, F. and Nieberg, H. (2000). Economic Performance of Organic Farms in Europe. In:Â€Organic Farming in Europe: Economics and Policy, Volume 5. University of Hohenheim, Department of Farm Economics, Stuttgart, Germany. Peeters, A. and Kopec, S. (1996). Production and productivity of cutting grasslands in temperate climates of Europe. The Proceedings of the 16th General Meeting European Grasslands Federation, pp. 59–73. Persson, P. A., Holmberg, R. and Lee, J. (1993). Rock Blasting and Explosives Engineering. CRC Press, Boca Raton, FL. Pettersson, C. G. and Eckersten, H. (2007). Prediction of grain protein in spring malting barley grown in northern Europe. European Journal of Agronomy, 27, 205–214. Peyraud, J. L. and Delaby, L. (2004). Can we minimize nitrogen excretion in dairy herds while maintaining performances? In:Â€Animal Production in Europe:Â€The Way Forward in a Changing World. Volume. 1. International Society for Animal Hygiène, SaintMalo, France. Pilbeam, C. J. (1996). Effect of climate on the recovery in crop and soil of N-15- labelled fertilizer applied to wheat. Fertilizer Research 45, 209–215. Pitre, F. E., Pollet, B., Lafarguette, F. et al. (2007). Effects of increased nitrogen supply on the lignification of poplar wood. Journal of Agriculture and Food Chemistry, 55, 10306–10314. Porteaus, F., Hill, J., Ball, A. S. et al. (2009). Effect of free air carbon dioxide enrichment (FACE) on the chemical composition and nutritive value of wheat grain and straw. Animal Feed Science and Technology, 149, 322–332. Poulsen, H. D. (2009). Danish Standard Values on Nutrient Content in ManureÂ€– 2009. (in Danish). Available at http://www.agrsci. dk/ny_navigation/institutter/institut_for_husdyrsundhed_ velfaerd_og_ernaering/husdyrernaering_og_miljoe/normtal, Aug 2010. Poulsen, H. D., Lund, P., Sehested, J., Hutchings, N. and Sommer, S.Â€G. (2006). Quantification of nitrogen and phosphorus in manure in the Danish normative system. In:Â€12th Ramiran International Conference:Â€Technology for recycling of manure and organic residues in a whole-farm perspective, DIAS report no. 123 from Danish Institute of Agricultural Sciences, vol. II, pp. 105–107. Powlson, D. S., Hart, P. B. S., Pruden, G. and Jenkinson, D. S. (1986a). Recovery of 15N-labelled fertilizer applied in autumn to winter wheat at four sites in eastern England. Journal of Agricultural Science, 107, 611–620. Powlson, D. S., Pruden, G., Johnston, A. E. and Jenkinson,Â€D.Â€S. (1986b). The nitrogen cycle in the Broadbalk wheat experiment:Â€recovery and losses of 15N-labelled fertilizer applied in spring and inputs of nitrogen from the atmosphere. Journal of Agricultural Science, 107, 591–609. Prins, W. H. (1983). Limits to nitrogen fertilizer on grassland. Ph. D. Thesis, Wageningen Agricultural University. Raison, C., Pflimlin, A. and Le Gall, A. (2006). Optimisation of environmental practices in a network of dairyfarms of the Atlantic Area. In: Proceedings of the Final Seminar of the Green Dairy Project Interreg. Atlantic Area III B No.100, Rennes, 13–14 December 2006,Â€pp.Â€43–65. Rathke, G. W., Christen, O. and Diepenbrock, W. (2005). Effects of nitrogen source and rate on productivity and quality of winter

60

oilseed rape (Brassica napus L.) grown in different crop rotations. Field Crops Research 94, 103–113. Raun, W. R., Johnson, G. V., Phillips, S. B. and Westerman, R. L. (1998). Effect of long-term N fertilisation on soil organic C and total N in continuous wheat under conventional tillage in Oklahoma. Soil and Tillage Research, 47, 323–330. Redwood, I. I. (2010). Theoretical and Practical Ammonia Refrigeration. General Books LLC. Reynolds, M., Foulkes, M. J., Slafer, G. A. et al. (2009). Raising yield potential in wheat. Journal of Experimental Botany, 60, 1899–1918. Ross, R. B. (1988). Handbook of Metal Treatments and Testing, 2nd edition. Springer, New York. Rudel, T. K., Schneider, L., Uriarte, M. et al. (2009). Agricultural intensification and changes in cultivated areas, 1970–2005. Proceedings of the National Academy of Sciences of the United States of America, 106, 20675–20680. Sastri, V. S. (1998). Corrosion Inihibitors: Principles and Applications. J.Â€Wiley, New York. Schepers, J. S. and Raun, W. R. (2008). Nitrogen in Agricultural Systems. Chicago, IL. Schils, R. L. M., van der Meer, H. G., Wouters, A. P., Geurink, J. H. and Sikkema, K. (1999). Nitrogen utilization from diluted and undiluted nitric acid treated cattle slurry following surface application to grassland. Nutrient Cycling in Agroecosystems, 53, 269–280. Schröder, J. J., Aarts, H. F. M., ten Berge, H. F. M., van Keulen, H. and Neeteson, J. J. (2003). An evaluation of whole-farm nitrogen balances and related indices for efficient nitrogen use. European Journal of Agronomy 20, 33–44. Schröder, J. J., Jansen, A. G. and Hilhorst, G. J. (2005). Long-term nitrogen supply from cattle slurry. Soil Use and Management, 21, 196–204. Sheriff, G. (2005). Efficient waste? Why farmers over-apply nutrients and the implications for policy design. Review of Agricultural Economics, 27, 542–557. Shewry, P. R. (2009). Improving the protein content and composition of cereal grain. Journal of Cereal Science, 46, 239–250. Shiel, R. S., El Tilib, A. M. A. and Younger, A. (1999). The influence of fertilizer nitrogen, white clover content and environmental factors on the nitrate content of perennial ryegrass and ryegrass/white clover swards. Grass and Forage Science, 54, 275–285. Sidlauskas, G. and Bernotas, S. (2003). Some factors affecting seed yield of spring oilseed rape (Brassica napus L.) Agronomy Research, 1, 229–243. Sieling, K. and Kage, H. (2008). The potential of semi-dwarf oilseed rape genotypes to reduce the risk of N leaching. Journal of Agricultural Science, 146, 77–84. Smil, V. (1999). Nitrogen in crop production:Â€an account of global flows. Global Biogeochemical Cycles, 13, 647–662. Smil, V. (2000). Feeding the World: A Challenge for the Twenty-First Century. MIT Press, Cambridge, MA. Smit, H. J., Metzger, M. J. and Ewert, F. (2008). Spatial distribution of grassland productivity and land use in Europe. Agricultural Systems, 98, 208–219. Sommer, S. G., Schjoerring, J. K. and Denmead, O. T. (2004). Ammonia emission from mineral fertilizers and fertilized crops. Advances in Agronomy, 82, 558–622. Stoecker, W. (1998) Industrial Refrigeration Handbook, McGraw-Hill, New York. Sutton, M. A., Howard, C. M., Erisman, J. W. et al. (2011). Assessing our nitrogen inheritance. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press.

Lars Stoumann Jensen and Jan K. Schjoerring Sylvester-Bradley, R. and Kindred, D. R. (2009). Analysing nitrogen responses of cereals to prioritize routes to the improvement of nitrogen use efficiency. Journal of Experimental Botany, 60, 1939–1951. Tatiya, R. R. (2005). Surface and Underground Excavations:Â€Methods, Techniques and Equipment. Taylor and Francis, London. ten Berge, H. F. M. and van Dijk, W. (2009). Management of Nitrogen inputs on farms within the EU regulatory framework. In: Proceedings 654, The International Fertilizer Society, York, UK. Tilman, D., Cassman, K. G., Matson, P. A., Naylor, R. and Polasky,Â€S. (2002). Agricultural sustainability and intensive production practices. Nature, 418, 671–677. Treacy, M., Humphreys, J., McNamara, K., Browne, R. and Watson,Â€C.Â€J. (2008). Farm-gate nitrogen balances on intensive dairy farms in the south west of Ireland. Irish Journal of Agricultural and Food Research, 47, 105–117. Turkington, R. (2009). Chemicals Used for Illegal Purposes: J. Wiley, New York. Turner, D. A., Edis, R. B., Chen, D. et al. (2010). Determination and mitigation of ammonia loss from urea applied to winter wheat with N-(n-butyl) thiophosphorictriamide. Agriculture, Ecosystems and Environment, 137, 261–266. UNFCCC (2008). United Nations Framework Convention on Climate Change:Â€Greenhouse Gas Inventory Data (online). http://unfccc.int Vellinga, T.V. and André, G. (1999). Sixty years of Dutch nitrogen fertilizer experiments, an overview of the effects of soil type, fertilizer input, management and of developments in time. Netherlands Journal of Agricultural Science, 47, 215–241. Vellinga, T. V., André, G., Schils, R. L. M. and Oenema, O. (2004). Operational nitrogen fertilizer management in dairy farming systems:Â€identification of criteria and derivation of fertilizer application rates. Grass and Forage Science, 59, 364–377. Vellinga, T. V., André, G., Schils, R. L. M., Kraak, T. and Oenema, O. (2010). Accounting for residual effects of previously applied nitrogen fertilizer on intensively managed grasslands. Grass and Forage Science, 65, 58–75. Velthof, G. L., Oudendag, D., Witzke, H. P. et al. (2009). Integrated assessment of nitrogen losses from agriculture in EU-27 using MITERRA-EUROPE. Journal of Environmental Quality, 38, 402–417. Vérité, R. and Delaby, L. (2000). Relation between nutrition, performances and nitrogen excretion in dairy cows. Annales de Zootechnie, 49, 217–230.

Virtanen, H. and Nousiainen, J. (2005). Nitrogen and phosphorus balances on Finnish dairy farms. Agricultural and Food Science, 14, 166–180. Vitousek, P. M., Naylor, R., Crews, T. et al. (2009). Nutrient imbalances in agricultural development. Science, 324, 1519–1520. von Witzke, H. and Noleppa, S. (2010). EU Agricultural Production and Trade:Â€Can More Efficiency Prevent Increasing “Land-Grabbing” Outside Of Europe? Report published by OPERA,Â€ Research Centre of Università Cattolica del Sacro Cuore. Available at http://www.opera-indicators.eu/eng/home. html Wang, X. T. and Below, F. E. (1996). Cytokinins in enhanced growth and tillering of wheat induced by mixed nitrogen source. Crop Science, 36, 121–126. Watson, C. J., Akhonzada, N. A., Hamilton, J. T. G. and Matthews, D. I. (2008). Rate and mode of application of the urease inhibitor N-(n-butyl) thiophosphoric triamide on ammonia volatilization from surface-applied urea. Soil Use and Management, 24, 246–253. Webb, J., Seeney, F. M., and Sylvester-Bradley, R. (1998). The response to fertilizer nitrogen of cereals grown on sandy soils. Journal of Agricultural Science, 130, 271–286. Webb, J., Anthony, S. G., Brown, L. et al. (2005). The impact of increasing the length of the cattle grazing season on emissions of ammonia and nitrous oxide and on nitrate leaching in England and Wales. Agriculture, Ecosystems and Environment, 105, 307–321. Witzke, H. P. and Oenema, O. (2007). Assessment of Most Promising Measures: Integrated Measures in Agriculture to Reduce Ammonia EmissionsÂ€– Task 3. Report, AlterraÂ€– Wageningen University Research, The Netherlands. Wohler, F. (2009). Handbook of Mineral Analysis. Bibliobazaar. Wojciechowska, M. and Lomnicki, S. (1999). Nitrogen oxides removal by catalytic methods. Clean Technologies and Environmental Policy, 1, 237–247. Xiccato, G., Schiavon, S., Gallo, L., Bailoni, L. and Bittante, G. (2005). Nitrogen excretion in dairy cow, beef and veal cattle, pig, and rabbit farms in Northern Italy. Italian Journal of Animal Science, 4,Â€103–111. Yara (2009). Yara Fertilizer Industry Handbook. www.yara.com Zimmer, Y. (2008). Cash Crop Report 2008. vTI Braunschweig. Available at http://www.agribenchmark.org.

61

Chapter

4

Chapter

Nitrogen in current European policies Lead author: Oene Oenema Contributing authors: Albert Bleeker, Nils Axel Braathen, Michaela Budňáková, KeithÂ€Bull, Pavel Čermák, Markus Geupel, Kevin Hicks, Robert Hoft, Natalia Kozlova, AdrianÂ€Leip, Â� Till Spranger, Laura Valli, Gerard Velthof and Wilfried Winiwarter

Executive summary Nature of the problem • Europe, and especially the European Union (EU), has many governmental policy measures aimed at decreasing unwanted reactive nitrogen (Nr) emissions from combustion, agriculture and urban wastes. Many of these policy measures have an ‘effects-based approach’, and focus on single Nr compounds, single sectors and either on air or waters. • This chapter addresses the origin, objectives and targets of EU policy measures related to Nr emissions, considers which instruments are being used to implement the policies and briefly discusses the effects of the policy measures.

Approaches • The chapter starts with a brief description of the basic elements of governmental policy measures. • A review of the main international conventions and EU policies related to emissions of Nr to air and water is then provided. • Finally the chapter provides a semi-quantitative assessment of the effectiveness and efficiency of European policy measures.

Key findings/state of knowledge • International conventions and other treaties have played a key role in raising awareness and establishing policy measures for Nr emissions abatement in EU through so-called Directives and Regulations. • There are many different EU Directives, often addressing individual Nr compounds from individual sectors (e.g. NOx emissions from combustion; NH3 emissions from agriculture, pollution of groundwater and surface water by nitrates from agriculture, discharge of total nitrogen from urban sewage to surface waters). • Many EU Directives have been revised following review and evaluation. There are increasing efforts to cluster single EU Directives into larger Framework Directives. • Compliance with, and effectiveness of, the Directives differs between sectors; it decreases in the order (i) reducing NOx emissions from combustion sources, (ii) reducing nitrogen (and especially Phosphorus) discharges to waters from industries and households, and (iii) reducing NH3 emissions and NO3 leaching from agriculture. • There is not much literature on the differences in the effectiveness and efficiencies of Directives; a number of factors seem to be involved in effectiveness and efficiency, but these have not yet been analysed in a coherent manner.

Major uncertainties/challenges • There is a huge diversity in Nr emission sources and pathways, while the number of policy instruments is limited. There is need to find the optimal mix of policy instruments targeted to the emission sources as well as the stakeholders involved. • It has been indicated that some EU Directives addressing emissions of nitrogen compounds from specific sources have antagonistic effects. The magnitude of these effects is not yet well known. • There is a delay in the environmental and ecological responses following the introduction of Directives; these are due to legislative delays, lack of enforcement and control, constraints in practice and because of biogeochemical hysteresis effects; these effects are not yet well understood quantitatively. • In general, only modest reductions in Nr emissions from agriculture have been achieved to date; this reflects the need for more effective and efficient policy measures and/or greater enforcement of current policies.

Recommendations • To examine further the differences between sectors of the factors that contribute to the effectiveness and efficiency of policy measures for the abatement of Nr emissions. • To explore further the effectiveness and efficiency of more integrated N management and integrated policy measures for the abatement of adverse impacts of Nr emissions. The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

62

Oene Oenema

4.1â•‡ Introduction This chapter discusses the nature and effects of governmental policies in Europe aimed at decreasing the unwanted emissions of reactive nitrogen (Nr) compounds into the wider environment. Policy is commonly defined as ‘a plan of actions to guide decisions’. Governmental policy is usually a response to unwanted developments or problems in society. Such policy is thus intended to change the developments in a desired direction and/or to solve problems, in this case related to excess Nr in the environment. Governmental policies are based on the premise that humans as individuals and/or as organizations change their behaviour and activities in response to such policies. This premise originates from the fact that humans prefer to live in communities (families, bands, tribes, chiefdoms, states), and that they accept vertical hierarchy (Diamond, 1997; Patterson, 2001). They are expected to follow rules from the top (in this case government) in return for services provided by government. The historian Fernard Braudel (1979, pp. 458–599) insightfully described the development of modern states in Europe and the main tasks of their governments:Â€(i) to secure obedience, (ii) to exert control over the market, which serves as a mechanism of exchange between the supply and demand of goods and services, and (iii) to strengthen the culture of the society. Evidently, governmental policies are directed to achieving the main tasks of the governments. Key governmental policies usually relate to national defence, food security, economic development, education, health care, spatial planning, infrastructure, traffic, etc. Environmental policy is a relatively new branch of governmental policy, with the theory borrowed initially from economic policy (Tinbergen, 1952). The general aim of environmental policy is to contribute to social welfare by protecting the environment through correcting societal failures, decreasing pollution, halting biodiversity loss and maintaining natural resources. The United Nations Conference on the Human Environment in Stockholm in 1972 is generally seen as having been a key step for increased political awareness in Europe about environmental problems created in part by N (UNEP, 1972), and subsequently for the establishment of environmental policies by governments. One of the main aims of the Conference was to put the issue of acid rain on the international agenda. Nitrogen oxides (NOx) and sulphur dioxide (SO2) are the main contributors to acid rain (Finlayson-Pitts and Pitts, 2000). They are formed during combustion processes and were linked initially to the acidification of Scandinavian lakes and streams. The 1972 Conference ultimately led to the establishment, in 1979, of the UNECE Convention on Long-range Transboundary Air Pollution (CLRTAP) (UNECE, 2010), which has been ratified by most countries in Europe. International treaties and conferences also played major roles in the establishment of water-related environmental policies. The first Convention on the Protection of the Marine Environment of the Baltic Sea was signed in Helsinki in 1974 (HELCOM, 2010). In 1992, a new convention was signed, aimed at protecting the Baltic Sea from all sources of pollution derived from land, shipping and atmospheric deposition

(HELCOM, 2010). The OSPAR Convention on the Protection of the Marine Environment of the North-East Atlantic was also signed in 1992 (OSPAR, 2010). One of the recommendations was the ‘substantial reduction (about 50%) of inputs of N and P into marine areas of the North-East Atlantic where these inputs are likely, directly or indirectly, to cause pollution’, between 1985 and 1995, using N (and P) balances as monitoring tools. The HELCOM and OSPAR Conventions have resulted in various national and EU policies on the protection of groundwater and surface waters, as discussed below. Justification of governmental policy to decrease Nr emissions is mainly based on the significant human health effects and biodiversity losses associated with increased amounts of various reactive N compounds in air, surface waters and groundwaters, and terrestrial ecosystems sensitive to eutrophication and acidification (Erisman et al. 2011, Chapter 2 this volume). Hence, the ultimate objective of governmental policy is ‘to decrease Nr emissions to a level where the value of marginal damages to human health and biodiversity is (approximately) equal to the marginal cost of achieving further reductions’ when considered from a cost–benefit point of view. An alternative formulation is ‘the ultimate objective of policies is to decrease Nr emission to levels that do not give rise to significant negative impacts on, and risks to human health and environment’. However, defining the objective of governmental policy is value-laden and often the subject of fierce political debate (Hajer, 1995; Baker et al., 1997). This debate is further complicated by the complexity of the cause–effect relationships of N compounds emissions and the multi-dimensional outcome of governmental policy, which affects different stakeholders, often with opposite interests, in different ways. This in turn often leads to compromises and delays in the implementation of governmental policy (Bressers and Huitema, 2001; Driessen and Leroy, 2007). The main sources of reactive N compound emissions distinguished by current governmental policy are: (i) combustion (mainly NOx by industry, power plants andÂ€traffic); (ii) waste waters (mainly dissolved and particulate N in discharges by industry and households); and (iii) agriculture (mainly NH3 and N2O to air, NO3 to groundwater and dissolved and particulate N to surface waters). The lack of full understanding of different emission sources, Nr compounds and loss pathways, and of different receptors with different sensitivities to Nr compounds (Hatfield and Follett, 2008) has led to a strong compartmentalization and (regional) differentiation of governmental policies. There are thus policies for specific sectors (energy, industry, households, waste waters and agriculture), N compounds (NOx, NO3, NH3, etc.), regions (countries, sensitive areas, vulnerable zones, etc.), and compartments or receptors (atmosphere, nature conservation areas, forests, groundwater, surface waters, soil, etc.). These complexities in part also reflect the compromises of fierce debates and diverging interests between stakeholders, for example, between industry and nature conservation organizations, and between the Departments of Economic

63

Nitrogen in current European policies Table 4.1 Possible policy instruments, with some examples

Regulatory instruments

Economic instruments

Communicative instruments

–â•‡ p ublic land use planning (zoning/ spatial planning) –â•‡ pollution standards and ceilings –â•‡ fertilization limits –â•‡ best available technique requirement

–â•‡ taxes –â•‡ subsidies (including price support) –â•‡ import/export tariffs –â•‡ tradable emission rights and quotas

–â•‡ extension services –â•‡ education and persuasion –â•‡ co-operative approaches

Development, Traffic and Agriculture on the one hand and the Departments for Environment and Nature Conservation on the other (Driessen and Leroy, 2007). The purpose of this chapter is to provide (i) some concepts of governmental policies, (ii) an overview of governmental policies in Europe (mainly EU) that influence N flows and emissions, and (iii) a preliminary assessment of the various policies, with the aim of identifying interactions between policies and critical success factors.

4.2â•‡ Concepts of governmental policy Basically, there are four principle drivers in organizing and governing societies, namely: • culture (human values, traditions, fashion and cultural habits); • market power and expertise (the ‘invisible hand’ of the free market); • public policy measures (state coercion, i.e. regulation pressure by governments); and • civic society pressure (pressure from non-governmental organizations (NGOs) and societal pressure and lobby groups). Public or governmental policy is a response to the identification of a societal problem, where culture, markets and civic society pressure collectively fail to solve that problem. Governmental policy aims at modifying human individual behaviour so as to achieve societal (public) objectives, i.e. to contribute to the total welfare of society (Tinbergen, 1952; Baumol and Oates, 1988). The fact that ‘public policy’ addresses societal objectives does not mean that everybody in the society equally accepts this policy and its consequences. There is often a strong divide in societies between those who believe in the cleansing mechanism of the market and in the ability of humans to act responsibly, and who therefore prefer a minimum of governmental policy, and those who emphasize the failures of markets and the need to help the less endowed in society, and therefore favour more extensive governmental policy. Policy instruments are the tools to implement the policy in practice. There are different type of instruments, the choices of which depend on the nature of the problem, the objectives of the policy and the competences and characteristics of the addressees (Baumol and Oates, 1988; Gunningham and Grabosky, 1998). Instruments can be divided into three categories:Â€ (i) regulatory or command-and-control instruments, (ii) economic or market-based instruments and (iii) communicative or persuasive instruments (Table 4.1).

64

Regulatory instruments (regulation) involve a restriction on the choice of agents, methods and actions. Regulations are compulsory measures imposing requirements on producers to achieve specific levels and standards of environmental quality, including environmental restrictions, bans, permit requirements, maximum rights or minimum obligations. They are the most common policy instrument used in EU environmental policy (e.g. Nitrates Directive). Economic instruments (stimulation) are meant to stimulate preferred production pathways. They are common in agricultural policy, for example, in the EU Common Agricultural Policy (CAP). Environmental taxes and tradable rights/quotas have only been implemented in a few countries. Subsidies are increasingly used as a policy instrument to promote environmentally friendly practices and the introduction of new technology. Communicative instruments (persuasion) include public projects to address environmental issues and measures to improve information flows to promote good practices and environmental objectives. This information can be provided to both producers, in the form of technical assistance and extension, and to consumers, e.g. via labelling. Technical assistance and extension are meant to provide users with information and technical assistance to implement environmentally friendly practices. This category also includes so-called voluntary approaches, e.g. codes of good agricultural practice (Sutton etÂ€al., 2007). Whether those addressed by policy then change their behaviour and contribute to achieving the objectives depends on the instrument and the decision environment of those addressed. A decision environment can be defined as ‘the collection of information, alternatives, values, and preferences available at the time of the decision’. An ideal decision environment would include all possible information, all of it accurate, and every possible alternative at the time. This is usually not the case and explains why the implementation of a policy in practice is far from complete. In short, compliance with a policy will depend on the knowledge and information held by the addressee (‘capability’), the availability of the appropriate tools and means (‘ability’) and on the persuasion (‘willingness’) of the addressee to implement the policy (Figure 4.1). The theoretical and empirical bases of governmental policy measures are still relatively small. This holds also for policy measures related to the abatement of unwanted Nr emissions. The relationships between ‘policy objectivesÂ€ – policy instrumentsÂ€ – change in human behaviourÂ€ – human health, ecological impacts and possible side-effects’ are complex, and to

Oene Oenema

Policy Monitoring

Tool box – Regulation – Stimulation – Persuasion

Instruments

Competences – Capability – Ability – Willingness

Change in humans’ behaviour

Societal objectives

Humans’ objectives

Figure 4.1 Simple representation of the intended working of governmental policy.

Driving forces

International conventions have played major roles in the establishment of governmental policies aimed at decreasing emissions of Nr to, and Nr concentrations in, the environment. Conventions and their protocols relevant to this chapter are summarized in Table 4.2 and further discussed in the supÂ� plementary information (Section 4.4). Intergovernmental organizations (IGOs), whilst not specifically legislative bodies, influence policy internationally (see Table 4.3). They are distinguished from treaties by virtue of their ‘international legal personality’. Further discussion on the inter-relationships of international conventions and IGOs and their interests in N control may be found in Bull et al. (2011, Chapter 25, this volume).

4.4â•‡ Policy measures affecting nitrogen in European Union

Responses

Pressures

4.3â•‡ International conventions and intergovernmental organizations

Impact

State

Figure 4.2 The Driving forcesÂ€– PressuresÂ€– StateÂ€– ImpactÂ€– Responses framework (DPSIR) for assessing cause–effect relationships and for developing a policy response (Source:Â€EEA, 1995.)

some extent based on trial and error. Further, the toolbox for implementing governmental policy measures is relatively small; choices have to be made between regulatory instruments, economic instruments and communicative/voluntary instruments, or a mix of these three. The available theoretical and empirical bases often do not help indicate, a priori, which combination of instruments will be most effective and efficient. The development of the so-called DPSIR framework (see Figure 4.2) and related frameworks by the Organisation for Economic Co-operation and Development (OECD) and the European Environmental Agency (EEA) in the 1990s has improved the understanding of the cause–effect Â�relationships of environmental pollution (see, for example, OECD, 1991; EEA, 1995). It has also provided a framework for responding to environmental problems via policy measures. According to the DPSIR framework, there is a chain of causal links starting with ‘driving forces’ (economic sectors, human activities) through ‘pressures’ (emissions, waste) to ‘states’ (physical, chemical and biological) and ‘impacts’ on ecosystems, human health and functions, eventually leading to political ‘responses’ (policy definition, prioritization, target setting, indicators).

In the following sections, current EU policy measures dealing with N are briefly summarized. Policies related to air and water are discussed first, followed by policies related to agriculture, biofuel and nature conservation. The final section (SectionÂ€ 4.4.6) provides a comprehensive overview. To facilitate access to the various EU policies documents, reference is made to the most recent websites (all policies are referenced as EC, 2010a–y). EU environmental policy is mostly established by means of Directives, imposing environmental objectives to be achieved by the Member States. EU Directives fix the framework in which Member States must create national legislation directed to industries/civilians in order to attain the environmental quality objectives laid down in the Directives. In contrast, EU agricultural policy is mostly established through so-called Regulations. These Regulations are directly binding for Member States and, depending on the issue, producers/stakeholders/ industries. Hence, EU Directives provide more flexibility than EU Regulations for Member States’ implementation. Note that EU Directives are commonly based on ‘regulatory instruments’ (Table 4.1) and that EU Regulations are often based on a mixture of ‘economic instruments’ and ‘regulatory instruments’. Understanding EU policy measures dealing with N emissions abatement requires insight into the understanding and perception by scientists and policy makers of the cause–effect relationships of these emissions. Many current policy measures dealing with N emissions reflect a simple ‘sourceÂ€ – receptor/ effect’ model of understanding. Combustion (mainly NOx by industry, power plants and traffic), waste waters (mainly dissolved and particulate N in discharges by industry and households) and agriculture (diffuse emissions of NH3 and N2O to air and NO3− to waters) are seen as the main N sources, while atmosphere, surface waters and groundwater are seen as the direct receptors. Thus, many policy measures focus on decreasing N compound emissions from specific sources and/or on decreasing N compound concentrations in receiving bodies (receptors) to below critical concentration levels.

65

Nitrogen in current European policies Table 4.2â•‡ Conventions and protocols addressing nitrogen emissions

1974 Helsinki Convention (HELCOM) on the Protection of the Baltic Sea in Helsinki 1974 OSPAR Convention (PARCOM) on the Protection of the North-East Atlantic 1976 Barcelona Convention on the Protection of the Mediterranean Sea 1979 UNECE Convention on Long-range Transboundary Air Pollution (CLRTAP) 1988 Sofia protocol on Nitrogen oxide (NOx) emissions 1992 Bucharest Convention on the Protection of the Black Sea 1992 Convention on Biological Diversity 1992 Convention on Transboundary Waters and International Lakes 1994 United Nations Framework Convention on Climate Change (UNFCCC) 1997 Kyoto Protocol 1999 Gothenburg Protocol on acidification, eutrophication and ground-level ozone Table 4.3â•‡ IGOs with linkages to nitrogen

1945 Food and Agriculture Organization (FAO) 1948 World Health Organization (WHO) 1950 World Meteorological Organization (WMO) 1972 United Nations Environment Programme (UNEP) 1988 Intergovernmental Panel on Climate Change (IPCC) 1996 Arctic Council

4.4.1â•‡ EU policy measures related to atmospheric Nr

Table 4.4 provides an overview of the three main EU Directives on nitrogen in the atmosphere. Following extensive reviews, the 1988 Directive on Large Combustion Plants (LCP; EC, 2010a), the 1996 Directive on Integrated Pollution Prevention and Control (IPPC; EC, 2010b), the 2000 Waste Incineration Directive (WID; EC, 2010c) and the 2005 Directive on Emission from Ignition Engines in Heavy-duty Vehicles (HDV; EC, 2010d), were incorporated into the 2008 Directive on Industrial Emissions concerning Integrated Pollution Prevention and Control (IPPC) (EC, 2010b). This 2008 IPCC Directive is now one of the cornerstones of EU Directives dealing with atmospheric Nr, and sets requirements and standards for NOx emissions from all kinds of combustion sources (Table 4.4). The IPPC Directive employs an integrated approach to the management of all types of pollution from industrial installations, including those for the intensive rearing of poultry or pigs. It requires these installations to have a permit and to minimize all kinds of pollution (including reactive N compounds emissions) by using Best Available Techniques (BAT). An essential part of the IPPC Directive is that the listed activities require a permit to operate, the approval and renewal of which is subject to cross-compliance with other European Community legislation.

66

The second cornerstone of EU Directives dealing with atmospheric Nr is the 2001 National Emission Ceilings Directive (NEC; EC, 2010e). This Directive sets upper limits (ceilings) for each Member State for the total emissions in 2010 and 2020 of the four pollutants responsible for acidification, eutrophication and ground-level ozone pollution (SO2, NOx, VOCs and NH3), but leaves it largely to the Member States to decide which measures to take in order to comply (Table 4.4). The Directive aims at achieving the long-term objectives of not exceeding critical levels and loads by establishing national emission ceilings, taking the years 2010 and 2020 as benchmarks. This Directive is currently (2010) under revision. The 1996 Framework Directive on Ambient Air addresses ambient air quality assessment and management (EC, 2010f). It includes a series of daughter directives, which set the numerical limit values for atmospheric pollutants. For example, the 1999 Air Quality Directive relates to limit values for, among others, nitrogen oxides (NOx), ozone (O3) and particulate matter (PM10) in ambient air. The main emphasis is human health in urban areas and on air pollutants from combustion sources. The most recent version of the Ambient Air Quality Directive was approved in 2008. It contains limit values for NOx, O3 and PM2.5, but not for NH3. Ozone is included as nitrogen oxides (NO and NO2) are important O3 precursor substances, and because of adverse effects of high O3 concentration on human health and crop growth. Particulate matter is included because of its close link to the N cycle (see Hertel et al., 2011; ChapterÂ€9 this volume), being formed as a result of the processing of ammonia, nitrogen oxide and other N-containing substances, and its effects on human health. The 2008 Ambient Air Quality Directive is now one of the three cornerstone Directives dealing with atmospheric Nr in the EU-27 (EC, 2010f).

4.4.2â•‡ EU policy measures related to N in waterÂ€bodies A number of EU policy measures exist which address the issue of Nr emissions and concentrations in water bodies, these are detailed below and summarized in Table 4.5. The 2000 Water Framework Directive (EC, 2010h) embraces all EU legislation for the protection of inland surface waters, transitional waters, coastal waters and groundwater. The Water Framework Directive (WFD) requires all waters to reach ‘good ecological status’ by 2015. It will do this by establishing a river-basin district structure within which demanding environmental objectives will be set, including ecological targets for surface waters and good chemical and quantitative status for groundwater bodies. It requires the implementation of measures from 11 other EU Directives, including the 1976 Bathing Water Directive (EC, 2010i), the 1990 Urban Waste-water Treatment Directive (EC, 2010j), the 1985 Environmental Impact Assessment Directive (EC, 2010k), the 1991 Nitrates Directive (EC, 2010l), the 1996 IPPC Directive (EC 2010b), the 1998 Drinking Water Directive (EC, 2010m) and the 2006 Groundwater Directive (EC, 2010n). The WFD includes an indicative list of main pollutant substances, including substances

Oene Oenema Table 4.4â•‡ Overview of main EU Directives related to N emissions to, and concentrations in, the atmosphere (see also EC, 2010g)

Directive

Description / objectives

Limit values

2008/50/EC

Ambient air quality: definitions, threshold values, targets and assessment, in relation to sulphur dioxide, nitrogen dioxide, particulate matter, lead, benzene and carbon monoxide.

• Critical level for NOx for vegetation (average over 1 year): 30 μg m−3 • Limit values for NOx for human health (averaged over 1 yr): 40 μg m−3 • Limit values for NOx for human health (averaged over 1 hr):Â€ 200 μg m−3 • Alert thresholds for NOx for human health (averaged over 3 hr):Â€ 400 μg m−3 • Target and limit values for PM2.5 in urban areas (average over 3 yr): 20–25 μg m−3.

2008/1/EC

Integrated Pollution, Prevention and Control (IPPC):Â€to prevent and control emissions from industrial activities into air, water or soil, in relation to polluting substances, including nitrogen

• Installations need a permit • Installations need to comply with environmental quality standards described in other Directives • Installations need to apply best available techniques (BATs)

2001/81/EC

National Emission Ceilings (NEC):Â€to limit emissions to protect the environment and human health against risks of adverse effects from acidification, eutrophication and groundlevel ozone, by establishing national emission ceilings, taking the years 2010 (and 2020) as benchmarks

• National emission ceilings for SO2, NOx, VOC and NH3, for each country to be attained by 2010, expressed in kilotonnes (Gg) • In regard of the long term objectives ‘not exceeding critical levels and loads and of effective protection of all people against recognized health risks from air pollution’ no ceilings have been yet set for 2020 though the Directive envisages ongoing review

which contribute to eutrophication (in particular, nitrates and phosphates). The WFD allows Member States the flexibility to define specific ambitions, targets and time frames, albeit under the constraints of proper underpinning and justifications. The most important linked Directives of the WFD as regards Nr emissions to groundwater and surface waters are the 1991 Urban Waste Water Directive and the 1991 NitratesÂ€Directive. The 1991 Urban Waste Water Directive (UWWD; EC, 2010j) concerning urban waste water treatment was adopted in 1991 to protect the water environment from the adverse effects of discharges of urban waste water and from certain industrial discharges. The UWWD has requirements for sewerage (or collection systems) to be established and sets standards for sewage treatment. The general principle of the Directive is to provide treatment of sewage from the largest discharges first, and to protect sensitive waters. It sets secondary treatment as the normal standard, but requires tertiary treatment where discharges affect sensitive areas identified under the Directive. It also requires that discharges from urban waste water treatment plants to sensitive areas do not contain more than 10–15 mg N per litre, depending on the size of the communities, and that the waste water treatment system removes 70%–80% of the initial amount of Nr in the sewage. The main objective of the 1991 Nitrates Directive is ‘to reduce water pollution caused or induced by nitrates from agricultural sources and prevent further such pollution’ (EC, 2010l). This Directive requires Member States to take the following steps:Â€ (i) water monitoring (with regard to nitrate concentration and trophic status); (ii) identification of waters that are polluted or at risk of pollution; (iii) designation of vulnerable zones (areas that drain into identified waters); (iv) the establishment of codes of good agricultural practices and

action programmes (a set of measures to prevent and reduce nitrate pollution); and (v) the review at least every four years of the designation of vulnerable zones and action programmes. Waters must be identified as polluted or at risk of pollution if nitrate concentrations in groundwater and surface waters contain or could contain more than 50 mg/l per litre if no action is taken, or if surface waters, including freshwater bodies, estuaries, coastal and marine waters are found to be eutrophic or in the near future may become eutrophic if no action is taken. The action programmes must contain mandatory measures relating to:Â€(i) periods when application of animal manure and fertilizers to land is prohibited; (ii) capacity of and facilities for storage of animal manure; and (iii) limits to the amounts of animal manure (170 kg/ha/yr) and fertilizers applied to land, which should ensure a balanced fertilization. The 2008 Marine Strategy Directive (EC, 2010p) aims to achieve good environmental status of the EU’s marine waters by 2020 and to protect the resource base upon which marinerelated economic and social activities depend. It covers the following marine regions:Â€ (a) the Baltic Sea; (b) the North-East Atlantic Ocean; (c) the Mediterranean Sea; and (d) the Black Sea. It contains an indicative list of characteristics, pressures and impacts which have to be monitored and assessed regularly, and for which environmental targets have to be set. The list of pressures and impacts includes inputs of fertilizers and other nitrogen- and phosphorus-rich substances (from point and diffuse sources, including agriculture, aquaculture and atmospheric deposition). Each Member State has to draw up a programme of cost-effective measures to address adverse characteristics, pressures and impacts. Impact assessments, including detailed cost–benefit analysis of the measures proposed, are required prior to the introduction of new measures.

67

Nitrogen in current European policies Table 4.5â•‡ Overview of main EU Directives related to N emissions and concentrations in water bodies (see also EC, 2010o)

Directive

Description / objectives

Requirements/Limit values

2000/60/EC

Water Framework Directive (WFD): to establish a framework for the protection of inland surface waters, transitional waters, coastal waters and groundwater from pollution and depletion

•

Urban Waste Water Treatment Directive (UWWD): to protect the environment from the adverse effects of waste water discharges from urban areas and certain industrial sectors

•

91/271/EEC

91/676/EEC

Nitrates Directive (ND): concerning the protection of waters against pollution caused by nitrates from agricultural sources

• • •

• •

•

• • •

• 2008/56/EC

Marine Strategy Framework Directive:Â€establishes a framework to take the necessary measures to achieve or maintain good environmental status in the marine environment by the year 2020 at the latest

• •

• •

2006/118/EC

Groundwater Directive: establishes a regime which sets underground water quality standards and introduces measures to prevent or limit inputs of pollutants into groundwater

•

•

•

The 2006 Groundwater Directive (EC, 2010n) complements the Water Framework Directive and requires Member States to:Â€ (i) establish groundwater quality standards by the end of 2008; (ii) carry out pollution trend studies; (iii) reverse pollution trends so that environmental objectives are achieved by 2015; (iv) operate measures to prevent or limit inputs of pollutants into groundwater; (v) make reviews of technical provisions of the Directive in 2013 and every six years thereafter; (vi) comply with good chemical status criteria (based on EU standards of nitrates and pesticides and on threshold values established by Member States).

68

Maintaining/establishing good ecological status in surface water bodies and good chemical and quantitative status in groundwater bodies Establishment of river basement management plans Designation of â•›‘protected areas’ For ‘limit values’ and ‘measures required’ reference is made to other Directives All agglomerations must be provided with collecting systems for urban waste water Identification of sensitive areas Requirements for discharges from urban waste water treatment plants to sensitive areas:Â€(i) a reduction of total Nr of 70%–80% of the influent; and (ii) maximum annual mean total N concentrations of 1.5–10 mg/l, depending on size of the urban area Establishment of a code of good agricultural practice, including balanced N fertilization, to be implemented by farmers on a voluntary basis Designation of Nitrate Vulnerable Zones Establishment of action programmes with mandatory measures in vulnerable zones, including N application limits Water quality trigger criteria:Â€(i) 50 mg nitrate per litre in groundwater and surface waters, and (ii) eutrophic status of surface waters Application limit for nitrogen from animal manure:Â€170 kg/ha/yr Determination of a set of characteristics for good environmental status Establishment of a comprehensive set of environmental targets for marine waters to guide progress towards achieving good environmental status Identification and implementation of measures needed to achieve or maintain good environmental status There are no prescribed limit values Groundwater quality standards for nitrate and active substances in pesticides, including their relevant metabolites, degradation and reaction products Threshold values for all pollutants and indicators of pollution which characterize groundwater as being at risk of failing to achieve good groundwater chemical status Establishes the 50 mg/l for nitrate as a binding maximum quality threshold

4.4.3â•‡ EU Common Agricultural Policy and itsÂ€reforms. The Common Agricultural Policy (CAP) of the EU was established in 1958 by the EEC. The CAP has contributed greatly to the modernization and productivity of agriculture and to food security in the EU (Ritson and Harvey, 1997). Indirectly, it has also contributed to increased inputs of N in agriculture via N fertilizers and to the import of animal feed from outside the EU, as well as to increased N losses from agriculture to the environment (Romstad et al., 1997). Following the recognition and increased awareness of the effects of surpluses of agricultural products and environmental

Oene Oenema

burden associated with the intensification of agricultural production, the CAP went through a series of reforms, notably in 1984 (implementation of milk quota), 1992 (set-aside regulations), 1997 (agenda 2000) and 2003 (fundamental change in the EU support to agriculture:Â€EC, 2010q; EC, 2010t; Meester et al., 2005; Blandford and Hill, 2006). In 2003, it was agreed that the CAP has two pillars:Â€(i) market policies and (ii) rural development policies. In 2008, agreement was reached to further modernize, simplify and streamline the CAP and remove restrictions on farming (the so-called ‘Health Check’). This agreement includes the phasing-out of the milk quota system, the abolition of set-aside regulations and a further shift from direct aid for production support to the Rural Development Programme (EC, 2010q; EC, 2010s.). The reforms of the CAP continue to have a significant influence on N use and its loss to the environment. ‘Cross-compliance’ is a main policy vehicle to implement the CAP reform. In this context, cross-compliance is the requirement that farmers in receipt of payments under the CAP are also shown to be meeting other relevant European Community legislation. In June 2003, cross-compliance became an obligatory element of the first pillar of CAP, thereby coupling existing environmental policies and other policies to agricultural income support, as implemented in the so-called ‘Single Farm Payments’ to farmers. There are two major aspects of cross-compliance in the Single Farm Payment (EC, 2010q):Â€(i) Compliance with 19 Statutory Management Requirements (SMRs) covering the environment, food safety, animal and plant health and animal welfare, including the provisions of the relevant directives; and (ii) Compliance with a requirement to maintain land in Good Agricultural and Environmental Condition (GAEC). Definitions of GAEC are specified at the national or regional level and address soil organic matter, soil erosion, maintenance of the land(scape) and avoidance of the deterioration of natural habitats. A few of the SMRs directly or indirectly address N inputs and N emissions in agriculture. These include, for example, the 1991 Nitrates Directive, the 1986 Sewage Sludge Directive, the 1992 Directive on the conservation of natural habitats and of wild flora and fauna (Habitats Directive), and the 1979 Directive on the conservation of wild birds (Birds Directive). Such cross-compliance with other environmental regulations has the potential to encourage the reduction of Nr losses from agriculture. However, this is not always the case. For example, emerging requirements for animal housing to meet new animal welfare standards (EC, 2010r) will in many cases contribute to increased emissions of NH3 and N2O. This interaction highlights the need to consider environmental regulation in the context of other societal pressures. The second pillar of the CAP is the Rural Development Policy, which for the period 2007 to 2013 is set out in Council Regulation No. 1698/2005 (EC, 2010t). Under this regulation, rural development policy is focused on three themes (known as ‘thematic axes’) plus the LEADER approach:Â€ (i) improving the competitiveness of the agricultural and forestry sector; (ii) improving the environment and the countryside;

(iii)Â€ improving the quality of life in rural areas and encouraging diversification of the rural economy; and (iv) mainstreaming the LEADER approach ‘Links between Activities Developing the Rural Economy’ (LEADER, ‘Liaison Entre Actions de Développement de l’Economie Rurale’). To help ensure a balanced approach to the rural development policy, Member States and regions are obliged to spread their rural development funding between all these thematic axes. Within each of the first three axes, various support mechanisms have been described in articles 20 to 35 for Axis 1, in articles 36 to 51 for axis 2 and in articles 52 to 59 for axis 3, which help with improving the agronomic and environmental performances of agricultural activities in the rural areas. These measures may include the setting up of advisory services, supporting modernization of agricultural holdings, supporting operations related to access to farm and forest land, land consolidation and improvement, energy supply and water management, and agri-environmental payments. Clearly, the Rural Development Policy can contribute to measures that decrease Nr losses from agriculture to the environment.

4.4.4â•‡ EU nature conservation policies The policy framework for preventing biodiversity loss in the EU is provided by the Birds and Habitats Directives, which are being implemented through Natura 2000, an EU-wide network of protected areas, which now covers some 18% of the territory of the EU. The 1979 Birds Directive (EC, 2010v) requires Member States to designate Special Protection Areas (SPAs) for endangered bird species. Currently, over 4000 SPAs have been designated, covering 8% of EU territory. The 1992 Habitats Directive (EC, 2010w) aims to protect other wildlife species and habitats. Each Member State is required to identify Special Areas of Conservation (SACs) and to put in place a special management plan to protect them. The SPAs and SACs together make up the Natura 2000 network. Member States are required to improve the ecological coherence of Natura 2000 by maintaining, and where appropriate developing, features of the landscape which are of major importance for wild fauna and flora. The Birds and Habitats Directives imply restrictions on human activities within and around the Natura 2000 areas. Widely established restrictions include infrastructural, industrial and agricultural activities in and near to Natura 2000 sites. The Directives also have implications for activities taking place that are not on the site itself. In addition, the Birds and Habitats Directives establish lists of designated species and habitats, with a commitment to monitoring the performance of these across the whole of the EU. This represents an important part of the overall objective of these Directives, though it should be noted that there is a lack of measures to protect such habitats and species outside of the Natura 2000 network. In principle, the Birds and Habitats Directives are drivers to safeguard biodiversity and to lower NH3 and NOx emissions, by virtue of the precautionary approach. However, this is still an area of ongoing development in learning to implement the

69

Nitrogen in current European policies

N input limits

SOURCES

Wastes

Combustion

Agriculture

Figure 4.3 Schematic overview of the N control mechanisms of European policy measures. For the N emission sources, there are N input limits and N compounds emission limits, for the N receptors, there are N compounds concentrations limits and N compounds exposure limits, including critical loads.

N emission limits

Air

N concentration limits

Water

RECEPTORS

N exposure limits Humans

Flora

Fauna

existing legislation, and in evaluating its limitations (COST 729, 2009).

4.4.5â•‡ EU bio-energy policy Current EU energy policy focuses on increasing the security of energy supply and reducing greenhouse gas (GHG) emissions, as set out in 2007 in ‘The Renewable Energy Road Map’ (COM(2006)848; EC, 2010u). In the Road Map, a mandatory target was set for achieving a 20% share of renewables in energy consumption in the EU by 2020 and a mandatory minimum target of 10% of all energy in transport from biofuels. The recent Directive on the promotion of the use of energy from renewable sources (2009/28/EC; EC, 2010y) amended the 2003 Biofuel Directive (2003/30/EC). Though ‘nitrogen’ is not mentioned explicitly in any of the energy policy documents (except for N2O as a greenhouse gas), the EU policy on bioenergy will have influence on N use in agriculture, as bio-energy crops require Nr for their growth and release various N compounds to the broader environment during and following their growth and utilization. The EU policy on bioenergy will also have influence on the total agricultural area used for the production of food, feed and fibres.

4.4.6â•‡ Summary of nitrogen control by EuropeanÂ€policies In summary, N flows and emissions in Europe are regulated by a broad variety of policy measures. These policy measures regulate N flows and emissions via (i) input control (e.g. N application limits in agriculture), (ii) emission control (e.g. Nr emission limits, discharge limits), (iii) concentration limits for Nr in air and water bodies, and (iv) Nr exposure limits and critical N loads (Figure 4.3). Input controls exist only for agriculture, via application limits for Nr from animal manure and fertilizers to agricultural land, and via provisions for the protein content of animal feed. Such limits do not apply for combustion and wastes. Emission controls exist for all major N compounds, for example via the national emission ceilings for NOx and NH3, NOx emission limits for stationary and mobile combustion

70

sources, discharge limits for industry and sewage treatment plants. Further, NH3 emissions abatement measures exists for animal housing, manure storages and manure application, and N fertilizer application to land. Table 4.6 provides a summary of quantitative EU limit values for various N compounds in air and water. In air, there are limit values for NOx (NO and NO2) and for substances that are formed in part through the presence of NOx in air, viz., ozone (O3) and fine particles (PM2.5 and PM10). Currently, there are no limit values for NH3 concentrations in air. In water, there are limit values for NO3−, NO2−, NH4+, and Ntotal. There are no limit values for N compounds in soil. Exposure limits for humans and N-sensitive flora and fauna are defined either via concentration limits or via input limits, such as critical loads. A critical load is defined by the CLRTAP (UNECE, 1999) and the NEC Directive (EC, 2010e) as ‘a quantitative estimate of an exposure to one or more pollutants below which significant harmful effects on specified sensitive elements of the environment do not occur, according to present knowledge’. Critical N-loads for ecosystems are determined following specific methodologies and criteria for mapping critical levels/ loads and geographical areas (ICP Modelling and Mapping, 2004). Critical loads form the basis for setting emission limits and ceilings (Sliggers and Kakebeeke, 2004).

4.5â•‡ Assessment of environmental policies inÂ€Europe Assessments of environmental policy usually include analyses of its compliance, expressed in terms of implementation of mandatory obligations, its effectiveness, expressed in terms of achieving policy objectives, and its efficiency, expressed in terms of the economic costs of its implementation. In addition, assessments may address the possible technical, technological, socio-economic, institutional and societal changes brought about by environmental policy. Assessment of compliance is usually the first step; it simply records whether the obligations of the policy (e.g. abatement measures, designation of specific areas, and monitoring and reporting obligations) have been satisfied. However, the effectiveness and efficiency of planned environmental policy may be

Oene Oenema Table 4.6â•‡ Summary of limit values for N compounds concentrations in air and water as set by EU policies

Effects

Indicators

Limit values / targets

Regulatory reference

Respiratory diseases of humans

NOx

40 μg m (annual mean) 200 μg m−3 (hourly mean) 400 μg m−3 (threshold, 3 hrs)

Ambient Air Quality Directive (2008/50/EC)

Ibid

PM2.5

20–25 μg m−3 (average of 3 yrs)

Ibid

Ibid

PM10

40–50 μg m (average of 3 yrs)

Ibid

Ibid

O3

180–240 μg m (hourly mean)

Ibid

Ibid

AOT40

120 μg m (hourly mean)

Ibid

Plant damage

NOx

30 μg m−3 (annual mean)

Ibid

Adverse effects on humans from nitrates

NO3−

50 mg/l in groundwater

Drinking Water Directive (98/83/EC) Nitrates Directive (91/676/EEC) Groundwater Directive (2006/118/EC)

Adverse effects on humans by nitrites

NO2−

0.5 mg/l in water used for drinking water. Further, [NO3−]/50 + [NO2−]/3 ≤ 1

Drinking Water Directive (98/83/EC)

Adverse effects on humans from ammonium

NH4+

Indicator value:Â€0.5 mg/l

Drinking Water Directive (98/83/EC)

Eutrophication of surface waters

NO3−

25–50 mg/l

Nitrates Directive (91/676/EEC)

Contamination of groundwater

NO3−

50 mg/l

Nitrates Directive (91/676/EEC) Goundwater Directive (2006/118/EC)

Eutrophication of surface waters

Ntotal

2–10 mg/l; discharge from sewage treatment plants

Urban Waste Water Directive (91/271/ EEC)

a

−3

−3

−3

a

−3

AOT40 stands for accumulated exposure over a threshold ozone concentration of 40 ppb.

assessed in advance (ex-ante) through simulation modelling and stakeholder consultation. Such ex-ante assessments provide a view of the effectiveness and efficiency of environmental policy prior to implementation (often assuming 100% compliance), and are instrumental for achieving political agreement for ratification and implementation. By contrast, retrospective (ex-post) assessments are usually based on analyses of data obtained through various monitoring programmes, censuses, inquiries and reviews. Assessments of environmental policy are sometimes heavily debated and also criticized. First, there are differences of views about the appropriateness of the objectives and targets that must be achieved, e.g. emission targets, concentration targets or ecological targets. Figure 4.4 shows that there is a large ‘separation’ between emissions targets and the human health and ecological impacts targets, there are also many possible interactions. Second, there is debate about the accuracy of, and uncertainties in, the data and the cause–effect relationships. For example, the NOx and NH3 emission estimates in Europe are thought to have an uncertainty range of 30% and 50%, respectively (EEA, 2005). Third, there are often discussions about the economic cost–benefit analyses and the effects on the competitiveness of sectors. Experiences over the past 20 years indicate that environmental policies in Europe have contributed to decreasing Nr losses to air, surface waters and groundwaters in Europe, but

that critical loads are still exceeded and that the environmental and ecological status of many groundwater bodies, surface waters and natural areas are still below the set targets (Erisman et al., 2011, Chapter 2 this volume). Many of these set targets reflect ecological targets and political compromises; few targets (if any) have been set at levels ‘where the value of marginal damages to human health and biodiversity is (approximately) equal to the marginal cost of achieving further reductions’, which would yield most societal benefit (see also Brink et al., 2011, Chapter 22 this volume). By contrast, there has been a tendency to go in one of two directions:Â€either to specify environmental targets based on their technical and political achievability or to set objectives for avoidance of adverse impacts. Many European environmental policies are based on regulatory instruments, with frequent use of BAT requirements and emission standards, and these appear to have a relatively low economic efficiency (OECD, 2007). EU environmental Directives leave little room for the use of more flexible economic instruments (e.g. taxes or trading systems for NOx emissions, taxes or trading systems for N-input to the agricultural sector). Economic instruments are not necessarily prohibited, but the Directives limit the flexibility these instruments could have offered (OECD, 2007). So far, policy measures aimed at decreasing Nr species emissions have achieved larger responses from combustion sources than from urban sources or from agricultural sources especially.

71


Emissions targets

Concentrations targets

Ecological process rate targets

Human health & ecological impacts targets

Acidification [NOX] in air Eutrophication NH3

[NH3] in air

N2O

[N2O] in air

Ground-level ozone (O3) formation

NO3

[NO3] in water bodies

Fine-particle matter (PMX) formation

NH4

[NH4] in drinking water

[Nt] surface waters Corrosion Other substances

Biodiversity loss

Food production

Toxification Radiation interception

Nt

Human health

Global Change

NOX

Climate change

Material loss

Interactions

Figure 4.4 Illustration of the major links between the multiple forms of reactive nitrogen emission and the resulting impacts on different concentration, process and impact targets.

This can be shown by the fact that relative emission reductions have been achieved in the following order:Â€ NOx emissions > Ntotal emissions from urban areas > NO3− leaching from agriculture > NH3 emissions from agriculture (see Erisman et al., 2011, Chapter 2, Figure 2.5). Emission reductions may also follow from changes in economic activities. For example, significant reductions in total NOx emissions to air in the EU-15 between 1990 and 2006 may be considered, to a large extent, influenced by environmental policies. By contrast, the decreases in total NOx emissions to air in the 12 new Member States of the EU (EU-12) between 1990 and 2006 mainly follow from the changes in the political and economic systems after 1989 (EEA, 2005; 2008), rather than the implementation of specific environmental policies. It can be seen from Figure 4.5 that the reductions in NOx emissions have been less successful than SO2 emission reductions, which is largely due to increased vehicle mileage offsetting the benefits of low NOx emission technologies (NEGTAP, 2001). By comparison, there has been only a small effect of environmental policies in reducing ammonia emissions. Figure 4.5 shows that NH3 emissions for the EU-15 only decreased by 10% between 1990 and 2006, while the larger decrease of 49% for the EU-12 was the result of

72

the political and economic changes following 1989, rather than due to specific environmental policies in the period. Within the EU-15, the differences in the effects of policies are also large. The extent to which, for example, the objectives of the policies to reduce ammonia emissions and nitrate leaching from agriculture have been achieved is variable across Member States (EEA, 2008). These variable results are ascribed to: • differences between Member States in their perceptions of EU Directives; • differences in economic sectors and systems and environmental conditions; • legislative delays and implementation delays; • economic costs of the measures and lack of enforcement; • continued economic growth, which has ‘neutralized’ some of the improvements in ‘eco-efficiency’ at the system level (e.g. increased car fleets offsetting projected NOx reductions from low emission vehicle technology); • ineffectiveness of some measures; • antagonisms between some of the measures; and • hysteresis effects, due to buffering reactions within the systems.

Oene Oenema EU-15 SO2

Emissions (Gg)

15000

NOX

NH3

12000 9000 6000 3000

NH3 emissions, relative to 1990 (%)

150

18000

Netherlands

EU-15

100

50

0

1990

1995

2000

2001

2002

2003

2004

2005

2006

Year

0 1995

2000

2001

2002

2003

2004

2005

2006

Year

EU-12

Latvia

Slovenia

EU-12 new

100

18000 SO2

15000

150

NH3 emissions, relative to 1990 (%)

1990

Emissions (Gg)

Spain

NOX

NH3

12000 9000 6000

50

0

1990

1995

2000

2001

2002 Year

2003

2004

2005

2006

Figure 4.6 Relative changes in total NH3 emissions to air in the EU-15, The Netherlands and Spain (top) and in the EU-12 (new), Latvia and Slovenia (Â�bottom) between 1990 and 2006 (source:Â€EEA, 2008).

3000 0 1990

1995

2000

2001

2002

2003

2004

2005

2006

Year

Figure 4.5. Changes in total NOX and NH3 emissions to air in EU-15 (top) and EU-12 (bottom) between 1990 and 2006. Emissions of SO2 to air are shown for comparison (source:Â€EEA, 2008).

4.5.1â•‡ Changes in NOx emissions from combustion

Combustion is a major source of NOx emissions and the basis for emissions abatement policy in Europe have been the 1988 Sofia Protocol on NOx emissions (Table 4.2) and the related EU Directives (Table 4.2). The transport and energy sectors are the main sources (Erisman et al., 2011, Chapter 2 this volume; EEA, 2008) and emissions of NOx in the EU-27 have decreased on average by about 31% between 1990 and 2005. Basically, reductions have occurred in all economic sectors and most countries have reported lower emissions of NOx in 2005 compared to 1990. The exceptions to this are Austria (7% increase), Cyprus (19%), Greece (6%), Portugal (13%) and Spain (26%). The three sectors ‘responsible’ for the vast majority of the decreased NOx emissions are road transport (contributing 53% of the total reduction in NOx emissions), energy industry (contributing 29%) and industry (energy) (contributing 15%). The significant reduction in NOx emissions from road transport (38% between 1990 and 2005) has been achieved despite the general increase in activity within this sector (EEA, 2008). Emissions of NOx have also declined in the energy industry (38% between 1990 and 2005), despite again an increase in activity (EEA, 2008). The decoupling of NOx emissions, transport and electricity and heat production has been due to (EEA, 2007; EMEP, 2007): • the introduction of catalytic converters in car engines; • the introduction of low-NOx combustion technology and flue gas treatment, which led to a 49% reduction;

• efficiency improvements, which resulted in a 14% reduction; • the switch in the fuel mix, away from coal and fuel oil towards natural gas, which led to an 8% reduction; • the lower share of nuclear and non-thermal renewable energy (i.e. excluding biomass) in 2004 compared to 1990, which actually increased emissions by 3%.

4.5.2â•‡ Changes in N losses from agriculture Agriculture in Europe contributes, on average, to about 80%–90% of the total emissions of NH3 into the atmosphere, to roughly 40%–60% of the Nr to surface waters, and to about 50%–70% of the emissions of N2O to the atmosphere (EEA, 2005; Oenema et al., 2007, 2009). Most of the NH3 originates from animal manure in stables, from manure storage systems and from the application of animal manure to agricultural land. Between 1990 and 2006, emissions of NH3 decreased by 12% in the EU-15 and by 47% in the EU-12 (Figure 4.5). In the EU-15, abatement policy and decreases in NH3 emissions were the greatest in the Netherlands and the least in Spain (FigureÂ€4.6). While the Netherlands is estimated to have had a 50% reduction in NH3 emissions between 1990 and 2006, NH3 emissions in Spain increased by 25% due to an expansion of the animal livestock sector. For the new Member States (EU-12), the contraction of the livestock herd and the decreased use of mineral fertilizer after 1989 resulting in decreases in NH3 emissions were greatest in Latvia (~70%) and least in Slovenia (~20%). Decreases in NH3 emissions in Hungary following the political and economic changes have been described by Horvath and Sutton (1998). There are a number of countries that report a decreasing trend of mean NO3− concentrations in shallow groundwaters following the implementation of the EU Nitrates Directive.

73


74

However, the decreases are modest and a significant number of monitoring stations show increasing NO3− concentrations (EC, 2007, 2010). Similarly, while 55% of the monitoring stations in surface waters in rural areas of the EU-15 had a decreasing trend in NO3− concentrations during the period 1996–2003, 31% of monitoring stations had stable NO3− concentrations and 14% of the stations showed increasing NO3− concentrations (EC, 2007, 2010x). Changes in NO3− concentrations have been related to changes in Nr surpluses. Surpluses of N of the soil surface balance in EU countries have on average decreased since 1990, in part in response to structural changes in agriculture following changes in the common agricultural policy, in part also in response to environmental policies, such as the Nitrates Directive. In the EU-15, mean Nr surplus decreased from 65 kg per ha in 1990 to 50 kg per ha in 2000 (EEA, 2005a). Surpluses (range 150–250kg per ha) and decreases in surpluses (range 30–50 kg per ha) were largest for the Netherlands, Belgium andÂ€Germany. The variable and slow responses of Member States to environmental policies in agriculture have been ascribed to (Romstad et al., 1997; Smith et al., 2007; MNP, 2007; Oenema et al., 2009; Mikkelsen et al., 2010): (i) the large differences in farming systems and environmental conditions in the EU-27 combined with the complexity of the N cycle; (ii) a variable interpretation by Member States of the targets and measures in environmental directives and regulations; (iii) hesitation in implementing measures, due to the perceived high costs to farmers and perceived low effectiveness; (iv) hesitation in introducing mechanisms to monitor compliance by farmers, due to the perceived high costs; (v) legislative delays; (vi) failure by farmers to implement measures, due to withinsystem constraints, perceived and actual costs, and the time needed for learning; and (vii) potential antagonisms between measures aimed at decreasing NH3 emissions and those aimed at decreasing NO3 leaching.

ammonia concentrations (Bleeker et al., 2009) and the envirÂ� onmental and ecological status of lakes, rivers and streams in rural areas have improved little yet (Stälnacke et al., 2004; Mourad et al., 2006).

Moreover, the recovery of the environmental and Â�ecological status of lakes, rivers and streams often takes more time than expected from the measures implemented and associated decrease of emissions. The same point has been made for atmospheric Nr compounds, including the question of why atmospheric ammonia levels did not decrease as fast as expected following implementation of emission reduction policies in Western Europe (Bleeker et al., 2009). Both of these findings point to the complexity of the systems and our limited understanding of the biogeochemical connectivity of systems and controls. There are ‘hysteresis’ effects and feedback mechanisms that are often overlooked and that lead to slow responses. This seems also to be the case for the new Member States in central Europe where fertilizer N inputs and livestock numbers decreased drastically following the political changes in the early 1990s, while the atmospheric

The treatment of urban waste water has also contributed to significant decreases in the Nr load to coastal waters and to the improvement of surface water quality in Europe in general. However, there are large spatial and temporal variations, and some contribution may have come from lower emissions from agriculture due to the implementation of the Nitrates Directive (EEA, 2005b). The 2005 OSPAR Assessment of Riverine Inputs (all sources) and Direct Discharges (urban waste water) for the period 1990–2002 noted significant decreases in total inputs of both N (up 32%) and P (up 135%) to the Arctic Waters and a significant reduction in total inputs of N (down 12%) in the Greater North Sea (OSPAR, 2005). Similarly, a downward trend in total riverine and direct point-source inputs of N and P has been observed for the Baltic Sea during the period 1994–2006, but again with large spatial and temporal variations (HELCOM, 2009). However, the overall policy target of a 50% reduction in

4.5.3â•‡ Changes in N losses from urban wasteÂ€waters The Urban Waste Water Treatment Directive (91/271/EEC; EC, 2010j) regulates discharges of municipal waste water from towns and larger villages and specifies which kind of treatment must be installed. The Directive requires that all European agglomerations (settlements) with a size of more than 2000 population equivalents (p.e.) are equipped with collecting and treatment systems for their waste waters. The basic level of treatment is so-called secondary treatment (i.e. removal of organic pollution). In sensitive areas (68% of the EU-27 territory), a more stringent treatment is required, for example, removal of a minimum of 75% of the N and P loads. Most EU Member States have designated their whole territory as a sensitive area, but some (e.g. United Kingdom, Spain, Hungary) have designated only a small area as sensitive (EC, 2009b). By the end of 2005, waste water collecting systems were in place for 93% of the total polluting load (in 83% of the agglomerations) (EEA, 2005b). Secondary treatment was in place for 87% of the load and was reported to work adequately for 78% of it. More stringent treatment was in place for 72% of the load and was reported to work adequately for 65% of it. The European Commission has concluded that considerable progress has been achieved in implementing the Directive, but that key challenges remain to align waste water treatment over the entire EU with the provisions of the Directive and the ‘good status’ environmental objective under the Water Framework Directive (EC, 2009). In particular, the secondary treatment and the more stringent treatment need to be improved, especially in the new Member States (EC, 2009b).

4.5.4â•‡ Changes in N pollution of marine watersÂ€byÂ€50%

Decrease in N and P losses (%)

Oene Oenema 100

have not decreased to the same extent. As a result, agriculture increasingly becomes a relatively large contributor to the loading of surface waters with N and P (EEA, 2005b).

N P

80 60

4.6â•‡ Assessment of factors crucial for effective nitrogen emission abatement

40 20 0 BE

DE

DK

NL

NO

SE

CH

Country Figure 4.7 Percentage reductions in anthropogenic discharges/losses of nitrogen and phosphorous to ‘eutrophication problem areas’ around the Eastern North Atlantic (source of data:Â€OSPAR, 2008b).

N and P input into marine surface waters (see Section 4.4.2) has not yet been achieved. The 2008 OSPAR Eutrophication Assessment (OSPAR, 2008a) shows that eutrophication is still a problem in many coastal areas of the Greater North Sea. The 2008 Report on the Implementation of PARCOM Recommendations 88/2 and 89/4 (OSPAR, 2008b) concludes that Contracting Parties contributing to N and P inputs to eutrophication problem areas have mostly achieved the 50% reduction target for discharges and losses of phosphorus (P), but not for N (see Figure 4.7). Modelling studies suggest that nutrient input reductions beyond the 50% target will be needed in some areas to eliminate all eutrophication problems (OSPAR, 2008a). Agriculture is the biggest contributor to discharges and losses of N to eutrophication problem areas (OSPAR, 2008b). Combustion in power plants and traffic (including road traffic and increasing emissions from maritime shipping in the North Sea and the Atlantic) are the main contributors to airborne NOx inputs to the OSPAR maritime area (OSPAR, 2005), while agriculture is the main contributor to atmospheric deposition of reduced nitrogen (mainly NH3). Eutrophication by N and P is also a major problem in the Baltic Sea (HELCOM, 2005, 2009). Total loads entering the Baltic Sea (as riverine and direct point-source discharges) amounted to 891 Gg N and 51 Gg P in 1990, and it was agreed to decrease these inputs by 50% by 1995 (HELCOM, 2010). In the Baltic Sea Action Plan, the maximum allowable nutrient input targets were set at 41% of the 1990 load of P and approximately 68% of that of N. Both targets have not yet been achieved; by 2006 the reduction for P was 45% and for N only 30%. Eutrophication by N and P inputs is less of a problem in the Mediterranean than in the North Sea and the Baltic Sea. In fact, the Mediterranean is one of the most oligotrophic regional seas in the world (Karydis and Chatzichristofas, 2003). Eutrophication is limited to coastal zones, especially in the western and northern half of the Mediterranean. However, N and P inputs to the Mediterranean marine environment have increased steadily over the past 20 years (UNEP, 2009). Summarizing, EU policy to treat municipal and industrial waste waters have been effective in decreasing N (and especially P) loadings to surface waters, though further improvements are needed (EC, 2009b). Diffuse N and P losses from agriculture

4.6.1â•‡ Differences between sectors So far, the most successful Nr emissions abatement policies have been on (see Section 4.5 and Erisman et al. 2011, Chapter 2 this volume):Â€(i) reducing NOx emissions to air from power plants and stationary combustion sources through catalytic converters, (ii) reducing emissions of NOx from mobile combustion sources to air (catalytic converters for gasoline cars, combustion optimization and NOx destruction by Selective Catalytic Reduction (SCR) with urea for diesel cars), and (iii) reducing N (and especially P) discharges to surface waters from industrial sources and households through sewage treatment plants. Though less spectacular than the decreases in SO2 emissions to air (see Figure 4.5), emission reductions for NOx to air and for Nr from human sewage to surface waters are larger than the emission reductions achieved for NH3 and NO3− from agriculture. The question is therefore:Â€‘why are certain policies more effective than others?’. So far, there has been little cross-sector comparison on the effectiveness and efficiency of policy measures aimed at decreasing Nr emissions. The success of the emissions abatement policies for NOx from combustion and Nr from human sewage may be ascribed to one or a combination of the following factors: (i) use of economic instruments (subsidies and taxes) to facilitate the implementation of the policy, which results in a high degree of compliance; (ii) availability of relatively straightforward and effective technologies to reduce the emissions effectively with few major side-effects; (iii) the limited number of addressees who must take action to implement the measures; (iv) the scale of investments required and the degree to which these are shared; (v) the cost of the compliance measures are relatively small and/or can be transferred to others; and (vi) enforcement and control, leading to a high degree of compliance with the policy measures. Theory and practice suggest that economic instruments or a mix of economic and regulatory and persuasive instruments tend to be more effective for the implementation of policy than a single regulatory or persuasive instrument (Gunningham and Grabosky, 1998; OECD, 2007). Subsidies, premiums and taxes often provide a strong incentive to adopt the provisions. Compliance with the obligations of a policy requires that all relevant stakeholders are informed and have the necessary knowledge, tools and will to implement the provisions. Subsidies on cars with catalytic converters to decrease NOx emissions, and

75


EU financial support for building sewage treatment plants, are indeed effective instruments for implementation of these emissions abatement technologies (OECD, 2007). The larger the number of addressees (stakeholders) of the policy, the larger the transaction costs of the policy and the less resource the government can allocate to supporting individual addressees. While cars with catalytic converters are driven by numerous drivers, few of these drivers know about the details of converter operation, as these are implemented by the car industry, which encompasses only few stakeholders. Similarly, while all humans in Europe produce Nr-containing wastes, few of them are involved in sewage collection and treatment. By contrast, all individual farmers in the EU (the percentage of farmers to the total work force ranges from 2% to 25% between the Member States) have to comply with the measures of the Nitrates Directive (especially those in Nitrate Vulnerable Zones) and other EU directives relevant to agriculture (see Section 4.4). The scale of investments in hardware and software needed to comply with policy obligations may differ greatly. Collection and treatment of urban sewage waters requires huge investments, but is done for a multitude of arguments, of which Nr emission abatement is only one, and the costs of the investments are transferred to and shared by numerous tax payers. Catalytic converters do not require much investment by the car industry (relative to other investments), although research costs may be significant. By contrast, building low-NH3-emission housing systems, manure storage systems and manure application techniques can require relatively large investments by individual farmers, though the Rural Development Programme may provide funds for subsidizing infrastructural modernizations (see Section 4.4.3). In the case of high-investment activities, such as new animal housing systems, much of the cost may be associated with other requirements, such as new animal welfare standards. For other techniques, such as low emission manure application, additional costs may be largely offset by saving more nitrogen in the system, thereby reducing fertilizer requirements (Webb et al., 2010). Compliance with the Nitrates Directive requires in principle relatively little investment, apart from the obligation of sufficient manure storage. However, the application limit of 170 kg N per ha per year can be a serious constraint to intensive livestock farms; they may have to export animal manure elsewhere (with or without prior processing) or will have to decrease livestock density. The costs of the catalytic converters or sewage treatment plants are all transferred to consumers (or tax payers), and therefore can be implemented easily by the car industry and communities, respectively. By contrast, farmers represent in many cases small businesses which have themselves to bear the cost of the measures for abating NH3 emissions and N leaching; they can less easily pass on costs to those further down the food production chain. For example, in a globalizing market for agricultural products, farmers in the EU may lose competitive power relative to farmers with less stringent environmental policies, unless other safeguards are put in place (such as the Rural Development Programme). There are nevertheless precedents for requiring investment in agriculture to meet policy requirements, such as animal welfare legislation. Such

76

environmental and welfare requirements come with associated costs which must, in the end, be born by governments and/ or consumers, or will have to be covered by increased income through up-scaling (larger farms). Summarizing, the relatively variable and slow implementation of environmental EU policy and measures in agriculture to decrease Nr emissions may be ascribed to: (1) ongoing incentives to maintain agricultural production levels and the limited ability of farmers to transfer the costs of environmental protection to consumers; (2) huge differences in farming systems and environmental conditions in the EU-27 and the complexities that arise when making the requirements of existing EU Directives farm-specific; (3) delays by Member States to implement measures in agriculture, fuelled by strong farm lobby groups, due to the perceived costs to farmers and the perceived low effectiveness; (4) delays by Member States to introduce effective control mechanisms to monitor compliance by all farmers, due to the difficulties in setting up such control systems as well as the perceived cost to a Member State; (5) failure by farmers to implement measures, due to within system constraints, perceived costs and the time needed for learning; and (6) the possibility for, and fear of, antagonisms between measures, due to lack of integration of measures aimed at abating NO3 leaching and measures aimed at abating NH3 and N2O emissions. Table 4.7 summarizes the results of a qualitative assessment of factors influencing the abatement of Nr emissions from different sectors. Various factors are different for agriculture compared to combustion and urban wastes, although it is unclear how much each of these contributes to differences in implementation of, and compliance with, the policies. Evidently, further studies are needed.

4.6.2â•‡ Differences between regions and EU Member States There are differences in the ways EU Member States and their regional governments implement environmental policies. These may relate to differences in the political need and political will, but also to differences in culture, environmental conditions, economic developments, institutional organization and in the availability of competent policy officers at regional and local levels. Such differences may change over time, for example, as a result of elections and changes in the political orientation of governments. Developments of civic society and pressure groups may also exert influence on the compliance to environmental policy (see Section 4.2). For example, farmers’ lobby groups were strong in delaying the implementation of the Nitrates Directive in the Netherlands during the 1990s, while green lobby groups greatly contributed to increasing the political pressure by the European

Oene Oenema Table 4.7â•‡ Qualitative assessment of factors that affect the implementation of EU policies to decrease Nr emissions:Â€NOx emissions from combustion, NH3 emissions and NO3 leaching from agriculture, and Ntot discharges from urban wastes

Factors

Combustion NOx to air

Policy instruments

Agriculture

Urban wastes

NH3 to air

NO3 to waters

Ntot to waters

Mixed

Regulation

Regulation

Mixed

Number of stakeholders

Few

Many

Many

Few

Technology level

Advanced

Modest

Modest

High

Level of standardization in production

High

Low

Low

High

Number of techniques involved

Few

Many

Many

Few

Development costs

High

High

High

High

Implementation costs

Modest

Modest for animal feeding and manure application; high for animal housings and manure storages

Low for optimizing fertilizer applications; high for adjusting farming systems

High

Who bears costs?

Manufacturers, but transferred effectively to consumers

Farmer

Farmer + public sector (RDP)

Water companies, but effectively transferred to consumers

Management activities & knowledge involved

Essentially no activities required by car drivers

Many activities, requires both proper techniques and information and knowledge

Many activities, requires information and knowledge

Many activities, requires both proper techniques and information and knowledge

Influence of climate & soil conditions

Absent

Large

Large

Negligible

Potential side-effects (apart from costs)

Increased N2O and NH3 emissions

Increased N2O emissions and energy use; fertilizer savings

Yield loss; fertilizer saving; increased / decreased NH3 emissions

Increased N2O emissions and energy use

Commission on the Netherland’s government to fully implement the Nitrates Directive (Bavel etÂ€ al., 2004). Within the context of the Nitrates Directive, changes in legislation are often under pressure of infringement procedures launched by the European Commission, indicating that enforcement of legislation is a key point. Scandinavian countries seem to have made most effort to comply with environmental policy. The effects of air pollution were already felt in the Scandinavian lakes and forests in the 1960s and 1970s, because these were highly sensitive to acidification and eutrophication. Though the origin of the air pollution largely came from outside Scandinavia, societal awareness of the effects led to the organization of the 1972 United Nations Stockholm Conference and to the foundation of CLRTAP (as discussed in Sections 4.1 and 4.3). These impacts also contributed to political will in Scandinavia to protect the environment from their own pollution sources. Western Europe has a high density of industrial and agricultural activities, with high emission densities. It has stakes in both continuation of economic activities and protection of the environment, and hence in the need to decrease the emission densities of economic activities. Southern Europe, in many locations, has a lower emission density than Western

Europe and an environment less sensitive to acidification than Scandinavia. Also, economic development and water harvesting are a societal priority in southern Europe. Finally, the 12 new Member States in central Europe had centralized political and economic systems until the early 1990s, with relatively low political priority for protecting the environment. These countries are now catching up following their accession to the EU in 2004 or 2007.

4.7â•‡ Conclusions • Environmental policy is a relatively new subject that emerged in the 1970s and 1980s. International agreements have given a strong impetus to the establishment of policy measures related to Nr emissions. The theoretical and empirical bases of policy measures related to Nr emissions are still small. • The toolbox for environmental policy instruments comprises regulatory instruments, economic instruments and communicative/voluntary instruments. Initially, there was a strong focus and emphasis on regulatory instruments; now there is increasing evidence that each environmental policy must have a specific mix of instruments, depending

77


•

•

•

•

•

78

on the capability, ability and willingness of the addressee to implement the environmental policy effectively and efficiently. Policy measures aimed at decreasing Nr emissions in the EU are effects-based or target-based, i.e. the policy measures aim to prevent well-defined human health effects or ecological effects or aim to meet specific threshold/target/ limit values. The policy measures aimed at decreasing Nr emissions in the EU have been implemented through Directives, which have to be addressed by all Member States through national legislation, and to a lesser extent Regulations, which have to implemented directly by all Member States. There is a large number of Directives, many of which have been revised following review and evaluation. There is also an increasing trend towards clustering specific Directives within Framework Directives. The Common Agricultural Policy (CAP) of the EU has a large influence on EU agriculture and indirectly also on Nr use and Nr emissions. Through a series of reforms of the CAP, there is increasing integration of agricultural, environmental and rural development objectives in agriculture, but the number of Directives and Regulations remains large. The EU Directives aimed at decreasing Nr emissions from the various sources have been developed and implemented while our understanding of the functioning of N in the biosphere, atmosphere and hydrosphere are still limited and evolving. Policies have been developed initially for single Nr compounds (NO3−, NH3, N2O and NOx), for single sectors (households, industries, traffic, crop production, animal production), for single environmental compartments (air, water, nature, humans), and for various specific impacts (e.g. human health, food security, climate change, eutrophication, acidification, biodiversity loss); this is partly because of our limited understanding of the complex N cycle, and partly because of the departmentalization of governments, These multicompound, multi-sector, multi-receptor, multi-impact approaches have contributed to a ‘wealth’ of policies, with some having interactive effects (both synergistic and antagonistic). As a result, there is an increasing quest for integrating environmental policy measures. Most successful Nr emission abatement policy measures, in terms of abatement of Nr emissions, have been on (i) reducing NOx emissions to air from power plants, stationary combustion sources and transport through catalytic converters, and (ii) reducing N (and especially P) to surface waters from industrial sources and households through sewage treatment plants. The success of these emission abatement policy measures has been ascribed to the availability of relatively straightforward technologies to reduce emissions, the limited number of addressees, the use of mixes of instruments and the level of governmental enforcement and control. However, there is not much literature on the comparison between

•

•

•

•

Directives or between sectors of the effectiveness and efficiencies of the various Directives related to Nr emissions abatement. Less successful, so far, have been policies on reducing Nr emissions from agriculture. In principle, the technologies and measures to reduce these emissions are available, but there are various reasons to explain why these have not been adopted and/or have not been effective. One of these reasons is the diversity and complexity of the farming systems involved and the complex, diffuse Nr pathways, which have resulted in many different regulatory obligations, but which are not equally effective for all farms. Further studies are needed to find out the optimal mix of packages of measures and incentives to decrease the diffuse Nr losses to air, soil and water. Based in part on the successful reduction of SO2 and NOx emissions from the energy, industry and transport sectors through technological measures, there is some belief that technology will reduce all unwanted emissions from all sectors. However, management and (changes in) economic activities may be equally important factors. So far, Scandinavian countries have done most on the implementation of environmental measures for nitrogen, perhaps because they felt the effects of air pollution on surface waters and forests most intensively. Current EU Directives on agriculture consider the threats from NO3 leaching, NH3 emissions (and N2O emissions) separately. However, when not combined with an integrated approach to N management, the policy measures may have the risk of antagonistic effects.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729. The authors are also especially grateful to Andrea Weiss for providing information on OSPAR, to Mark Sutton for many critical comments and suggestions and to two anonymous reviewers for helpful comments.

Supplementary materials Supplementary materials (as referenced in the chapter) areÂ� available online through both Cambridge University Press:Â€ www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Baker, S., Kousis, M., Richardson, D. and Young, S. (eds.) (1997). The Politics of Sustainable Development. Routledge, London. Baumol, W. J. and Oates, W. E. (1988). The Theory of Environmental Policy. Cambridge University Press, Cambridge. Bavel, M. van, Frouws, J. and Driessen, P. (2004). The Netherlands and the Nitrates Directive. Ostrich or strategist? Wageningen University, Wageningen.

Oene Oenema Blandford D. and Hill, B. (eds.) (2006). Policy Reform and Adjustment in the Agricultural Sectors of Developed Countries. CAB International, Wallingford, UK. Bleeker, A., Sutton, M. A., Acherman, B. et al. (2009). Linking ammonia emission trends to measured concentrations and deposition of reduced nitrogen at different scales. In:Â€Atmospheric Ammonia, eds. M. Sutton, S. Reis and S. M. H. Baker, Springer, New York, pp. 123–180. Braudel, F. (1979). Civilization and Capitalism 15th–18th Century, Volume II. The Wheels of Commerce. William Collins, London. Bressers, H. T. A. and Huitema, D. (2001). What the doctor should know; politicians are special patients. The impact of the policy making process on the design of economic instruments. In:Â€Market-Based Instruments for Environmental Management: Politics and Institutions, ed. M. S. Andersen and R.-U. Sprenger. Edward Elgar, Cheltenham, UK. Brink, C., van Grinsven, H., Jacobsen, B. H. et al. (2011). Costs and benefits of nitrogen in the environment. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Bull, K., Hoft, R. and Sutton, M. (2011). Co-ordinating European nitrogen policies between directives and international conventions. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. CONSSO (2002). Progress Report, Fifth International Conference on the Protection of the North Sea 20–21 March 2002. Committee of North Sea Senior Officials (CONSSO). Bergen, Norway. http:// www.regjeringen.no/md/html/nsc/progressreport2002/Progress_ Report.pdf. COST 729 (2009). Nitrogen Deposition and Natura 2000:Â€ScienceÂ€and Practice in Determining Environmental Impacts. Summary for Policy Makers. http://cost729.ceh.ac.uk/webfm_send/18 (accessed 4 January 2009). Diamond, J. (1997). Guns, Germs, and Steel:Â€The Fates of Human Societies. W.W. Norton, New York. Driessen, P. and Leroy, P. (2007). Environmental Policy. Analysis and Perspectives. Coutinho, Bussum, NL. EC (2007). Report from the Commission to the Council and the European Parliament. On implementation of Council Directive 91/676/EEC concerning the protection of waters against pollution caused by nitrates from agricultural sources for the period 2000– 2003 {SEC(2007)339} COM (2007) 120 final. EC (2009). Fifth Commission Summary on the Implementation of the Urban Waste Water Treatment Directive Brussels, Commission Staff Working Document SEC (2009) 1114 final, 23Â€pp. http://ec.europa.eu/environment/water/water-urbanwaste/ implementation/implementationreports_en.htm EC (2010a). Directive on Large Combustion Plants (LCP; 2001/80/ EC) http://ec.europa.eu/environment/air/pollutants/stationary/ lcp.htm EC (2010b). Directive on Integrated Pollution Prevention and Control (IPPC; 1996/61/EC) http://ec.europa.eu/environment/air/ pollutants/stationary/ippc/index.htm EC (2010c). Directive on Waste Incineration (WID; 2000/76/EC) http:// ec.europa.eu/environment/air/pollutants/stationary/wid.htm EC (2010d). Directive on Emission from Ignition Engines in Heavy-duty Vehicles (HDV; 2005/55/EC) http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/directive2005–55-ec_en.htm EC (2010e). Directive on National Emission Ceilings (NEC, 2001/81/ EC) http://ec.europa.eu/environment/air/pollutants/ceilings.htm

EC (2010f). Framework Directive on Ambient Air (96/62/EC / 2008/50/EC) http://ec.europa.eu/environment/air/quality/ legislation/existing_leg.htm EC (2010g). Existing legislation related to air quality in the EU. http:// ec.europa.eu/environment/air/legis.htm EC (2010h). Water Framework Directive (WFD; 2000/60/EC) http://ec.europa.eu/environment/water/water-framework/ index_en.html EC (2010i). Bathing Water Directive (BWD; 2006/7/EC) http:// ec.europa.eu/environment/water/water-bathing/index_en.html EC (2010j). Urban Waste-water Treatment Directive (91/271/EEC) http://ec.europa.eu/environment/water/water-urbanwaste/ index_en.html EC (2010k). Environmental Impact Assessment Directive (85/337/ EEC) http://ec.europa.eu/environment/eia/full-legal-text/85337. htm EC (2010l). Council Directive 91/676/EEC concerning the protection of waters against pollution caused by nitrates from agricultural sources. http://ec.europa.eu/environment/water/water-nitrates/ index_en.html EC (2010m). Drinking Water Directive (98/83/EC) http://ec.europa. eu/environment/water/water-drink/index_en.html EC (2010n). Groundwater Directive (2006/118/EC) http://ec.europa. eu/environment/water/water-framework/groundwater.html EC (2010o). Water in EU legislation. http://ec.europa.eu/ environment/water/index_en.htm EC (2010p). A Marine Strategy Directive to save Europe’s seas and oceans (2008/56/EC) http://ec.europa.eu/environment/water/ marine/index_en.htm EC (2010q). Agriculture; meeting the needs of farmers and consumers. http://europa.eu/pol/agr/index_en.htm EC (2010r). Animal Welfare legislative references. http://ec.europa.eu/ food/animal/welfare/references_en.htm EC (2010s). CAP Health Check. http://ec.europa.eu/agriculture/ healthcheck/index_en.htm EC (2010t). Rural Development legislations. http://ec.europa.eu/ agriculture/rur/leg/index_en.htm EC (2010u). EU Energy Policy. http://ec.europa.eu/agriculture/ bioenergy/index_en.htm EC (2010v). Birds Directive (79/409/EEC) http://ec.europa.eu/ environment/nature/legislation/birdsdirective/index_en.htm EC (2010w). Habitats Directive (92/43/EEC) http://ec.europa.eu/ environment/nature/legislation/habitatsdirective/index_en.htm EC (2010x). Report from the Commission to the Council and the European Parliament. On implementation of Council Directive 91/676/EEC concerning the protection of waters against pollution caused by nitrates from agricultural sources based on Member State reports for the period 2004–2007 COM(2010)47 final. http://ec.europa.eu/environment/water/water-nitrates/pdf/ com_2010_47.pdf EC (2010y). Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. http://europa.eu/ legislation_summaries/energy/renewable_energy/en0009_en.htm EEA (1995). Europe’s Environment:Â€the Dobris Assessment. European Environment Agency, Copenhagen. EEA (2005a). The European Environment, State and Outlook 2005. European Environment Agency, Copenhagen.

79

Nitrogen in current European policies EEA (2005b). Source Apportionment of Nitrogen and Phosphorus Inputs into the Aquatic Environment. European Environmental Agency, Copenhagen. EEA (2007). EN09 Emissions (CO2, SO2 and NOx) from public electricity and heat productionÂ€– explanatory indicators. (http:// ims.eionet.europa.eu/Sectors_and_activities/energy/indicators/ EN09%2C2007.04) EEA (2008). Annual European Community LRTAP Convention Emission Inventory Report 1990–2006, Submission to EMEP through the Executive Secretary of the UNECE. European Environmental Agency, Copenhagen. EMEP (2007). Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe in 2005. Norwegian Meteorological Institute, Oslo. Erisman, J. W., van Grinsven, H., Grizzetti, B. et al. (2011). The European nitrogen problem in a global perspective. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.Â€W. Erisman et al. Cambridge University Press. Finlayson-Pitts, B. J. and Pitts, J. N. (2000). Chemistry of the Upper and Lower Atmosphere:Â€Theory, Experiments and Applications. Academic Press, London. Gunningham, N. and Grabosky, P. (1998). SMART Regulation: Designing Environmental Policy. Clarendon Press, Oxford. Hajer, M. (1995). The Politics of Environmental Discourse: Ecological Modernization and the Policy Process. Oxford University Press. Hatfield, J. L. and Follett, R. F. (eds.) (2008). Nitrogen in the Environment:Â€Sources, Problems and Management, 2nd Edition. Elsevier, Amsterdam. HELCOM (2005). Nutrient Pollution to the Baltic Sea in 2000. Baltic Sea Environment Proceedings No. 100. Helsinki Commission, Baltic Marine Environment Protection Commission, Helsinki. http://www.helcom.fi/publications/bsep/en_GB/ bseplist/ HELCOM (2009). Eutrophication in the Baltic Sea, an integrated thematic assessment of the effects of nutrient enrichment in the Baltic Sea region. Baltic Sea Environment Proceedings No. 115. Helsinki Commission, Baltic Marine Environment Protection Commission, Helsinki, 152 pp. http://www.helcom.fi/publications/ bsep/en_GB/bseplist/ HELCOM (2010). Helsinki Commission/Baltic Marine. Environment Protection Commission. http://www.helcom.fi/ (accessed 15–02–2010). Hertel, O., Reis, S., Ambelas Skjøth, C. et al. (2011). Nitrogen processes in the atmosphere. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Horvath L. and Sutton M. A. (1998). Long term record of ammonia and ammonium concentrations at K-puszta, Hungary. Atmospheric Environment, 32, 339–344. ICP Modelling and Mapping (2004). Manual on Methodologies and Criteria for Modelling and Mapping Critical Loads and Levels and Air Pollution Effects, Risks and Trends. (Available at:Â€http:// icpmapping.org/cms/zeigeBereich/5/manual-und-downloads. html. Accessed 4 January 2009) Karydis, M. and Chatzichristofas, F. (2003). Coastal eutrophication in the Mediterranean marine environment:Â€a short review. In: Proceedings 8th International Conference on Environmental Science and Technology, Lemnos Island, Greece, 8–10 September 2003, pp.Â€426–430. Meester, G., Oskam, A. and Silvis, H. (eds.) (2005). EU Policy for Agriculture, Food and Green, From Policy to Practice. Wageningen Academic Publishers, Wageningen (in Dutch).

80

Mikkelsen, S. A., Iversen, T. M., Jacobsen, B. H. and Kjær, S. S. (2010). Denmark–European Union:Â€reducing nutrient losses from intensive livestock operations, in:Â€Livestock in a Changing Landscape. Volume 2. Experiences and Regional Perspectives, eds. P.Â€Gerber, H. Mooney, J. Dijkman, S. Tarawali and C. de Haan. Island Press, Washington, pp 140–153. MNP (2007). Efficacy of the Manure Policy in the Past (1998–2005), Present (2006–2007) and Future (2008–2015), Report 500124001, Environmental Assessment Agency MNP, Bilthoven, The Netherlands (in Dutch). Mourad, D. S. J., van der Perk, M. and Piirimae, K. (2006). Changes in nutrient emissions, fluxes and retention in a North-Eastern European lowland drainage basin. Environmental Monitoring and Assessment, 120, 415–448. NEGTAP (National Expert Group on Transboundary Air Pollution) (2001). Transboundary Air Pollution:Â€Acidification, Eutrophication and Ground-Level Ozone in the UK. NEGTAP, London. OECD (1991). Environmental Indicators: A Preliminary set. OECD, Paris. OECD (2007). Instrument Mixes for Environmental Policy. OECD, Paris. Oenema, O., Oudendag, D. A., Witzke, H. P. et al. (2007). Integrated Measures in Agriculture to reduce Ammonia Emissions. Alterra, Wageningen, The Netherlands. Oenema, O., Witzke, H. P., Klimont, Z., Lesschen, J. P. and Velthof, G.Â€L. (2009). Integrated assessment of promising measures to decrease nitrogen losses from agriculture in EU-27. Agriculture, EcosystemsÂ€and Environment, 133, 280–288. OSPAR (2005). 2005 Assessment of Data Collected Under the OSPAR Comprehensive Study on Riverine Inputs and Direct Discharges for the Period 1990–2002. Publication No.Â€2005/233. OSPAR Commission. OSPAR (2008a). Second OSPAR Integrated Report on the Eutrophication Status of the OSPAR Maritime Area, OSPAR Publication No. 372/2008. OSPAR Commission. OSPAR (2008b). Nutrients in the OSPAR Convention area: Assessment of the Implementation of PARCOM Recommendations 88/2 and 89/4. OSPAR publication No. 344/2008, OSPAR Commission. OSPAR (2010). Protecting and Conserving the North-East Atlantic and its Resources. http://www.ospar.org (accessed 15 February 2010). Patterson, M. (2001). Understanding Global Environmental Politics. Domination, Accumulation, Resistance. Palgrave, Basingstoke, UK. Ritson, C. and Harvey, D. (eds.) (1997). The Common Agricultural Policy. CAB International, Wallingford, UK. Romstad, E., Simonsen, J. and Vatn, A. (1997). Controlling Mineral Emissions in European Agriculture: Economics, Policies and the Environment. CAB International, Wallingford, UK. Sliggers, J. and Kakebeeke, W. (2004). Clearing the Air: 25 Years of the Convention on Long-range Transboundary Air Pollution. United Nations, New York. Smith, R. J., Glenn, G. A., Parkinson, R. and Richards, J. P. (2007). Evaluating the implementation of the Nitrates Directive in Denmark and England using an actor-orientated approach. European Environment, 17, 124–144. Stälnacke, P., Vandsemb, S. M., Vassiljev, A., Grimvall, A. and Jolankai, G. (2004). Changes in nutrient levels in some Eastern European rivers in response to large-scale changes in agricultural production. Water Science and Technology, 49, 29–36. Sutton, M. A., Erisman, J. W. and Oenema. O. (2007). Strategies for controlling nitrogen emissions from agriculture:Â€regulatory, voluntary and economic approaches. In:Â€Fertilizer

Oene Oenema Best Management Practices: General Principles, Strategies for their Adoption and Voluntary Initiatives vs Regulations. International Fertilizer Industry Association, Paris, France, pp.Â€245–259. Tinbergen, J. (1952). On the Theory of Economic Policy. NorthHolland, Amsterdam. UNECE (1999). Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution to Abate Acidification, Eutrophication and Ground-Level Ozone. http://www.unece.org/env/lrtap/ multi_h1.htm UNECE (2010). Convention on Long-Range Transboundary Air Pollution. http://www.unece.org/env/lrtap

UNEP (1972). Report of the United Nations Conference of the Human Environment. United Nations Environment Programme, Stockholm, 1972. UNEP (2009). State of the Environment and Development in the Mediterranean, United Nations Environment Programme/ Mediterranean Action Plan (UNEP/MAP- Plan Bleu), Athens, Greece. Webb, J., Pain, B., Bittman, S. and Morgan, J. (2010). The impacts of manure application methods on emissions of ammonia, nitrous oxide and on crop response:Â€a review. Agriculture, Ecosystems and Environment, 137, 39–46.

81

Chapter

5

Chapter

The challenge to integrate nitrogen science and policies:Â€the European Nitrogen Assessment approach Lead author: Mark A. Sutton Contributing authors: Clare M. Howard, Jan Willem Erisman, William J. Bealey, Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti

Executive summary Nature of the problem • Anthropogenic releases of reactive nitrogen (Nr) can disturb natural systems and affect human health and welfare in many different ways. Scientific and policy views of the nitrogen cycle have typically addressed these problems from separate perspectives, looking in each case at only part of the overall issue. • Given the multi-faceted nature of the nitrogen cycle, it is a major challenge to develop a more-integrated understanding of how different areas of nitrogen science and policies fit together.

Approaches • Observations from the first part of the European Nitrogen Assessment (ENA Part I) are summarized, considering the distinctive character of Nr in Europe, the benefits and threats, and the current policies. Approaches to developing the following parts of the Assessment are discussed with an emphasis on how to draw out the key issues.

Key findings • Recognizing the multi-pollutant, multi-phase complexity of the nitrogen cycle, it is concluded that it is essential to focus on a limited set of priority issues to allow effective communication between nitrogen scientists and policy makers. • A pathway is developed for prioritization of the key environmental concerns of excess Nr. Starting with around twenty environmental effects, the list is reduced down, first to nine main concerns, and then to five key societal threats. • The five key threats of excess Nr in Europe are identified as:Â€Water quality, Air quality, Greenhouse gas balance, Ecosystems and biodiversity, and Soil quality. These headingsÂ€– which are easily remembered by the acronym ‘WAGES’Â€– provide a basis for summarizing societal concern about excess Nr in later chapters of the Assessment. • The selection of five key threats represents a conscious balancing of complexity and simplification. The division also lends itself to developing communication models, as illustrated by its analogy to the classical cosmology of Empedocles and Aristotle.

Major challenges • Ongoing efforts must focus on linking scientific communities between nitrogen form (N2O, NH3, NOx, NO3−, etc.), environmental compartment (air, water, plant, soil, etc.), and spatial scale (farm, landscape, region, continent, etc.), tracing the Nr cascade from the main source sectors, especially agriculture and combustion. • For policy makers, major infrastructural constraints limit the connection of different threats and media (climate, air pollution, water pollution, etc.) through spatial scales (local, regional, global policies). Political positions often require a deliberate separation between issues, making it harder to negotiate joined-up approaches. • Ongoing efforts are needed to simplify further the nitrogen story. Multiple communication models should be used, matching the needs of different audiences.

The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

82

Mark A. Sutton

Recommendations • Ongoing research at the scale of nitrogen processes (ENA Part II) is essential as this provides the foundation for developing mechanistic understanding. • Additional efforts are needed to quantify nitrogen flows through different spatial scales (ENA Part III). A new emphasis on rural and urban landscapes provides an important link between the plot and regional scales, while nitrogen budgeting should be further developed as an integration tool identifying the main Nr pathways. • The key societal threats of excess Nr (ENA Part IV) are not equally well quantified. While understanding of the water and air threat is rather mature, the threat of Nr on greenhouse gas balance is still at the stage of early quantification. Further efforts are needed to quantify fully the impacts of all the threats. • The first multi-pollutant approaches to Nr cost–benefit analysis, future scenarios and integrated Nr management (ENA Part V) should be further developed to underpin the development of integrated abatement strategies. • Efforts at communicating Nr to policy makers should highlight how integrated Nr management can help meet multiple environmental targets. For the general public, efforts should emphasize the responsibility we all have to manage our own nitrogen footprint.

5.1â•‡ Introduction As introduced in ChapterÂ€1, the European Nitrogen Assessment (ENA) has been structured in five parts, each of which deals with a different stage of the ENA process (Sutton etÂ€al., 2011, Chapter 1 this volume). The previous chapters in Part I provide the background upon which we here develop the vision and approach for the rest of the Assessment. It is thus clear from the continental context explained in ChapterÂ€2, that, compared with many other parts of the world, Europe is characterized by an excess of reactive nitrogen (Nr) leading to many environmental problems (Erisman etÂ€al., 2011, Chapter 2 this volume). This is true equally of the Nr which is deliberately produced for food production, as for the Nr that is produced inadvertently through high temperature combustion processes. This concern about excess Nr provides a central theme running through the Assessment, which may be contrasted with some other parts of the world (such as parts of Africa and South America) where there is still a societal shortage of Nr. Based on the analysis of ChapterÂ€3 (Jensen etÂ€al., 2011, this volume), it is equally evident that there are huge societal benefits associated with fixing atmospheric N2 into Nr. The focus of Jensen etÂ€ al. is on adding up the benefits of deliberate N2 fixation in fertilizer production and biological nitrogen fixation, which are together essential for healthy functioning of the European economy. If the most obvious benefit is the supply of Nr-containing fertilizers to agriculture, it is equally obvious that modern society would not be possible without the concomitant production of Nr explosives (essential for all mining activity in the world), many plastics (such as nylon) and a huge diversity of other Nr-containing chemicals. By contrast, we must exclude formation of Nr due to the combustion of fossil fuels from the list of benefits, since the Nr produced is immediately lost to the environment, while control efforts (such as catalytic converters) focus on its chemical denitrification back to N2, rather than capture and use of the Nr produced. In economic terms, it is obvious that a better management of nitrogen has substantial benefits. The aim to improve nitrogen use efficiency means that less Nr fertilizers are needed, potentially saving costs and the energy-use associated with their production (Jensen etÂ€ al., 2011). However, it is equally clear that Nr as a commodity is currently rather cheap, its price

being largely set by the fuel costs associated with its production in the Haber–Bosch process. As a result, no policies appear to have been needed in recent decades to ensure adequate supply of Nr in Europe. Against this, it is clear from the analysis of Oenema etÂ€al. (2011a, ChapterÂ€4 this volume) that there has been a plethora of different government policies related to the release of different Nr forms into the environment. As Nr is emitted, its conversion into many different chemical forms and accumulation of stocks in each of air, land and water, makes it obvious that several environmental effects should be expected. One of the clear conclusions of Oenema etÂ€al. (2011a) is that policies designed to address these environmental effects have in most cases been conducted separately according to Nr form and environmental media in which the pollution form occurs (e.g., freshwater pollution, urban air pollution, marine pollution). There is thus a lack of a joined-up view of how Nr is being lost into the environment and how holistic approaches may be implemented to manage the nitrogen cycle. Based on ChapterÂ€4, it appears that one of the most problematic sectors for managing Nr threats on the environment is agriculture. Oenema etÂ€ al. (2011a) explain how policies to reduce atmospheric emissions from large combustion sources, such as power stations, and from vehicles have been relatively successful. By focusing on a few key actors (e.g., large power generating companies, vehicle manufacturers), with a clear ability to transfer any costs to consumers, it has been possible to achieve a high take-up of technical measures. The challenges in those sectors have been the offsetting of reductions, achieved through technical mitigation measures, by increased consumption patterns (e.g., increased vehicle miles per person), as well as by some chemical trade-offs (e.g., catalytic converters increasing N2O and NH3 emissions). By comparison, the challenges to control Nr loss from agriculture have been rather harder. In this sector, Nr emissions are generated by many independent actors (individual farmers), operating a high diversity of processes, in a rather open system (i.e., open to air, soil and water). The sector, especially with the smaller operators, is frequently characterized by conservatism, reflected in a caution to take up new technical approaches, particularly if there is uncertainty on how any perceived costs might be transferred to consumers.

83

The challenge to integrate nitrogen science and policies

in piecemeal fashion, and that is far too complex to be easily explained to the general public. In this chapter, we reflect on these challenges to develop the basis for subsequent Parts II–V of the European Nitrogen Assessment. We outline a vision for integration both across nitrogen science domains and across policy domains, as well as the links between the two. In particular, we focus on how to communicate the issues in a way that balances complexity with an easily understandable list of priority concerns.

5.2â•‡ Integrating nitrogen science 5.2.1â•‡ Linking nitrogen forms, processes and scales

FigureÂ€5.1 Temporal changes in annual reactive nitrogen (Nr) inputs to Europe (EU-27) and the emissions to the environment. Top panel:Â€crop biological nitrogen fixation, N in imported feed and food, land application of mineral fertilizers (mainly from N fixation by the Haber–Bosch process) and N fixed in industry and traffic combustion processes. Bottom panel:Â€leaching and run-off to water, especially as nitrate, emission to air as ammonia (NH3) mainly from agriculture, emission to air as nitrogen oxides (NOx) mainly from industry and traffic. Inferred from Leip etÂ€al. (2011, Chapter 16 this volume) and Bouwman etÂ€al. (2011).

Such differences between sectors are clearly reflected in the trends of Nr pollution to air and water as summarized by Oenema etÂ€ al. (2011a), with only modest reductions in the agricultural emissions. If these effects of recent policies over the last two decades are put into historical context, it is clear that European use and emissions of Nr from all sources are still greatly in excess of natural rates. The massive increase in fertilizer and manure use in Europe since 1860 is shown in ChapterÂ€2 by Erisman etÂ€ al. (2011). This can be compared with the amounts Nr inputs through N fixation (crops and NOx formation) mineral fertilizer application feed and food imports, and the consequent emissions of Nr through leaching and run-off, and as ammonia (NH3) and nitrogen oxides (NOx) emission to air (FigureÂ€5.1). Overall, there has been a major up-Â�regulating of the European nitrogen cycle by human activities. Recent policies over the last two decades have made some progress in reducing both inputs and emissions, but represent only the first step in developing optimized strategies for Nr management. The message that emerges is that there is a huge diversity of Nr pollutant forms, including NOx, nitrous oxide (N2O), NH3, NO3–, leading to many secondary pollutants (including many organic nitrogen forms in water and in air), and an even longer list of environmental effects. The problem of Nr in the environment provides a degree of complexity that few scientists are able cover in full, that policy makers have so far tackled

84

The complexity and many facets of the nitrogen cycle are clearly reflected in the structure and relationships of the scientific communities that have developed to study it. This is illustrated in FigureÂ€5.2, for the research area of atmosphere– biosphere exchange of Nr and interacting compounds. In this area, in recent decades, the degree of specialization has become such that the individual compounds have become the focus for whole research communities, and with the main focus of integration being between components listed in the vertical in FigureÂ€5.1. Examples are recent major collaborations, such as the GRAMINAE project, which focused on the exchange of ammonia, also considering interactions with nitric acid and inorganic aerosol dynamics (Sutton etÂ€ al., 2001, 2009a), and the GREENGRASS project, which investigated the integration between N2O, CH4 and CO2 exchange processes with European grasslands (Soussana etÂ€al., 2007). At a similar scale, the NOFRETETE project addressed the dynamics of N2O and NO fluxes with European forest soils (Pilegaard etÂ€ al., 2006). Given the different measurement tools relevant for each of these chemical species and diverse set of biogeochemical proÂ� cesses (e.g., plant processes, soil processes, atmospheric chemistry interactions), it therefore becomes a major challenge to integrate our understanding of all of the different components listed in FigureÂ€ 5.2. Such integration efforts have become an important focus in recent years, such as within the NitroEurope and ACCENT research communities (Sutton etÂ€al., 2007; Fowler etÂ€al., 2009), which have started to develop the links across the whole suite of pollutant forms shown in FigureÂ€5.2. It is evident, however, that these efforts are part of addressing an even larger challenge to integrate communities, not just between Nr forms, but between environmental media and spatial contexts. At present, there is still only limited connection between many Nr research communities, such as between interests in stratospheric chemistry, biosphere–atmosphere exchange, freshwater and marine pollution. In aquatic research for instance, the community dealing with nitrate groundÂ�Â�water contamination is rather distinct from that concerned with nitrogen river fluxes causing coastal zone eutrophication. A similar tension is repeated across many domains within the nitrogen cycle, as the tendency to specialization of recent decades is reflected by the need to assess the wider perspective. From the perspective of environmental management, it is clear that much more linkage between science communities

Mark A. Sutton

Net GHG

Nitric Acid (HNO3)

Ozone (O3)

Carbon dioxide (CO2)

Wet deposition + – NH4 and NO3

Aerosol NH4NO3

Nitrogen dioxide (NO2)

Methane (CH4)

Particle Organic Nitrogen (PON)

Ammonia (NH3)

Nitric oxide (NO)

Nitrous oxide (N2O)

Volatile Organic Nitrogen (VON)

across the N cycle is needed (Galloway etÂ€ al., 2008; Sutton etÂ€al., 2009b). This is essential to deal with issues of trade off ’s between different Nr forms and linked biogeochemical cycles. Such management issues apply over different temporal and spatial scales. For example, ways need to be understood of how Nr migrates through individual natural or agricultural ecosystems, and subsequently moves from one ecosystem type to another, either through human transfers (such as agricultural products) or by dispersion through water courses and the atmosphere. With the centre of gravity of recent European nitrogen research tending toward individual Nr forms at a series of different scales, the question that arises is how fast and how far should a greater level of integration be developed. Here it must be recognized that the process of scientific integration is a slow one, and that an effort such as the European Nitrogen Assessment represents only the first steps on the path toward integration. It is with this in mind that the following group of chapters (Part II) retain a current focus on the point and process scale reflecting the expertise of the current scientific communities (Butterbach-Bahl etÂ€al., 2011a; Durand etÂ€al., 2011; Voß etÂ€al., 2011; Hertel etÂ€al., 2011, Chapters 6–9 this volume). The ‘integration target’ for these chapters is to ensure that all the Nr forms and their major interactions are considered, focusing on what we know about Nr processing in each of these systems, as well as the major knowledge gaps. While the complexity of these systems will always make it hard to take a fully integrated approach at the process level, such a focus is nevertheless essential as the foundation for understanding the component mechanisms underlying the system. In developing the ENA though a series of workshops as explained in ChapterÂ€1 (Sutton etÂ€al., 2011, this volume), the degree of integration achieved in Part II of the Assessment, provided the basis to develop the subsequent stages. Hence Part III of the Assessment develops the next steps of integration, in examining the nitrogen cycle of Europe through successive spatial scales.

FigureÂ€5.2 Specialization in Nr and greenhouse gas fluxes over recent decades has led to separation of research communities on biosphere– atmosphere exchange according to chemical compound. The vertical arrows indicate first stages of emerging integration, while the vertical blocks indicate a stronger separation between research communities.

5.2.2â•‡ Integrating through the nitrogen cascade A useful concept in upscaling nitrogen processes that has informed recent research is the ‘nitrogen cascade’ (Galloway etÂ€al., 2003). In the classical concept of the nitrogen cycle, the emphasis is on recycling between atmospheric di-nitrogen (N2) and the many Nr forms. In a natural system, the nitrogen cycle is seen as being in balance, as nitrogen fixation from N2 to Nr is ultimately matched by denitrification returning Nr to N2. Under anthropogenic modification, the amounts of Nr in circulation are increased in multiple directions. The concept of the nitrogen cascade emphasizes a rather different view of the same system under human influence. Substantial energy is needed to fix nitrogen from N2 into Nr forms, be it fuels needed for industrial production of ammonia or the photosynthetic energy needed for biological nitrogen fixation. The fixation process therefore raises nitrogen from a low to a high energy state, providing a starting point for the subsequent cascade. The important point of the cascade is that, as this energy is gradually dissipated, the Nr converts between a multiplicity of different forms, with different environmental consequences at every stage. Each molecule of Nr can therefore be expected to be involved in several environmental effects before it is finally denitrified back to unreactive N2. The complexity of the nitrogen cascade means that any graphical description is necessarily a simplification. An extremely simplified version is shown in ChapterÂ€ 1 (Sutton etÂ€al., 2011), illustrating the cascade of fertilizer Nr produced by the Haber–Bosch process. This can be compared with FigureÂ€5.3, which provides a more completeÂ€– but still very simplifiedÂ€– summary of the cascade, accounting for each of the main anthropogenic influences on nitrogen fixation. For simplicity, the only numbers shown in FigureÂ€5.3 are the amounts of N2 fixed to Nr by human activities, comparing global estimates (Galloway etÂ€al., 2008; Erisman etÂ€al., 2011) with European estimates, as developed in ChapterÂ€16 for the EU-27 by Leip etÂ€al. (2011, this volume). At both Global and European

85


FigureÂ€5.3 Simplified view of the nitrogen cascade, highlighting the major anthropogenic sources of reactive nitrogen (Nr) from atmospheric di-nitrogen (N2), the main pollutant forms of Nr (orange boxes) and nine main environmental concerns (boxes outlined with blue). Estimates of N fixation for the world (Tg /yr for 2005, in black; Galloway etÂ€al., 2008) are compared with estimates for Europe (Tg /yr for 2000, in blue italic; Leip etÂ€al., 2011, ChapterÂ€16 this volume). Energy is needed to fix N2 to Nr, which is gradually dissipated through the cascade with eventual denitrification back to N2. Blue arrows represent intended anthropogenic Nr flows; all the other arrows are unintended flows.

levels, industrial production is by far the largest source of new Nr. Industrial production accounts for 63% and 70%, of total anthropogenic nitrogen fixation at Global and European scales, respectively, of which use of the Haber–Bosch process to make fertilizers is the largest component. Compared with the global average, Europe has a higher contribution of Nr production from combustion processes (power generation, transport, other industry) (13%, 21%, respectively), reflecting the fact that Europe is an industrialized region with a high intensity of energy use and motorized transport. By contrast, compared with the global average, Europe has a smaller contribution from crop biological nitrogen fixation (24%, 8%, respectively), owing to the high use of mineral fertilizers in Europe. It is clear from the magnitude of these numbers that understanding the pathways and fate of industrially produced nitrogen fertilizers must be a priority for Europe. Secondly, the Nr formed from high temperature combustion is of particular interest as it represents a completely unintended production of Nr. As can be seen from FigureÂ€ 5.3, each of the forms of Nr produced can be inter-converted, having several environmental effects, before eventually being denitrified to N2 at the end of the cascade. For the purpose of this summarized view, nine main environmental concerns are shown in FigureÂ€5.3, as further discussed in Section 5.4.

86

It must be acknowledged that, to date, the nitrogen cascade represents a purely conceptual framework. So far, models have yet to be constructed that fully trace the pathway of anthropogenically fixed Nr though all forms and stages, showing how many times on average each Nr atom contributes to different environmental effects. The cascade concept nevertheless provides a stimulus for integrating Nr research at different spatial scales, as well as for identifying control points (Erisman etÂ€al., 2001; Galloway etÂ€al., 2008; see Section 5.5). Based on the need to track Nr through the different environmental compartments and impacts in the cascade, Part III of the Assessment emphasizes the building up of understanding between the different scales. Each of the chapters in PartÂ€III necessarily addresses the lateral flows of Nr, including, as relevant, those by direct human transfers (movement of fertilizer, manure, feed and food), by atmospheric transport and by water flow in catchments. FigureÂ€5.4 illustrates the conceptual upscaling of Part III, putting into context the following parts of the assessment. The components scale up from the farm level (Jarvis etÂ€al., 2011, ChapterÂ€10 this volume) to complete multimedia Nr integration across Europe (de Vries etÂ€al., 2011; Leip etÂ€al., 2011, Chapters 15 and 16 this volume). Traditionally, much effort has been put into addressing regional scale transfers in watersheds and in the atmosphere,

Mark A. Sutton

Upscaling & integration

Part V

European Nitrogen policies & future challenges Water quality

Air quality

Greenhouse balance

Ecosystems & biodiversity

Soil quality

Part IV

FigureÂ€5.4 Main structure of the following parts of the European Nitrogen Assessment, highlighting the upscaling elements addressed in Part III and the five key societal threats of excess nitrogen addressed in Part IV.

Integrating nitrogen fluxes at the European scale Geographic variation in terrestrial nitrogen budgets across Europe Nitrogen flows from European watersheds to coastal marine waters Nitrogen flows and fate in rural landscapes

Atmospheric transport & deposition of nitrogen in Europe Nitrogen flows and fate in urban landscapes

Part III

Nitrogen flows in farming systems across Europe

Processes & mechanisms

Nitrogen processing in the biosphere

and these scales are addressed by the chapters of Billen etÂ€al. (2011, ChapterÂ€ 13 this volume) and Simpson etÂ€ al. (2011, ChapterÂ€14 this volume). Much is known at these scales, which represent relatively mature, though still challenging, areas of research. As complex areas, there remain many unknowns, including tracing the fate of nitrogen through compartments and the chemical exchanges between inorganic and the many organic Nr forms. Nevertheless, by comparison, full assessment of the nitrogen cascade at the landscape scale has been much less studied compared with regional Nr transfers in either air or water. A full view of the Nr cascade in landscapes integrates spatially explicit transfers between source and sink landscape elements (e.g., farms, fields, roads, forests, mountains) and between each of the air, land and water compartments. The landscape scale is particularly important as it is the linking scale between the plot and region, being the scale in which many local decisions on Nr management take place. Two chapters address these new scales of integration, contrasting Nr flows in rural and urban landscapes (Cellier etÂ€al., 2011; SvirejevaHopkins etÂ€al., 2011, Chapters 11 and 12 this volume). If the integration of nitrogen science across Nr forms and spatial scales already represents a complex challenge, it cannot be forgotten that nitrogen also interacts with many other biogeochemical cycles. Given the complexity and resources, it is important to limit the scope of this assessment to focus on the nitrogen interactions, rather than assess all biogeochemical cycles simultaneously. In order to optimize the scientific effort, the approach taken here centres on the interactions between nitrogen forms, while not forgetting the main links with other element cycles where these occur. These main links tend to be different according to the environmental compartment being considered. In atmospheric chemistry, the main biogeochemical links are with sulphur chemistry and organic carbon chemistry (e.g., in aerosol and photochemical oxidant transformation processes, Hertel etÂ€ al., 2011; Simpson etÂ€al., 2011). In terrestrial ecosystems, key relationships exist

Part II

between nitrogen supply, turnover of organic matter and net carbon storage, both through influences on primary production and decomposition (Butterbach-Bahl etÂ€ al., 2011a) and with phosphorus through the use of manures in agriculture (Jarvis etÂ€al., 2011). Finally, in freshwater and marine systems, the relationships between nitrogen and phosphorus are especially important, as well as interactions with silica (Durand etÂ€al., 2011; Billen etÂ€al., 2011). The Assessment thus addresses the interactions with key other element cycles according to the priorities in the different environmental compartments.

5.3â•‡ The challenge to integrate nitrogen policies Similar to the scientific perspective, Oenema etÂ€ al. (2011a, ChapterÂ€4 this volume) explain how current environmental policies in Europe have taken a rather fragmented approach to the nitrogen cycle. The reasons for this appear to relate both to this historical separation between the supporting scientific communities, and the fact that such policies have been driven by perceived environmental problems in different contexts. This is equally reflective of the way in which policy portfolios are typically separated in government departments (e.g., between water and air, between urban and rural, between agriculture and nature, etc.). It can even be the case that political positions require a deliberate separation between issues, making it harder to negotiate joined-up approaches. An example here is the desire of some parties to the UN Framework Convention on Climate Change to ensure that climate related policy issues are not addressed in other conventions. This makes it challenging to address multi-effect interactions in international conventions, which is compounded by the fact that the relevant multi-lateral environmental agreements are made up of different memberships (e.g., UN, UN-ECE, EU-27, etc.). In response to the present position, there is a strong case to be considered for developing more joined-up, integrated

87


approaches to managing nitrogen in the environment (see also Section 5.5 in relation to Part V of the Assessment). Aside from the procedural difficulties, a key challenge that emerges is to find the optimum level of integration. For example a balance should be identified between the simplicity of single-issue approaches, versus the inefficiencies that occur as a result of not considering the major interactions. Put in another way, it must be recognized that joined-up approaches take longer to develop and the advantages (in optimization, improved delivery, synergies, avoidance of trade-offs, etc.) must be shown to outweigh the risk of additional complexity (Oenema etÂ€ al., 2011b; Bull etÂ€al., 2011, Chapters 23 and 25 this volume). In particular, the benefits of integration must be shown to be achievable in order to address the potential concern that the complexity of integration becomes an excuse for inaction. However, an integrated approach can also lead to simpler policies if integration is used to weigh and prioritize the many N-issues. In this context, it is extremely important to identify the priority issues that should be integrated. If a short list of key issues can be established, this would provide a foundation on which to identify an achievable level of integration, at least between the key issues. The short-list can then inform the discussion on to what extent specific multi-sector, multi-issue policies on nitrogen are needed, or to what extent existing policies should be further developed to be nitrogen aware, what may be called ‘nitrogen proofing’.

5.4â•‡ Distilling complexity to integrate and communicate nitrogen It is not surprising that something as multi-faceted as the nitrogen cascade should be associated with an extremely large number of environmental concerns. As explained above, establishing a short-list of priority issues is therefore important as a basis for informing the integration of environmental and other

policies. At the same time, such a short-list is needed in order to communicate the nitrogen problem more effectively to the general public. The Nitrogen in Europe (NinE) networking programme (funded by the European Science Foundation), in its aims to link together for the first time the environmental problems related to Nr, established a short list of priorities. The identification of the priorities was carried out in two stages. Starting with a list of around 20 environmental issues related to nitrogen, NinE agreed a list of 9 major environmental concerns, as a basis for communication of its efforts to link issues across the nitrogen cycle. The outcome is clearly expressed in the NinE logo, including the mnemonic acronym ‘ACT AS GROUP’ (FigureÂ€5.5). While this first listing of nine concerns was useful to illustrate the challenge faced by the NinE network, it was recognized it needed further simplification to allow effective communication to a wider audience. In the second stage of simplification, the full list of around 20 environmental concerns was ranked by a group of European experts as a contribution of the activities NinE, NitroEurope and COST Action 729 to an expert workshop under the Convention on Long-range Transboundary Air Pollution (Saltsjöbaden-3, Gothenburg, April 2007), with the short-list being tested through further consultation with experts and other stakeholders during the COST Action 729 Workshop on integrated assessment modelling for nitrogen (Laxenburg, November 2007). The full lists of nitrogen effects, as developed in this process (Erisman etÂ€al., 2007) are shown in TablesÂ€5.1 to 5.3. In estimating scores of the ‘relevance and link to nitrogen’, the experts used a scale of 1 (highest relevance) to 5 (unimportant). This prioritization combined consideration of the relevance of the issue to society and the extent to which nitrogen contributed to that issue.

Stratospheric chemistry and ozone

88

Acidification

Greenhouse gas

of soils & waters

& global warming

Terrestrial

Ozone

eutrophication & biodiversity

vegetation & health

Coastal

Urban

& marine eutrophication

air quality & health

Aquatic

Particles

eutrophication & water quality

health, visibility & global dimming

FigureÂ€5.5 Graphical representation of the main nitrogen concerns addressed by the Nitrogen in Europe (NinE) networking programme of the European Science Foundation. The nine environmental concerns together provide the mnemonic ‘ACT AS GROUP’.

Mark A. Sutton TableÂ€5.1â•‡ Summary of the direct effects of excess nitrogen on humans in relation to currently used indicators, the current existence of limit values (legally binding and/or broadly established for scientific assessment) in Europe, and the link to the nitrogen cascade. The relevance and link to N provides a prioritization estimated by an expert group for international action to mitigate the effects of excess nitrogen (based on Erisman etÂ€al., 2007)

Direct effects on humans

Indicators

Limit?

Link to N cascade

Relevance and link to N

– ozone

O3 conc. values including SOMO35

Yes

NOx emission

3

– other photochemical oxidants

Organic NO3, PAN

No

NOx emissions

5

– fine particulate aerosol

PM10, PM2.5

Yes

NH3, NOx emissions

1

– direct toxicity of NO2

NO2

Yes

NOx emissions

2

Nitrate contamination of drinking water

NO3 conc (aq.)

Yes

NO3 leaching

2

Increase allergenic pollen production, and several parasitic and infectious human diseases

—

No

N fertilizer and N deposition

5

Blooms of toxic algae and decreased swimmability of in-shore water bodies

Chlorophyll A NO3 (aq.)

No

Run-off, N deposition

1

Respiratory disease in people caused by exposure to high concentrations of:

Relevance and link to nitrogen qualitatively incorporates the societal priority of the issue and the N contribution to that issue:Â€(1) highest relevance, (2) high relevance, (3) significant relevance, (4) some relevance, (5) unimportant. Indicators:Â€SOMO35:Â€sum of ozone concentration above 35 parts per billion in air; PAN, peroxyacetyl nitrate; PM10, PM2.5:Â€particulate matter in air having a median diameter larger than 10 μm and 2.5 μm, respectively. TableÂ€5.2â•‡ Summary of the effects of excess nitrogen on ecosystems in relation to currently used indicators, the current existence of limit values (legally binding and/or broadly established for scientific assessment) in Europe and the link to the nitrogen cascade. The relevance and link to N provides a prioritization for future international action to mitigate the effects of excess nitrogen (based on Erisman etÂ€al., 2007)

Direct effects on ecosystems

Indicators

Limit?

Link to N cascade

Ozone damage to crops, forests, and natural ecosystems

O3 flux, AOT40


Yes

NOx emission

2

Acidification effects on forests, soils, ground waters, and aquatic ecosystems

Critical loads

Yes

N deposition

2

Eutrophication of freshwaters, lakes (incl. Biodiversity)

BOD, NO3 (aq) Critical loads

Yes No


3

Eutrophication of coastal ecosystems inducing hypoxia (incl. Biodiversity)

BOD, NO3 (aq) Critical loads

Yes No


1

Nitrogen saturation of soils (incl. effects on GHG balance)

Critical loads

Yes

N deposition

1

Biodiversity impacts on terrestrial ecosystems (incl. Pests and diseases)

Critical loads, critical level (NH3 in air)

Yes

N deposition

1

Relevance and link to nitrogen qualitatively incorporates the societal priority of the issue and the N contribution to that issue:Â€(1) highest relevance, (2) high relevance, (3) significant relevance, (4) some relevance, (5) unimportant. Indicators:Â€AOT40:Â€accumulated ozone concentration above a threshold of 40 parts per billion in air; BOD:Â€biological oxygen demand in water.

TablesÂ€ 5.1–5.3, show that the following issues were given the highest scores: (1) Respiratory disease caused by fine particulate matter in the atmosphere. (2) Blooms of toxic algae and decreased swimmability of in-shore water bodies. (3) Eutrophication of coastal ecosystems inducing hypoxia (incl. their biodiversity).

(4) Nitrogen saturation of soils (incl. effects on GHG balance). (5) Biodiversity impacts on terrestrial ecosystems (including pests and diseases). (6) Global climate warming induced by excess nitrogen. (7) Regional climate cooling induced by aerosol.

89

The challenge to integrate nitrogen science and policies TableÂ€5.3â•‡ Summary of the effects of excess N on other societal values in relation to currently used indicators, the current existence of limit values (legally binding and/or broadly established for scientific assessment) in Europe, and the link to the nitrogen cascade. The relevance and link to N provide a prioritization for future international action to mitigate the effects of excess nitrogen (based on Erisman etÂ€al., 2007)

Effects on other societal values

Indicators

Limit?

Link to N cascade


Odour problems associated with animal agriculture

NH3 concentration

No

NH3 emission

5 (in Europe)

Effects on monuments and engineering materials

Precipitation acidity, O3, PM10, PM2.5 concentrations

Yes

NOx, NH3

3

Regional hazes that decrease visibility at scenic vistas and airports

PM2.5

No

NOx, NH3

4 (for Europe)

Depletion of stratospheric ozone

NOx, N2O concentrations

No

NOx, N2O

3

Global climate warming induced by excess nitrogen

N2O, CH4, CO2 concentrations

No

N2O (direct & indirect sources), CH4, CO2

1

Regional climate cooling induced by aerosol

PM2.5 concentration

No

NOx, NH3

1

Relevance and link to nitrogen qualitatively incorporates the societal priority of the issue and the N contribution to that issue:Â€(1) highest relevance, (2) high relevance, (3) significant relevance, (4) some relevance, (5) unimportant. Indicators:Â€PM10, PM2.5:Â€particulate matter in air having a median diameter larger than 10 μm and 2.5 μm, respectively.

Based on these and recognizing some overlap, a group of five key societal threats of nitrogen was identified: • Air quality (including respiratory disease concerns), especially as affected by NOx, O3 and particulate matter. • Water quality (including ecosystems and human health concerns). • Greenhouse balance (including effects on trace gases and atmospheric aerosol). • Ecosystems and biodiversity (including pests and diseases), especially as affected by atmospheric Nr deposition. • Soils quality (including effects on nitrogen saturation and acidification). In regard of the last threat, feedback from stakeholders at the Laxenburg workshop, strongly argued for the inclusion of soils, given the need to include soil acidification and to consider soils as an integrator of different pressures. Subsequent reflection of these five key societal threats of excess nitrogen has shown several interesting features. Firstly, with these headings, many of the other environmental concerns become automatically incorporated into the overall framework. For example, the air quality threat of Nr includes both ozone and particulate matter, while the soils threat includes acidification and alteration of soil organic matter storage. Secondly, this subdivision into five threats has allowed analysis of the negative effects caused by nitrogen together with some potential benefits. This is the case for the effect of nitrogen on greenhouse balance, where warming effects due to Nr (e.g., nitrous oxide, ozone) are at least partly offset by several cooling effects (e.g., aerosol, carbon sequestration). In addition to highlighting the key issues for policy makers, the key societal threats also lend themselves to developing wider communication approaches. The selection of five issues highlights the complexity of the nitrogen problem (i.e. it is multi-issue), while focusing on a list which is sufficiently short to remember. The threats can, for example, be considered as the WAGES of excess nitrogen, being a mnemonic for the

90

five key threats to:Â€Water, Air, Greenhouse balance, Ecosystems andÂ€Soils. Somewhat more surprising is the observation that this list of five threats also falls neatly into another ancient communication framework. The Greek philosopher Empedocles is famous for having presented the fundamental components of matter as four ‘elements’:Â€water, air, fire and earth (Wright, 1995), to which Aristotle subsequently added aether as the quintessence, or fifth element (de Caelo I.2, Guthrie, 1986; Wilderg, 1988). FigureÂ€5.6 illustrates the analogy between these elements of the Greek cosmos and the five key threats of excess nitrogen. In this model of nitrogen in the environmental macrocosm, the allocation of water, air and soil is straightforward, while greenhouse balance is linked to fire, with ecosystems and biodiversity, placed as the quintessence. FigureÂ€5.6 also shows how it is possible to apply the system to highlight the key Nr forms for each threat. Such an assignment is naturally open to much debate, and must be considered loosely. However, any such controversy should not hinder the use of this model to communicate nitrogen issues, just as the longstanding debate in identifying Empedocles’ elements (Wright, 1995) did not Â�preventÂ€– or even encouragedÂ€– acceptance of his approach. Part IV of the ENA considers each of these five societal threats of excess nitrogen in turn, water quality (Grizzetti etÂ€al., 2011, ChapterÂ€17), air quality (Moldanová etÂ€al., 2011, ChapterÂ€18), greenhouse balance (Butterbach-Bahl etÂ€al., 2011b, ChapterÂ€19), ecosystems and biodiversity (Dise etÂ€al., 2011, ChapterÂ€20) and soil quality (Velthof etÂ€ al., 2011, ChapterÂ€ 21). A deliberately sectoral approach is taken in each of these chapters, providing the basis to show the key issues that need to be linked in developing more integrated approaches. As far as possible, trends in Nr threats over time and across Europe are explored to show how the problem has arisen and to highlight the outlook in the light of existing policies. Although these five chapters were initially developed in parallel using a common framework, it quickly became apparent that different approaches were needed according to each

Mark A. Sutton

A

FIRE

B

Hot

Dry

AETHER QUINTESSENCE

AIR

GREENHOUSE BALANCE

AIR QUALITY

EARTH

Moist

Dr y

Hot

Moist

Cold

SOIL QUALITY

ECOSYSTEMS & BIODIVERSITY

Cold WATER QUALITY

WATER

C

Nitrous oxide (N2O) & Nitrogen-GHG interactions Atmospheric chemistry

Nitrogen oxides (NOx) particles (PM2.5) & ozone (O3) in air

Storage of organic matter

Ammonia & organic N in Ecosystems

Soil Organic Nitrogen (SON)

Aqueous Transformations

Human health Aqueous nitrate (NO3–) & other dissolved N

FigureÂ€5.6 Analogy of nitrogen concerns in the environment to the Empedoclean–Aristotelian framework of the macrocosm:Â€(A) visualization of Empedocles’ elements, indicating their common properties, together with aether, the Aristotelian ‘quintessence’; (B) the five key societal threats of reactive nitrogen visualized in macrocosmic framework; (C) key chemical forms and issues typifying each of the key societal threats. In each model, the diagonals represent shared properties of the adjacent elements.

issue. Thus some of the key threats like water quality and air quality represent mature areas, where data on spatial patterns and trends are well established. By contrast, the threat of Nr on greenhouse gas balance represents a much less well developed research and policy area. In this case, the chapter provides a first examination, drawing together the evidence needed to guide assessment of the net effects and the future policy development. The use of these five threats was also tested as a communication tool by the NinE programme in collaboration with the BBC ‘Green Room’ (NinE, 2008; Sutton, 2008). In exploring the idea of the ‘NitroNet’, the web of interlinked challenges related to nitrogen, members of the public visiting the web-site were asked in the ‘NitroNet Poll’ how they would rate the different societal threats, on a scale of 1 to 5, from unimportant to important. The results, summarized in FigureÂ€5.7, show

that, while there were some significant differences between mean scores, members of the public had a wide range of views over which problems were the priority. The clear message was that all the issues need to be addressed. Feedback also showed that while some respondents accepted the challenge to prioritize between issues, others rejected the whole idea, considering that it is impossible to set such priorities. Whatever view one takes, the very debate on the validity of such a comparison again serves its main purpose of encouraging people to start thinking about how to manage the multiple threats of excess nitrogen. An important communication tool is the use of visual images. For example, a further simplification of the Greek cosmological analogy (FigureÂ€ 5.6) provides the basis to summarize visually the five key societal threats of excess nitrogen (FigureÂ€5.8). Of the five threats, it is notable that the two that scored highest in

91


completely eradicated, being replaced by a thick algal slime. Of course there are many stages between the conditions shown by these two photographs, but the comparison powerfully demonstrates the way in which reactive nitrogen supply is having major effects on terrestrial biodiversity. The second (lower) image shows the results of nutrient enrichment in coastal marine waters. Again, excess reactive nitrogen encourages algal growth, forming harmful ‘blooms’ which reduce oxygen availability and threaten fish and other species populations. Such algal blooms can have negative effects on bathing water quality, reducing water visibility and causing high levels of foam, as shown here, resulting from released gelatinous substances.

a

Water quality

bc

Air quality Greenhouse gas balance

abd

Ecosystems & biodiversity

ad

Soil quality

bc 0

1

2

3

4

5

Score

FigureÂ€5.7 Outcome of scores from members of the public regarding the relative priorities between the five key societal threats of excess nitrogen, from the NitroNet Poll of NinE (2008) in cooperation with the BBC ‘Green Room’. Members of the public were asked to allocate a score for each issue ranging from 1 (low priority) to 5 (high priority). The error bars are standard errors (n = 175), with the different letters (a to d) indicating significant Â�differences (P = 0.05) based on paired t-tests (a shared letter indicates no significant difference between bars).

FigureÂ€5.8 Summary of the five key societal threats of excess reactive nitrogen, with the visualization based on the analogy presented in FigureÂ€5.6. PhotoÂ€sources: Shutterstock.com and garysmithphotography.co.uk.

the NitroNet PollÂ€– Water quality, and Ecosystems and biodiversityÂ€– are also highly amenable to visual images. To illustrate this, FigureÂ€5.9 shows two examples of the effects of excess reactive nitrogen in the environment. The first (top) shows some of the consequences of high levels of atmospheric ammonia deposition on the epiphyte flora of birch woodland in the United Kingdom. On the left, at a clean site, the birch trunk shows a rich diversity of lichens and bryophytes. On the right, at a site near an intensively managed livestock farm, the natural flora has been

92

5.5â•‡ Integrating European nitrogen policies and future challenges The analysis of key societal threats in Part IV of the Assessment provides the platform to develop more integrated science and policy approaches. With the main issues clearly identified, Part V of the Assessment brings these threats together and relates them to the benefits of Nr for food security and industrial production. As the same time, such approaches can be informed by the development of more integrated approaches, such as the establishment of comprehensive nitrogen budgets and maps for Europe (de Vries etÂ€al., 2011; Leip etÂ€al., 2011). Three chapters in Part V address different elements of integration followed by two chapters on how to communicate the outcomes with policy makers and the general public. It is evident that the simple comparison of the ‘NitroNet Poll’ (FigureÂ€5.7) represents only a first step, and it is a major challenge to develop more rigorous approaches to bring together the nitrogen issues and set priorities. In fact, the NitroNet Poll captures significant differences of opinion, as the respondents were answering based on a wide mix of perspectives and degrees of knowledge. In developing a more formal approach to compare the issues, economic methods may be used. In the subsequent assessment, ChapterÂ€22 applies economic approaches to assess the environmental costs and societal benefits of reactive nitrogen in Europe (Brink etÂ€ al., 2011, Chapter 22 this volume). Using willingness-to-pay approaches, the authors estimate societal damage costs as euro per kg Nr emission for each of Nr to water, NH3 to air, NOx to air and N2O to air. The quantified uncertainty bounds estimated by Brink etÂ€al. highlight the major challenges of such an approach. Nevertheless, they provide a foundation for discussion with policy makers, showing the substantial financial benefits of mitigating Nr emissions. The different elements of managing Nr are brought together by Oenema etÂ€al. (2011b, ChapterÂ€23 this volume). Based on the foregoing contributions, they analyze what it means to develop ‘integrated approaches’ to nitrogen management. They examine several dimensions of integration, linking scales, issues, stakeholders etc. Based on these reflections, they identify a package of key actions that together provide an integrated perspective for overall management of anthropogenic Nr emissions and their effects. Such a short list builds on the key actions previously identified by Galloway etÂ€al. (2008) and warrants further consideration as a foundation for developing future European

Mark A. Sutton FigureÂ€5.9 Visual illustrations of the effect of excess nitrogen on the natural environment. Top:Â€effect of atmospheric ammonia on epiphyte biodiversity in birch woodland in the UK:Â€left, clean conditions showing a rich array of lichens and bryophytes (photo:Â€Ian Leith); right, replacement of the natural epiphyte flora under high ammonia by a thick algal slime (photo:Â€Mark Sutton). Bottom:Â€nitrogen input into coastal seas in excess over silica, can cause severe algal blooms, in this case with Phaeocystis globosa, leading to a build up of gelatinous foam on a Dutch beach (photo: Gilles Billen).

policies. While some of the key actions are already being implemented in existing policies, it is equally clear that most of them require much more attention. In developing future perspectives, it is vital to have a clear idea on the trajectories of Nr emissions and effects that can be expected. This places an important role on scenario development, considering both future economic development and current plans for Nr mitigation. Such a first assessment for the major Nr emissions in Europe is brought together by Winiwarter etÂ€al. (2011, ChapterÂ€24 this volume). A major achievement is the combination of both

short- and medium-term scenarios, though further efforts will be needed to develop scenarios of integrated packages for Nr management, including the long-term (e.g., 2100). In developing the European Nitrogen Assessment, it has become clear how nitrogen issues cut across all global change threats. At the same time, the connected nature of the nitrogen cycle has clearly not been fully addressed by policy makers or recognized by the general public. A key task for the Assessment must therefore be to consider how better to communicate the nitrogen issues to these audiences.

93


Bull etÂ€al. (2011, ChapterÂ€25 this volume) address the issue of how to communicate the Nr challenge with policy makers, highlighting the possibilities for more effective coordination between multi-lateral environmental agreements. They demonstrate the complexity of the international landscape for nitrogen, involving agreements between many different sets of national parties (e.g., UN, UN-ECE, EU, other groupings). The key challenge they raise is to develop mechanisms that ensure joined up approaches to nitrogen management, linking all of the key societal threats. They assess the possibility and relative merits different options, ranging from coordination actions, ‘nitrogen proofing’ existing policies, to the establishment of an international convention on nitrogen. An intermediate option is to develop the basis for a multi-media protocol on nitrogen, drawing on the work of the existing international conventions. The analysis of Bull etÂ€al. (2011) reflects the tension to ensure streamlined approaches that avoid substantial additional burden on the current conventions, while maximizing the synergies. In particular, it remains a challenge to develop a framework that develops sufficient ‘gravity’ to ensure that the different Nr related problems are drawn together. Finally, Reay etÂ€al. (2011, ChapterÂ€26 this volume) address the issue of how to communicate the European nitrogen challenge to the general public. They highlight how insufficient recognition has been given to the different barriers to optimizing future human use of nitrogen and its environmental consequences. They assess these barriers and relate them to the key societal levers, drawing on experience from the societal and policy challenge to manage climate change. In particular, they highlight the role of societal choice both in raising awareness of the nitrogen challenge, and in making a significant contribution towards meeting mitigation targets. Patterns of societal consumptionÂ€– one of the seven key actions listed by Oenema etÂ€al. (2011b, ChapterÂ€23 this volume)Â€– are identified as a key focus relevant for the nitrogen cycle, especially related to diets, food choice and food waste. Based on the dominance of the food chain in the nitrogen cascade (FigureÂ€5.3), human food choices have major effects on the overall amounts of Nr processed and lost to the environment. Reay etÂ€ al. (2011) discuss how a ‘segmented strategy’ can be used to reach different stakeholders, including the use of proven communication tools, with involvement of the social sciences.

5.6â•‡ Conclusions A key emerging feature of the European Nitrogen Assesment (ENA) is the challenge to develop holistic approaches that integrate across science disciplines, across policy domains and build the links between the science and policy communities. Until now the multi-media, multi-impact nature of the nitrogen cycle has been much broader than the individual science communities, and several steps are needed to provide the basis to address the whole. This chapter reflects at the end of Part I of the ENA on how to develop subsequent parts of the Assessment. Based on the particular characteristics of the European nitrogen problem, on the major benefits of reactive nitrogen to the European

94

economy and on the many existing policies that address parts of the nitrogen cycle, it is evident that Europe has a long history and experience of actively managing its nitrogen cycle. At the same time, it is clear that the complexity of the system has hindered the development of a broad perspective that would aim to optimize European nitrogen management. The need for such a perspective is fully justified by the multiple, non-linear interacting impacts illustrated by the nitrogen cascade. In seeking to optimize future European nitrogen management, the first steps must be to agree on the scientific foundations. For this reason, the following parts of the ENA focus on our understanding of nitrogen processes (Part II) and how these can be upscaled from the farm to the European level (PartÂ€III). The next step must be to establish the basis for prioritization of the key issues, which is necessary to set the framework for judging the optimum level of integration. Given the complexity of integration, the components need to be kept as simple as possible to be understandable and implementable in policy. Without such an issue building, the complexity of nitrogen management risks becoming seen in policy circles as an excuse for inaction. Based on these reflections, this chapter reports a distillation of the many different nitrogen threats facing society. In a first cut, the Nitrogen in Europe (NinE) programme identified a network of nine main concerns of excess Nr. Such a grouping is well fitted to the scientific community, emphasizing the need to cooperate between media and disciplines. However, it was concluded that this listing remains too complicated to be the basis for developing integrated approaches with policy makers or to communicate the issues with society. The chapter therefore describes a second distillation, into five key societal threats of excess nitrogen. This listing of five key threats was derived from a prioritization and clustering of around twenty environmental concerns, with the number five reflecting a deliberate balancing between complexity and simplification. The five key societal threats:Â€ Water quality, Air quality, Greenhouse gas balance, Ecosystems and biodiversity, and Soil quality turn out to be well suited to developing communication models. In mnemonic form these threats can be seen as the ‘WAGES of excess nitrogen’, and be easily illustrated by analogy to each of the ‘elements’ of classical Greek philosophy. Part IV of the Assessment analyzes each of the five key threats, clearly highlighting the main reasons why society should be concerned about excess Nr in the European environment. The initial aim was to ensure that these five chapters were closely streamlined in approach, highlighting the magnitude, spatial distribution, temporal trends and current efforts to manage each issue. However, it quickly became clear that the knowledge base for the five threats is very different, with the result that the greenhouse gas and soils threat chapters focus much more on problem quantification, while the water, air, and ecosystems chapters are able to provide more detail on trends and patterns. Finally, the key societal threats provide the basis to inform the development of more holistic approaches to nitrogen

Mark A. Sutton

management in Part V of the Assessment. A first examination of the costs of Nr pollution on the European environment and the benefits of Nr mitigation is conducted, scenarios of future Nr use and pollution are brought together, and integrated approaches to Nr management are developed. In particular, a package of seven key actions is identified, which would together provide the basis for integrated management of the European Nr resource, minimizing the environmental threats. These messages provide the foundation for further communication and application by policy makers and by society at large. Here major challenges remain. In the context of policy development, it is evident that more holistic approaches are needed, but much more work is needed by policy makers to agree the optimum degree of policy integration, and the right framework within which it should be conducted. There are major opportunities for closer working between existing multi-lateral environmental agreements in Europe, such as the different conventions of the UN-ECE and across EU policy domains. Such action needs to develop sufficient ‘gravity’ to pull together the key nitrogen concerns, while being sufficiently streamlined in order to have a chance of making effective progress. The bottom line of Section V of the Assessment is the challenge to involve European citizens in recognizing and taking action on nitrogen. Major efforts still need to be devoted to simplifying the nitrogen story to make it understandable to the general public, developing the key messages. One of the key hooks identified is the importance of personal food choice to the whole nitrogen cascade. This illustrates the need for future efforts to quantify the impacts and mitigation potential, as well as to quantify the co-benefits of eating patterns that are both healthy for the individual and for the environment.

Acknowledgements The authors gratefully acknowledge funding support from the European Commission for the NitroEurope Integrated Project and the COST Action 729 programme, the European Science Foundation for the NinE programme and the UK Department for Environment Food and Rural Affairs for support of the UN-ECE Task Force on Reactive Nitrogen, together with underpinning support from the UK NERC Centre for Ecology and Hydrology. We are grateful to David Leaver for support in implementing the NitroNet Poll.


References Billen, G., Silvestre, M., Grizzetti, B. etÂ€al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:Â€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press.

Bouwman, A. F., Klein Goldewijk, K., van der Hoek, K. W. etÂ€al. (2011). Exploring global changes in nitrogen and phosphorus cycles in agriculture induced by livestock production for the period 1900–2050. Proceedings of the National Academy of Sciences of the USA (submitted). Brink, C., van Grinsven, H., Jacobsen, B. H. etÂ€al. (2011). Costs and benefits of nitrogen in the environment. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Bull, K., Hoft, R. and Sutton, M.A. (2011). Co-ordinating European nitrogen policies between directives and international conventions. In:Â€The European Nitrogen Assessment, ed. M.Â€A.Â€Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Butterbach-Bahl, K., Gundersen, P., Ambus, P. etÂ€al. (2011a). Nitrogen processes in terrestrial ecosystems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. etÂ€al. (2011b). Nitrogen as a threat to the European greenhouse balance. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Cellier, P., Durand, P., Hutchings, N. etÂ€al. (2011). Nitrogen flows and fate in rural landscapes. In:Â€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. de Vries, W., Leip, A., Reinds, G. J. etÂ€al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:Â€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Dise, N., Ashmore, M., Belyazid, S. etÂ€al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Durand, P., Breuer, L., Johnes, P. J. etÂ€al. (2011). Nitrogen processes in aquatic ecosystems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Erisman, J. W., De Vries, W., Kros, H. etÂ€al. (2001). An outlook for a national integrated nitrogen policy. Environmental Science and Policy, 4, 87–95. Erisman, J. W., Spranger, T., Sutton, M. A. etÂ€al. (2007). Working Group 5:Â€NitrogenÂ€– integrated environmental policies. In:Â€Air Pollution and its Relations to Climate Change and Sustainable DevelopmentÂ€– Linking Immediate Needs with Long-Term Challenges “Saltsjöbaden 3” (12–14 March 2007, Gothenburg, Sweden). http://asta.ivl.se/Workshops/Saltsjobaden3/Conclusions/ WG5.pdf (last accessed 15 September 2010). Erisman, J. W., van Grinsven, H., Grizzetti, B. etÂ€al. (2011). The European nitrogen problem in a global perspective. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Fowler, D., Pilegaard, K., Sutton, M. A. etÂ€al. (2009). Atmospheric composition change:Â€ecosystemsÂ€– atmosphere interactions. Atmospheric Environment, 43, 5193–5267. Galloway, J. N., Aber, J. D., Erisman, J. W. etÂ€al. (2003). The nitrogen cascade. BioScience, 53, 341–356. Galloway, J. N., Townsend, A. R., Erisman, J. W. etÂ€al. (2008). Transformation of the nitrogen cycle:Â€recent trends, questions and potential solutions. Science, 320, 889–892. Grizzetti, B., Bouraoui, F., Billen, G. etÂ€al. (2011). Nitrogen as a threat to European water quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press.

95

The challenge to integrate nitrogen science and policies Guthrie, W. K. C. (1986). Aristotle: On the Heavens, Loeb Classical Library, Harvard University Press,Â€Cambridge, MA. Hertel, O., Reis, S., Ambelas Skjøth, C. etÂ€al. (2011). Nitrogen turnover processes in the atmosphere. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Jarvis, S., Hutchings, N., van der Hoek, K., Brentrup, F. and Olesen, J. (2011). Nitrogen flows in farming systems across Europe. In:Â€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. etÂ€al. (2011). Benefits of nitrogen for food fibre and industrial production. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Leip, A., Achermann, B., Billen, G. etÂ€al. (2011). Integrating nitrogen fluxes at the European scale. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Moldanová, J., Grennfelt, P., Jonsson, Å. etÂ€al. (2011). Nitrogen as a threat to European air quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. NinE (2008). The NitroNet Poll. Nitrogen in Europe programme of the European Science Foundation. www.nine-esf.org/?q=nitronet_poll (last accessed 15 September 2010). Oenema, O., Bleeker, A., Braathen, N. A. etÂ€al. (2011a). Nitrogen in current European policies. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Oenema, O., Salomez, J., Branquinho, C. etÂ€al. (2011b). Integrated approaches to nitrogen management. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Pilegaard, K., Skiba U., Ambus A. etÂ€al. (2006). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 651–661. Reay, D. S., Howard, C.M., Bleeker, A. etÂ€al. (2011). Societal choice and communicating the European nitrogen challenge. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Simpson, D., Aas, W., Bartnicki, J. etÂ€al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:Â€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Soussana, J. F., Allard, V., Pilegaard, K. etÂ€al. (2007). Full accounting of the greenhouse gas (CO2, N2O, CH4) budget of nine European

96

grassland sites. Agriculture, Ecosystems and Environment, 121, 121–134. Sutton, M. (2008) Snared in a homemade ‘NitroNet’. The Green Room, BBC News, 8 July 2008, http://news.bbc.co.uk/ 1/hi/sci/tech/7496036.stm (last accessed 15 September 2010). Sutton, M. A., Milford, C., Nemitz, E. etÂ€al. (2001). Biosphere–atmosphere interactions of ammonia with grasslands:Â€experimental strategy and results from a new European initiative. Plant and Soil, 228,Â€131–145. Sutton, M. A., Nemitz, E., Erisman, J. W. etÂ€al. (2007). Challenges in quantifying biosphere–atmosphere exchange of nitrogen species. Environmental Pollution, 150, 125–139. Sutton, M. A., Nemitz, E., Milford, C. etÂ€al. (2009a). Dynamics of ammonia exchange with cut grassland:Â€synthesis of results and conclusions of the GRAMINAE Integrated Experiment. Biogeosciences (GRAMINAE Special Issue), 6, 2907–2934. Sutton, M. A., Oenema, O., Erisman, J. W. etÂ€al. (2009b) Managing the European Nitrogen Problem:Â€A Proposed Strategy for Integration of European Research on the Multiple Effects of Reactive Nitrogen. Centre for Ecology and Hydrology / Partnership for European Environmental Research, Edinburgh, UK. http://www.clrtap-tfrn. org/european-research-strategy (last accessed 15 September 2010). Sutton, M. A., Howard, C. M., Erisman J. W. etÂ€al. (2011). Assessing our nitrogen inheritance. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Svirejeva-Hopkins, A., Reis, S., Magid, J. etÂ€al. (2011). Nitrogen flows and fate in urban landscapes. In:Â€The European Nitrogen Assessment, eds. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Velthof, G. etÂ€al. (2011). Nitrogen as a threat to European soil quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Voß, M., Baker, A., Bange, H. W. etÂ€al. (2011). Nitrogen processes in coastal and marine systems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Wildberg, C. (1988). John Philoponus’ Criticism of Aristotle’s Theory of Aether, De Gruyter,Â€Berlin. Winiwarter, W., Hettelingh, J. P., Bouwman, L. etÂ€al. (2011). Future scenarios of nitrogen in Europe. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C.M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Wright, M. R. (1995). Empedocles:Â€The Extant Fragments. Gerald Duckworth, London.

Part

II

Nitrogen processing in the biosphere

Chapter

6

Nitrogen processes in terrestrial ecosystems Part II

Lead authors: Klaus Butterbach-Bahl and Per Gundersen Contributing authors: Per Ambus, Jürgen Augustin, Claus Beier, Pascal Boeckx, Michael Dannenmann, Benjamin Sanchez Gimeno, Andreas Ibrom, Ralf Kiese, Barbara Kitzler, Robert M. Rees, Keith A. Smith, Carly Stevens, Timo Vesala and Sophie Zechmeister-Boltenstern

Executive summary Nature of the problem • Nitrogen cycling in terrestrial ecosystems is complex and includes microbial processes such as mineralization, nitrification and denitrification, plant physiological processes (e.g. nitrogen uptake and assimilation) and physicochemical processes (leaching, volatilization). In order to understand the challenges nitrogen puts to the environment, a thorough understanding of all these processes is needed.

Approaches • This chapter provides an overview about processes relating to ecosystem nitrogen input and output and turnover. On the basis of examples and literature reviews, current knowledge on the effects of nitrogen on ecosystem functions is summarized, including plant and microbial processes, nitrate leaching and trace gas emissions.

Key findings/state of knowledge • Nitrogen cycling and nitrogen stocks in terrestrial ecosystems significantly differ between different ecosystem types (arable, grassland, shrubland, forests). • Nitrogen stocks of managed systems are increased by fertilization and N retention processes are negatively affected. • It is also obvious that nitrogen processes in natural and semi-natural ecosystems have already been affected by atmospheric Nr input. • Following perturbations of the N cycle, terrestrial ecosystems are increasingly losing N via nitrate leaching and gaseous losses (N2O, NO, N2 and in agricultural systems also NH3) to the environment.

Major uncertainties/challenges • Due to their complexity, ecosystem nitrogen stocks and nitrogen cycling processes are not well studied, as compared to those of carbon. However, strong ecosystem feedbacks to global changes have to be expected, especially with regard to nitrate leaching, C sequestration and emissions of the primary and secondary greenhouse gases N2O and NO.

Recommendations • In view of the still limited knowledge on nitrogen and carbon interactions at ecosystem and landscape scales and effects of global changes (climate, N deposition, landuse, land management) on C and N cycling, multi-disciplinary research needs to be initialized, encouraged and supported. Interdisciplinary and multi-scale studies should focus on simultaneous and comprehensive measurements of all major Nr fluxes at site and landscape scales including plant uptake/release of organic and inorganic N compounds as well as microbial Nr conversion. • Based on an in-depth understanding of nitrogen cycling processes, best management options need to be developed to minimize negative environmental impacts of global change on N cycling, C/N interactions and biosphere–atmosphere–hydrosphere exchange processes.


99

Nitrogen processes in terrestrial ecosystems

6.1â•‡ Introduction Nitrogen (N) is a key nutritional element for any life form on earth. Living organisms use N to form a number of complex organic compounds such as amino and nucleic acids, chitin and proteins. In unperturbed terrestrial systems N is a factor which limits net primary ecosystem production (Vitousek and Howarth, 1991). However, this situation has changed dramatically during the last decades. In most regions of Europe there is little limitation of biomass production by N, due to intensive use of fertilizers in agricultural systems and increasing N deposition to natural and semi-natural systems. The effects of increased N availability include changes in biosphere–atmosphere exchange (such as increased aerosol formation and emissions of the greenhouse gas N2O from terrestrial systems), eutrophication of terrestrial and aquatic systems (with consequences for species composition and richness, carbon sequestration, surface and drinking water quality), or acidification of soils and water bodies following the deposition of reactive nitrogen (Nr). Nitrogen cycling in terrestrial ecosystems and landscapes is mainly driven by microbiological and plant processes, with physico-chemical processes such as diffusion, emission, volatilization, leaching or erosion leading to displacement of N on site, regional and global scales (Galloway et al., 2003; Erisman et al., 2008). To better understand how N is affecting ecosystem functioning and to predict future ecosystem responses to increased N availability, it is necessary to have detailed knowledge of the processes involved in N cycling in terrestrial systems. This chapter provides an overview of current knowledge on N stocks in terrestrial ecosystems (Section 6.2), sources of N inputs (Section 6.3), N cycling at the ecosystem scale (Section 6.4), N loss pathways (Section 6.5) and N effects at the ecosystem scale (Section 6.6). The chapter has a particular focus on forest ecosystems, while agricultural systems are more thoroughly considered in Jarvis et al., 2011 (ChapterÂ€10 this volume).

6.2â•‡ Nitrogen in terrestrial ecosystems 6.2.1â•‡ Nitrogen as a key element in biogeochemistry Nitrogen is a key element for global biogeochemistry and its cycling is closely linked to the carbon cycle. Nitrogen availability often limits net primary production in agricultural as well as natural and semi-natural ecosystems (Vitousek and Howarth, 1991; De Vries et al., 2006). Nitrogen bound in organic compounds is an essential part of all proteins and enzymes and thus, N is driving the key metabolic processes involved in growth and energy transfer. Furthermore, N is a part of chloroÂ� phyll, the green pigment of the plant that is responsible for photosynthesis. A large fraction of N in primary producers is utilized directly in capturing energy in photosynthesis (Evans, 1989). Nitrogen has different oxidation states from −3 in NH3 to +5 in NO3− and a series of microbial processes such

100

as nitrification (autotrophic, heterotrophic), denitrification or anaerobic ammonium oxidation (anammox) have evolved to either gain energy from the oxidation process or to use oxidized N compounds as an alternative electron acceptor when growing anaerobically. In contrast to global C cycling, where the largest fluxes are associated with the net primary production of terrestrial and marine systems, the global biogeochemical cycle of N is dominated by microbial processes in soils, sediments and water bodies (Seitzinger et al., 2006). This is also a major difference from the global cycling of phosphorus, which becomes available to the biosphere mostly through mineral weathering. Depending on ecosystem type and land use, N cycling and N storage in soils and vegetation varies considerably. In agricultural systems, N cycling is dominated by N fertilization and crop removal, while in natural and semi-natural systems N cycling is largely affected by climatic, edaphic and landscape conditions and the sum of N inputs via N deposition and biological N fixation. Across differing climates, shrublands in Europe have been predominantly N-limited systems, i.e. plant production was suboptimal due to shortages in available N. In such systems N availability is mostly hampered by poor soil properties, e.g. high sand content and thus reduced ion exchange capacities and low organic matter content, which negatively affects retention of reactive nitrogen (Nr; here and in the following:Â€all organic N forms plus inorganic N forms except N2) in the system. Also the human use of these systems (e.g. grazing, fodder collection) over centuries has depleted the nutrient reservoirs. At the present time, shrublands and heathlands exposed to high N deposition are often showing indications of N saturation, such as changes in species composition or nitrate leaching (Schmidt et al., 2004). Wetlands are also mostly N-limited systems, due to accelerated losses of Nr via the denitrification pathway. Forests are naturally N-limited systems, at least for European climatic conditions, whereas tropical rainforests are often N-rich systems. This situation has changed markedly during the last decades due to atmospheric N deposition (De Vries et al., 2007). Forest foliage is more effective at receiving N deposition compared to other vegetation types, resulting in an increased N deposition. Signs of N saturation of forests have been widely reported and include accelerated growth and significant Nr losses via nitrate leaching and N trace gas emissions (Dise et al., 2009; Pilegaard et al., 2006).

6.2.2â•‡ Distribution of nitrogen stocks in the soil and plant system Based on a detailed database of soil properties, Batjes (1996) estimated global amounts of soil N to be 133–140 Pg N for the upper 100 cm of the soil profile. By comparison, about 10 Pg N is held in the plant biomass and about 2 Pg N in the microbial biomass (Davidson, 1994). This shows that on an ecosystem scale, soils are the main reservoir for N. We have compiled data for representative European terrestrial ecosystems on N pools and fluxes in representative

Klaus Butterbach-Bahl and Per Gundersen

Hyytiälä, Finland Pine Forest (Boreal)

N input

1.5

0

2.6

2

0

40

BNF 0

170

N-Fertils.

Höglwald, Germany Spruce Forest (Temperate)

N-Deposit.

40

Maulde, Belgium arable land, wheat (Temperate)

Plant N stock

175 Harvest

not considered

90

192

100

22

0 1?

0.3

9000

1570

ca. 550 (gross mineral.)

20

26

0

5-6

4-5

N-leaching ??

25 NH3

110 (net mineral)

? N2

4660

3.8 N2O+NO

Soil N Stock

1500

17

190

Plant N uptake

175

Plant N litter

not considered

??

30

25–30

< 25

Soil N flux density proxy (litterfall + throughfall) (kg N ha–1 yr–1)

< 60

60–80

>80

Proportion of input leached (%)

25) for growth response to N fertilization in conifers (Hyvönen et al., 2008). N content in needles correlates with the forest floor C:N ratio and is also a good indicator of N status (Kristensen et al., 2004). At N contents in needles below 1.4%, no nitrate leaching occurs and the system appears N-limited (Table 6.2), whereas above 1.4% N in needles leaching often occurs. This level corresponds surprisingly well with the threshold of 1.3%–1.4% N above which conifer stands show no growth response to N fertilizer additions (Sikström et al., 1998). In accordance with the definition of N saturation, nitrate leaching occurs if the soil flux density of mineral N (defined as N deposition plus net mineralization) exceeds the capacity of N uptake by plants. Net mineralization, which is a measurable parameter, includes (by definition) the microbial demand, i.e. the microbial immobilization. Datasets combining N deposition and net mineralization suggest a threshold in N flux density of 100 kg N ha−1 yr−1 for elevated nitrate leaching (Andersson et al., 2002; Fisk et al., 2002). Net mineralization is not routinely measured, but may be strongly correlated to the N flux with litterfall. Hence, total aboveground N input to the soil (throughfall N plus litterfall N) excluding belowground root litter input can be taken as a proxy for N flux density. Using this parameter the threshold is around 60 kg N ha−1 yr−1 for conifers (TableÂ€6.2). Retention of deposited N in soils mainly occurs in the forest floor (Nadelhoffer et al., 1999). This retention may slowly change the organic matter quality and decrease the C:N ratio of forest floors with time. There is no documentation whether this has occurred, but an indication is given by decreasing C:N (increasing forest floor per cent N) with increasing N deposition (Aber et al., 2003; Kristensen et al., 2004). On the other hand, Moldan et al. (2006) found no change in C:N ratio after a decade of chronic N addition to a coniferous forest, but an increased accumulation of both C and N in the organic mor layer. The European soil inventory of Vanmechelen et al. (1997) showed that currently approximately 40% of the c. 4000 sites had forest floor C:N ratios below the threshold of 25, below which elevated nitrate leaching often occurs. With current N loads, many forest sites may move towards N saturation. At this condition, the ecosystem is very responsive to changes in N deposition (Gundersen et al., 1998b). Reductions in N deposition will, with only a short time delay, translate into a proportional decrease in nitrate leaching. This was shown in experiments where N inputs in throughfall water were removed by roofs. In these experiments, nitrate leaching was reduced and N content in older needles decreased, indicating reversibility of the N-saturated condition (Boxman et al., 1998a; Bredemeier et al., 1998).

6.6â•‡ Effects of nitrogen 6.6.1â•‡ Nitrogen enrichment and saturation Since N is considered the major limiting element in terrestrial ecosystems, increased N deposition would be expected to be retained in the ecosystem and stimulate growth. However, N

114

deposition in large areas in Europe exceeds the normal growth requirement of forests and other (semi-natural) ecosystems and they may thus over time become enriched in N. Central to our understanding of N enrichment and its effects is the concept of ‘nitrogen saturation’, that describes a series of temporal changes in ecosystem functioning in response to increased N input. Nitrogen saturation can be defined in several ways (Ågren and Bosatta, 1988; Aber et al., 1989). The most widely used definition is in conditions where ‘availability of mineral N may exceed the combined nutritional demands of plants and microbes’ (Aber et al., 1989) which then can be determined as elevated nitrate leaching from the rooting zone. In this situation, what also occurs is that N (derived from deposition) leaves terrestrial ecosystems and may potentially affect downstream ecosystems through the ‘N cascade’ (Galloway etÂ€ al., 2003). Ågren and Bosatta (1988) have defined an N-saturated system as ‘an ecosystem where N losses approximate or exceed the inputs of N’ which implies an accumulation in the system close to zero. From a theoretical point of view, this may be the most proper use of the term ‘saturation’. However, this type of ‘true saturation’ may be seen as the end point of the N-saturation continuum where there is practically no biological control over N retention. Agricultural systems are often N-saturated according to the first definition, since it is usually economical to add more fertilizer N than can be taken up by the crop. On the other hand, they will never reach true saturation since a major fraction of the N input will be removed in harvested crops. The development of N saturation by increased N inputs involves a complex interaction of the processes in the N cycle (Aber et al., 1989, 1998). The progression from N limitation to N excess and the potential effects of N deposition may be explained as follows. In the first phase primary production increases, plants and microbes effectively absorb added N and the N content of plants increases. Retranslocation of N from senescent foliage (and roots) may decrease leading to higher N contents in litter materials and thus increased litterfall N flux. The internal cycling of N is accelerated through increases in litterfall N, net mineralization and tree N uptake. As N availability is increased, the composition of the forest floor vegetation may gradually change towards more nitrophilic species and other essential resources (P, K, Ca, Mg, or water) may at least periodically limit growth. In the accelerated N cycle, net nitrification becomes important and nitrate starts to appear in soil water. When elevated nitrate leaching becomes chronic, soil acidification resulting from N transformations becomes significant. Destabilization and potential forest decline from excess N deposition have been shown in case studies where the nutritional imbalance was important (Roelofs et al., 1985). Recent synthesis efforts support this general scheme although the understanding of processes and interactions has become more complex and detailed (Aber et al., 1998, 2003; Emmett etÂ€al., 1998; Gundersen et al., 1998b). The timing of the changes in ecosystem N function at a certain N deposition load is still not well understood, nor is the cumulative load required to change the N status of a low-N ecosystem known.


6.6.2â•‡ Biodiversity and vegetation change Nitrogen deposition has the potential to impact on biodiversity in a wide range of ways (see also Dise et al., 2011, ChapterÂ€20, this volume). Aboveground impacts on vegetation include Â�direct toxicity to sensitive species (Britto and Kronzucker, 2002), reduced resistance to environmental stresses such as frost (Caporn et al., 2000), increased susceptibility to pests and disease (Brunstig and Heil, 1985) and soil-mediated effects of acidification and eutrophication (Stevens et al., 2006). The degree to which N deposition has either a positive or negative impact on an individual species depends on the tolerances and requirements of the respective species. At a community level, the impact is usually negative, resulting in a change in the species composition from that typically found at a given location. Additionally, declines in species richness have been widely reported in a number of vegetation communities, both from experimental studies (Bobbink et al., 1998; Suding etÂ€al., 2005), studies using natural deposition gradients (Stevens etÂ€al., 2004; Maskell et al., 2009) and studies of species change over time (Dupré et al., 2010). Changes in species composition of the vegetation are likely to have impacts on N cycling. Species may have differential abilities to use N or a preference for either nitrate or ammonium. Consequently, changing species composition may impact on the relative rates or form of N uptake (Britto and Kronzucker, 2002). Other feedbacks between plant community change and the N cycle may result from different concentrations of N stored in plant tissues, plant growth rates and life-span and ease of decomposition of dead materials, all resulting in changes in the availability of N or residence times in N pools.

6.6.3â•‡ Carbon cycle Effects on plant growth and aboveground C sequestration In agricultural systems, N additions from fertilizer usually lead to increased plant growth even though the system is N-saturated as defined above, but in this case other nutrients (P, K, and micronutrients) are added as well. When NPK fertilizers are used in forest ecosystems strong growth responses are usually observed there as well (Jarvis and Linder, 2000), but when N alone is added growth response is more modest and dependent on N status prior to fertilization (Hyvönen et al., 2008). Growth responses to repeated annual N fertilization in northern Europe decrease with soil C:N ratio, with an indicated threshold for a growth response at C:N ratio 25 (Hyvönen et al., 2008). Under high loads of N fertilization the growth response can be reversed, as other nutrients become limiting (Högberg et al., 2006). Experiments more closely simulating chronic N inputs from deposition (lower doses split in several additions over the growing season or the whole year) show quite variable growth responses:Â€ the European NITREX sites showed no growth response to N addition (Emmett et al., 1998), whereas a longterm experiment at four sites in Michigan, USA, did show an increase in woody biomass (500 kg C ha−1 yr−1) after adding 30 kg N ha−1 yr−1 over 10 years (Pregitzer et al., 2008). In an

Nr-enriched site, a growth increase after trees were relieved from excess N deposition has also been observed (Boxman etÂ€ al., 1998a). The experiments simulating chronic N inputs were criticized for adding N to the soil, rather than to the canopy as happens with N deposition (Sievering, 1999). Recently it was shown that canopy uptake of N stimulated plant growth at a low-Â�deposition (3 kg N ha−1 yr−1 total) site in the USA (Sievering etÂ€al., 2007) and a modelling analysis indicate that direct canopy N uptake, by-passing the competition for nutrients exerted by soil microbes, could have a significant effect on the sensitivity of tree growth (Dezi et al., 2009). However, a study by Dail et al. (2009) revealed a limited canopy N uptake (see Section 6.3.2). Thus the experimental evidence is somewhat inconclusive. Recent Â�multi-factorial analysis of forest growth at nearly 400 plots across Europe showed significant responses to N deposition at 1%–2% growth increase per kg N ha−1 yr−1 when analysed at stand level (Solberg et al., 2009) as well as at individual tree level (Laubhann et al., 2009). A similar analysis of deposition effects on forest C storage, using, North American data showed comparable results (Thomas et al., 2010). Stronger responses were seen for sites with high C:N ratios (Solberg et al., 2009) as also observed in the fertilizer experiments (Hyvönen et al., 2008). Forest growth has increased in Europe by more than 60% over the past 50 years (Ciais et al., 2008), resulting in a strong increase in C sequestration. N deposition has been suggested as the main driving factor for this increase in forest growth (Karjalainen et al., 2008). On the one hand, it has been argued that N only makes a minor contribution to this sink since most of the N is retained in soils at a C:N ratio of approx. 30 (Nadelhoffer et al., 1999). Recently, Magnani et al. (2007) suggested that a large fraction of ecosystem C sequestration potential in temperate and boreal forests could be attributed to the effects of atmospheric N deposition. This study has fuelled a new debate on the effect of elevated N deposition on C sequestration (de Vries et al., 2008, 2009; Sutton et al., 2008). Sutton et al. (2008) suggested that the proposed sensitivity of about 200 kg C per kg N (Magnani et al., 2008) could be an artefact resulting from the parallel effects of other environmental factors and should be reduced to 50–75:1. De Vries et al. (2009) compiled data from a range of different approaches (observational, experimental and modelling) to analyse the impact of N deposition on C sequestration. The results of the various studies were in close agreement with regard to the effect of N addition for aboveground C sequestration (i.e. 15–30 kg C:1 kg N), but more variable for soil values (i.e. 5–35 kg C:1 kg N). All together these data indicate a total C sequestration range of 30–70 kg C per kg N deposition. The discrepancy in estimated C–N sensitivity of forest ecosystems between different studies could stem from several sources of error. On the one hand, studies based on regional N deposition datasets (Magnani et al., 2007) ignore the fact that higher dry N deposition rates are generally observed over forest canopies, as a result of their greater roughness. By underÂ�estimating N deposition, they would overestimate the response to a unit dose of added N. On the other hand, ecosystem manipulation studies could also be affected by artefacts, as they neglect the potentially important role of canopy N uptake and often apply

115


doses (up to 100 kg N ha−1 yr−1) well in excess of natural N deposition over most forest ecosystems (Dise et al., 2009). Whatever the sensitivity, the question is whether this apparent C sequestration from N deposition will be sustained in the long-term if forest ecosystems become N-saturated. At true N saturation (i.e. no or very low N accumulation) the ecosystem may also be C-saturated (no C sequestration) at least in the soil compartment (trees in managed systems may still grow and sequester N, but with no further response to N). The long-term fertilizer experiments in Sweden (Hyvönen et al., 2008) show that C sequestration per unit of added N decreases with dose (i.e. the response levels of above c. 30 kg N ha−1 yr−1). It remains to be seen if the same is true for increasing N deposition or for cumulative loads. An important question could be if there is an optimum level of N deposition for C sequestration or if a critical threshold exists for the effect of N on C sequestration.

Effect on soil C processes and C sequestration Although there is observational as well as experimental evidence for increased wood production, there is no evidence for an increase in leaf litter production due to elevated N deposition. Despite the relatively unsophisticated sampling of this ecosystem flux, it has not been part of the European forest monitoring programme and no data compilation seems to be available which could allow an analysis of the effect of N on leaf litter fluxes. N addition experiments show no response in leaf litter mass but instant increases in N concentrations of litter and thereby in the aboveground litterfall N flux (Gundersen et al., 1998b; Pregitzer et al., 2008). As the litterfall N flux increases, N mineralization also increases (Gundersen et al., 1998b; Nave et al., 2009). The N content in all compartments increases and major internal N fluxes increase. The responses of soil C processes are more complex and less well understood. Early stages of litter decomposition may also respond positively to elevated N deposition, since microorganisms on high C:N ratio litter material need to immobilize N for the decomposition (Berg, 2000). At later stages, on the other hand, when the easily decomposable organic matter has been processed, decomposition may actually decrease with N availability (Berg and Matzner, 1997). Thus with a constant leaf litter C input, SOM is expected to accumulate due to reduced decomposition at elevated N deposition. However, the mechanism whereby N addition is accompanied by a decrease in decomposition and increase in soil C stocks is still unclear (Janssens et al., 2010). A prevailing hypothesis is the lower production of lignolytic enzymes and phenol oxidases (see review in Janssens et al., 2010). There is growing evidence of reduced soil respiration from chronic N addition experiments (Burton et al., 2004; Hagedorn et al., 2003; Janssens etÂ€ al., 2010) in parallel with a decline in soil microbial biomass (Treseder, 2008). Since soil respiration, however, includes not only a component from decomposition but also respiration components from roots and mycorrhyzae, we cannot conclude that reduced soil respiration indicates increased SOM-C accumulation (C sequestration), as it could also indicate reduced root respiration.

116

A critical gap in our knowledge is the contribution of belowground litter input (roots, mycorrhyzae and exudates) to SOM formation versus that from aboveground plant litter (leaves, etc.) (Rasse et al., 2005). Likewise the responses of these belowground components to elevated N deposition are not well known, but total belowground C allocation is known to decline with increasing productivity and N availability (Palmroth et al., 2006). Root biomass (Boxman et al.,1998b; Nadelhoffer, 2000) and turnover (Majdi, 2004) may decrease with N availability. Likewise ectomycorrhyzal mycelial growth was also negatively affected by N additions after N fertilization (Nilsson and Wallander, 2003; Parrent and Vilgalys, 2007), as an effect of N deposition (Nilsson et al., 2000) and with increasing N availability in natural nutrient fertility gradients (Nilsson et al., 2005). Alternatively, accumulation of SOM by reduced decomposition could be outweighed by reduced contributions from other sources of SOM-C (roots, mycorrhyzae and exudates). The integrated response of these processes to elevated N in the form of increased SOM accumulation may be difficult to measure due to the size and the variability in this pool (Yanai et al., 2003). However, long-term forest N fertilization experiments from Sweden and Finland did reveal an increase in SOM-C from N addition (Hyvönen et al., 2008) and for the first time a significant response of SOM-C was shown for a chronic N addition experiment at four forest sites in Michigan, USA (Pregitzer et al., 2008). A long-term N addition experiment on heathland also revealed increased SOM-C accumulation (Evans et al., 2006). Further measurements of increased organic layer thickness over 40 years in an intensive network of sites across Sweden (Berg et al., 2009) indicate that C sequestration in organic layers could be a widespread phenomenon. A metadata analysis of studies on the response of CO2 flux from N additions in multiple terrestrial and wetland ecosystem types by Liu and Graever (2009) showed a large variation of net ecosystem CO2 exchange (NEE) for non-forest ecosystems (grassland, wetland and tundras), thus leading to a statistically insignificant effect.

6.6.4â•‡ N leaching associated effects Nitrate leaching is an acidifying process in soils and an important process in acidification of lower soil horizons (Velthof etÂ€ al. 2011, Section 21.4 this volume). The input may be as nitric acid or as ammonia/ammonium that can be nitrified in the soil and release protons. Soils have an ability to Â�neutralize acids through the supply of base cations from weathering and cation exchange reactions. Increased acidification of forest soils has been observed during recent decades (FalkengrenGrerup et al., 1987; Wesselink et al., 1995) with pH declining by up to 1 unit, which in part may be caused by air pollution including deposition of N compounds. Both proton (H+)producing and proton-consuming processes including N species occur in soils, but a net acidification only occurs when nitrate is leached from the system (Gundersen and Rasmussen, 1990). Each 14 kg N ha−1 yr−1 of nitrate leached is equivalent to the production of 1 kmol H+ ha−1 yr−1. Depending on the acid status of the soil, base cations and/or Al will be leached


with the nitrate. An increasing fraction of the acidity in acidsensitive surface waters is related to nitrate (Stoddard et al., 1999; Durand et al., 2011, Chapter 7, this volume). In Europe, approximately 30% of monitored forest sites leach between 7 and 50 kg N ha−1 yr−1 (De Vries et al., 2007; MacDonald et al., 2002), equivalent to an acid production of 0.5–3.5 kmol H+ ha−1 yr−1. At the majority of these sites this was buffered by Al release (Dise et al., 2001; De Vries et al., 2007) and nitrate and aluminium are usually positively correlated in acid soil and surface waters. In the long-term this may lead to significant nutrient loss, impairment of base cation uptake by Al toxicity and potentially reduced forest production (Velthof et al., 2011, Chapter 21, this volume) as well as Al toxicity in surface water (Havas and Rossland, 1995).

6.6.5â•‡ N-driven vulnerability to disturbances Nitrogen deposition may play a significant role in increasing forest susceptibility to wildfires, as documented in mixed coniferous forests of southern California (USA). High levels of ozone and nitrogenous compounds derived from regional urbanization and industrialization cause specific changes in forest tree C, N, and water balances that enhance individual tree susceptibility to drought, bark beetle attack, and disease, and when combined contribute to the whole ecosystem susceptibility to wildfire (Grulke et al., 2009). Similar findings have been documented for deserts and coastal sage scrub formations in southern California where N deposition and frequent fire promotes increased grass biomass from invasive species and increases the risk of fire further as it provides the fuel for subsequent fires (Brooks et al., 2004; Rao et al., 2009).

6.7â•‡ Summary This section summarizes the main conclusions on the terrestrial N cycling processes and their importance. Major uncertainties and gaps in knowledge are also highlighted.

Nitrogen pools and N availability (1) On the ecosystem scale, soils are the main reservoir for N. This is more pronounced for agricultural systems, with more than 90%–95% of Nr being stored in the soil as compared to forest systems, where N storage in soil is 50–70%. (2) N availability varies with temperature and humidity gradients in Europe. (3) The contemporary global biogeochemical cycle of N in terrestrial ecosystems is dominated by microbial processes in soils.

Nitrogen inputs in non-agricultural systems (4) An understanding of biological N2 fixation in a few legume crop plants is relatively advanced, but much less is known about biological N2 fixation in non-agricultural legumes or in other N2-fixing organisms. The difficulties of measuring rates of biological N2 fixation accurately at the ecosystem scale have so far hampered a better understanding of the

importance of biological N2 fixation for most terrestrial ecosystems. (5) The estimation of N deposition inputs at the site scale is affected by neglecting the input of dissolved organic nitrogen (DON) and by the uncertainty in atmosphere canopy interaction of N species.

Nitrogen cycling â•‡ (6) The understanding of N cycling in terrestrial ecosystems has undergone a paradigm shift since 1990. Until then, the perception was that (i) N mineralization is the limiting step in N cycling, (ii) plants take up inorganic N, and (iii)Â€plants poorly compete for N against microbes and use only the N which is ‘left over’ by microbes. Consequently, net N mineralization has been assessed to measure plantavailable N. Since then studies have shown that plants effectively compete for N with microorganisms and take up organic N, in a broad range of ecosystems. â•‡ (7) Nitrogen mineralization/ammonification is the dominant control of gross nitrification. â•‡ (8) Microbial nitrate immobilization is a significant process of Nr-retention in a wide range of terrestrial ecosystems that depend largely on gross nitrification. â•‡ (9) The importance of other recently recognized processes (DNRA, anammox, nitrifier denitrification) for Nr-cycling in ecosystems is not well developed. (10) Denitrification rates in soils are highly uncertain, despite more than eight decades of research, partly because of our lack of understanding and partly due to the large spatial and temporal variability. (11) Factors controlling Nr fluxes include moisture content (water-filled pore space) and soil temperature, soil properties, such as clay content, carbon content, C:N ratio and pH and vegetation factors, which are all affected by land use change and climatic change.

Nitrogen outputs (12) Burning is a major pathway of Nr loss for ecosystems exposed to high fire frequencies. The influence of Nr deposition on fuel N build-up should be considered in order to estimate wildfire NOx emissions. This effect is neglected at present. (13) Major outputs in ecosystems not exposed to fires are nitrate leaching and gaseous losses (N2O, NO, N2 and in agricultural systems also NH3) to the environment. N2 emissions are most uncertain due to the difficulties to quantify N2 emissions and to constrain what is driving denitrification on site and regional scales. (14) The C:N ratio of the forest floor or the top mineral soil is a good indicator of N status related to NO3– leaching. At C:N above 25, mineral N is usually retained, whereas below 25, NO3– leaching often occurs and increases with increasing N deposition.

Nitrogen effects (15) Atmospheric Nr input has caused N saturation in terms of a decline in the soil C:N ratio in the forest floor, associated

117


with an increase in (i) Nr (nitrate) leaching and in gaseous N losses (N2O, NO, N2), (ii) N availability and related biodiversity change, and (iii) the occurrence of pests and diseases. (16) Atmospheric Nr input has led to an increase in aboveground C sequestration but the impact on soil C sequestration is less clear. A critical gap in our knowledge is the impact of Nr deposition on belowground litter input (roots, mycorrhyzae and exudates) and through that on soil C sequestration.

Future studies on terrestrial N cycling (17) Interdisciplinary and multi-scale studies should focus on simultaneous and comprehensive measurements of all major Nr fluxes at site and landscape scale, including plant uptake/release of organic and inorganic N compounds as well as microbial Nr conversion. (18) Linking plant physiological and soil microbial Nr cycling, as well as soil hydrological Nr transport, to more reliable estimates of ecosystem N fluxes will be a major research challenge for coming years. In particular, this will include further development of methodological approaches and experimental assays that allow direct assessment of N turnover processes in the larger context of intact plant– soil-systems, where competitive mechanisms between microorganisms and plants persist. This will require an intensified interdisciplinary cooperation between plant scientists and soil ecologists/soil microbiologists.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), the COST Action 729 and the Villcum Foundation, Denmark.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press:Â€ www.cambridge.org/ena and the Nitrogen in Europe website:Â€ www.nine-esf.org/ena.

References Aber, J. D., Nadelhoffer, K. J., Steudler, P. and Melillo, J. M. (1989). Nitrogen saturation in northern forest ecosystems. BioScience, 39, 378–386. Aber, J. D., McDowell, W., Nadelhoffer, K. et al. (1998). Nitrogen saturation in temperate forest ecosystems:Â€hypothesis revisited. BioScience, 48, 921–934. Aber, J. D., Goodale, C. L., Ollinger, S. V. et al. (2003). Is nitrogen deposition altering the nitrogen status of northeastern forests? BioScience, 53, 375–389. Accoe, F., Boeckx, P., Busschaert, J., Hofman, G. and Van Cleemput,Â€O. (2004). Gross N transformation rates and net N mineralisation rates related to the C and N contents of soil organic matter fractions in grassland soils of different age. Soil Biology and Biochemistry, 36, 2075–2087. Ågren, G. I. and Bosatta, E. (1988). Nitrogen saturation of terrestrial ecosystems. Environmental Pollution, 54, 185–198.

118

Alexis, M. A., Rasse, D. P., Rumpel, C. et al. (2007). Fire impact on C and N losses and charcoal production in a scrub oak ecosystem. Biogeochemistry, 82, 201–216. Ambus, P., Mosier, A. and Christensen, S. (1992). Nitrogen turnover rates in a riparian fen determined by 15N dilution. Biology and Fertility of Soils, 14, 230–236. Andersson, P., Berggren, D. and Nilsson, I. (2002). Indices for nitrogen status and nitrate leaching from Norway spruce (Picea abies (L.) Karst.) stands in Sweden. Forest Ecology and Management, 157, 39–53. Azam, F., Mahmood, T. and Malik, K. A. (1988). Immobilisationremineralisation of NO3-N and total N balance during decomposition of glucose, sucrose and cellulose in soil incubated at different moisture regimes. Plant and Soil, 107, 159–163. Baggs, E. M. (2008). A review of stable isotope techniques for N2O source partitioning in soils:Â€recent progress, remaining challenges and future considerations. Rapid Communication in Mass Spectrometry, 22, 1664–1672. Bais, H. P., Weir, T. L., Perry, G., Gilroy, S. and Vivanco, J. M. (2006). The role of root exudates in rhizosphere interactions with plants and other microorganisms. Annual Review of Plant Biology, 57, 233–266. Ball, B. C., Bingham, I., Rees, R. M., Watson, C.A. and Litterick, A. (2005). The role of crop rotations in determining soil structure and crop growth conditions. Canadian Journal of Soil Science, 85, 557–577. Barbosa, P., Camia, P., Kucera, J. et al. (2009). Assessment of forest fire impacts and emissions in the European Union based on the European Forest Fire Information System. In:Â€Wildland Fires and Air Pollution. ed. A. Bytnerowicz, M. Arbaugh, A. Riebau and C.Â€Andersen, Elsevier: New York, pp. 197–208. Barton, L., McLay, C. D. A., Schipper, L. A. and Smith, C. T. (1999). Annual denitrification rates in agricultural and forest soils:Â€a review. Australian Journal of Soil Research, 37, 1073–1093. Bateman, E. J. and Baggs, E. M. (2005). Contributions of nitrification and denitrification to N2O emissions from soils at different waterfilled pore space. Biology and Fertility of Soils, 41, 379–388. Batjes, N. H. (1996). Total carbon and nitrogen stocks in the soils of the world. European Journal of Soil Science, 47, 151–163. Beier, C., Emmett, B. A., Peñuelasc, J. et al. (2008). Carbon and nitrogen cycles in European ecosystems respond differently to global warming. Science of the Total Environment, 407, 692–697. Berg, B. (2000). Litter decomposition and organic matter turnover in northern forest soils. Forest Ecology and Management, 133, 13–22. Berg, B. and Matzner, E. (1997). Effect of N deposition on decomposition of plant litter and soil organic matter in forest systems. Environmental Reviews, 5, 1–25. Berg, B., Johansson, M.-B., Nilsson, Å., Gundersen, P. and Norell,Â€L. (2009). Sequestration of Carbon in humus layers in Swedish forestsÂ€– direct measurements. Canadian Journal of Forest Research, 39, 962–975. Binkley, D., Stottlemyer, R., Suarez, F. and Cortina, J. (1994). Soil nitrogen availability in some arctic ecosystems in northwest Alaska:Â€responses to temperature and moisture. Ecoscience, 1, 64–70. Bobbink, R., Hornung, M. and Roelofs, J. G. M. (1998). The effects of air-borne nitrogen pollutants on species diversity in natural and semi-natural European vegetation. Journal of Ecology, 86, 717–738. Bollmann, A. and Conrad, R. (1997). Acetylene blockage technique leads to underestimation of denitrification rates in oxic soils

Klaus Butterbach-Bahl and Per Gundersen due to scavenging of intermediate nitric oxide. Soil Biology and Biochemistry, 29, 1067–1077. Bonin, P. (1996). Anaerobic nitrate reduction to ammonium in two strains isolated from costal marine sediment:Â€A dissimilatory pathway. FEMS Microbiology Ecology, 19, 27–38. Booth, M. S., Stark, J. M. and Rastetter, E. (2005). Controls on Nitrogen cycling in terrestrial ecosystems:Â€a synthetic analysis of literature data. Ecological monographs, 75, 139–157. Bouwman, A. (1998). Nitrogen oxides and tropical agriculture. Nature, 392, 886–887. Bouwman, A. F., Boumans, L. J. M. and Batjes, N. H. (2002). Estimation of global NH3 volatilization loss from synthetic fertilizers and animal manure applied to arable lands and grasslands. Global Biogeochemical Cycles, 16, 1024–1038. Bouwman, A. F., Lee, D. S., Asman, W. A. H. et al. (1997). A global high-resolution emission inventory for ammonia. Global Biogeochemical Cycles, 11, 561–587. Boxman, A. W., van der Ven, P. J. M. and Roelofs, J. G. M. (1998a). Ecosystem recovery after a decrease in nitrogen input to a Scots pine stand at Ysselsteyn, the Netherlands. Forest Ecology and Management, 101, 155–163. Boxman, A. W., Blanck, K., Brandrud, T.-E. et al. (1998b). Vegetation and soil biota response to experimentally-changed nitrogen inputs in coniferous forest ecosystems of the NITREX project. Forest Ecology and Management, 101, 65–79. Bredemeier, M., Blanck, K., Dohrenbusch, A. et al. (1998). The Solling roof experimentsÂ€– site characteristics, experiments and results. Forest Ecology and Management, 101, 281–293. Breuer, L., Kiese, R. and Butterbach-Bahl, K. (2002). Temperature and moisture effects on nitrification rates in tropical forest soils. Soil Science Society of America Journal, 66, 834–844. Britto, D. T. and Kronzucker, H. J. (2002). NH4+ toxicity in higher plants:Â€a critical review. Journal of Plant Physiology, 159, 567–584. Brooks, M. L., D’Antonio, C. M., Richardson, D. M. et al. (2004). Effects of invasive alien plants on fire regimes. Bioscience, 54, 677–688. Brookshire, E. N. J., Valett, H. M., Thomas, S. A. and Webster, J. R. (2007). Atmospheric N deposition increases organic N loss from temperate forests. Ecosystems, 10, 252–262. Brumme, R., Borken, W. and Finke, S. (1999). Hierachical control on nitrous oxide emission in forest ecosystem. Global Biogeochemical Cycles, 13, 1137–1148. Brunstig, A. M. H. and Heil, G. W. (1985). The role of nutrients in the interactions between a herbivorous beetle and some competing plant species in healthlands. Oikos, 44, 23–26. Burger, M. and Jackson, L. E. (2003). Microbial immobilisation of ammonium and nitrate in relation to ammonification and nitrification rates in organic and conventional cropping systems. Soil Biology and Biochemistry, 35, 29–36. Burton, A. J., Pregitzer, K. S., Crawford, J. N., Zogg, G. P. and Zak, D. R. (2004). Simulated chronic NO3 deposition reduces soil respiration in northern hardwood forests. Global Change Biology, 10, 1080–1091. Butterbach-Bahl, K., Rothe, A. and Papen, H. (2002). Effect of tree distance on N2O and CH4-fluxes from soils in temperate forest ecosystems. Plant and Soil, 240, 91–103. Butterbach-Bahl, K., Kahl, M., Mykhayliv, L. et al. (2009). A European wide inventory of soil NO emissions using the biogeochemical models DNDC/ Forest DNDC. Atmospheric Environment, 43, 1392–1402.

Butterbach-Bahl, K., Kock, M., Willibald, G. et al. (2004). Temporal variations of fluxes of NO, NO2, N2O, CO2 and CH4 in a tropical rain forest ecosystem. Global Biogeochemical Cycles, 18, doi:10.1029/2004GB002243. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. et al. (2011). Nitrogen as a threat to the European greenhouse balance. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Campbell, J. L., Hornbeck, J. W., McDowell, W. H. et al. (2000). Dissolved organic nitrogen budgets for upland, forested ecosystems in New England. Biogeochemistry, 49, 123–142. Campbell, J. L., Rustad, L. E., Boyer, E. W. et al. (2009). Consequences of climate change for biogeochemical cycling in forests of northeastern North America. Canadian Journal of Forest Soils, 39, 264–284. Cape, J. N., Anderson, M., Rowland, A. P. and Wilson, D. (2004). Organic nitrogen in precipitation across the United Kingdom. Water, Air, and Soil Pollution:Â€Focus, 4, 25–35. Caporn, S. J. M., Ashenden, T. W. and Lee, J. A. (2000). The effect of exposure to NO2 and SO2 on frost hardiness in Calluna vulgaris. Environmental and Experimental Botany, 43, 111–119. Carlsson, G. and Huss-Danell, K. (2003). Nitrogen fixation in perennial forage legumes in the field. Plant and Soil, 253, 353–372. Cassman, K. G. and Harwood, R. R. (1995). The nature of agricultural systemsÂ€– food security and environmental balance. Food Policy, 20, 439–454. Castaldi, S., Carfora, A., Fiorentino, A. et al. (2009). Inhibition of net nitrification activity in a Mediterranean woodland:Â€possible role of chemicals produced by Arbutus unedo. Plant and Soil, 15, 273–283. Chapman, S. K., Langley, J. A., Hart, S. C. and Koch, G. W. (2006). Plants actively control nitrogen cycling:Â€uncorking the microbial bottleneck. New Phytologist, 169, 27–34. Chapuis-Lardy, L., Wrage, N., Liemetay, A., Chotte, J. L. and Bernoux, M. (2007). Soils, a sink for N2O? A review. Global Change Biology, 13, 1–17. Christensen, S. and Tiedje, J. M. (1990). Brief and vigorous N2O production by soil at spring thaw. Journal of Soil Science, 41, 1–4. Ciais, P., Schelhaas, M. J., Zaehle, S. et al. (2008). Carbon accumulation in European forests. Nature Geoscience, 1, 425–429. Cleveland, C. C., Townsend, A. R., Schimel, D. S. et al. (1999). Global patterns of terrestrial biological nitrogen (N2) fixation in natural ecosystems. Global Biogeochemical Cycles, 13, 623–645. Cliff, J. B., Bottomley, P. J., Gaspar, D. J. and Myrold, D. D. (2002). Exploration of inorganic C and N assimilation by soil microbes with time of flight secondary ion mass spectrometry. Applied and Environmental Microbiology, 68, 4067–4073. Colloff, M. J., Wakelin, S. A., Gomez, D. and Rogers, S. L. (2008). Detection of nitrogen cycle genes in soils for measuring the effects of changes in land use and management. Soil Biology & Biochemistry, 40, 1637–1645. Conrad, R. (2002). Microbiological and biochemical background of production and consumption of NO and N2O in soil. In:Â€Trace Gas Exchange in Forest Ecosystems, ed. R. Gasche, H. Papen and H. Rennenberg. Dordrecht:Â€Kluwer Academic Publishers, pp.Â€3–33. Corré, M. F., Beese, F. O. and Brumme, R. (2003). Soil nitrogen cycle in high nitrogen deposition forest:Â€changes under nitrogen saturation and liming. Ecological Applications, 13, 287–298. Costa, E., Perez, J. and Kreft, J. U. (2006). Why is metabolic labour divided in nitrification? Trends in Microbiology, 14, 213–219.

119

Nitrogen processes in terrestrial ecosystems Crutzen, P. J., Mosier, A. R., Smith, K. A. and Winiwarter, W. (2008). N2O release from agro-biofuel production negates global warming reduction by replacing fossil fuels. Atmospheric Chemistry and Physics, 8, 389–395. Dail, D. B., Hollinger, D. Y., Davidson, E. A. et al. (2009). Distribution of nitrogen-15 tracers applied to the canopy of a mature sprucehemlock stand, Howland, Maine, USA. Oecologia, 160, 589–599. Dannenmann, M., Gasche, R., Ledebuhr, A. and Papen, H. (2006). Effects of forest management on soil N cycling in beech forests stocking on calcareous soils. Plant and Soil, 287, 279–300. Davidson, E. A. (1994). Climate change and soil microbial processes:Â€secondary effects are hypothesised from better known interacting primary effects. In:Â€Soil Response to Climate Change, ed. M. D. A. Rounsevell and P. J. Loveland. Springer,Â€Berlin: pp.Â€156–168. Davidson, E. A. (2009). The contribution of manure and fertilizer nitrogen to atmospheric nitrous oxide since 1860. Nature Geoscience, 2, 659–662. Davidson, E. A., Hart, S. C. and Firestone, M. K. (1992). Internal cycling of nitrate in soils of a mature conferous forest. Ecology, 73, 1148–1156. Davidson, E. A., Hart, S. C., Shanks, C. A. and Firestone, M. K. (1991). Measuring gross nitrogen mineralisation, immobilisation, and nitrification by 15N isotopic pool dilution in intact soil cores. Journal of Soil Science, 42, 335–349. Davidson, E. A., Matson, P. A., Vitousek, P. M. et al. (1993). Processes regulating soil emissions of NO and N2O in a seasonally dry tropical forest. Ecology, 74, 130–139. De Vries, W., Reinds, G. J., Gundersen, P. and Sterba, H. (2006). The impact of nitrogen deposition on carbon sequestration in European forests and forest soils. Global Change Biology, 12, 1151–1173. De Vries, W., van der Salm, C., Reinds, G. J. and Erisman, J. W. (2007). Element fluxes through European forest ecosystems and their relationships with stand and site characteristics. Environmental Pollution, 148, 501–513. De Vries, W., Solberg, S., Dobbertin, M. et al. (2008). Ecologically implausible carbon response. Nature, 441, E1–E3. De Vries, W., Solberg, S., Dobbertin, M. et al. (2009). The impact of nitrogen deposition on carbon sequestration by terrestrial ecosystems. Forest Ecology and Management, 258, 1814–1823. Dezi, S., Medlyn, B., Tonon, G. and Magnani, F. (2009). Effect of nitrogen deposition on forest carbon sequestration:Â€model analysis. Global Change Biology, 16, 1470–1486. Dise, N. B., Matzner, E., Armbruster, M. and MacDonald, J. (2001). Aluminum output fluxes from forest ecosystems in Europe:Â€a regional assessment. Journal of Environmental Quality, 30, 1747–1756. Dise, N. B., Rothwell, J. J., Gauci, V., van der Salm, C. and de Vries, W. (2009). Predicting dissolved inorganic nitrogen leaching in European forests using two independent databases. Science of the Total Environment, 407, 1798–1808. Dise, N. B., Ashmore, M., Belyazid, S. et al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Dobbie, K. E., McTaggart, I. P. and Smith, K. A. (1999). Nitrous oxide emissions from intensive agricultural systems:Â€variations between crops and seasons, key driving variables, and mean emission factors. Journal of Geophysical Research, 104, 26891–26899. Dragosits, U., Theobald, M. R., Place, C. J. et al. (2002). Ammonia emission deposition and impact assessment at the field scale:Â€a case

120

study of sub-grid spatial variability. Environmental Pollution, 117, 147–158. Dupré, C., Stevens, C. J., Ranke, T. et al. (2010). Changes in species richness and composition in European acidic grasslands over the past 70 years:Â€the contribution of cumulative atmospheric nitrogen deposition. Global Change Biology, 16, 344–357. Durand, P., Breuer, L., Johnes, P. et al. (2011). Nitrogen processes in aquatic ecosystems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Emmett, B. A., Gundersen, P., Kjønaas, O. J. et al. (1998). Predicting the effects of atmospheric nitrogen deposition in conifer stands:Â€evidence from the NITREX ecosystem-scale experiments. Ecosystems, 1, 352–360. Eno, C. (1960). Nitrate production in the field by incubating the soil in polyethylene bags. Soil Science Society of America Proceedings, 24, 277–279. Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. and Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. Erisman, J. W., van Grinsven, H., Grizzetti, B. et al. (2011). The European nitrogen problem in a global perspective. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Eskew, D. L., Eaglesham, A. R. J. and App, A. A. (1981). Heterotrophic (N-15)2 fixation and distribution of newly fixed nitrogen in a riceflooded soil system. Plant Physiology, 68, 48–52. Evans, C. D., Caporn, S. J. M., Carroll, J. A. et al. (2006). Modelling nitrogen saturation and carbon accumulation in heathland soils under elevated nitrogen deposition. Environmental Pollution, 143, 468–478. Evans, J. R. (1989). Photosynthesis and nitrogen relationships in leaves of C3 plants. Oecologia, 78, 9–19. Faegerli, H. and Aas, W. (2008). Trends of nitrogen in air and precipitation:Â€Model results and observations at EMEP sites in Europe, 1980–2003. Environmental Pollution, 154, 448–461. Falkengren-Grerup, U., Linnermark, N. and Tyler, G. (1987). Changes in acidity and cation pools of south Swedish soils between 1949 and 1985. Chemosphere, 16, 2239–2248. Fisk, M. C., Zak, D. R. and Crow, T. R. (2002). Nitrogen storage and cycling in old- and second-growth northern hardwood forests. Ecology, 83, 73–87. Fowler, D., O’Donoghue, M., Muller, J. B. A. et al. (2004). A chronology of nitrogen deposition in the UK between 1900 and 2000. Water, Air and Soil Pollution, 4, 9–23. French, S., Levy-Booth, D., Samarajeewa, A. et al. (2009). Elevated temperatures and carbon dioxide concentrations:Â€effects on selected microbial activities in temperate agricultural soils. World Journal of Microbiology Biotechnology, 25, 1887–1900. Freppaz, M., Williams, B. L., Edwards, A. C., Scalenghe, R. and Zanini, E. (2007). Labile nitrogen, carbon and phosphorous pools and nitrogen mineralisation and immobilisation rates at low temperatures in seasonally snow-covered soils. Biology and Fertility of Soils, 43, 519–529. Gaige, E., Dail, D. B., Hollinger, D. Y. et al. (2007). Changes in canopy processes following whole-forest canopy nitrogen fertilisation of a mature spruce-hemlock forest. Ecosystems, 10, 1133–1147. Galloway, J. N., Aber, J. D., Erisman, J. W. et al. (2003). The nitrogen cascade. BioScience, 53, 341–356. Galloway, J. N., Dentener, F. J., Capone, D. G. et al. (2004). Nitrogen cycles:Â€past, present, future. Biogeochemistry, 70, 153–226.

Klaus Butterbach-Bahl and Per Gundersen Gimeno, B. S., Yuan, F., Fenn, M. E. and Meixner, T. (2009). Management options for mitigating nitrogen (N) losses from N-saturated mixed-conifer forests in California. In:Â€Wildland Fires and Air Pollution. ed. A. Bytnerowicz, M. Arbaugh, A. Riebau and C. Andersen, Elsevier, New York: pp. 425–455. Glass, A. D. M., Britto, D. T., Kaiser, B. N. et al. (2002). The regulation of nitrate and ammonium transport systems in plants. Journal of Experimental Botany, 53, 855–864. Griffin, T. S., He, Z. and Honeycutt, C.W. (2005). Manure composition affects net transformation of nitrogen from dairy manures. Plant and Soil, 273, 29–38. Groffman, P. M., Altabet, M. A., Böhlke, J. K. et al. (2006). Methods for measuring denitrification:Â€diverse approaches to a difficult problem. Ecological Applications, 16, 2091–2122. Groffman, P. M., Butterbach-Bahl, K., Fulweiler, R. W. et al. (2009). Incorporating spatially and temporally explicit phenomena (hotspots and hot moments) in denitrification models. Biogeochemistry, 93, 49–77. Grulke, N. E., Minnich, R. A., Paine, T. D. et al. (2009). Air pollution increases forest susceptibility to wildfires:Â€A case study in the San Bernardino Mountains in southern California. In: Wildland Fires and Air Pollution, ed. A. Bytnerowicz, M. Arbaugh, A. Riebau and C. Andersen, New York:Â€Elsevier, pp. 365–403. Gundersen, P. and Rasmussen, L. (1990). Nitrification in forest soils:Â€effects of nitrogen deposition on soil acidification and aluminium release. Reviews of Environmental Contamination Toxicology, 13, 1–45. Gundersen, P., Callesen, I. and de Vries, W. (1998a). Nitrate leaching in forest ecosystems is controlled by forest floor C/N ratio. Environmental Pollution, 102, 403–407. Gundersen, P., Emmett, B. A., Kjønaas, O. J., Koopmans, C. and Tietema, A. (1998b). Impact of nitrogen deposition on nitrogen cycling:Â€a synthesis of NITREX-data. Forest Ecology and Management, 101, 37–55. Gundersen, P., Boxman, A. W., Lamersdorf, N., Moldan, F. and Andersen, B. R. (1998c). Experimental manipulation of forest ecosystemsÂ€– lessons from large roof experiments. Forest Ecology and Management, 101, 339–352. Gundersen, P., Schmidt, I. K. and Raulund-Rasmussen, K. (2006). Leaching of nitrate from temperate forests: effects of air pollution and forest management. Environmental Reviews, 14, 1–57. Gundersen, P., Sevel, L., Christiansen, J. R. et al. (2009). Do indicators of nitrogen retention and leaching differ between coniferous and deciduous forests in Denmark? Forest Ecology and Management, 258, 1137–1146. Hagedorn, F., Spinnler, D. and Siegwolf, R. (2003). Increased N deposition retards mineralization of old soil organic matter. Soil Biology and Biochemistry, 35, 1683–1692. Harrison, K. A., Bol, R. and Bardgett, R. D. (2007). Preferences for different nitrogen forms by coexisting plant species and soil microbes. Ecology, 88, 989–999. Hart, S. C., Nason, G. E., Myrold, D. D. and Perry, D. A. (1994). Dynamics of gross nitrogen transformations in an old-growth forest:Â€the carbon connection. Ecology, 75, 880–891. Havas, M. and Rossland, B. O. (1995). Response of zoopankton, benthos and fish to acidification:Â€an overview. Water, Air and Soil Pollution, 85, 51–62. Hayatsu, M., Tago, K. and Saito, M. (2008). Various players in the nitrogen cycle:Â€diversity and functions of the microorganisms

involved in nitrification and denitrification. Soil Science and Plant Nutrition, 54, 33–45. Herridge, D. F., Peoples, M. B. and Boddey, R. M. (2008). Global inputs of biological nitrogen fixation in agricultural systems. Plant and Soil, 311, 1–8. Hertel, O., Reis, S., Skjøth, C. A. et al. (2011). Nitrogen processes in the atmosphere. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erismar et al. Cambridge University Press. Hofstra, N. and Bouwman, A. F. (2005). Denitrification in agricultural soils:Â€summarizing published data and estimating global annual rates. Nutrient Cycling in Agroecosystems, 72, 267–278. Högberg, P. and Read, D. J. (2006). Towards a more plant physiological perspective on soil ecology. Ecology and Evolution, 21, 548–554. Högberg, P., Fan, H., Quist, M., Binkley, D. and Tamm, C. O. (2006). Tree growth and soil acidification in response to 30 years of experimental nitrogen loading on boreal forest. Global Change Biology, 12, 489–499. Houlton, B. Z. and Bai, E. (2009). Imprint of denitrifying bacteria on the global terrestrial biosphere. Proceedings of the National Academy of Sciences of the USA, 106, 21713–21716. Humbert, S., Tarnawski, S., Fromin, N. et al. (2010). Molecular detection of anammox bacteria in terrestrial ecosystems:Â€distribution and diversity. ISME Journal, 4, 450–454. Hyvönen, R., Persson, T., Andersson, S. et al. (2008). Impact of longterm nitrogen addition on carbon stocks in trees and soils in northern Europe. Biogeochemistry, 89, 121–137. Ingwersen, J., Butterbach-Bahl, K., Gasche, R., Richter, O. and Papen,Â€H. (1999). Barometric process separation:Â€New method for quantifying nitrification, denitrification, and nitrous oxide sources in soils. Soil Science Society of America Journal, 63, 117–128. IPCC (1997). Revised 1996 Guidelines for National Greenhouse Gas Inventories, Bracknell, UK: IPCC. IPCC (2007). 2006 Guidelines for National Greenhouse Gas Inventories, Vol. 4, Chap. 11, Hayama, Japan: IGES. Jackson, L. E., Burger, M. and Cavagnaro, T. R. (2008). Roots, nitrogen transformations and ecosystem services. Plant Biology, 59, 341–363. Janssens, I. A., Dieleman, W., Luyssaert, S. et al. (2010). Reduction of forest soil respiration in response to nitrogen deposition. Nature Geoscience, 3, 315–322. Jansson, S. L., Hallam, M. J. and Bartholomew, W. V. (1955). Preferential utilization of ammonium over nitrate by microorganisms in the decomposition of oat straw. Plant and Soil, 4, 382–390. Jarvis, P. G. and Linder, S. (2000). Constraints to growth of boreal forests. Nature, 405, 904–905. Jarvis, S. C., Stockdale, E. A., Shepherd, M. A. and Powlson, D.S. (1996). Nitrogen mineralisation in temperate agricultural soils:Â€processes and measurement. Agronomy, 57, 187–235. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. et al. (2011). Benefits of nitrogen for food fibre and industrial production. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Johnson, D. W., Murphy, J. D., Walker, R. F., Miller, W. W. and Glass, D. W. (2007). Wildfire effects on forest carbon and nutrient

121

Nitrogen processes in terrestrial ecosystems budgets:Â€comparison of reconstructed versus measured values. Ecological Engineering, 31, 183–192. Karjalainen, T., Schuck, A., Prietzel, J. et al. (2008). Causes and Consequences of Forest Growth Trends in Europe, Results of the RECOGNITION Project, EFI Report No. 21. Joensuu:Â€European Forest Institute. Kesik, M., Blagodatsky, S., Papen, H. and Butterbach-Bahl, K. (2006a). Effect of pH, temperature and substrate on N2O, NO and CO2 production by Alcaligenes faecalis p. J. Applied Microbiology, 101, 655–667. Kesik, M., Brüggemann, N., Forkel, R. et al. (2006b). Future scenarios of N2O and NO emissions from European forest soils. Journal of Geophysical Research, 97, 9777–9783. Klemedtsson, L., von Arnold, K., Weslien, P. and Gundersen, P. (2005). Soil N ratio as scalar parameter to predict nitrous oxide emissions. Global Change Biology, 11, 1–6. Kreutzer, K., Butterbach-Bahl, K., Rennenberg, H. and Papen, H. (2009). The complete nitrogen cycle of a N-saturated spruce forest ecosystem. Plant Biology, 11, 694–700. Kriebitzsch, W. U. (1978). Stickstoffnachlieferung in sauren Waldböden Nordwestdeutschlands. Scripta Geobotanica (Goltze, Göttingen, Germany), 14, 1–66. Kristensen, H. L., Gundersen, P., Callesen, I. and Reinds, G. J. (2004). Relationships between soil nitrate concentrations and environmental factors. Ecosystems, 7, 180–192. Laubhann, D., Sterba, H., Reinds, G. J. and de Vries, W. (2009). The impact of changes in atmospheric deposition and climate on forest growth in European monitoring plots:Â€An empirical tree growth model. Forest Ecology and Management, 258, 1751–1761. Li, C., Frolking, S. and Butterbach-Bahl, K. (2005). Carbon sequestration can increase nitrous oxide emissions. Climatic Change, 72, 321–338. Liu, L. and Greaver, T. L. (2009). Nitrogen addition stimulates emissions of biogenic greenhouse gases in terrestrial and wetland ecosystems. Ecology Letters, 12, 1103–1117. Lovett, G. M. and Lindberg, S. E. (1993). Atmospheric deposition and canopy interactions of nitrogen in forests. Canadian Journal of Forest Research, 23, 1603–1616. Ludwig, J., Meixner, F. X., Vogel, B. and Förstner, J. (2001). Soil-air exchange of nitric oxide:Â€an overview of processes, environmental factors, and modeling studies. Biogeochemistry, 52, 225–257. MacDonald, J. A., Dise, N. B., Matzner, E. et al. (2002). Nitrogen input together with ecosystem nitrogen enrichment predict nitrate leaching from European forests. Global Change Biology, 8, 1028–1033. Machefert, S. E., Dise, N. B., Goulding, K. W. T. and Whitehead, P. G. (2002). Nitrous oxide emission from a range of land uses across Europe. Hydrology and Earth System Sciences, 6, 325–337. Magnani, F., Mencuccini, M., Borghetti, M. et al. (2007). The human footprint in the carbon cycle of temperate and boreal forests. Nature, 447, 848–850. Magnani, F., Mencuccini, M., Borghetti, M. et al. (2008). Ecologically implausible carbon response? Reply. Nature, 451, E3–E4. Majdi, H. (2004). Root and needle litter decomposition responses to enhanced supplies of N and S in a Norway spruce forest in southwest Sweden. Plant Biosystems, 138, 225–230. Maskell, L. C., Smart, S. M., Bullock, J. M., Thompson, K. and Stevens, C. J. (2009). Nitrogen deposition causes widespread loss of species richness in British habitats. Global Change Biology, 16, 671–679. McClain, M. E., Boyer, E. W., Dent, C. L. et al. (2003). Biogeochemical hotspots and hot moments at the interface of terrestrial and aquatic ecosystems. Ecosystems, 6, 301–312.

122

McDowell, W. H., Magill, A. H., Aitkenhead-Peterson, J. A. et al. (2004). Effects of chronic nitrogen amendment on dissolved organic matter and inorganic nitrogen in soil solution. Forest Ecology and Management, 196, 29–41. McLauchlan, K. (2006). The nature and longevity of agricultural impacts on soil carbon and nutrients:Â€a review. Ecosystems, 9, 1364–1382. Meyer, A., Grote, R., Polle, A. and Butterbach-Bahl, K. (2009). Simulating mycorrhiza contribution to forest C- and N cycling – the MYCOFON model. Plant and Soil, 327, 493–517. Michalzik, B., Kalbitz, K., Park, J.-H., Solinger, S. and Matzner, E. (2001). Fluxes and concentrations of dissolved organic carbon and nitrogenÂ€– a synthesis for temperate forests. Biogeochemistry, 52, 173–205. Miranda, A. I., Marchi, E., Ferretti, M. and Millán, M. (2009). Forest fires and air quality issues in southern Europe. In:Â€Wildland Fires and Air Pollution. ed. A. Bytnerowicz, M. Arbaugh, A. Riebau and C. Anderse, New York: Elsevier, pp. 209–231. Misselbrook, T. H., Shepherd, M. A. and Pain, P. F. (1996). Sewage sludge applications to grassland:Â€Influence of sludge type, time and method of application on nitrate leaching and herbage yield. Journal of Agricultural Science, 126, 343–352. Moldan, F., Kjønaas, O. J., Stuanes, A. and Wright, R. F. (2006). Increased nitrogen in runoff and soil following thirteen years of experimentally-increased nitrogen deposition to a coniferousforested catchment at Gårdsjön, Sweden. Environmental Pollution, 144, 610–620. Moritsuka, N., Yanai, J., Mori, K. and Kosaki, T. (2004). Biotic and biotic processes of nitrogen immobilisation in the soil-residue interface. Soil Biology and Biochemistry, 36, 1141–1148. Müller, C., Kaleem Abbasi, M., Kammann, C. et al. (2004). Soil respiratory quotient determined via barometric process separation combined with nitrogen-15 labeling. Soil Science Society of America Journal, 68, 1610–1615. Müller, C., Rütting, T., Kattge, J., Laughlin, R. J. and Stevens, R. J. (2007). Estimation of parameters in complex 15N tracing models by Monte Carlo sampling. Soil Biology and Biochemistry, 39, 715–726. Murty, D., Kirschbaum, M. U. F., McMurtrie, R. E. and McGilvray, A. (2002). Does conversion of forest to agricultural land change soil carbon and nitrogen? A review of the literature. Global Change Biology, 8, 105–123. Myrold, D. D. and Posavatz, N. R. (2007). Potential importance of bacteria and fungi in nitrate assimilation in soil. Soil Biology and Biochemistry, 39, 1737–1743. Nadelhoffer, K. J. (2000). The potential effects of nitrogen deposition on fine-root production in forest ecosystems. New Phytologist, 147, 131–139. Nadelhoffer, K., Emmett, B. A., Gundersen, P. et al. (1999). Nitrogen deposition makes a minor contribution to carbon sequestration in northern forests. Nature, 398, 145–148. Näsholm, T., Kielland, K. and Ganeteg, U. (2009). Uptake of organic nitrogen by plants. New Phytologist, 182, 31–48. Nave, L. E., Vance, E. D., Swanston C. W. and Curtis, P. S. (2009). Impacts of elevated N inputs on north temperate forest soil C storage C/N, and net N-mineralization. Geoderma, 153, 231–240. Neill, C., Piccolo, M., Melillo, J., Steudler, P. and Cerri, C. (1999). Nitrogen dynamics in Amazon forest and pasture soils measured by 15N pool dilution. Soil Biology and Biochemistry, 31, 567–572. Nijburg, J. W., Coolen, M. J. L., Gerards, S., Gunnewiek, P. J. A. K. and Laanbroek, H. J. (1997). Effects of nitrate availability and the presence of Glyceria maxima on the composition and activity of

Klaus Butterbach-Bahl and Per Gundersen the dissimilatory nitrate-reducing bacterial community. Applied and Environmental Microbiology, 63, 931–937. Nilsson, L. O. and Wallander, H. (2003). Production of external mycelium by ectomycorrhyzal fungi in a Norway spruce forest was reduced in response to nitrogen fertilisation. New Phytologist, 158, 409–416. Nilsson, L. O., Baath, E., Falkengren-Grerup, U. and Wallander, H. (2000). Growth of ectomycorrhyzal mycelia and composition of soil microbial communities in oak forest soils along a nitrogen deposition gradient. Oecologia, 153, 375–384. Nilsson, L. O., Giesler, R., Baath, E. and Wallander, H. (2005). Growth and biomass of mycorrhyzal mycelia in coniferous forests along short natural nutrient gradients. New Phytologist, 165, 613–622. Nishio, T., Komada, M., Arao, T. and Kanamori, T. (2001). Simultaneous determination of transformation rates of nitrate in soil. Japan Agriculture Research Quarterly, 35, 11–17. Oenema, O., Witzke, H. P., Klimont, Z., Lesschen, J. P. and Velthof, G. L. (2009). Integrated assessment of promising measures to decrease nitrogen losses from agriculture in EU-27. Agriculture, EcosystemsÂ€and Environment, 133, 280–288. Ormeci, B., Sanin, S. L. and Peirce, J. J. (1999). Laboratory studies of NO flux from agricultural soil:Â€effects of soil moisture, pH and temperature. Journal of Geophysical Research, 104, 1621–1629. Palmroth, S., Oren, R., McCarthy, H. R. et al. (2006). Aboveground sink strength in forests controls the allocation of carbon below ground and its [CO2]-induced enhancement. Proceedings of the National Academy of Sciences of the USA, 103, 19362–19367. Papen, H., Gessler, A., Zumbusch, E. and Rennenberg, H. (2002). Chemolithoautotrophic nitrifiers in the phyllosphere of a spruce ecosystem recieving high atmospheric nitrogen input. Current Microbiology, 44, 56–60. Park, J. H. and Matzner, E. (2006). Detrial control on the release of dissolved organic nitrogen (DON) and dissolved inorganic (DIN) from the forest floor under chronic N deposition. Environmental Pollution, 143, 178–185. Parkin, T. B. (1987). Soil microsites as a source of denitrification variability. Soil Science Society of America Journal, 51, 1194–1199. Parrent, J. L. and Vilgalys, R. (2007). Biomass and compositional responses of ectomycorrhyzal fungal hyphae to elevated CO2 and nitrogen fertilisation. New Phytologist, 176, 164–174. Perakis, S. S. and Hedin, L. O. (2002). Nitrogen loss from unpolluted South American forests mainly via dissolved organic compounds. Nature, 415, 416–419. Pilegaard, K., Skiba, U., Ambus, P. et al. (2006). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 615–661. Pregitzer, K. S., Zak, D. R., Burton, A. J., Ashby, J. A. and MacDonald, N. W. (2004). Chronic nitrate additions dramatically increase the export of carbon and nitrogen from northern hardwood ecosystems. Biogeochemistry, 68, 179–197. Pregitzer, K. S., Burton, A. J., Zak, D. R. and Talhelm, A. F. (2008). Simulated chronic nitrogen deposition increases carbon storage in Northern Temperate forests. Global Change Biology, 14, 142–157. Raastad, I. A. and Mulder, J. (1999). Dissolved organic matter (DOM) in acid forest soils at Gårdsjön (Sweden):Â€natural variabilities and effects of increased input of nitrogen and of reversal of acidification. Water, Air and Soil Pollution, 114, 199–219. Rao, L. E., Allen, E. B. and Meixner, T. (2009). Risk-based determination of critical nitrogen deposition loads for fire spread in southern California deserts. Ecological Applications, 20, 1320–1335.

Rasse, D. P., Rumpel, C. and Dignac, M. F. (2005). Is soil carbon mostly root carbon? Mechanisms for a specific stabilisation. Plant and Soil, 269, 341–356. Recous, S., Mary, B. and Faurie, G. (1990). Microbial immobilisation of ammonium and nitrate in cultivated soils. Soil Biology and Biochemistry, 22, 913–922. Rennenberg, H., Dannenmann, M., Gessler, A. et al. (2009). Nitrogen balance in forest soils:Â€nutritional limitation of plants under climate change stresses. Plant Biology, 11, 4–23. Rice, C. W. and Tiedje, J. M. (1989). Regulation of nitrate assimilation by ammonium in soils and in isolated soil microorganisms. Soil Biology and Biochemistry, 21, 597. Roelofs, J. G. M., Kempers, A. J., Houdijk, A. L. F. M. and Jansen, J. (1985). The effect of air-borne ammonium sulphate on Pinus nigra var. maritima in the Netherlands. Plant and Soil, 84, 45–56. Ross, D. S., Lawrence, G. B. and Fredriksen, G. (2004). Mineralisation and nitrification patterns at eight northeastern USA forested research sites. Forest Ecology and Management, 188, 317–335. Rothe, A., Huber, C., Kreutzer, K. and Weis, W. (2002). Deposition and soil leaching in stands of Norway spruce and European beech:Â€results from the Höglwald research in comparison with other case studies. Plant and Soil, 240, 33–45. Roy, A. H. and Hammond, L. L. (2004). Challenges and opportunities for the fertilizer industry. In:Â€Agriculture and the Nitrogen Cycle (SCOPE Book 65), ed. A. Mosier, K. J. Syers and J. R. Freney, Washington DC:Â€Island Press, pp. 233–243. Rustad, L. E., Campbell, J. L., Marion, G. M. et al. (2006). A meta-analysis of the response of soil respiration, net nitrogen mineralisation and aboveground plant growth to experimental ecosystem warming. Oecologia, 126, 543–562. Samarkin, V. A., Madigan, M. T., Bowles, M. W. et al. (2010). Abiotic nitrous oxide emission from the hypersaline Don Juan Pond in Antarctica. Nature Geoscience, 3, 341–344. Scarascia-Mugnozza, G., Bauer, G. A., Persson, H., Matteucci, G. and Masci, A. (2000). Tree biomass, growth and nutrient pools. In:Â€Carbon and Nitrogen Cycling in European Forest Ecosystems, ed. E. D. Schulze. Berlin:Â€Springer, pp. 49–62. Schimel, J. P. and Bennett, J. (2004). Nitrogen mineralisation:Â€challenges of a changing paradigm. Ecology, 85, 591–602. Schmidt, I. K., Emmett, B. A., Tietema, A. et al. (2004). Soil solution chemistry and element fluxes in three European heathlands and their responses to warming and drought. Ecosystems, 7, 638–649. Seitzinger, S., Harrison, J. A., Böhlke, J. K. et al. (2006). Denitrification across landscapes and waterscapes:Â€a synthesis. Ecological Applications, 16, 2064–2090. Sexstone, A., Revsbech, N., Parkin, T. and Tiedje, J. (1985). Direct measurement of oxygen profiles and denitrification rates in soil aggregates. Soil Science Society of America Journal, 49, 645–651. Shaw, M. R. and Harte, J. (2001). Response of nitrogen cycling to simulated climate change:Â€differential responses along a subalpine ecotone. Global Change Biology, 7, 193–210. Shepherd, M. A., Hatch, D. J., Jarvis, S. C. and Bhogal, A. (2001). Nitrate leaching from reseeded pasture. Soil Use and Management, 17, 97–105. Sievering, H. (1999). Reply to Nitrogen deposition makes a minor contribution to carbon sequestration in temperate forests. Nature, 400, 629–630. Sievering, H., Tomaszewski, T. and Torizzo, J. (2007). Canopy uptake of atmospheric N deposition at a conifer forest:Â€part I – canopy N budget, photosynthetic efficiency and net ecosystem exchange. Tellus, 59B, 483–492.

123

Nitrogen processes in terrestrial ecosystems Sikström, U., Nohrstedt, H. Ö., Petterson, F. and Jacobson, S. (1998). Stem-growth response of Pinus sylvestris and Picea abies to nitrogen fertilisation as related to needle nitrogen concentration. Trees, 12, 208–214. Silver, W. L., Herman, D. J. and Firestone, M. K. (2001). Dissimilatory nitrate reduction to ammonium in upland tropical forest soils. Ecology, 82, 2410–2416. Simek, M., Jisova, L. and Hopkins, D. W. (2002). What is the so-called pH-optimum of denitrification? Soil Biology and Biochemistry, 34, 1227–1234. Sjöberg, G., Bergkvist, B., Berggren, D. and Nilsson, S. I. (2003). Long term N addition effects on the C mineralisation and DOC production in mor humus under spruce. Soil Biology and Biochemistry, 35, 1305–1315. Smil, V. (1999). Nitrogen in crop production:Â€an account of global flows. Global Biogeochemical Cycles, 13, 647–662. Smith, K. A. (1997). The potential for feedback effects induced by global warming on emissions of nitrous oxide by soils. Global Change Biology, 3, 327–338. Solberg, S., Dobbertin, M., Reinds, G. J. et al. (2009). The impact of changes in atmospheric deposition and climate on forest growth in European monitoring plots:Â€an empirical stand growth model. Forest Ecology and Management, 258, 1735–1750. Spangenberg, A. and Kölling, C. (2004). Nitrogen deposition and nitrate leaching at forest edges exposed to high ammonia emissions in Southern Bavaria. Water, Air and Soil Pollution, 152, 233–255. Sparks, J. P. (2009). Ecological ramifications of the direct foliar uptake of nitrogen. Oecologia, 159, 1–13. Staelens, J., Houle, D., De Schrijver, A., Neirynck, J. and Verheyen, K. (2008). Calculating dry deposition and canopy exchange with the canopy budget model:Â€review of assumptions and application to two deciduous forests. Water, Air and Soil Pollution, 191, 149–169. Stark, J. M. and Firestone, M. K. (1995). Mechanisms for soil moisture effects on activity of nitrifying bacteria. Applied and Environmental Microbiology, 61, 218–221. Stevens, R. J., Laughlin, R. J. and Maline, J. P. (1998). Soil pH affects the processes reducing nitrate to nitrous oxide and di-nitrogen. Soil Biology and Biochemistry, 30, 1119–1126. Stevens, C. J., Dise, N. B., Gowing, D. J. and Mountford, J. O. (2006). Loss of forb diversity in relation to nitrogen deposition in the UK:Â€regional trends and potential controls. Global Change Biology, 12, 1823–1833. Stevens, C. J., Dise, N. B., Mountford, J. O. and Gowing, D. J. (2004). Impact of nitrogen deposition on the species richness of grasslands. Science, 303, 1876–1879. Stockdale, E. A., Hatch, D. J., Murphy, D. V., Ledgard, S. F. and Watson, C. J. (2002). Verifying the nitrification to immobilisation ratio (N/I) as a key determinant of potential nitrate loss in grassland and arable soils. Agronomie, 22, 831–838. Stoddard, J. L., Jeffries, D. S., Lükewille, A. et al. (1999). Regional trends in aquatic recovery from acidification in North America and Europe. Nature, 401, 575–578. Subbarao, G. V., Ishikawa, T., Ito, O. et al. (2006). A bioluminescence assay to detect nitrification inhibitors released rom plant roots:Â€a case study with Brachiaria humidicoa. Plant and Soil, 288, 101–112. Suding, K. N., Collins, S. L., Gough, L. et al. (2005). Functional- and abundance-based mechanisms explain diversity loss due to N fertilisation. Proceedings of the National Academy of Sciences of the USA, 102, 4387–4392.

124

Sutton, M. A., Simpson, D., Levy, P. E. et al. (2008). Uncertainties in the relationship between atmospheric nitrogen deposition and forest carbon sequestration. Global Change Biology, 14, 2057–2063. Thomas, R. Q., Canham, C. D., Weathers, K. C. and Goodale, C. L. (2010). Increased tree carbon storage in response to nitrogen deposition in the US. Nature Geoscience, 3, 13–17. Tiedje, J. M. (1988). Ecology of denitrification and dissimilatory nitrate reduction to ammonium. In:Â€Biology of Anaerobic Microorganisms, ed. A. J. B Zehnder.Â€John Wiley and Sons, NewÂ€York: pp. 179–244. Tiedje, J. M., Sexton, A. J., Myrold, D. D. and Robinson, J. A. (1982). Denitrification:Â€ecological niches, competition and survival. Antonie van Leeuwenhoek, 48, 569–583. Tiessen, H., Stewart, J. W. and Bettany, J. R. (1982). Cultivation effects on the amounts and concentrations of carbon, nitrogen, and phosphorus in grassland soils. Agronomy Journal, 74, 831–835. Tietema, A. and Wessel, W. W. (1992). Gross nitrogen transformations in the organic layer of acid forest ecosystems subjected to increased atmospheric nitrogen input. Soil Biology and Biochemistry, 24, 943–950. Treseder, K. K. (2008). Nitrogen additions and microbial biomass:Â€a meta-analysis of ecosystem studies. Ecology Letters, 11, 1111–1120. Trinsoutrot, I., Recous, S., Mary, B. and Nicolardot, B. (2000). C and N fluxes of decomposing 13C and 15N Brassica napus L.:Â€effects of residue composition and N content. Soil Biology and Biochemistry, 32, 1717–1730. Vall, P. M. and Vidal, C. (1999). Nitrogen in Agriculture. ec.europa.eu/ agriculture/envir/report/en/nitro_en/report.htm Van Breemen, N., Boyer, E. W., Goodale, C. L. et al. (2002). Where did all the nitrogen go? Fate of nitrogen inputs to large watersheds in the northeastern U.S.A. Biogeochemistry, 57/58, 267–293. Van Cleemput, O. (1998). Subsoils:Â€chemo- and biological denitrification, N2O and N2 emissions. Nutrient Cycling in Agroecosystems, 52, 187–194. Van Egmond, N. D., Bresser, A. H. M. and Bouwman, A. F. (2002). The European nitrogen case. Ambio, 31, 72–78. Vanmechelen, L., Groenemans, R. and Van Ranst, E. (1997). Forest Soil Conditions in Europe: Results of a Large-Scale Soil Survey. Brussels: EC, UN/ECE, Ministry of the Flemish Community. Velthof, G., Barot, S., Bloem, J. et al. (2011). Nitrogen as a threat to European soil quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al. Cambridge University Press. Verchot, L. V., Holmes, Z., Mulon, L., Groffman, P. M. and Lovett, G. M. (2001). Gross vs net rates of N mineralisation and nitrification as indicators of functional differences between forest types. Soil Biology and Biochemistry, 33, 1889–1901. Vinten, A. J. A., Vivian, B. J., Wright, F. and Howard, R. S. (1994). A comparative study of nitrate leaching from soils of differing textures under similar climatic and cropping conditions. Journal of Hydrology, 159, 197–213. Vitousek, P. M. and Howarth, R. W. (1991). Nitrogen limitation on land and in the sea:Â€how can it occur? Biogeochemistry, 13, 87–115. Vitousek, P. M., Aber, J. D., Howarth, R. W. et al. (1997). Human alteration of the global nitrogen cycle:Â€Sources and consequences. Ecological Applications, 7, 737–750. Vitousek, P. M., Cassman, K., Cleveland, C. et al. (2002). Towards an ecological understanding of biological nitrogen fixation. Biogeochemistry, 57, 1–45.

Klaus Butterbach-Bahl and Per Gundersen Wan, S., Hui, D. and Luo, Y. (2001). Fire effects on nitrogen pools and dynamics in terrestrial ecosystems:Â€a meta-analysis. Ecological Applications, 11, 1349–1365. Watson, C. J. and Mills, C. L. (1998). Gross nitrogen transformations in grassland soils as affected by previous management intensity. Soil Biology and Biochemistry, 30, 743–753. Wesselink, L. G., Meiwes, K.-J., Matzner, E. and Stein, A. (1995). Longterm changes in water and soil chemistry in spruce and beech forests, Solling, Germany. Environmental Science and Technology, 29,Â€51–58. Wolf, B., Zheng, X., Brüggemann, N. et al. (2010). Grazing-induced reduction of natural nitrous oxide release from continental steppe. Nature, 464, 881–884.

Wrage, N., Velthof, G. L., Beusichem van, M. L. and Oenema, O. (2001). Role of nitrifier denitrification in the production of nitrous oxide. Soil Biology and Biochemistry, 33, 1723–1732. Yamulki, S., Harrison, R. M., Goulding, K. W. T. and Webster, C. P. (1997). N2O, NO and NO2 fluxes from a grassland:Â€effect of soil pH. Soil Biology and Biochemistry, 29, 1199–120. Yanai, R. D., Stehman, S. V., Arthur, M. A. et al. (2003). Detecting change in forest floor carbon. Soil Science Society of America Journal, 67, 1383–1593. Zumft, W. G. (1997). Cell biology and molecular basis of denitrification. Micobiology and Molecular Biology Reviews, 61, 533–616.

125

Chapter

7

Nitrogen processes in aquatic ecosystems Lead authors: Patrick Durand, Lutz Breuer and Penny J. Johnes Contributing authors: Gilles Billen, Andrea Butturini, Gilles Pinay, Hans van Grinsven, Josette Garnier, Michael Rivett, David S. Reay, Chris Curtis, Jan Siemens, Stephen Maberly, Øyvind Kaste, Christoph Humborg, Roos Loeb, Jeroen de Klein, Josef Hejzlar, Nikos Skoulikidis, Pirkko Kortelainen, Ahti Lepistö and Richard Wright

Executive summary Nature of the problem • Freshwater ecosystems play a key role in the European nitrogen (N) cycle, both as a reactive agent that transfers, stores and processes N loadings from the atmosphere and terrestrial ecosystems, and as a natural environment severely impacted by the increase of these loadings.

Approaches • This chapter is a review of major processes and factors controlling N transport and transformations for running waters, standing waters, groundwaters and riparian wetlands.

Key findings/state of knowledge • The major factor controlling N processes in freshwater ecosystems is the residence time of water, which varies widely both in space and in time, and which is sensitive to changes in climate, land use and management. • The effects of increased N loadings to European freshwaters include acidification in semi-natural environments, and eutrophication in more disturbed ecosystems, with associated loss of biodiversity in both cases. • An important part of the nitrogen transferred by surface waters is in the form of organic N, as dissolved organic N (DON) and particulate organic N (PON). This part is dominant in semi-natural catchments throughout Europe and remains a significant component of the total N load even in nitrate enriched rivers. • In eutrophicated standing freshwaters N can be a factor limiting or co-limiting biological production, and control of both N and phosphorus (P) loading is often needed in impacted areas, if ecological quality is to be restored.

Major uncertainties/challenges • The importance of storage and denitrification in aquifers is a major uncertainty in the global N cycle, and controls in part the response of catchments to land use or management changes. In some aquifers, the increase of N concentrations will continue for decades even if efficient mitigation measures are implemented now. • Nitrate retention by riparian wetlands has often been highlighted. However, their use for mitigation must be treated with caution, since their effectiveness is difficult to predict, and side effects include increased DON emissions to adjacent open waters, N2O emissions to the atmosphere, and loss of biodiversity. • In fact, the character and specific spatial origins of DON are not fully understood, and similarly the quantitative importance of indirect N2O emissions from freshwater ecosystems as a result of N leaching losses from agricultural soils is still poorly known at the regional scale. • These major uncertainties remain due to the lack of adequate monitoring (all forms of N at a relevant frequency), especiallyÂ€– but not onlyÂ€– in the southern and eastern EU countries.

Recommendations • The great variability of transfer pathways, buffering capacity and sensitivity of the catchments and of the freshwater ecosystems calls for site specific mitigation measures rather than standard ones applied at regional to national scale. • The spatial and temporal variations of the N forms, the processes controlling the transport and transformation of N within freshwaters, require further investigation if the role of N in influencing freshwater ecosystem health is to be better understood, underpinning the implementation of the EU Water Framework Directive for European freshwaters.


126

Patrick Durand, Lutz Breuer and Penny J. Johnes

7.1â•‡ Introduction The scope of this chapter is to document the extent of knowledge regarding the fate of nitrogen (N) in European freshwater systems (wetlands, standing and running waters, the hyporheic zone and groundwaters). Another aim is to highlight those areas where knowledge is currently sparse and future research is required to underpin the development of sound evidencebased environmental policy for the wide range of nutrient enriched waters across Europe. The key processes and controls of nitrogen turnover in freshwaters are discussed to understand the observed trends and the impacts of these processes on the ecological status and societal value of European freshwaters. The definition of groundwater used in this chapter includes the vadose zone beyond the reach of the root system of terrestrial vegetation, running waters are considered down to the limit of their tidal influence, wetlands are restricted to riparian areas receiving surface and groundwater but including the hyporheic zone, and standing waters include all freshwater lakes, ponds, pools and reservoirs. This chapter first describes the main factors controlling nitrogen cycling in freshwater systems, the distribution of N forms in waters and their origin, and the role of N in the ecology of those systems. The specific characteristics of N cycling in different types of freshwater systems are then highlighted.

7.2â•‡ Factors controlling N cycling in freshwaters 7.2.1â•‡ The water cycle Fresh water is defined as water with less than 0.5 g/l of dissolved salts. Some 3% of the water on Earth is freshwater. About two thirds of it is frozen in polar caps and glaciers, most of the remainder is present as groundwater, and only 0.3% is surface water. Freshwater ecosystems occupy over 3% of the Earth’s surface. In Europe, freshwaters cover 1% of the surface and wetlands 0.8% (European Environment Agency, 2005). The classical figure of the water cycle (FigureÂ€ 7.1) illustrates that one main feature of aquatic ecosystems is their

interconnectivity. Water infiltrating from terrestrial ecosystems recharges groundwater. In flat, low lying areas the groundwater table reaches the surface, determining the extent of wetlands. These wetlands are often located close to the streams or lakes (riparian wetlands). Streams and lakes are fed by groundÂ�water discharge, but also by surface overland flow and subsurface interflow in variable proportions. The direction of fluxes between groundwater, wetlands, streams and lakes varies according to hydrological conditions. Owing to natural or artificial obstacles to flow, surface waters create standing water bodies that are generally connected to the hydrological network. They can be located at the source of the streams or along the main course, and vary widely in extension and depth. Standing waters also occur in topographic depressions in the landscape including, for example, kettle holes in some post-glacial landscapes which are often less well connected to the running water network. The main hydrological driver is the excess rainfall, defined here as the amount of water available for groundwater recharge or runoff after interception and evapotranspiration. In Europe, this excess rainfall varies from a few mm/yr in the driest Mediterranean zones to more than 1000 mm in NorthWestern coasts. In most of Europe, it is between 150Â€mm and 500Â€mm (FigureÂ€7.2). The seasonality of lotic (running water) ecosystems is largely determined by the hydrological regime. In Europe, three major hydrological regimes exists (FigureÂ€ 7.3):Â€ (1) the oceanic temperate regime, with moderate variations in mean rainfall distribution over the year, and increased evapotranspiration during summer, leading to low summer discharge and winter flooding; (2) the Mediterranean regime with low rainfall and high evapotranspiration in summer leading to extremely low summer discharge and flooding in spring and autumn; and (3) the snowmelt controlled regime (mountainous and Nordic regions) with high discharge during spring or early summer. The first two regimes are characterized by low discharge during the periods with the highest temperature and light intensity, while in the snowmelt regime, the most productive period occurs before the summer flood, under sub-optimal light and temperature conditions.

FigureÂ€7.1 The global water cycle (courtesy of Sandra Süß).

127

Nitrogen processes in aquatic ecosystems

Lower reach systems (stream order >7) are still deeper and wider. They are no longer influenced by substantial lateral hydraulic dilution and show a decline of autochthonous primary production because of increased depth and turbidity which limits primary production capacity. They are again often heterotrophic, with stream metabolism strongly dependent on organic material brought in from upstream reaches. Sedimentation of fine material is possible and sediments are often rich in organic material.

7.2.3â•‡ Residence times

FigureÂ€7.2 Map of excess rainfall in Europe, based on long term rainfall data and computed actual evapotranspiration (Mulligan etÂ€al., 2006) (© 2006, JRC, European Commission).

7.2.2â•‡ Stream order The fate of nitrogen in freshwater depends strongly on the stream network geometry, which can be described in a synthetic and meaningful way by the stream order concept. The Strahler stream order system (Strahler, 1952) was proposed to define stream sizes based on a hierarchy of tributaries. Headwaters are first-order streams. When two first-order streams connect they form a second-order stream and so forth. The ecological functioning of water bodies, and therefore the N cycling, varies according to Strahler’s order as follows. Streams of Strahler’s order 1 to 3 are characterized by shallow depth and narrow width, steep slope and a relatively high contribution of lateral inputs of water with respect to the volume of the reach. Most inputs of energy are in the form of coarse organic material from riparian vegetation. Shading by riparian trees is common in these reaches, limiting light availability for stream flora. The overall metabolism of the system is typically heterotrophic, dominated by fungi and shredder invertebrates. Mid-reach river systems are wider, deeper and less strongly influenced by dilution. They receive more light so that autoÂ� chthonous primary production can occur either through macrophyte (typically stream order 4–5) or planktonic (typically stream order 5–7) development. The overall metabolism of the system becomes autotrophic. Organic matter, either of autochthonous origin or transferred from upstream systems, is dominated by fine particulate organic material and supports a community of collector or grazer invertebrates.

128

In all aquatic ecosystems, N cycling is controlled by the energy sources (light, organic matter and reduced inorganic compounds such as sulphur (S) or ferrous minerals), redox conditions (oxygen availability) and the nutrient loads. The main factor differentiating N turnover rates in the different types of freshwater systems, however, is the residence time of the water. The mean residence time of a well mixed reservoir is defined as the ratio between the volume of water (V) and the flux that goes through it (Q). Depending on the lines of flux, the actual residence time may vary considerably within a given water body:Â€ for example, in groundwater and lakes the residence time increases with depth, and in streams it is higher near the banks and the bed than in the middle of the stream. For running waters residence times are often defined for reaches (the portion of the stream between two confluences). For lakes and wetlands, residence time may be very short along the primary flow channels, but very long in areas less well connected to the primary flow channels. This is particularly evident in lakes formed over flooded river valleys, and in wetlands with a clear point of inflow and outflow. Typical residence times increase from order 1 streams (minutes to hours) to larger streams and wetlands (weeks), standing waters (weeks to decades) and groundwaters (decades to centuries). The highest variability in residence time is found in standing waters (from days to decades) and for groundwater (from months to thousands of years). Residence time also varies over time within a system. Bearing in mind the general definition of residence time as V/Q, since the flux is far more variable than the volume, the variability of residence time is controlled by the variations in discharge from the catchment. Therefore, Â�rivers with low flows in summer have residence times similar to riverine lakes, and numerous lakes and wetlands are subjected to flushing episodes during high flow events.

7.2.4â•‡ Nitrogen delivery to freshwaters Nitrogen can reach freshwaters through a number of pathways:Â€by atmospheric deposition on the catchment or directly on the water body; by leaching from diffuse sources within the catchment, such as those resulting from fertilizer and manure application; by sediment erosion of N rich soils and surface applications of manure in catchments; and by direct input from point sources such as sewage treatment works. A further source of reactive N (Nr) is nitrogen fixation. Some prokaryotes, such as some cyanobacteria, also possess the nitrogenase enzyme that allows atmospheric nitrogen to be converted into ammonia

Patrick Durand, Lutz Breuer and Penny J. Johnes

FigureÂ€7.3 Examples of the three major hydrological regimes of European rivers. Blue bars:Â€rainfall; yellow bars:Â€evapotranspiration; brown lines:Â€temperature; dark lines:Â€specific discharge.

and thus can exploit dissolved nitrogen gas in freshwaters. A few freshwater diatoms such as Epithemia and Rhapalodia also possess this ability via endosymbiotic inclusions believed to be derived from a cyanobacterium related to Cyanothece (Prechtl etÂ€al., 2004), although this mechanism is not widely found in European freshwaters. The nutrient load delivered to aquatic ecosystems from diffuse catchment and atmospheric sources depends strongly on the hydrological processes, particularly the relative importance of different water pathways in the transfer of the various N forms from terrestrial to aquatic systems (FigureÂ€7.4). Overland flow is responsible for the transport of particulate forms of N and, to a lesser extent, of dissolved organic N (DON) and ammonium. In agricultural areas, dissolved inorganic N (DIN) concentrations in overland flow are usually low compared to those of subsurface water, causing dilution of DIN in streamwater during flood events (Durand and Juan Torres, 1996; Durand etÂ€al., 1999; Kemp and Dodds, 2001). This is not always true in situations where subsurface waters are

low in DIN and where surface accumulation of N occurs, for example:Â€during snowmelt events when atmospheric N deposition has accumulated in the snowpack; shortly after fertilizer applications; in intensively farmed outdoor stock enterprises; in Mediterranean forested zones during floods (Bernal etÂ€ al., 2006; Johnes, 2007a). Most N leaving the soil in the form of nitrate will be transported to adjacent water bodies by water that has infiltrated in the soil, but the conditions controlling this transfer vary depending on the relative importance of shallow pathways (interflow, return flow, shallow groundwater seepage) and of deep infiltration (Creed etÂ€ al., 1996; Molenat etÂ€ al., 2002). DON is mostly delivered during storm events, with N-rich soil porewater flushed to the hyporheic zone and adjacent surface waters. Less DON is transported to groundwater stores in permeable areas with deep aquifers, where the pathway is dominated by nitrate flux. In these areas unsaturated vertical flow can be very long and lead to significant accumulation of nitrate in the vadose zone. The temporal variation of nitrate fluxes

129

Nitrogen processes in aquatic ecosystems FigureÂ€7.4 Schematic of the different flows in a headwater catchment. The colours of the arrows symbolize the age of water from the most recent (light blue) to the oldest (purple) (courtesy of Sandra Süß).

from groundwater to rivers is damped, and the mean residence time of water and solutes in these areas can be in the order of several decades (Wade etÂ€al., 2006).

7.3â•‡ Nitrogen forms and sources in freshwater 7.3.1â•‡ Particulate and dissolved N components Nitrogen cycling in aquatic ecosystems is complex and involves a variety of N forms and associated oxidation states. Both the oxidized and reduced inorganic N species (NO2−, NO3−, NH4+, NH3) and organic N fractions (DON, PON) are commonly found in all freshwater, estuarine and coastal waters across Europe. Nitrate, nitrite, ammonium and DON are directly available for plant uptake, supporting production in both the algal and higher plant communities. In addition the gaseous forms (N2, N2O, NO) are exchanged with the atmosphere. N speciation in streams varies along a gradient of N enrichment. Nitrate is the dominant form in highly enriched rivers, whereas DON is the dominant form in less enriched rivers. DON can also be an important secondary constituent of the TN load even in the most enriched rivers. FigureÂ€7.5 presents nitrogen species concentrations in different European streams ranging from oligotrophic to hypertrophic. Data were collated for the present work from all available national databases and both published and unpublished research on the relative proportion of total N present in the various N species forms in European rivers. Data were only included where at least the inorganic N species and total N had been determined at high sampling frequency (typically weekly to daily sampling frequency). For some sites only total organic N (TON) concentrations were reported (shown yellow on FigureÂ€ 7.5) alongside inorganic N species concentrations. For the majority of sites, however, the full species range had been determined, including both DON and PON fractions. A total of 84 separate annual records for 57 streams were found which fulfilled these criteria. Data have been plotted along a gradient of increasing total N concentrations, from the lowest concentrations typically found in boreal headwater streams to the highest concentrations found in rivers draining the intensively farmed lowland regions of Europe (FigureÂ€7.5).

130

FigureÂ€7.5 Concentrations of N species in 57 European streams and rivers (an explanation of data sources is given in the text. Stream ID, sampling locations and observation periods are provided in the Supplementary Material of Chapter 7).

Two clear patterns emerge from FigureÂ€ 7.5:Â€ nitrate concentrations increase in absolute terms and also as a proportion of total N along a gradient from ultra-oligotrophic to hyper-trophic waters, and mean annual DON concentrations also increase along this gradient but as a decreasing proportion of total N. Concentrations of nitrate range from 40% of the emissions at the global scale (Davidson and Kingerlee, 1997; IGAC, 2000), and >10% for some European countries (Butterbach-Bahl etÂ€al., 2004; Skiba etÂ€al., 1997; Stohl etÂ€al., 1996). Emissions resulting from fertilizer use could represent 40% of soil emissions at global scale (IGAC, 2000) and up to 65% for the USA (Hall etÂ€al., 1996). Rural agricultural areas receiving N fertilizers in countries with long dry periods are likely the largest sources of soil NO. The NitroEurope Integrated Project (NEU, 2010) and the NOFRETETE (Nitrogen oxides emissions from European Forest Ecosystems) project point at Europe forests as large sources of NO (Pilegaard etÂ€ al., 2006). The coniferous forest at Höglwald,

185

Nitrogen processes in the atmosphere

DE, receiving high atmospheric N deposition is a large source of NO, whereas the boreal forest at Hyytiälä, FI, and moorland sites in FI and UK have very small emissions (Skiba etÂ€al., 1997). In semi-natural/Â�natural ecosystems that do not receive N from fertilization or grazing, atmospheric N deposition significantly affect NO and N2O emissions. The NOFRETETE project showed correlation between N deposition rates and NO emissions from coniferous forest soils (Pilegaard etÂ€ al., 2006). Along a wet deposition gradient in Cumbria (17–40 kg N ha−1 yr−1), a linear relationship was observed between wet deposition of N, KCl extractable NH4+ and NO3− and NO and N2O emissions from semi-natural grassland on peat (Skiba etÂ€al., 2007).

9.3â•‡ Transformation of N compounds in theÂ€atmosphere The following section provides a description of atmospheric transformation processes of Nr and highlights where these processes play a significant role.

9.3.1â•‡ Reactions between NH3 and acid gases andÂ€aerosols In the reactions between gas phase NH3 and gas phase acids, new aerosol particles are formed. However, NH3 may also condense onto existing atmospheric particles. Gaseous NH3 will practically always react with sulphuric acid (H2SO4) in gas or aerosol phase, if H2SO4 is present. H2SO4 is formed from gas phase oxidation of SO2 by hydroxyl (OH) radical or from aerosol phase conversion by hydrogen peroxide (H2O2) and ozone (O3). The later process is pH dependent, and may in fact be catalysed by NH3, since uptake of NH3 increases the pH of aerosols (Apsimon etÂ€al., 1994; Junge and Ryan, 1958). The reaction between NH3 and H2SO4 is usually considered irreversible. In traditional CTM, it is thus common to describe the reaction as irreversible and taking place in two steps forming ammonium bisulphate (NH4)HSO4 and ammonium sulphate (NH4)2SO4 (Hov etÂ€al., 1994), respectively:

NH3 + H2 SO4 → NH4 HSO 4

(9.2)

NH3 + NH4 HSO4 → (NH4 )2 SO4 .

(9.3)

The rate of reaction between NH3 and H2SO4 has been analysed in detail in laboratory studies (Baldwin and Golden, 1979; Gupta etÂ€al., 1995; Huntzicker etÂ€al., 1980; McMurry etÂ€al., 1983). At high RH, the limiting factor for the transformation is the molecular diffusion of NH3 to the acid particles, whereas at low RH only between 10% and 40% of the collisions between NH3 gas molecules and H2SO4-containing particles lead to reaction (Huntzicker etÂ€ al., 1980; McMurry etÂ€ al., 1983). For small particles, the relatively large surface area makes the diffusion process more efficient. Organic material on the surface of the particles may, however, limit the uptake of NH3 (Daumer etÂ€al., 1992).

186

Whereas the NH3 reaction with H2SO4 may be considered irreversible, this is not the case for the reactions with other acid gaseous compounds. Presence of nitric acid (HNO3) and/or hydrochloric acid (HCl) together with NH3 lead to equilibrium between these gases and their aerosol phase reaction productsÂ€– the ammonium salts:Â€ammonium nitrate (NH4NO3) and ammonium chloride (NH4Cl). For the reaction with HNO3 this may be expressed as:

NH3 + HNO3 ↔ NH4 NO3 .

(9.4)

Experimental studies show that to a good approximation an equilibrium product, keq = [NH3][HNO3], of the gas phase concentrations of NH3 and HNO3 at saturation of the air, may be expressed by a function depending solely on temperature and humidity (Stelson etÂ€al., 1979; Stelson and Seinfeld, 1982). The RH at the point of deliquescence RHd [%] = 856.23/T + 1.2306, and ln( K ) = ln(K eq ) = 0.78 −

ln( K ) = ln( K eq ) −

24, 220 ln(T ) − 6.1 RH < RH d T 298

(9.5)

20, 75 + ln(K eq )  RH − RH d    RH ≥ RH d 101 − RH  100 − RH d  (9.6)

where T is in K. Besides the reactions with H2SO4 and HNO3, NH3 may also take part with HCl and form NH4Cl (Pio and Harrison, 1987a):

HNO3 + NaCl → NaNO3 + HCl

(9.7)

NH3 + HCl ↔ NH4 Cl.

(9.8)

Whereas HCl is a primary pollutant emitted by coal burning and incineration, HNO3 is the main secondary pollutant from oxidation of NOx emissions (see SectionÂ€ 9.3.3). New measurement data indicate that in NW Europe, HCl concentrations are similar to those of HNO3 in summertime, in terms of mixing ratio. However, NH4Cl concentrations are much lower than NH4NO3 concentrations. HCl is emitted from anthropogenic sources, but it is also released in displacement reaction in sea spray particles when these take up HNO3 (Wall etÂ€al., 1988): â•‡â•‡â•‡ NaCl + HNO3 → NaNO3 + HCl.

(9.9)

In the first EMEP model this reaction was accounted for by a first order decay of HNO3 of 10–5 s−1, and a reverse reaction rate coefficient of half this size (Hov etÂ€al., 1994). Measurements in California showed that NO3− in the coarseparticle mode is Â�primarily associated with high Na+ levels in marine air (Wall etÂ€ al., 1988), and that NO3− in course fraction particles has a peak at 3 μm diameter, where the product of the Na and mass distribution also peaks. This displacement reaction is thus the most likely explanation for

Ole Hertel

HCl concentrations of up to 250Â€ pptv (Harris etÂ€ al., 1992) observed in the marine boundary layer. Experimental studies have determined an equilibrium product at saturation of the air with the two gases NH3 and HCl (Pio and Harrison, 1987b). See also FigureÂ€9.13. NH4NO3 and NH4Cl are semi-volatile and the salts are deliquescent under most tropospheric conditions in northern Europe and may dissolve in pre-existing aerosol droplets or absorb onto the surface of any pre-existing aerosol particles. Thereby NO3−, Cl− and NH4+ are incorporated in suspended particulate matter in the particle size range, mainly in the submicron accumulation mode size-range, and therefore contribute to PM2.5 and PM10, the metrics used for human health assessment (Moldanova etÂ€ al., 2011, ChapterÂ€ 18 this volume). It may be reasonable to assume equilibrium of NH3 and HNO3, and NH3 and HCl. However, observations of particle size distribution of inorganic N, S, and Cl species in maritime air over the North Sea show products of partial pressures of [NH3][HNO3] and [NH3][HCl] that often fall below the theoretical lines of equilibrium (Ottley and Harrison, 1992), and one should therefore be careful when applying the assumption of equilibrium. This is due to sources and sinks, but mainly because the theoretical lines are for pure salts, while the co-existence of sulphate in the aerosol can dramatically decrease the equilibrium vapour concentration product.

9.3.2â•‡ Changed NH3 to NH4+ conversion rate due to changes in S emissions Early experiments carried out by Mckay (1971) showed that 50% of the available NH3 is converted into ammonium sulphate in about 35 minutes, based on concentrations present in the atmosphere at that time (20 µg m−3 SO2 and 2.7 µg m−3 NH3). Models like the EMEP Unified Model assume an instantaneous and irreversible formation of ammonium sulphate, only limited by the availability of the least abundant of NH3 and SO42−. Any excess NH3 may then react with HNO3, forming NH4NO3. Over the last decades, a dramatic decrease in SO2 emissions occurred (FigureÂ€9.14). Especially in the East European countries, the SO2 emissions dropped by approximately 60% in the late 1980s/early 1990s. This drop in SO2 emissions and resulting ambient concentrations has affected the formation of (NH4)2SO4. Measurements show that in the Netherlands in the early 1980s the NH3/NH4+ conversion rate was 28.8% per hour, while at present it is about 5% per hour (Van Jaarsveld, 2004). Trends in observations and EMEP model results for wet deposited N are in compliance with trends in emissions (Fagerli and Aas, 2008). For air concentrations less information is available, since most of EMEP sites did not start measuring TIA (Total Inorganic NHx = gas phase NH3 + aerosol phase NH4+) and TIN (Total Inorganic NO3− = HNO3 + aerosol phase NO3−) until the end of the 1980s and only a few

2007 Figure 9.13 HCl concentrations 2006 (left) and 2007 (right) based on 30 sites with monthly monitoring. Source:Â€Pollutant deposition (CEH, 2010).

187


sites (~20) have reported continuously. Moreover, the gas and particulate phases have very different chemical (e.g. their role in the NH4+–NH3–HNO3–NO3–SO42− equilibriums) and physical properties (e.g. the aerosols have a much longer residence time and are transported over longer distances), and the trend in the gas and particulate phase may thus be different. In some Eastern European countries NH3 emissions have declined by 30%–60%, but NHx concentrations decreased only by 20%–30% (EMEP, 2010). In Germany, however, NHx concentrations declined by 20%–30%, whilst emission reductions are reported to have been 10%–20%. The explanation is a combination of a less efficient formation of NH4+ aerosol (due to decreasing SOx) and less efficient dry deposition of NH3 due to less acidic surfaces; both effects leading to a shift towards gaseous NH3 relative to particulate NH4+.

9.3.3â•‡ NOy chemistry in the troposphere

The emission of NOx takes place mainly in the form of nitrogen monoxide (NO) and to a lesser extent (usually 5%–20%) as nitrogen dioxide (NO2). The fraction of directly emitted NO2 from road traffic in western countries has increased in recent years as a result of the use of catalytic converters. However, in the tropospheric boundary layer the distribution between NO and NO2 is governed to a large degree by O3 that reacts very fast with NO to form NO2. In sunlight NO2 photo dissociates (wavelengths 200–420 nm) to form NO and the very shortlived oxygen (O(3P)) radical. The latter will in most cases again form O3 in a reaction with free oxygen (O2).

NO + O3 → NO2 + O2

(9.10)

NO2 + hv → NO + O( 3 P)

(9.11)

O( 3 P) + O2 + M → O3 + M

In the above reactions M is a third body (either an N2 or O2 molecule) that absorbs excess vibrational energy and thereby stabilizes the formed O3 molecule. These reactions all have time scales of seconds to minutes. The reaction rate of reaction (9.9) is temperature dependent, but has a typical value about 4 × 10–4 ppbv−1 s−1. Under typical atmospheric boundary layer conditions, reaction (9.9) will either lead to the complete conversion of all the O3 to NO2, or to the conversion of all NO to NO2 (Clapp and Jenkin, 2001). In highly polluted atmosphere (e.g. an urban area) or close to pollution sources, the former behaviour is usually observed because although O3 is widely distributed in the lower atmosphere, its concentration is not usually high compared with NO in the highly polluted atmosphere, and hence O3 concentrations become rapidly depleted. During daylight, the main fate of NO2 is to undergo photolysis (9.10), reforming O3 (9.11) and NO (Dickerson etÂ€al., 1982). This reaction has a typical rate coefficient under summer Â�conditions in the mid afternoon at mid altitudes of about 7 × 10–3 s−1. Reaction (9.12) is the only production path for O3 in the atmosphere. FigureÂ€9.15 illustrates the NO–NO2–O3 chemistry in urban streets using a highly simplified module (Palmgren etÂ€al., 1996) in the Operational Street Pollution Model (OSPM) (Berkowicz, 2000). This module includes the reactions (9.10) and (9.11) and an assumption of reaction (9.12) being instantaneous. In adÂ�dition the model includes a ventilation rate between the street canyon and the surrounding air, and a distribution of the direct emission of NO and NO2 from street traffic (Palmgren etÂ€al., 1996).

80-03 80-98

–40

–20

0

20

Change in emission (in %)

188

40

(9.12)

60

80

Figure 9.14 EMEP emission changes for different European countries for two periods:Â€1980–1998 and 1980–2003. Source:Â€EMEP (2010).

Ole Hertel

Whenever NO is present, the most important atmospheric reaction of the hydroperoxy radical (HO2) is the conversion of NO to NO2:

NO + HO2 → NO2 + OH.

(9.16)

The hydroperoxy radical is one of many peroxy radicals that take part in the conversion of NO to NO2. Organic peroxy radicals (RO2) play likewise an important role and are mainly formed by the attack of the OH radical on the organic compounds ubiquitously present in the polluted atmosphere. These reactions follow a similar path as the CO oxidation, and may in a simplified form be presented as:

OH + RXH → R + H2 O

(9.17)

R + O 2 + M → RO2 + M.

(9.18)

RXH represents the organic compound, whereas R is an organic radical such as the alkyl radical and RO2 an alkyl peroxy radical. The only important atmospheric pathway of the alkyl radical is reaction (9.18) with O2 to form alkyl peroxy radicals (Finlayson-Pitts and Pitts, 1986). The RO2 radical may subsequently covert NO to NO2 in the same way as the HO2 radical (reaction (9.16)). During combustion processes at high temperatures, e.g. inside the motor of a petrol or diesel-driven vehicle, NO is formed from ambient N2. However, in the very NO rich air inside the exhaust pipe of vehicles and inside emitting chimneys, another oxidation path than (9.11) takes place:

Figure 9.15 The chemistry of NOx in urban streets. (Top) the observed relationship between NO2 and NOx. For NOx concentrations below about 20 ppb, all NOx is in the form of NO2, since the air contains sufficient O3 for converting all NO to NO2. For higher NOx concentrations, only the direct emission of NO2 contributes to further increase in NO2 concentrations. (Bottom) Comparison between observed and calculated hourly mean concentrations of NO2. All data are from Jagtvej in 2003, and calculations performed with OSPM. Only working days during daytime (800–1600) are included. Correlation coefficient (R2) = 0.7 (Hertel and Brandt, 2009).

The OH radical initiates the oxidation of a wide range of compounds in the atmospheric boundary layer. OH interacts with peroxy radicals that are responsible for the formation of excess concentrations of photo oxidants like O3. In the background troposphere, carbon monoxide (CO) plays a role in this system:

CO + OH → CO2 + H

(9.13)

H + O2 + M → HO2 + M.

(9.14)

This conversion is so rapid that for all practical purposes reaction (9.13) may actually be written as:

2 CO + OH + M O → CO2 + HO2 + M.

(9.15)

NO + NO + O2 → 2NO2 .

(9.19)

This reaction is a third order reaction with a second order dependence of the NO concentration, and has a reaction rate coefficient of 2.3 × 10–38 cm6 molecules−2 s−1 (Hampson and Gavin, 1978). The further transformation of NO2 to HNO3 takes place with a typical rate of about 5% per hour in the troposphere:

NO2 + OH → HNO3.

(9.20)

The hydroxyl radical (OH) is formed in the daytime in the presence of sunlight (Finlayson-Pitts and Pitts, 1986). The photo dissociation of O3 leads to the formation of both O(3P) and O(1D) radicals, a fraction of the latter reacts with water vapour to form two OH radicals:

O3 + hv → O2 + O( 3 P)

(9.21)

O3 + hv → O2 + O(1 D)

(9.22)

O( 1 D) + H2 O → 2OH.

(9.23)

Reaction (9.17) takes place in competition with O(1D)’s reaction with third body O2 or N2 molecules to form O(3P), that in turn reform O3 in reaction (9.12). The OH radicals initiate

189


most of the degradation of hydrocarbons in the atmosphere, a chain of reactions that e.g. lead to the formation of high O3 concentrations during summer. During night-time the nitrate (NO3) radical has a less important but still somewhat similar role for the degradation of hydrocarbons in the atmosphere as the OH radical in daytime. Despite the considerably lower reactivity compared with OH, its higher peak concentrations in the night-time troposphere allow the NO3 radical to play a major role in transformation of organic compounds. The NO3 radical is formed during nighttime in reaction with NO2. The dinitrogen pentoxide (N2O5) is a reservoir compound for the NO3 radical at low temperatures, but it is broken down to its precursors NO2 and NO3 at higher temperatures:

NO2 + O3 → NO3 + O2

(9.24)

NO3 + NO2 + M → N 2 O5 + M

(9.25)

N2 O5 + M → NO2 + NO3 + M.

(9.26)

The typical night-time NO3 radical concentrations in the atmospheric boundary layer are in the order 107 to 108 molÂ� ecules m−3 (which is the pptv range). During daytime both NO3 and N2O5 photo dissociate so fast that the concentrations of these compounds are insignificant. In the tropospheric boundary layer the photolysis of NO3 radical (with a typical noon lifetime of about 5 s) follow two different paths (λ represents here the wavelength):

NO3 + hv (ν < 700nm ) → NO + O2

(9.27)

NO3 + hv (ν < 580nm ) → O2 + O(3 P).

(9.28)

Close to pollution sources from combustion processes, e.g. road traffic or power plants, the NO3 radical is quickly removed by reaction with NO:

NO3 + NO → 2NO2 .

(9.29)

During night-time the heterogeneous conversion of N2O5 to HNO3 is an important process:

N 2 O5 + H2 O → 2HNO3 .

(9.30)

The lifetime of N2O5 with respect to this removal is on the order of minutes in the tropospheric boundary layer. This production of HNO3 may in winter be equally important as daytime conversion of NO2 by OH radical. As already described, particulate nitrate (NO3−) is formed when HNO3 reacts with NH3 and form new aerosol particles, and when it sticks to existing particles in the atmosphere. In addition organic NO3− may be formed from gaseous NO2 on the surfaces of aerosols in other heterogeneous reactions (see SectionÂ€9.3.4). The NO3 radical attacks alkanes by hydrogen abstraction in a similar way as the reactions of the OH radical:

190

RH + NO3 → R + HNO3 .

(9.31)

Followed by formation of a peroxy radical (RO2) that may oxidize an NO molecule to NO2. Also for alkenes, the attack of the NO3 radical is similar to the reactions of the OH radical; the NO3 radical adds to the double bond. This reaction is followed by rapid O2 addition which leads to the production of a peroxy radical. Reaction of soil emissions of NO with atmospheric OH has been suggested to provide an in-canopy source of HNO3 (Farmer etÂ€al., 2006). HNO3 concentrations usually peak during the day, regulated by the emissions of NOx, photochemical activity and the gas/aerosol equilibrium of NH4NO3 shifting towards the gas phase with increasing temperature and decreasing relative humidity. A reaction similar to reaction (9.22), but less important, is the reaction between NO and OH radical forming nitrous acid (HONO):

NO + OH + M → HONO + M.

(9.32)

During daytime HONO photo dissociate (λ < 400 nm) rapidly back to the reactants:

HONO + hv → NO + OH.

(9.33)

Therefore, HONO formed in the late evening may serve as a night-time reservoir of OH and NO, which are subsequently liberated again the following morning, when sunlight starts up reaction (9.23) again. Studies in the highly polluted Po Valley in Northern Italy have shown an interrelation between simultaneous peaks in NOx concentrations and aerosol surfaces and peak HONO concentrations during foggy periods (Notholt etÂ€al., 1992). This was taken as evidence for heterogeneous Â�conversion on aerosol surfaces through either of the reactions:

NO + NO2 + H2 O → 2HONO

(9.34)

2NO2 + H2 O → HNO3 + HONO.

(9.35)

Probably this type of conversion plays an important role also in many urban areas over Europe, but so far only few studies have been carried out.

9.3.4â•‡ Organic N compounds in the atmosphere The presence of atmospheric organic N compounds has been evident for years (Cornell etÂ€ al., 2003; Neff etÂ€ al., 2002), but direct measurements of individual species are rather sparse. The evidence of organic N comes from analysis of precipitation samples for total N and comparison with inorganic N content, to give ‘dissolved organic N’ (DON) by difference. This approach has been prone to several analytical artefacts (Cape etÂ€al., 2001), but DON may in fact contribute up to half of total water-soluble N in precipitation. The fraction depends highly on location, and on whether air masses are of marine or terrestrial origin. DON has been ignored in estimating environmental consequences of N deposition, although there is reason to believe that many, if not all, components of DON are biologically available (Krab etÂ€ al., 2008; Lipson and Nasholm,

Ole Hertel

2001; Paerl and Whitall, 1999; Qualls and Haines, 1992). There appears to be a DON contribution from marine air masses (Cornell etÂ€al., 1995, 2001), and DON proportions are consistently high in unpolluted air, especially in the tropics. For continental/terrestrial samples, annual average concentrations of DON in precipitation appear to correlate better spatially with NH4+ than with NO3−, suggesting an agricultural source, but the seasonal variation is not correlated with NH4+ concentrations, implying that different sources are involved (Cape etÂ€al., 2004). There is limited evidence available from sampling of air diÂ�rectly using a nebulizing mist sampler that both gas phase and particle phase components contribute to water-soluble organic N (WSON) in the atmosphere, which leads to the occurrence of DON in precipitation. Organic N is measured in fog (Zhang and Anastasio, 2001) and cloud water (Hill etÂ€ al., 2007), but there is some concern that most analyses for DON are made on bulk rainfall samples (i.e. collected using an open funnel) and that a significant fraction of the measured DON might have been dry-deposited on the funnel surface. This presents problems of interpretation, but does not remove the problem of identifying the source, composition and fate of organic N compounds (see also FigureÂ€9.16).

nitrates that serve as important reservoirs of NO2. The most abundant of these nitrates is the peroxy acetyl nitrate (PAN):

CH3 CHO + hv → CH 3C(O)

(9.36)

CH3 CHO + OH → CH 3 C(O) + H2 O

(9.37)

CH3 C(O) + O2 → CH3 C(O)OO

(9.38)

CH3 C(O)OO + NO2 + M → CH3 C(O)OONO2 + M (9.39) Reaction (9.38) is very fast, and reactions (9.38) and (9.39) may for many practical purposes be regarded as taking place in one step. PAN is thermally unstable and equilibrium between peroxy acetyl radical and NO2 on one side and PAN on the other side is established in the boundary layer. High PAN and O3 concentrations are often observed together during photo chemical smog episodes. The thermal degradation gives PAN a lifetime of ~ 1.7 h at 273 K and 50 h at 263 K. The PAN formation is competing with NO degradation of peroxy acetyl radical: CH3 C(O)OO + NO → CO2 + NO2 + CH3O2

Evidence for reduced organic N in the atmosphere Some measurements of individual components of reduced organic N in gas, particulate and aqueous phases have been reported and indicate potential sources and fates of these compounds (TableÂ€9.3), but in most cases rather small concentrations are measured and these cannot account for the rather high proportions of DON in precipitation.

Formation of organic oxidized N compounds When aldehydes are photo dissociated or react with OH, an alkyl radical is formed, which in turn may form peroxy alkyl

This reaction is totally dominating at ppbv levels of NO meaning that PAN and other peroxy alkyl nitrates are formed only in the background atmosphere, and i.e. not inside urban areas. However, substantial PAN concentrations may still be observed in urban areas, especially at relatively low temperatures. The peroxy alkyl nitrates include compounds Â�produced in a similar way as PAN, but generated from biogenic isoprene emissions that may be of importance in southern Europe, and have similar thermal degradation pathways as PAN.

Figure 9.16 Illustration of the interaction between the various nitrogen oxide (NOy) compounds in the tropospheric boundary layer. The symbol Δ represents energy leading to thermal degradation, hν solar radiation leading to photo dissociation and RH a hydrocarbon reacting with the specie in question. PPN is a notation for peroxy propionyl nitrate, but also other peroxy nitrates than PPN and PAN play a role in this context (Derwent and Hertel, 1998).

HONO OH, HONO, NO2

H, H

O

NO

hυ

O

NO

hυ

hυ

,N

O

2

O3, HO2, RO2, NO3

ROO2

NO2

∆ HO2

∆ , hυ

HO2NO2

∆,

hυ

NO

NO

3

NO3

hυ, O3, NO

N2O5

2

∆,

hυ

RH

PAN PPN

(9.40)

OH

HNO3

191

Nitrogen processes in the atmosphere TableÂ€9.3 Reported data on reduced organic N compounds in the atmosphere

Species

Sources

Atmospheric reactions and fate

Other direct measurements

Amines and Urea

Direct emissions from some industrial processes and from agricultural activity such as slurry spreading (Kallinger and Niessner, 1999). Some data exist for the latter process relative to ammonia emissions, which may be helpful in modelling emissions, but release rates are c. 1% of NH3. Ocean water may be source or sink (Hatton and Gibb, 1999).

Water soluble (Calderon etÂ€al., 2007; Gibb etÂ€al., 1999). Likely acid–base reactions and form particles (cf. NH4+) (Angelino etÂ€al., 2001; Blando and Turpin, 2000). Photo degradation possible (McGregor and Anastasio, 2001). O3 reaction (Tuazon etÂ€al., 1994). Likely high dry deposition rates (low surface resistanceÂ€– ‘sticky’ molecules).

PTR-MS data on trimethylamine from slurry (Twigg, 2006). Published data (Beddows etÂ€al., 2004; Gorzelska etÂ€al., 1992; Gorzelska etÂ€al., 1994; Gronberg etÂ€al., 1992; Gundel etÂ€al., 1993; Palmiotto etÂ€al., 2001; Schade and Crutzen, 1995) (Cornell etÂ€al., 1998; Mace etÂ€al., 2003b; Mace etÂ€al., 2003c; Mace etÂ€al., 2003a; Mace and Duce, 2002).

Amino acids, proteins and peptides

Ocean surface (Milne and Zika, 1993), grassland (soil) (Scheller, 2001).

Photodestruction (Anastasio and McGregor, 2000; McGregor and Anastasio, 2001; Milne and Zika, 1993; Saxena and Hildemann, 1996).

Measurement data (Gorzelska etÂ€al., 1992; Kieber etÂ€al., 2005; Mace etÂ€al., 2003b; Mace etÂ€al., 2003c; Mace etÂ€al., 2003a; Matsumoto and Uematsu, 2005; Scheller, 2001; Zhang and Anastasio, 2001; Zhang and Anastasio, 2003).

Hydrogen cyanide and methyl cyanide (Acetonitrile)

Biogenic emissions (terrestrial) (Shim etÂ€al., 2007) and biomass burning (Bertschi etÂ€al., 2003; Cicerone and Zellner, 1983; Holzinger etÂ€al., 1999; Li and Tan, 2000); anthropogenic (Holzinger etÂ€al., 2001).

(Bange and Williams, 2000; Cicerone and Zellner, 1983; Karl etÂ€al., 2004). Other direct measurements:Â€(Sprung etÂ€al., 2001; Warneke etÂ€al., 2001).

9.4â•‡ Dry deposition and bi-directional fluxes of N compounds Nr compounds are being monitored in many regional networks across the world, such as the European EMEP programme (EMEP, 2010), the NitroEurope Integrated Project (NEU, 2010) the US National Atmospheric Deposition Network (NADP, 2010), the Acid Deposition Monitoring Network in East Asia (EANET, 2010) and several others. However, these networks measure air concentrations rather than fluxes, and deposition is estimated using inferential modelling approaches, which are underpinned by often sparse databases of campaign based process studies with limited geographical coverage. This is partly due to the fact that instrumentation to measure fluxes of sticky compounds such as NH3, HNO3 or HONO are expensive and labour intensive to operate. The measurement of each individual Nr compound is technically more challenging than that of CO2 fluxes, for example. Robust low cost flux measurement approaches are lacking, although recent developments of a Conditional Time-Averaged Gradient (COTAG) method (Famulari etÂ€al., 2010) show promise for wide-scale deployment over long periods for short vegetation. A first regional flux measurement network for Nr compounds is established within the European NitroEurope IP. This network takes a three-tier approach, where selected Nr compounds

192

are measured at a network of 13 supersites, using advanced micrometeorological flux measurement techniques. At a further nine regional sites the novel COTAG systems are being deployed, while deposition is derived at a further 50+ sites from concentration measurements, using inferential techniques (Tang etÂ€al., 2009). Spatial coverage of Nr deposition can only be achieved through numerical modelling. The gaseous Nr compounds most commonly considered for dry deposition are NH3, HNO3 and NO2. Their relative contributions to N deposition depend on the pollution climate. In agricultural areas NH3 may dominate the atmospheric N loading, while in more industrial and urban areas HNO3 and NO2 may be more important. In addition, NH3 deposition depends on the N status of the receiving surface, with fertilized vegetation and vegetation receiving high atmospheric N deposition inputs acting as a less efficient sink or even net source of NH3. In dry regions, stomatal deposition may make a larger relative contribution to net exchange than in wet regions, where leaf cuticles provide a very efficient sink for water soluble gases (NH3 and HNO3). In the UK, dry deposition of NH3, HNO3 and NO2 is estimated to have contributed 48 (14.5%), 57 (17.3%) and 9 (0.03%) Gg, respectively, to the total N deposition of 330 Gg N in 2004, with the rest originating from wet and cloud deposition (211 Gg, 63.9%) and aerosol deposition (16 Gg N, 0.05%) (Fowler

Ole Hertel

etÂ€ al., 2009). The UK deposition model uses detailed knowlÂ� edge of land use to estimate the vegetation-dependent deposition velocities and fluxes as a function of land use in each 5â•›km × 5â•›km grid square of the country, combined with longterm measurements of air concentrations which are unique in Europe in terms of spatial coverage.

9.4.1â•‡ The dry deposition process Dry deposition is the direct deposition of gases or aerosols at terrestrial or marine surfaces. The dry deposition of gases and particles is a continuous process and governed by their air concentrations, turbulent transport processes in the boundary layer, the chemical and physical nature of the depositing species, and the capability of the surface to capture or absorb the species. In relation to deposition transport, the boundary layer may be considered to consist of two layers:Â€the fully turbulent layer and the quasi-laminar layer. The quasi-laminar layer is introduced to quantify the way in which pollutant transfer differs from momentum transfer in the immediate vicinity of the surface (Hicks etÂ€al., 1987). In this layer, the transport is dominated by molecular diffusion. Once at the surface, the chemical, biological and physical nature of the surface determines the capture or absorption of the gases and particles. Deposition to water surfaces (oceans or fresh waters) may thus be very different from deposition to vegetated surfaces on land. The deposition process may be considered as a series of resistances, by analogy with an electrical circuit (Monteith and Unsworth, 2008). The resistances refer to the transport proÂ� cesses through the various ‘layers’ defined above:Â€ turbulent transfer (usually denoted Ra), quasi-laminar (Rb) and surface (Rc). For a complex surface with several potential absorption sinks (e.g. vegetation) the resistance Rc may be viewed as a network of parallel resistances, representing transfer to the external leaf surface, through stomata, to water on the surface, or through the canopy to the underlying soil surface. The total resistance (RT) is the sum of all the series and parallel resistances (RaÂ€+Â€RbÂ€+Â€Rc), and is usually expressed in units of sÂ€m−1. The inverse of the total resistance (1/RT) is known as the deposition velocity (vd) and has units of m s−1. The turbulent transfer resistance (Ra) depends upon the height at which the deposition flux is measured, so the total resistance (RT) and deposition velocity (vd) also vary with height above the surface. The transfer flux (F) is defined as the product of the air concentration of a gas or particles at height z, multiplied by the deposition velocity at height z, and (in the absence of competing chemical reactions (Sorensen etÂ€ al., 2005)) does not vary with height, provided that the air concentration is horizontally uniform. This formulation assumes that the surface concentration of the gas is zeroÂ€– where this is not the case (see below) the effect can be described either as a decreased driving force for deposition (concentration difference between height z and the surface) or as an increased surface resistance. The deposition velocity (vd) is often reported as a constant even though it depends on a set of variables, e.g. wind

speed, surface roughness and atmospheric stratification. Joffre (1988) has suggested a parameterization which depends on the meteorological conditions, roughness length and the molecular diffusion coefficient for the compound of interest. The various components of the total transport resistance can be estimated from meteorological data if several assumptions are made concerning spatial and temporal homogeneity. The atmospheric turbulent resistance (Ra) can be calculated from:

â•‡â•‡â•‡

Ra ( z ) =

1  z  z z0    ln −ψ  ,   , κ u *  z0  L L 

(9.41)

where z is the reference height, z0 is the roughness length, u* is the friction velocity, κ is the von Karman constant (≈ 0.4) and L is the Monin–Obukhov length. For the diabatic surface layers (Businger, 1982) a stability function ϕ is introduced (Businger etÂ€al., 1971). For neutral conditions ϕ = 1 and ϕ is greater/less than unity for stable/unstable stratifications. In the above equation, ψ is the integrated stability function. The resistance of the underlying thin molecular lamÂ� inar sub-layer is given by (Kramm, 1989; Kramm etÂ€al., 1991; Kramm and Dlugi, 1994): Rb = ∫

z

zs

Di + K

=

  uz + Bi   u u 

(9.42)

â•‡â•‡â•‡ where uz0 is a characteristic velocity for the layer zs < z < z0, Bi is the sub-layer Stanton number, which is a function of the roughness Reynolds number Re* = u* z/ν and the Schmidt number, Sci = ν/Di. The Stanton number can be estimated as (Kramm and Dlugi, 1994): Bi−1 = aSc ib Re *c + ε , (9.43) where the following values are suggested for smooth surfaces a = 13.6, b = 2/3, c = 0 and ε = −15.5 and for rough surfaces, a = 7.3, b = 0.5, c = 0.25 and ε = −5. The surface resistance term depends on the physical and chemical nature of the absorbing surface, and parameterizations should be adapted to the surface concerned. The value of vd is often expressed as annual or seasonal averages, for the purpose of calculating deposition fluxes as the product of air concentrations and deposition velocities. Deposition velocities and concentrations should refer to the same heightÂ€– usually the height at which the concentrations are measured. Tall vegetation causes increased atmospheric turbulence, so Ra values are smaller, and deposition velocities are larger, than for short vegetation. Consequently, estimating deposition of different components to the countryside requires knowledge about land use as well as the spatial pattern of air concentrations. The air–sea gas exchange of the very soluble gases HNO3 and NH3 is rate limited by the vertical transport in the boundary layer, because the uptake at the water surface is very fast relative to other commonly studied gases. Of the two very soluble N-gases, HNO3 exchange rates are larger than NH3 due

193


to the higher solubility. The less soluble NO2 and NO gases, deposit much slower to the marine surface. The surface resistance is the most important resistance for slightly soluble gases and relates to the transfer velocity Kc, which is also used for air-sea exchange of other gases like CO2, DMS and CH4. The surface resistance is a key parameter for the deposition of a gas to a water surface, and may be expressed as:

F 1 = Kc = c . ∆c w Rc

(9.44)

Here Fc is the flux across the surface and Δcw is the concentration gradient across the laminar sub-layer in the water. The resistance across the water surface is controlled by the Henry’s law coefficient (H), which describes the solubility of different gases, and is a strong function of temperature. The effective overall surface resistance is therefore:

Rc ,e ff = Rc H *,

(9.45)

where H* is the dimensionless Henry’s law coefficient (TableÂ€9.4). The process of dry deposition of particles differs from that of gases in two respects. • Deposition depends on particle size, since transfer to the surface involves Brownian diffusion, inertial impaction/ interception and sedimentation (all of which are a strong function of particle size) (Slinn, 1982). • It is assumed that the surface resistance for particles less than 10 μm diameter (Hicks and Garland, 1983) is negligibly small to all surfaces. For submicron particles, the transport through the boundary layer is more or less the same as for gases. However, transport of particles through the quasi-laminar layer can differ. For particles with a diameter 10 μm is more controlled by sedimentation. Deposition of particles with a diameter between 0.1 μm and 1 μm is determined by the rates of impaction and interception and depends heavily on the turbulence intensity. Transfer through the quasi-laminar layer close to the surface presents a considerable restriction on the deposition of 0.1–1.0 µm diameter particles. Uptake of particles by surfaces is thus largely controlled by micro-structures and turbulence intensity. Most of the theory and measurements of particle fluxes have focused on sulphate particles (SO42−), which mostly occur in the submicron size range as (NH4)2SO4. Other submicron aerosol particles are expected to behave similarly, although semi-volatile particles may form or evaporate depending on the local equilibrium with the constituent gases (e.g. NH4NO3 and NH3/HNO3). The most widely used model is an empirical parameterization (Wesely etÂ€ al., 1985), which is based upon flux measurements of SO42− over grass. In this model, vd is represented as a function of the friction velocity u* and the Monin–Obukhov length L. Then for SO42− particles and low vegetation, vd can be calculated by using (Erisman and Draaijers, 1995):

194

vd = vd =

2 /3 u*   300   ⋅ 1 +    500   − L  

u* 500

L0

(9.46)

Ruijgrok proposed another parameterization derived from measurements over coniferous forest (Ruijgrok etÂ€ al., 1997). In this approach, which is simplified from the Slinn model (Slinn, 1982), vd is not only a function of u*, but also of relative humidity (RH) and surface wetness. Inclusion of RH allows for particle growth under humid conditions and for reduced particle bounce when the canopy is wet. Dry deposition velocity is expressed as: 1 1 = Ra + , vd vds

(9.47)

where Ra is the aerodynamic resistance, which is the same as for gaseous species, and vds is the surface deposition velocity. For tall canopies vds is parameterized by (Ruijgrok etÂ€ al., 1997) as vds = E ⋅

u 2* , uh

(9.48)

where uh is the wind speed at the top of the canopy, which is obtained by extrapolating the logarithmic wind profile from ZR to the canopy height h. Now uh can be expressed as:

uh =

u* k

  10 ⋅ z0 − d   z0    10 ⋅ z0 − d   ln   −ψ h  h  + ψh    . (9.49) z L  L    0   

Note that E is the total efficiency for canopy capture of particles, parameterized for dry and wet surface separately (Erisman etÂ€al., 1997). For dry surfaces, for SO42− particles (Brook etÂ€al., 1999): 0.005 u 0.28 *  E = 0.28 0.005 u * 

RH − 80   ⋅ 1 + 0. 18 ⋅ exp 20  

RH ≤ 80%

(9.50)

RH > 80% .

For wet surfaces, for SO42− particles (Brook etÂ€al., 1999): 0.08 u*0.45  E = RH − 80  0.45  0.08 u* ⋅ 1 + 0.37 ⋅ exp 20    

RH ≤ 80%

RH > 80% ,

where RH is taken at the reference height.

(9.51)

Ole Hertel

Erisman and Draaijers used the following general form for the calculation of vd (Erisman and Draaijers, 1995):

vd =

1 Ra +

1 vds

+ vs ,

Compound

(9.52)

where vs is the deposition velocity due to sedimentation, to represent deposition of large particles, and vds can be estimated from (9.48). Relations for E for different components and conditions may be derived from model calculations and multiple regression analysis (Erisman and Draaijers, 1995). For larger supermicron particles (Na+, Ca2+ and Mg2+), and therefore for some NO3− particles, and for low vegetation (for all particles), the sedimentation velocity has to be added:

vs = 0.0067 m ⋅ s −1 vs = 0.0067 ⋅ e

0.0066⋅ RH 1.058 − RH

RH ≤ 80 m ⋅ s −1

RH > 80%.

TableÂ€9.4 Summary of Henry’s law coefficients of various gaseous nitrogen compounds

(9.53)

For sulphate deposition velocity, observations suggest that there is a distinct upper limit which depends on land use type. As a result, it is required that

vds ≤ vm , (9.54) where vm is the observed maximum deposition velocity (Walcek etÂ€al., 1986).

9.4.2â•‡ Bi-directional fluxes of N-containing gases Plant fixation of N2 provides the single largest atmospheric N input to the biosphere worldwide. However, since it is not associated with acidifying effects, controlled by the plants themselves and its rate is not altered through human activity (other than through land-use change), it is not usually considered in atmospheric N deposition budgets. Direct measurement approaches of N2 fixation are lacking as the flux to the biosphere is very small compared with ambient N2 concentrations. Instead N2 fixation is measured in laboratory, e.g. with isotope techniques. Nitrous oxide (N2O), an important greenhouse gas with a lifetime of 114 years, is usually assumed to be emitted by terrestrial surfaces (see Jarvis etÂ€al., 2011, ChapterÂ€10, this volume). Although reports of transient N2O deposition fluxes in the literature are increasing in number, see, for example, Flechard etÂ€al. (2007), the magnitude of N2O uptake is small and negligible compared with the main contributors to atmospheric N deposition. For the other N containing gases there are several parallel pathways of pollutant exchange with vegetation, which include adsorption to the leaf cuticles, exchange through the stomata with the sub-stomatal cavity and exchange with the soil. All these processes are potentially bi-directional, depending on the relative magnitude of the air concentration and the gaseous concentrations in chemical equilibrium with the leaf surface, the apoplastic fluid and the soil solution, respectively. The likelihood for uptake increases with the water solubility

Henry’s law coefficient at 25 °C in water [mol kg−1 bar−1]

NH3

61

HNO3

2.6 × 106

HONO

49

NO

0.0019

NO2

0.012

N2O

0.025

PAN

4.1

and Henry’s law coefficient of the gas, which vary over Â�several orders of magnitude (TableÂ€ 9.4). A database of Henry’s law coefficients is available (Mainz, 2010).

Nitric acid Because of its high deposition rate, HNO3 makes a significant contribution to Nr deposition in areas exposed to air containing emitted NOx. HNO3 is highly water soluble and commonly assumed to deposit at the maximum rate permitted by turbulence, i.e. surface resistance is negligible. While this is probably a reasonable approximation for most situations, several authors have observed emission gradients or reduced uptake rates of HNO3, probably owing to non-zero HNO3 surface concentrations in equilibrium with NH4NO3 aerosol deposited to leaf surfaces (Neftel etÂ€al., 1996; Nemitz etÂ€al., 2004; Zhang etÂ€al., 1995). In the case of trace gases with negligible surface resistance, the deposition velocity is very sensitive to the atmospheric resistances (Ra and Rb), which over aerodynamically rough surfaces are small (5–10 s m−1). In such conditions, even a very small surface resistance for HNO3 would strongly influence deposition rates. Currently there are insufficient field data to show whether HNO3 deposition is subject to a surface resistance, and this remains a research priority.

Ammonia NH3 dominates atmospheric N deposition to semi-natural vegetation in agricultural areas, especially in Northern Europe where NH3 deposition is favoured at high humidity and cold temperatures, although, these conditions also favour conversion to ammonium aerosol. NH3 is less water soluble than HNO3. Thus, NH3 previously absorbed to wet leaf surfaces may more readily be desorbed (re-emitted) as leaf water layers dry out again (Flechard etÂ€al., 1999). Another complication is that plants under certain conditions may release NH3. Generally plants contain inorganic N in the form of NH4+ and NO3−. These nutrients are mainly present in the liquid part (apoplast) between the cells of the plant. NH4+ is an important by-product of plant biochemical pathways resulting in non-zero NH4+ concentrations in the leaf apoplast, which results in non-zero gas-phase concentrations (stomatal compensation points, χs) in equilibrium with this NH4+apo

195


concentration at the apoplastic pH, for example. Current evidence suggests that NH4+apo increases with increasing N supply to the plant, either through fertilization or high atmospheric N inputs. The compensation point χs is the product of a temperature function describing the Henry’s Law equilibrium and the ratio of Γs = [NH4+apo]/[H+apo]. Values of Γs range from 10 000 after fertilization. At 10 ºC, this equates to values of χs of < 0.15, 2.3 and > 15 µg m−3, respectively. Emission potentials of fertilized soils can be even larger. This large range illustrates that the direction of NH3 exchange is often difficult to estimate a priori. Several papers have recently reviewed the literature on bidirectional NH3 exchange and compiled extensive database on compensation points (Massad etÂ€ al., 2010b; Zhang etÂ€ al., 2010) in order to provide the necessary input for application in atmospheric transport models. The compensation points increase with N input as it is the main driver of apoplast and bulk leaf NH4+ concentrations (Massad etÂ€al., 2010a), but the compensation point also vary between different plant species and with growth stage and season (Riedo etÂ€ al., 2002). The decomposition of litter has been found to play a dominant role (Zhang etÂ€al., 2010). The stomatal pathway for NH3 exchange is only available when stomata are open during daytime, and thus deposition to (often wet) leaf surfaces is the dominant pathway during the night, unless soil surfaces provide a major source and are well exposed to the atmosphere. Deposition fields of NH3 are particularly uncertain, due to (i) uncertainties in the overall magnitude as well as spatial and temporal patterns of agricultural NH3 emissions and (ii) the large variability of NH3 deposition rates to different surfaces. Specific dry deposition sub-models for the surface resistance that include the description of a canopy compensation point for NH3 have been derived and implemented in connection with the analysis of different plant surfaces, e.g. for beans (Farquhar etÂ€ al., 1980), oilseed rape plants (Husted etÂ€al., 2000), and heather (Calluna vulgaris) (Schjørring etÂ€al., 1998). It is common to apply a two-pathway process description (Fowler etÂ€al., 2009; Loubet etÂ€al., 2001):Â€(a) a stomatal pathway, which is bi-directional and modelled using a stomatal compensation point, and (b) a plant surface pathway, which denotes exchange with water surfaces or waxes on the plant surface. The stomatal compensation point may be calculated from knowledge of the aqueous phase chemistry. The equilibrium NH3 ambient air concentration for the stomatal compensation point has been expressed as (Sorteberg and Hov, 1996):  NH4+  NH 3 ( g )  = χ cp = 10 (1.6035 − 4207.62 /T )  +  ,  H  (9.55) where χcp is the compensation point concentration of NH3, and [NH4+] and [H+] are the concentrations of ammonium and hydrogen ion in stomatal cavity, respectively.

196

The leaf surface may work as a capacitance for NH3 and SO2 uptake, and this capacitance increases with humidity (Van hove etÂ€ al., 1989). This transport is independent of solar radiation and contrary to the uptake through stomata, it also takes place during night. Sutton etÂ€al. (1998) defined the canopy compensation point as: ′ (9.56) where χz09 is the canopy compensation point, χ is the NH3 ambient air concentration, z is the height above ground, d is the displacement height, and Fg is the vertical flux. The vertical flux Fg may be divided into a flux towards the leaf surface Fw and a flux through stomata Fs:

Fg = Fw + Fs .

(9.57)

And these fluxes may be written as: ′

′

where Rw and Rs are leaf surface and stomatal resistances, and expressions for these may be found in the work by Sutton etÂ€al. (1993, 1998). From the above two equations the total flux Fg may be expressed as: cp

Fg =

−

′ z0

−

′ z0

.

(9.58)

Similarly the total flux may be derived from the expression of the canopy compensation point: s

w

− ( z) . Ra ( z ) + Rb ′

(9.59) Combining these two equations and eliminating the total flux Fg provides a general expression for the canopy compensation point: Fg =

′

z0

z0

 ( z )  + cp    Ra ( z ) + Rb  Rs = . −1 ( Ra ( z ) + Rb ) + Rs −1 + Rw −1

(9.60)

Several generalized parameterizations of bi-directional NH3 exchange have recently been developed for inclusion in regional CTMs (Gore etÂ€al., 2009; Massad etÂ€al., 2010b; Zhang etÂ€al., 2010), but these have not yet been tested in the spatial modelling environments. In an earlier study, Sorteberg and Hov implemented a simpler parameterization of bi-directional fluxes of NH3 into a Lagrangian long-range transport model, assuming pH to be a constant value of 6.8 (Sorteberg and Hov, 1996). Concerning the concentration of NH4+(aq), they assumed this to be 150 and 50 μmol l−1 for crop and grassland, respectively. The model with these relatively crude assumptions was applied for the European area for the year 1993, and compared with basic scenario without bi-directional flux

Ole Hertel

parameterization. The results indicated a reduction of 0%–20% in total sulphur deposition and a 0%–25% increase in NH3 deposition compared with a simple flux model. The emission through stomata was found to account approximately 0.1% of the total NH3 emission. Loubet etÂ€al. applied a 2D local scale model with the above parameterization of bi-directional fluxes of NH3 based on the canopy compensation point approach to a moorland area (Loubet etÂ€al., 2001). With the FIDES (Flux Interpretation by Dispersion and Exchange over Short Range) model they simulated transport and dispersion to a moorland placed 260 m downwind from a pasture grazed with sheep. Experimental studies have shown that over the sea the atmospheric fluxes of NH3 may also be upward or downward (Lee etÂ€ al., 1998; Quinn etÂ€ al., 1988; Sørensen etÂ€ al., 2003) depending on the meteorological conditions and the relationship between the pH and contents of NH4+ in the upper surface waters on the one side, and the NH3 concentrations in ambient air just above the water surface on the other side. The bidirectional NH3 flux over sea is expressed as an exchange with the water surface: â•‡

F = Ve ( Ceq − C air ) ,

(9.61)

where Ve is the exchange velocity between air and sea (that equals 1/(Ra + Rb)), Ceq is the NH3 concentration in the air at equilibrium with the NHx in the water, and Cair is the actual ambient air concentration of NH3. F is the flux of NH3; the flux is positive when the sea emits NH3 and negative when deposition takes place. The ambient air NH3 concentration at equilibrium is expressed as (Asman etÂ€al., 1994): M NH3 [ NH xs ]

Ceq =

R ×T × H NH3

,  1 10 − pHs  +    γ NH3 γ NH4 × K NH4 

 1 56 EX P 4092  T 

1  , 298 .15  

 1  1 K NH4 = 5.67 ×10 −10 EX P −6286  − .  T 298 .15   

Nitric oxide NO is rather water-insoluble and there is no efficient mechanism for NO to react on the surface or inside leaves, so its deposition rate is rather slow. By contrast, soils commonly act as a source for NO. Some of these soil emissions of NO are oxidized to NO2 (and possibly HNO3) within plant canopies, and taken up more efficiently than NO and thus the behaviour of NO still needs to be taken into account in surface–atmosphere exchange. Nitrogen dioxide Plant uptake of NO2 is slower than that of the more water soluble gases (HNO3, NH3), but it is a significant contributor to N deposition. The NO2 deposition to vegetation is primarily regulated by stomata, and for most plants the internal resistance is negligible, and NO2 deposition velocities may thus be computed from a knowledge of stomatal resistance or conductance (Thoene etÂ€ al., 1991). Studies indicate a small effective stomatal compensation point for NO2 for some plant species, in the range of > 0 to 2 ppb; e.g. an American experimental study found a value of 1.5 nmol mol−1 for the canopy compensation point for NO2 over deciduous forest (Horii etÂ€al., 2004). However, the underlying process is not currently understood, and some laboratory work has failed to reproduce the field observations. Because of its low water solubility, deposition to (and reaction with) surface water, including sea water, is also slow (Cape etÂ€al., 1993).

Nitrous acid (9.62)

where Ceq is in [μg m−3], [NHxs] is the NHx concentration in the sea [μm], MNH3 is the molecular mass of NH3 [g mol−1], γNH3 is the activity coefficient of NH3×H2O, γNH4 is the activity coefficient of NH4+ in sea water, R is the gas constant (8.2075×10–5 atm. m3 mol−1 K−1) and HNH3 is the Henry’s law coefficient for NH3 [m atm−1], pHs is the pH in sea water, which is a measure of the activity of H+ in sea water, and KNH4 is the dissociation constant for NH4+ [M]. The values for HNH3 and KNH4 are expressed as: H NH3

The above formulation was developed for computing the impact of bi-directional fluxes over the North Sea and applied to measured data (Asman etÂ€al., 1994). The formulation has since been applied in the Lagrangian ACDEP model (Sørensen etÂ€ al., 2003), where the results showed a redistribution of N deposition in the coastal region off the coast of the Netherlands.

The biosphere/atmosphere exchange of HONO is generally bi-directional, and daytime concentrations of HONO are low, as it is rapidly photolysed in sunlight. With solubility similar to NH3, HONO is deposited to vegetation under most conditions. Observations of HONO emission have been attributed to production of HONO at surfaces, e.g. through the reaction of NO2 with NO on wet surfaces (Harrison and Kitto, 1994) or NO2 reduction on humic acid (Stemmler etÂ€ al., 2006). In connection with an experimental study, a parameterization of bi-directional fluxes of both NH3 and HONO was applied for estimating dry deposition of N compounds to the Amazon Basin from measured ambient air concentrations (Trebs etÂ€al., 2006).

(9.63)

Organic nitrogen compounds

(9.64)

Organic N compounds account for approximately 20%–30% of the total N deposition in precipitation (Cape etÂ€ al., 2001; Cornell etÂ€al., 2003; Holland etÂ€al., 1999) although this is often not included in N deposition estimates. Much of this organic contribution is presumably due to scavenging of organic N

197

Nitrogen processes in the atmosphere FigureÂ€9.17 The processes of capture of pollutants by cloud and rain.

cloud droplet

AB

A CDEF

SO2

CDEF

NO3–

SO42–

HNO3 NO2 NO

precipitation A - dissolution E - impaction

B - oxidation C - diffusiophoresis D - Brownian diffusion F - cloud condensation nuclei pathway

compounds in the aerosol phase and cloud water. However, the contribution of gaseous organic N compounds to N deposition is even less studied. PAN is considered an important N reservoir species, responsible for much of the N transport in remote regions. PAN is thought to deposit slowly and remains stable at cold temperatures. At warmer tropospheric temperatures PAN decomposes quickly. Newly developed instruments have resulted in new measurements indicating deposition rates of PAN (and other PAN-like compounds) that are significantly larger than classical predictions (Turnipseed etÂ€al., 2006; Wolfe etÂ€al., 2009), especially to wet vegetation. Thus the lifetime of PAN with respect to deposition may be shorter than previously thought. In addition, PAN is water insoluble and the comparably large deposition fluxes to wet surfaces indicate that the current mechanistic understanding of the deposition process is incomplete. There are parallels to the deposition of O3, which also appears to exhibit larger deposition rates to wet surfaces than can be explained by its solubility (Fowler etÂ€al., 2001). The importance of alkyl nitrates has recently been demonstrated for Blodgett Forest, Sierra Nevada, USA (Farmer etÂ€al., 2006), although it appears that the pollution climate of their site is quite unique. Nevertheless, information is lacking to form a robust picture of the importance of these compounds across the full range of European conditions. Although amines have been measured as emitted from agricultural activities (Schade and Crutzen, 1995), there is currently no information on their dry deposition.

198

9.4.3â•‡ Deposition of N containing aerosols Deposition of particles containing SO42−, NO3−, Cl− and NH4+ contributes to the potential acidification and eutrophication (N components) of ecosystems. Compared to gaseous deposition of acidifying compounds onto low vegetation, particle deposition fluxes are usually found to be small. However, in difference from wet deposition it takes place all the time and furthermore it is believed that the fluxes of small particles are currently underestimated for very rough surfaces like forests. Erisman etÂ€al. (1997) found that deposition of aerosols to the Speulder forest contributed 20% and 40% to the total dry deposition of S and N, respectively. Parameterizations of aerosol dry deposition velocities to forests differ greatly between models (Tang etÂ€al., 2009).

9.5â•‡ Wet scavenging of N compounds from the atmosphere Wet deposition or scavenging is defined as the removal of gases and aerosol from the atmosphere by precipitation snow, rain. Unlike dry deposition, the wet deposition processes are indirect; rain, hail and snow are the vectors for transport of the pollutant to the surface. The apparent simplicity of the measurement approach for wet deposition, a simple precipitation collector placed on the ground contrasts appreciably with the underlying physical and chemical pathways of solutes into the collected precipitation

Ole Hertel

sample. There is also significant uncertainty in the relative magnitudes of dry deposition of trace chemical species as gases and aerosols onto the collecting equipment. The incorporation of pollutants in clouds and precipitation include many different processes, which will be considered in turn. The Nr compounds are present in aerosols or as gases. Regarding aerosols, the N is mainly present as NH4+ or NO3− (although some organic N is also present). The bulk of the aerosol mass is present in the size range 0.1–1.0 μm (diameter). These aerosols are removed through interception by falling rain or snow, a process known as washout or by incorporation of the aerosol into cloud droplets within clouds, a process known as rainout (FigureÂ€9.17). Washout is responsible for 10%–20% of the N in wet deposition on average, but depends naturally on the relative amounts of N present in cloud water and in the air through which the precipitation falls. The aerosol scavenging within cloud occurs through a number of physical and chemical pathways (FigureÂ€9.17) as C, D, E and F while the gases are incorporated through solution and oxidation processes (A and B). The phoretic process includes diffusiophoresis, in which aerosol particles are transported in the direction of a mean flux of vapour molecules. In the case of a cloud droplet growing by vapour

Figure 9.18 Orographic enhancement of precipitation in the UK; an East–West transect.

Figure 9.19 The seeder–feeder process of orographic enhancement of precipitation. Source: Fowler and Battarbee, 2005.

diffusion of water molecules towards the droplet surface, aerosols would move along the vapour flux towards the growing droplet. Additional phoretic mechanisms are presented by electrical and thermal gradients (electrophoresis and thermophoresis respectively). The phoretic processes contribute relatively small amounts of the solute in cloud water (Goldsmith etÂ€al., 1963). Aerosols may also be captured by cloud droplets following Brownian diffusion (D) to the droplet surface and rates of Brownian diffusion vary strongly with particle size, being significant for particles smaller than 100 nm in diameter. However, diffusion rates are very small relative to molecular diffusion and diffusional mechanisms make only minor contributions to the wet removal pathway. The remaining minor process leading to capture of aerosols by cloud droplets is impaction and interception (E). As implied in the name these processes lead to the capture of aerosols by droplets when one is unable to follow the streamlines of airflow around the other and the aerosol and droplet collide. The bulk of the aerosol N in cloud water is incorporated through the activation of aerosols containing NO3− or NH4+ into cloud droplets. The N containing aerosols are effective cloud condensation nuclei and are readily incorporated into cloud droplets through the nucleation scavenging pathway. Thus the main route is nucleation scavenging for aerosol NO3−, and NH4+ (Pruppacher and Jaenicke, 1995). The pathway for below wet scavenging of the gaseous N compounds depends on the solubility and reactivity of the specific gas. In the case of NH3 and HNO3, which are highly soluble, clouds and rain remove these gases effectively from the air. The contribution of NO and NO2 to dissolved N in precipitation is very small as these gases are not very soluble (relative to NH3 or HNO3). Wet deposition is monitored by simple methods (precipitation collectors) analysed for major anthropogenic ions SO42−, NO3−, NH4+, H+ and marine ions Cl−, Na+, Mg2+. The networks of collectors for precipitation chemistry are much less dense than precipitation collectors for the national meteorological services, mainly because of the costs of chemical analysis. Furthermore, precipitation chemistry collectors are located a height above ground to reduce contamination from ground based sources, and the practice of locating collectors above the ground reduces the capture of small droplets due to aerodynamic screening by the collector. The relative contributions to deposition from dry and wet deposition change with distance from source as primary pollutant concentrations decline and oxidation from gas to particle remove gas phase species which dry deposit quickly. Thus the areas more than a few hundred km from sources receive most of their N deposition in precipitation. In regions in which the amounts of precipitation are large, wet deposition dominates the N loads, as in most of the uplands of Europe. However, it is not simply the precipitation amount that needs to be considered in assessing the relative contributions of wet and dry deposition. The processes leading to orographic enhancement of rainfall amount have a profound effect on the overall scavenging of pollutants from the atmosphere. The meteorological process which enhances precipitation in much of maritime northern Europe is the seeder–feeder

199

Nitrogen processes in the atmosphere Figure 9.20 The incorporation of pollutant aerosols into orographic cloud. Source: Fowler etÂ€al., (1991).

mechanism, in which orographic cloud, formed over hills and mountains is washed out by precipitation falling from higher levels in the troposphere, as shown in FigureÂ€ 9.18 and first described by Bergeron (1965). The process occurs widely and is responsible for most of the enhancement of precipitation over uplands in the UK and Scandinavia. The process has been extensively studied in the UK, where, especially in the West of the country, annual rainfall is in the range 1000–3000 mm with the amounts in excess of 100 mm being mainly generated through seeder-feeder scavenging. The mountains are very effective in increasing rainfall (FigureÂ€9.18) and wet deposition by the seeder–feeder process (FigureÂ€9.19) in which low level hill cloud droplets are washed out by falling precipitation from higher level. The hill cloud is more polluted than higher level cloud because boundary layer aerosols are effectively activated into cloud droplets as they are forced to rise and cool over the hills and mountains. The seeder–feeder effects on precipitation amount have been simulated in process-based models and are able to simulate observed spatial patterns in precipitation (Carruthers and Choularton, 1983). Models have also been used to simulate the wet deposition of pollutants over mountains (Dore etÂ€al., 1990) and compared with detailed campaign measurements in an upland area. Extending the modelling of orographic enhancement of wet deposition to the country scale has enabled detailed spatially resolved wet deposition maps to be generated (Dore etÂ€al., 1990). As orographic enhancement of wet deposition has been shown to be a major contributor to the total deposition in upland Britain the explicit inclusion of the process in deposition maps has been regarded as a routine component of wet deposition mapping (NEGTAP, 2001). The resulting wet, and total N deposition maps show a strong influence of altitude and requires a grid resolution on the same scale as the complex topography to reproduce (90%), while separate collection of slurries and solids dominate in UK, France and Central/Eastern Europe (40

N2O emission (kg N2O-N/ha) 0-5 5 - 10 10 - 15 15 - 25 >25

0

2

4

0

6 Kilometers

2

4

6 kilometers

(c) NO3 groundwater (mg NO3/I) 0-5 5 - 10 10 - 25 25 - 50 >50

0

2

4

6 Kilometers

Figure 11.6 Maps calculated using Initiator2 for the NFW region in the Netherlands for 2004:Â€(a) annual NH3 emissions from manure application; (b) total annual N2O emissions; (c) nitrate concentrations in the upper groundwater (from Kros et al., 2007).

slopes). In the scenarios, source areas were placed upstream or downstream of sink areas, as well as spread in checkerboard patterns throughout the catchments. When applied to real cases, the model output compared favourably with real catchments in Brittany (NW France). More recently, a consortium of European research groups has started an ambitious project on landscape analysis of N interactions, as a component of the EU NitroEurope Integrated Project (see also www.nitroeurope.eu; Sutton etâ•¯al., 2007). One of the aims of the landscape component of the project is the

240

joint development of an integrated landscape scale model, NitroScape, to simulate the flows of N between all components of rural landscapes. The NitroScape model is a framework coupling suitable existing component models for the atmosphere, ecosystems and hydrological components, as well as farm scale processes, with a spatial database (Cellier etâ•¯al., 2006; S. Duretz, personal communication, 2010) (Figure 11.8). The approach is similar to the one used in the LANAS project described above, but with a more sophisticated model coupler, which allows interaction between the component models

Pierre Cellier

Figure 11.7 Modelled example of landscape scale mitigation of nitrate leaching via the introduction of non-N-fertilized set aside grassland in a drinking water borehole catchment (boundaries in blue line), situated in the Tyrebæk stream watershed, Central Jutland, Denmark. The ‘before’ and ‘after’ maps show results from crop rotation, manure, farm and hydro-geological models, before and after introducing extensive farming systems in the borehole catchment (after Hutchings et al., 2004; Dalgaard, 2009) with permission.

during run-time and minimum adaptation of existing models. NitroScape will consider the majority of the components of N transfer at landscape scale. It will be tested and verified over a range of rural landscapes under different climatic conditions, with different farming systems including livestock. Each landscape has a specific topic and includes natural areas where impacts of N can be predicted.

11.5.3â•‡ Conclusion on landscape modelling Progress has been made by a number of recent and current studies exploring landscape scale modelling from different starting points and for different purposes, whether to

investigate strategies for the provision of clean drinking water or to protect sensitive semi-natural areas from excess atmospheric N deposition. This required the consideration of both natural and anthropogenic processes and modelling them with sufficient levels of detail in a spatial context. A clear challenge emerges of how to implement the interaction between different component models, using the right tools. These models have also improved the understanding of the relative importance of transfer and transformation processes in rural landscapes. However, there is still much to learn about the interactions of the different elements in the landscape and the development of new models can help with this.

241

Nitrogen flows and fate in rural landscapes

Figure 11.8 Schematic of the NitroScape modelling framework to provide a fully integrated treatment of N exchange fluxes at the landscape scale. The landscape is envisaged as integrating farms, fields, seminatural land and non-agricultural sources, with lateral and vertical dispersion fluxes through the atmosphere and hydrosphere (from Sutton et al., 2007).

11.6â•‡ The importance of integrating the landscape perspective into N assessment and management 11.6.1â•‡ N mitigation at the landscape scale The following examples of land use management, often referred to as ‘spatial abatement’ or ‘spatial planning’ (Bleeker and Erisman, 1998; Lekkerkerk, 1998; Theobald etâ•¯al., 2001; Sutton etâ•¯al., 2004; Dragosits etâ•¯al., 2006; Schou etâ•¯al., 2006) highlight the relevance of the landscape scale for mitigating N impacts on the environment. • Establishing tree belts around NH3 sources (e.g. animal housing) or sensitive areas has been suggested as an efficient tool to diminish deposition to sensitive ecosystems and could be used as a tool for their protection (Sutton etâ•¯al., 2004; Dragosits etâ•¯al., 2006). • Constructing wetlands or more generally restoration and management of wetlands (Haycock etâ•¯al., 1997; Woltemade, 2000; Tanner etâ•¯al., 2003; Viaud etâ•¯al., 2004) have proved to significantly decrease the NO3− concentration in surface waters and thus are an efficient buffering element, protecting the river course from the impact of N (Haycock etâ•¯al., 1997; Viaud etâ•¯al., 2004). This has led eco-engineers to the implementation of constructed wetlands for water quality objectives. It is typically a landscape issue because their efficacy and their management depend on the catchment (including hydrological functioning, hedgerow network and grassed strips) that contains the wetlands and on the farming systems (Haycock etâ•¯al., 1997). • On-farm spatial planning provides means to help protect sensitive areas by locating certain activities in the most suitable location. This can include locating farmsteads, crops

242

and grasslands, as well as high emission activities, such as manure spreading to locations that reduce emissions and/or impacts of the emissions. Such strategies can also help protect fresh water by decreasing NO3− leaching and groundwater contamination (see Figure 11.7; Dalgaard, 2009; Deffontaines etâ•¯al., 1994), as well as help protect sensitive ecosystems such as Natura 2000 sites from NH3 deposition. For example, Dragosits etâ•¯al. (2005) modelled the effect of burning poultry manure for power generation (instead of spreading it on fields) or moving poultry houses away from a nature reserve on NH3 and N2O emission, N deposition (Figure 11.9) and NO3− leaching. These measures can exploit spatial relationships to reduce emissions (e.g. arranging activities to reduce N2O emissions) as well as use the source–sink relationship to decrease local impacts of NH3 on sensitive ecosystems (Loubet etâ•¯al., 2009). These approaches can be considered as extending the vision of ‘precision farming’ from the field to the landscape scale. In all cases, practitioners are faced by the complexity of the landscape because it involves not only the studied system (wetlands, tree belt, etc.), but also the surrounding landscape. Modifying crop spatial allocation needs to consider the whole farming system for consistency and its interactions with the landscape. All these measures, therefore, must be placed in a landscape perspective and consider long-term interactions.

11.6.2â•‡ Using landscape-scale interactions to improve regional models Air pollution or climate models at regional or national scale often use a grid size of between 5 × 5 km2 and 50 × 50 km2, limiting simulations of atmospheric concentration or deposition

Pierre Cellier Figure 11.9 Difference in N deposition (NH3 dry deposition) due to moving of poultry from two sets of buildings in the immediate vicinity of a nature reserve (hatched area) to a more distant location (approx. 1.5 km east/right) (from Dragosits et al., 2005). With permission from Elsevier.

to this resolution. In reality, atmospheric deposition of N, especially NH3 dry deposition, can vary by several orders of magnitude within a grid square of a national or regional model (Dragosits etâ•¯al., 2002). This variability is mainly due to the localized nature of NH3 emission sources and the high dry deposition velocity for NH3 for semi-natural vegetation (sinks). Using data from a regional model could, therefore, significantly underestimate (or overestimate) the environmental impacts since the actual deposition at a particular location could be much higher (or lower) than the model simulation. Landscape-scale atmospheric models can take into account the sub-grid short-scale interactions between sources and sinks and should therefore be used to better assess the uncertainty of national or regional models by estimating the statistical distribution of deposition values within the grid square. This would help to assess local deposition and impacts on conservation areas at a regional scale (see e.g. Loubet etâ•¯al., 2009; Hertel etâ•¯al., 2009). Similarly, in regional scale water quality models, diffuse sources of nutrients from agricultural areas are most often estimated either from empirically determined export coefficients or from an additive approach based on the output of separately run plant/soil/water models at the plot scale. In the best case, they use an arbitrary reduction coefficient accounting for ‘landscape’ or ‘riparian’ retention (see e.g. Billen etâ•¯al., 2009). None of these approaches are able to simulate the effect of changes in the spatial structure or functioning at the landscape scale. Landscape-scale transfer models can help draw a more complete picture by quantifying the storage/release of N pools in soils and groundwater, which are, per se, an important issue for N management, and by describing the intra-annual dynamics of the N delivery to the streams. These models are also better suited to complex scenario analyses, especially to quantify the effects of management practices on N losses. Such results could be aggregated as input to larger scale models, based on the catchment/subcatchment aggregation.

11.6.3â•‡ Role of the landscape scale in environmental N policy measures A number of policies and measures in the EU and various Member States (see Oenema etâ•¯al., 2011, Chapter 4 this volume) address the importance of landscape structure and functions in relation to N. The potential for considering the landscape scale in these policies depends in part on the level of detail that can be used by Member States to implement them. • Water related policies (Water Framework Directive, Nitrates Directive, Urban Waste Water Directive and Groundwater Directive):Â€the Water Framework Directive applies a river basin and a catchment approach, while the Nitrates Directive distinguishes Nitrate Vulnerable Zones and various areas at farm level (near water courses, sloping areas, wet soils, etc.). In the case of the Groundwater Directive, groundwater bodies or aquifers are distinguished. Member States have some degree of freedom to interpret the spatial variability within landscapes according to these directives. • Air related policies (Air Quality Directive, NEC and IPPC Directives):Â€the main environmental targets relate to emission ceilings at national level, concentration levels in the air and the implementation of best available techniques at farm, car, machine and company level. Its spatial component depends on envisaged measures and the target. The landscape scale is to some extent addressed in the case of the protection of sensitive areas (e.g. permit for farm extension close to a Special Area of Conservation). However, there is potential for greater consideration of landscape planning approaches as a means to maximize the environmental benefit for any given national emissions ceiling. • Nature protection policies (Habitats Directive, Birds Directive):Â€these policies have a strong spatial component through the identification of high nature value areas

243

Nitrogen flows and fate in rural landscapes

(Special Areas of Conservation and Special Protection Areas, making up the Natura 2000 network). It is left to the Member States to identify and to prescribe conditions and measures applicable to these areas, and also around these areas. For example, some Member States have restrictions on farming activities, especially on the intensification of farming activities within and near Natura 2000 areas (see e.g. Hertel etâ•¯al., 2009). Up to now, the assessments generally only consider the location of point sources, e.g. animal housing. Diffuse sources, such as fertilizer and organic manure application, are rarely considered, but have significant local impacts. There is potential for further use of buffer zones in source areas for both atmospheric and water based nitrogen inputs. • Rural Development Regulations and Agri-Environmental Regulations have a strong spatial component. Farmers in less favourable areas and/or near high nature value areas may be supported in exchange for landscape maintenance and forbearance of intensification of farming activities. Farmers may also receive support for introducing low-NH3 emissions techniques for manure storage and application. The landscape perspective also provides the means to link EU agri-environment support more effectively through ‘cross compliance’ with other Directives. For example, where farm management plans associated with support payments are considered as ‘plans or projects’ under the Habitats Directive, landscape analysis provides the means to optimize spatial Nr management. In summary, there are a large number of opportunities provided by EU Directives and Regulations to address the landscape scale. These are needed to better account for local conditions in relation to the wide variety of farming systems and environmental conditions. As yet, there is a huge difference in the interpretation of the EU Directives and Regulations between Member States, and this also is also the case for addressing the landscape scale. This is notably the case with the Nitrates Directive (Smith etâ•¯al., 2007) and protection of the Natura 2000 areas (COST 729, 2009). Our analysis suggests that there are ample possibilities to address the landscape scale, with so far only limited use being made of this scale. Up to now, the Â�policy-maker is faced with a lack of practical tools for supporting this type of analysis, such as user-friendly landscape models. Moreover, there is a need for case studies and improved databases for analysis at this scale.

11.6.4â•‡ The importance of detailed and simple tools for landscape assessment All the cases described above have highlighted the relevance of the landscape scale for N assessment and management. However, no simple rule exists of how to make an assessment of an environmental measure or abatement technique at the landscape scale. Depending on the level of detail to be applied, this may need to consider a large number of N sources and sinks, with complex and changing relationships between them. Hence it is not straightforward to identify similarities between

244

situations and thus to extrapolate a conclusion for one location directly to another location/situation or to derive simple rules that are generally applicable at the landscape scale. It is clear that comprehensive modelling will be the privileged approach to investigate potential strategies and make an assessment of measures and scenarios at the landscape scale. This requires detailed modelling of processes, as described in Section 11.4. Application of such models to multiple cases and/or regional or larger scales would need detailed landscape databases and the development of landscape typologies. Nevertheless, it is clear that simple practical tools are also needed. While detailed approaches are needed to understand and quantify the interactions, the outcomes of such models also need to be generalized. In this respect, the development of publicly accessible screening tools provides an important step forward. These simpler models can be based on simplifying assumption allowing analytical relationships to be derived or on simpler numerical schemes. This makes it possible to investigate with reasonable accuracy the flows (including input to sensitive ecosystems) and concentration fields of N species. For example instead of using complicated atmospheric transport models Rihm and Kurz (2001) used a function of deposition vs. distance that was developed for the Netherlands (10-year average, averaged over all wind directions) and applied it to Switzerland. It was coupled to a spatially detailed NH3 emission inventories (200 × 200 m2 or less), that formed the input for the calculations of NH3 concentration fields. Although this should not be done in principle as the Swiss climate differs from the Dutch climate, a good correlation was obtained between modelled and measured values for 17 sites. Later Thöni etâ•¯al. (2004) refined the method adjusting the function deposition vs. distance, so that an optimum correlation was obtained for the Swiss situation. Similar examples can be found in other countries (e.g. the SCAIL model in the UK; Theobald etâ•¯al., 2009) and for hydrological modelling (e.g. Durand and Torres, 1996) or ecosystem models (e.g. Strayer etâ•¯al., 2003).

11.7â•‡ Future challenges The examples above have shown that analysing the N cascade at the landscape scale make it possible to integrate the major processes that modify the N flows and balance. To this extent, the landscape scale also appears to be a very practical scale for implementing and assessing environmental measures. However, it is also highlighted that analysing and modelling landscape interaction for N is a complex task and that no approach has yet been found to be completely satisfactory for the complete analysis. At the same time, there is a parallel need for the development of simple practical tools that can support landscape level decision making in the rural environment. The major questions faced for the coming years include the following. How do we best account for the interactions between farming systems and landscape? Spatial heterogeneity, as well as interactions with farm management, is shown to have strong effects on N flows and transformation at landscape scale. As exemplified in Figure 11.1 farm activity may determine the

Pierre Cellier

spatial arrangement of fields, roads and hedgerows. Moreover farm activity and hence N application to land is not only organized according to the distance from the farm (Figure 11.5) but also to the topography (e.g. grasslands are often located in wetter and less productive areas). These interactions are complex and dependent on local conditions. Hence, there is a need for more study and analysis on the interrelationship between farming systems and landscape features. How can we develop a landscape typology to describe landscape variety in modelling at European scale? European rural landscapes present a wide range of variability, due to climate, physical environment (e.g. topography, soils) and history. Moreover, experiences in landscape modelling have shown that it requires detailed local data, including spatial data on activities/environmental variables, etc. National average data are usually not sufficient to represent local spatial and temporal (diurnal, seasonal or inter-annual) variability. Consequently, there is a need to develop methods to derive a landscape typology giving a limited number of landscape classes based on landscape features and farming systems. These could be based on either real landscape description using aerial photography or remote sensing, or on a farming system approach (see e.g. Figure 11.7) or both. Such a landscape typology would allow landscape processes to be treated more effectively in largerscale operation models. Is it feasible to derive scenarios of future landscapes at 2030 or 2100 horizon? Due to different drivers such as climate change, population increase, extension of urban areas or changes in agricultural and environmental policies, European rural landscapes are expected to change significantly in the next few decades. This could have a significant effect on N flows and efficiency of policy measures. There is a need to examine potential scenarios for future landscape structure and dynamics in order to account for this in climate change and land use change scenarios. How do we develop and test monitoring approaches to assess N flows and budgets at the landscape scale? While modelling at the landscape scale is now becoming firmly established, as illustrated by the studies described above, monitoring approaches for landscape level assessment also need to be developed further, at least to enable the validity of the landscape modelling to be tested. This monitoring should integrate measurement of the spatial and temporal variability of NH3, N2O, NOx and NO3− including the role of hot-spots. Further testing and verification of bioindicators of N responses could be integrated with the physicochemical monitoring activities. How do we best account for landscape issues in environmental N policies? Landscape scale models should be adapted for practical use by landscape planners, farm advisers or policy-Â�makers. This effort will also need databases based on case studies which could be used as a basis for analysis. The use of a landscape typology (see above) would make it possible to integrate and make assessments at a larger scale. There is an ongoing need for simple tools to support the implementation of landscape scale N policies, complementing the detailed models.

How do we assess pollution swapping? In the frame of environmental policies, the risk of pollution swapping (within or beyond the landscape) is increasingly important and must be further explored. The landscape scale is especially relevant, as N transformations often occur in locations different from where N has been applied. Landscape scale modelling can help to understand the origin and magnitude of these transformations by linking together the processes between landscape elements, allowing the synergies and trade-offs to be better quantified.

Acknowledgements We are grateful for support for this work through the NitroEurope Integrated Project of the European Commission, and for coordination funds through the COST 729 and ESF Nitrogen in Europe (NinE) programmes.

References Baudry, J., Bunce, R. G. H. and Burel, F. (2000). Hedgerows:Â€an international perspective on their origin, function and management. Journal of Environmental Management, 60, 7–22. Beaujouan, V., Durand, P. and Ruiz, L. (2001). Modelling the effect of the spatial distribution of agricultural practices on nitrogen fluxes in rural catchments. Ecological Modelling, 137, 93–105. Beven, K. J. (ed.) (1997). Distributed Modelling in Hydrology:Â€Applications of the TOPMODEL Concept. Wiley, Chichester, UK. Beven, K. J. and Kirkby, M. J. (1979). A physically based variable contributing areas model of basin hydrology. Hydrological Sciences Bulletin, 24, 43–69. Billen, G. and Garnier, J. (2000). Nitrogen transfers through the Seine drainage network:Â€a budget based on the application of the Riverstrahler model. Hydrobiologia, 410, 139–150. Billen, G., Thieu, V., Garnier, J. and Silvestre, M. (2009). Modelling the N cascade in regional watersheds:Â€the case study of the Seine, Somme and Scheldt rivers. Agriculture, Ecosystems and Environment, 133, 234–246. Billen, G., Silvestre, M., Grizzetti, B. etâ•¯al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.Â€W.Â€ErismanÂ€etâ•¯al. Cambridge University Press. Bleeker, A. and Erisman, J. W. ( 1998). Spatial planning as a tool for decreasing nitrogen loads in nature areas. Environmental Pollution, 102, 649–655. Boehlke, J. K. and Denver, J. M. (1995). Combined use of groundwater dating, chemical, and isotopic analyses to resolve the history and fate of nitrate contamination in two agricultural watersheds, Atlantic Coastal Plain, Maryland. Water Resource Research, 31, 2319–2339. Bouma, J., Varallyay, G. and Batjes, N. H. (1998). Principal land use changes anticipated in Europe. Agriculture Ecosystems and Environment, 67, 103–119. Brisson N., Mary B., Ripoche D. etâ•¯al. (1998). STICS:Â€a generic model for the simulation of crops and their water and nitrogen balances. I. Theory and parameterization applied to wheat and corn. Agronomie, 18, 311–346. Butterbach-Bahl, K., Gundersen, P., Ambus, P. etâ•¯al. (2011). Nitrogen processes in terrestrial ecosystems. In:Â€The European Nitrogen

245

Nitrogen flows and fate in rural landscapes Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Cellier, P., Bleeker, A., Bog, E. et al. (2006). Landscape scale assessment of nitrogen interactions. In:Â€The GHG Cycle in the Northern Hemisphere, Open Science Conference, European Commission, Sissi-Lassithi, Crete (GRC), Nov. 14–18/11/2006, p. 149 (oral communication, abstract). Cirmo, C. P. and McDonnell, J. J. etâ•¯al. (1997). Linking the hydrologic and biogeochemical controls of nitrogen transport in near-stream zones of temperate-forested catchments:Â€a review. Journal of Hydrology, 199, 88–120. Clarisse, L., Clerbaux, C., Dentener, F., Hurtmans, D. and Coheur, P. F. (2009). Global ammonia distribution derived from infrared satellite observations. Nature Geoscience, 2, 479–483. Colbach, N., Lucas, P. and Meynard, J. M. (1997). Influence of crop management on take-all development and disease cycles on winter wheat. Phytopathology, 87, 26–32. COST729 (2009). Nitrogen deposition and Natura 2000 workshop. See http://cost729.ceh.ac.uk/n2kworkshop Dahm, C. N., Grimm, N. B., Marmonier P., Valett H. M. and Vervier, P. (1998). Nutrient dynamics at the interface between surface waters and groundwaters. Freshwater Biology, 40, 427–451. Dalgaard, T. (2009). Landscape agroecology:Â€managing interactions between agriculture, nature and socio-economy. In Â€Proceedings from the Multi-level Processes of Integration and Disintegration, ed. Nagel etâ•¯al. Green Week Scientific Conference, International Congress Centre, Berlin, Jan. 14–15 2009. Margraf Publishers. Dalgaard, T., Heidman, T. and Mogensen, L. (2002). Potential N-losses in three scenarios for conversion to organic farming in a local area of Denmark. European Journal of Agronomy, 16, 207–221. Deffontaines J.-P., Brossier, J., Benoît, M. etâ•¯al. (1994). Agricultural practices and water quality: a research development project. In:Â€Brossier,Â€J., de Bonneval, Laurence, Landais, E. (eds.), Coll. Science Update, pp. 31–61. Deffontaines, J. P., Thenail, C. and Baudry, J. (1995). Agricultural systems and landscape patternsÂ€– how can we build the relationship? Landscape and Urban Planning, 31, 3–10. De Vries, W., Kros, J. and Velthof, G. (2005). Integrated evaluation of agricultural management on environmental quality with a decision support system. In 3rd International Nitrogen Conference, Oct.Â€12–16 2004, Nanjing. China, pp. 859–870. Dinnes, D. L., Karlen D. L., Jaynes D. B. etâ•¯al. (2002). Nitrogen management strategies to reduce nitrate leaching in tile-drained midwestern soils. Agronomy Journal, 94, 153–171. Dragosits, U., Theobald, M. R., Place, C. J. etâ•¯al. (2002). Ammonia emission, deposition and impact assessment at the field scale:Â€a case study of sub-grid spatial variability. Environmental Pollution, 117, 147–158. Dragosits U., Theobald M. R., Place C. J. etâ•¯al. (2005). Interaction of nitrogen pollutants at the landscape level and abatement strategies. In 3rd International Nitrogen Conference, Nanjing, China, Oct. 12–16 2004, pp. 30–34. Dragosits, U., Theobald, M. R., Place, C. J., ApSimon, H. M. and Sutton, M. A. (2006). The potential for spatial planning at theÂ€landscape level to mitigate the effects of atmospheric ammonia deposition. Environmental Science and Policy, 9, 626–638. Durand, P. and Juan Torres, J. L. (1996). Solute transfer in agricultural catchments:Â€the interest and limits of mixing models. Journal of Hydrology, 181, 1–22.

246

Durand, P., Breuer, L., Johnes, P. etâ•¯al. (2011). Nitrogen processes in aquatic ecosystems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Forman, R. T. (1997). Land Mosaics: The Ecology of Landscapes and Regions. Cambridge University Press. Forman, R.T. and Godron, M. (1981). Patches and structural components for a landscape ecology. BioScience, 31, 733–740. Forman, R. T. and Godron, M. (1986). Landscape Ecology. Wiley, NewÂ€York. Fowler, D., Pitcairn, C. E. R., Sutton, M. A. et al. (1998). The mass budget of atmospheric ammonia in woodland within 1 km of livestock buildings. Environmental Pollution, 102, 343–348. Galloway J. N., Aber, J. D., Erisman, J. W. etâ•¯al. (2003). The Nitrogen Cascade. BioScience, 53, 341–356. Gras, F. and Benoît, M. (1998). Influence des systèmes de culture et des pratiques agricoles sur la qualité de l’eau minérale de Vittel. Le programme de recherches AGREV. Comptes Rendus de l’Académie d’Agriculture de France (FRA), 5, 166–168. Haag, D. and Kaupenjohann, M. (2001). Landscape fate of nitrate fluxes and emissions in Central Europe:Â€a critical review of concepts, data, and models for transport and retention. Agriculture, Ecosystems and Environment, 86, 1–21. Haycock, N. E., Burt, T. P., Goulding, K. W. T. and Pinay, G. (1997). Buffer Zones:Â€Their Processes and Potential in Water Protection. Quest Environmental, Harpenden, UK. Hertel, O., Reis, S. and Ambelas Skjøth, C. (2011). Nitrogen processes in the atmosphere. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al. Cambridge University Press. Hill, J. (1998). Application of computational modelling to ammonia dispersion from agricultural sources. PhD thesis, University of London. Hutchings, N., Dalgaard, T., Rasmussen, B. M. etâ•¯al. (2004). Watershed nitrogen modelling. In:Â€Controlling Nitrogen Flows and Losses, ed. D. J. Hatch etâ•¯al.,Wageningen Academic Publishers, Wageningen, The Netherlands, pp. 47–53. Jarvis, S., Hutchings, N., Brentrup, F., Olesen, J. and van der Hoek, K. (2011). Nitrogen flows in farming systems across Europe. In:Â€The European Nitrogen Assessment. ed. M. A. Sutton, C. M. Howard, J.Â€W. Erisman etâ•¯al. Cambridge University Press. Kristensen, S. P., Thenail, C. and Kristensen, L. (2001). Farmers’ involvement in landscape activities:Â€An analysis of the relationship between farm location, farm characteristics and landscape changes in two study areas in Jutland, Denmark. Journal of Environmental Management, 61, 301–318. Kros, J., de Vries, W., Voogd, J. C. H., Gies, T. J. A. and Roelsma,Â€J. (2007). Meervoudige milieumonitoring Noordelijke Friese Wouden: gebiedsstatus van emissie en depositie van ammoniak in relatie tot gebiedsdoelstellingen. Alterra, Wageningen, The Netherlands. Leibowitz, L. G., Loehle, C., Li, B. L. and Preston, E. M. (2000). Modeling landscape functions and effects:Â€a network approach. Ecological Modelling, 132, 77–94. Lekkerkerk, L. (1998). Implications of Dutch ammonia policy on the livestock sector. Atmospheric Environment, 32, 581–587. Lipiec, J. and Pniewski, W. (1995). Effects of soil compaction and tillage systems on uptake and losses of nutrients. Soil and Tillage Research, 35, 37–52. Liu, J. and Taylor, W. W. (2002). Integrating Landscape Ecology into Natural Resource Management. Cambridge University Press.

Pierre Cellier Loubet, B., Cellier, P., Milford, C. and Sutton, M. A. (2006). A coupled dispersion and exchange model for short-range dry deposition of atmospheric ammonia. Quarterly Journal of the Royal Meteorological Society, 132, 1733–1763. Loubet, B., Asman, W. A., Theobald, M. R. etâ•¯al. (2009). Ammonia deposition near hot spots:Â€processes, models and monitoring methods. In:Â€Atmospheric Ammonia: Detecting Emission Changes and Environmental ImpactsÂ€– Results of an Expert Workshop under the Convention on Long-range Transboundary Air Pollution, ed. M.Â€A. Sutton, S. Reis and S. M. H. Baker, Springer, New York. Marshall, E. (2002). Editorial:Â€introducing field margin ecology in Europe. Agriculture, Ecosystems and Environment, 89, 1–4. McCollin, D. (2000). Editorial:Â€hedgerow policy and protectionÂ€– changing paradigms and the conservation ethic. Journal of Environmental Management, 60, 3–6. Meeus, J. H. A. (1993). The transformation of agricultural landscapes in Western Europe. Science of the Total Environment, 129, 171–190. Molenat, J., Gascuel-Odoux, C., Ruiz, L. and Gruau, G. (2008). Role of water table dynamics on stream nitrate export and concentration. in agricultural headwater catchment (France). Journal of Hydrology, 348, 363–378. Mosier, A., Kroeze, C., Nevison, C. etâ•¯al. (1998). Closing the global N2O budget:Â€nitrous oxide emissions through the agricultural nitrogen cycle. Nutrient Cycling in Agroecosystems, 52, 225–248. Mourier, B., Walter, C. and Merot, P. (2008). Soil distribution in valleys according to stream order. CATENA, 72, 395–404. Oehler, F., Durand, P., Bordenave, P., Saadi, Z. and SalmonMonviola,Â€J. (2009). Modelling denitrification at the catchment scale. Science of the Total Environment, 405, 1726–1737. Oenema, O., Bleeker, A., Braathen, N. A. etâ•¯al. (2011). Nitrogen in current European policies. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al. Cambridge University Press. Pitcairn, C. E. R., Fowler, D., Leith, I. D. etâ•¯al. (2003). Bioindicators of enhanced nitrogen deposition. Environmental Pollution, 126, 353–361. Rapport, D. J., Gaudet, C., Karr, J. R. etâ•¯al. (1998). Evaluating landscape health:Â€integrating societal goals and biophysical process. Journal of Environmental Management, 53, 1–15. Reay, D. S., Smith, K. A. and Edwards, A. C. (2003). Nitrous oxide emission from agricultural drainage waters. Global Change Biology, 9, 195–203. Rihm, B. and Kurz, D. (2001). Deposition and critical loads of nitrogen in Switzerland. Water, Air and Soil Pollution, 130, 1223–1228. Rounsevell, M. D. A., Annetts, J. E., Audsley, E., Mayr, T. and Reginster,Â€I. (2003). Modelling the spatial distribution of agricultural land use at the regional scale. Agriculture Ecosystems and Environment, 95, 465–479. Scholefield, D., Lockyer, D. R. Whitehead, D. C. V. and Tyson, K. C. (1991). A model to predict transformation and losses of N in UK pastures grazed by beef cattle. Plant and Soil, 131, 165–178. Schou, J. S., Tybirk, K., Løfstrøm, P. and Hertel, O. (2006). Economic and environmental analysis of buffer zones as an instrument to reduce ammonia loads to nature areas. Land Use Policy, 23, 533–541. Skiba, U., Pitcairn, C., Sheppard, L. etâ•¯al. (2004). The influence of atmospheric N deposition on nitrous oxide and nitric oxide fluxes and soil ammonium and nitrate concentrations. Water, Air and Soil Pollution, Focus, 4, 37–43.

Smith, J. U., Bradbury, N. J. and AdT, M. (1996). SUNDIAL:Â€a PC-based system for simulating nitrogen dynamics in arable land. Agronomy Journal, 88, 38–43. Smith, R. J. N., Glenn, G. A., Parkinson, R. and Richards, J. P. (2007). Evaluating the implementation of the Nitrates Directive in Denmark and England using an actor-orientated approach. European Environment, 17, 124–144. Sovik, A. K., Augustin, J., Heikkinen, K. etâ•¯al. (2006). Emission of the greenhouse gases nitrous oxide and methane from constructed wetlands in Europe. Journal of Environmental Quality, 35, 2360–2373. Spieles, D. J. and Mitsch, W. J. (2000). The effects of season and hydrologic and chemical loading on nitrate retention in constructed wetlands:Â€a comparison of low- and high-nutrient riverine systems. Ecological Engineering, 14, 77–91. Stockle, C. O., Donatelli, M. and Nelson, R. (2003). CropSyst, a cropping systems simulation model. European Journal of Agronomy, 18, 289–307. Strayer, D., Beighley R. E., Thompson, L. C. etâ•¯al. (2003). Effects of land cover on stream ecosystems:Â€roles of empirical models and scaling issues. Ecosystems, 6, 407–423. Sutton, M. A., Dragosits, U., Theobald, M. R. etâ•¯al. (2004). The role of trees in landscape planning to reduce the impacts of atmospheric ammonia deposition. In:Â€Landscape Ecology of Trees and Forests, ed. R. Smithers, IALE (UK)/ Woodland Trust, Grantham, UK, pp.Â€143–150. Sutton, M. A., Nemitz, E., Erisman, J. W. etâ•¯al. (2007). Challenges in quantifying biosphere-atmosphere exchange of nitrogen species. Environmental Pollution, 150, 125–139. Sutton, M. A., Howard, C. M., Erismar, J. W. etâ•¯al. (2011). The challenge to integrate nitrogen science and policies. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.Â€W. Erisman etâ•¯al. Cambridge University Press. Tanner, C. C., Nguyen, M. L. and Sukias, J. P. S. (2003). Using constructed wetlands to treat subsurface drainage from intensively grazed dairy pastures in New Zealand. Water Science and Technology, 48, 207–213. Thenail, C. (2002). Relationships between farm characteristics and the variation of the density of hedgerows at the level of a micro-region of bocage landscape: study case in Brittany, France. Agricultural Systems, 71, 207–230. Thenail, C. and Baudry, J. (2004). Variation of farm spatial land use pattern according to the structure of the hedgerow network (bocage) landscape:Â€a case study in northeast Brittany. Agriculture Ecosystems and Environment, 101(1), 53–72. Thenail, C. and Baudry, J. (2005). Farm riparian land use and management:Â€driving factors and tensions between technical and ecological functions. Environmental Management, 36, 640–653. Theobald, M. R., Milford, C., Hargreaves, K. J. etâ•¯al. (2001). Potential for ammonia recapture by farm woodlands:Â€design and application of a new experimental facility. The ScientificWorld, 1(S2), 791–801. Theobald, M. R., Dragosits, U., Place, C. J. etâ•¯al. (2004). Modelling nitrogen fluxes at the landscape scale. Water, Air and Soil Pollution, Focus, 4, 135–142. Theobald, M. R., Bealey, W. J., Tang, Y. S., Vallejo, A. and Sutton, M. A. (2009). A simple model for screening the local impacts of atmospheric ammonia. Science of The Total Environment, 407, 6024–6033. Thöni, L., Brang, P., Braun, S., Seitler, E. and Rihm, B. (2004). Ammonia monitoring in Switzerland with passive samplers:Â€patterns, determinants and comparison with modelled concentrations. Environmental Monitoring Assessment, 98, 93–107.

247

Nitrogen flows and fate in rural landscapes Turner, B. L., Meyer, W. B. and Skole, D. L. (1994). Global land-use land-cover change – towards an integrated study. Ambio, 23, 91–95. Van Jaarsveld, J. A. (1995). Modelling the long-term atmospheric behaviour of pollutants on various spatial scales. PhD Thesis, Universiteit Utrecht, Utrecht. Van Pul, W.A.J., van Jaarsveld, J.A., Vellinga, O.S., van den Broek, M. and Smits, M.C.J. (2008). The VELD experiment:Â€An evaluation of the ammonia emissions and concentrations in an agricultural area. Atmospheric Environment, 42, 8086–8095 Viaud, V., Merot, P. and Baudry, J. (2004). Hydrochemical buffer assessment in agricultural landscapes:Â€from local to catchment scale. Environmental Management, 34, 559–573. Vogt, J., Puumalainen, J., Kennedy, P. and Folving, S. (2004). Integrating information on river networks, catchments and major forest types:Â€towards the characterisation and analysis of European landscapes. Landscape Urban Planning, 67, 27–41.

248

Weathers, K., Cadenasso, M. and Pickett, S. (2001). Forest edges as nutrient and pollutant concentrators:Â€Potential synergisms between fragmentation, forest canopies, and the atmosphere. Conservation Biology, 15, 1506–1514. Westergaard, K. (2006). The landscape composition of organic and conventional, dairy and crop farms in two different geological regions in Denmark. Agriculture Ecosystems and Environment, 117, 63–70. Willems, E., Vandevoort, C., Willekens, A. and Buffaria, B. (2000). Landscape and land cover diversity index. In:Â€From Land Cover to Landscape Diversity in the European Union, European Commission:Â€DG AGRI, EUROSTAT and the Joint Research Centre (Ispra). Woltemade, C. J. (2000). Ability of restored wetlands to reduce nitrogen and phosphorus concentrations in agricultural drainage water. Journal of Soil and Water Conservation, 55, 303–312.

Chapter

12

Nitrogen flows and fate in urban landscapes Lead authors: Anastasia Svirejeva-Hopkins and Stefan Reis Contributing authors: Jakob Magid, Gabriela B. Nardoto, Sabine Barles, Alexander F. Bouwman, Ipek Erzi, Marina Kousoulidou, Clare M. Howard and Mark A. Sutton

Executive summary Nature of the problem • Although cities take only 1.5%–2% of the Earth’s land surface, due to their dense population, settlement structure, transportation networks, energy use and altered surface characteristics, they dramatically change the regional and global nitrogen cycle. Cities import and concentrate Nr in the form of food and fuel, and then disperse it as air and water pollution to other ecosystems covering much larger areas.

Approaches • A mass-balance approach was used in order to quantify the fluxes of reactive nitrogen (Nr) in and out of cities. • Cities can be characterised either as a source of Nr (i.e. emitting large amounts as liquid or solid household waste, automobile exhaust, air pollution from power plants) or a sink of Nr (through importing more food, fossil fuels, etc., and having fewer emissions to the air and water). • Paris metropolitan area is used as a case study, which represents an evolving European capital with much available data.

Key findings/state of knowledge • The Paris Metropolitan Area changed from being a sink in the eighteenth and nineteenth centuries to a source of Nr today. Major changes in the city functioning occurred before 1950, but especially recent decades have been characterised by an unprecedented amplification of those changes. • The major part of Nr output is attributed to the combustion of fossil fuels for transport and energy, which converts both atmospheric N2 and fossil Nr to reactive NOx. The second largest Nr contribution comes from incineration of solid waste, and third highest emissions come from sewage water treatment plants. • Urban wastewater discharge into rivers largely contributes to N contamination of the aquatic environment, although sophisticated and expensive tertiary treatment techniques are now available to drastically reduce Nr emissions. • Denitrification in urban landscapes is controlled by the presence of water bodies and green areas. These areas have lower biomass and decomposition rates than natural ecosystems.

Major uncertainties/challenges • A better understanding is needed regarding the following uncertainties:Â€(a) the mechanisms of dry-deposition in urban systems with patchy vegetation; (b) the complex patterns of air flow in densely built-up areas; (c) the fate of Nr in urban soils with altered water regimes and impermeable surfaces.

Recommendations • To achieve sustainability of urban systems in relation to the N cycle, road transport of goods and passengers has to be reduced, household waste generation minimised, and wastewater treatment improved. • More attention should be given to future sewage processing systems that process Nr (and other nutrients) for reuse as a fertiliser rather than losing the Nr resource by denitrifying it back to N2. • Such measures could eventually turn urban areas from sources of Nr to N-neutral or even N-sink areas. Regional adaptation measures should be specifically tailored to the individual urban ecosystems of Europe.


249

Nitrogen flows and fate in urban landscapes

12.1â•‡ Introduction 12.1.1â•‡ Problem setting and approach In this chapter we ask and try to answer the following questions:Â€ what are the specific features of urban landscapes in Europe and how does the local urban N cycle differ from that for natural ecosystems? What are the important issues concerning N management in cities in Europe as a region? As a case study we chose Paris for a more in-depth quantification of terrestrial and atmospheric fluxes, input and output to the system-city. But first of all, why is a chapter dedicated to nitrogen fluxes in urban systems required for the European Nitrogen Assessment (ENA)? Although cities take only 1.5%–2% of the Earth’s land surface, due to their dense population, settlement structure, transportation networks, energy use and altered surface characteristics, they change the N cycle substantially. Firstly, they fix substantial amounts of atmospheric N2 to Nr as NOx through the high temperature combustion of fossil fuels. Secondly, they drive the industrial fixation of Nr to fertilisers, importing the Nr produced in food for burgeoning urban populations, subsequently dispersing it in air and water pollution to other ecosystems over much larger areas than the cities themselves. In other words, they act as concentrators, transformers and dispersers of N, representing new entities of the Earth System. We use an ecosystem approach in this chapter. Any ecosystem is an open system, whose functioning is supported by inand out-fluxes of matter and energy. These fluxes constitute the system itself and determine its boundaries (Odum, 1973).

12.1.2â•‡ The city as an ecosystem Any city, and especially an industrial one, represents an incomplete or ‘heterotrophic’ system, receiving energy, food, materials, water and other substances from outside of its boundaries. The city could be considered as a specific heterotrophic ecosystem that differs very much from a natural heterotrophic ecosystem. In fact, a city has a more intensive metabolism per area unit, requiring a significant inflow of artificial energy. Its consumption per urban area unit may be three to four orders of magnitude higher than for a same-sized rural non-agricultural area. One hectare of the city area may use 1000 times more energy than the same area of the rural territory. During the process of its own metabolism, a city consumes large amounts of various materials:Â€food, water, wood, metals, etc., all that we call ‘grey energy’. Products of city’s metabolism are large volumes of substances that are more toxic than those produced by natural ecosystems. Most cities have wide green belts (consisting of trees, bushes, lawns, as well as ponds and lakes), so it could be said that some autotrophic component is also present. Wakernagel and Rees (1997) introduced the concept of the ecological footprint, which provides an account of the total area of productive surfaces required to produce, under prevailing technology, the resources consumed by a country or a city. Feeding modern cities is associated with one-third to half of the global ecological footprint, ranging world-wide from 0.8 to

250

3.8 global-hectares per inhabitant (WWF, 2002). This simple ‘footprint’ indicator, however, neither fully describes the complexity of the relationships that a city establishes with its rural hinterland for the supply of food, nor how, over time, these relationships impact upon the development of both the city and the countryside. Billen etâ•¯al. (2009) introduced the so-called ‘food-print’ indicator that takes into account the area required for producÂ� ing agricultural goods, being expressed in terms of the effective surface of the surrounding territory needed to support the food requirements of a city. Billen etâ•¯al. (2009) found that despite a 50-fold increase in the population of Paris since the fifteenth century, the food-print of the city barely increased twofold. In contrast, the further doubling of the population in the twentieth century was paradoxically accompanied by a fivefold decrease in the food-print, because of the intensification of agricultural production. While emphasising the scale of the changes for Paris, this example also illustrates that the ‘food-print’ is only a partial indicator of the impact of a city. For example, it does not include a comprehensive account of all resources required, such as energy, fertilisers and waste disposal, which might be converted into additive ‘global hectare equivalent’ units. From an ecological point of view, a modern city could be defined as a ‘parasite’ of its rural surroundings. As things are at present, a city does not produce significant quantities of food or other organic substances, it does not purify its air and it does not return water or inorganic substances to the pristine state of their respective biogeochemical cycles. However, theoretically, one could consider a city from a different perspective, for example looking at it as being in a symbiotic relationship with its environment, because a city produces and exports goods and services, money and cultural values, receiving in return goods and services as well.

12.1.3â•‡ The urbanisation processes Urbanisation is considered to be one of the most powerful and characteristic anthropogenic forces on Earth in the twenty-first century. Although, as noted above, cities occupy only 1.5%–2% of the Earth’s land surface, they are home to over 50% of the world’s population. The number of city dwellers has increased from a mere 14% of world population in 1900 and will further increase to an estimated 60% by 2030 (UNCHS, 2003). The total number of people living in major European cities in the 1400s was only 1.1 million, increasing to 3.5 million in the 1700s, and 3.8 million in the 1800s. However, during the twentieth century, an exponential growth of urban dwellers took place, accompanied by a large increase in the number of large cities (Hohenberg and Hollen, 1985). At present, urbanisation proÂ�cesses in Western Europe have reached the so-called ‘stagnation period’, giving an example of saturated growth. This is, however, accompanied by the expansion of medium-sized cities (~1–2 million). The dynamics of European urbanisation, the differences between urban patterns in Western and Eastern Europe and the implications for the terrestrial, water and atmospheric parts of the N cycle are discussed in greater detail in the following sections.

Anastasia Svirejeva-Hopkins and Stefan Reis

The N cycle, one of the main cycles in ‘Biosphera Machina’, is closely interconnected with the carbon, water and oxygen cycles. Urbanisation-related disturbances in the main driving cycles, caused by urban human activities, lead to global, regional and local environmental problems, such as global warming, photochemical smog, soil acidification, and eutrophication pollution of surface, ground and coastal waters. Even though in some cities urban population might stabilise or even slightly decrease, actual urban areas as such are expected to continue to expand in the twenty-first century, accompanied by growing energy production, increased food demand, and expanding transportation and industrialisation. The demands of high production to feed the urban population alter land cover, biodiversity, and hydrology, both locally and regionally. Similarly, urban waste discharge affects local to global biogeochemical cycles and climate. Although agricultural production is by far the largest influence which has caused the amount of Nr entering the biospheric cycle to double compared to pre-industrial conditions (Smil, 1999), today more than half the crops produced in rural areas are consumed in urban zones. Transportation and industry are concentrated in urban centres, making them point sources of N containing greenhouse and other trace gases such as NOx, N2O and many organic Nr forms (Pataki etâ•¯al., 2006). Air and in particular water pollution influence nutrient cycling and primary production in adjacent ecosystems, especially as many major European cities are located along rivers and coastlines. The Nr in solid wastes generated in cities also eventually enters the air and water, affecting biogeochemical cycles, while the extent of influence depends on the vectors by which materials are carried (e.g. rubbish disposed of by landfill or incineration, dispersal of emitted NOx and NH3, etc.). In addition, it is important to study urban areas in the context of the ENA, since two of the five key threats identified in ENA (Sutton etâ•¯al., 2011, Chapter 5, this volume), namely water quality and air quality, are important aspects of the functioning of urban landscapes. Moreover, pollution generation by cities is a matter of increasing concern, especially when urbanisation outpaces societal capacity to implement pollution-control measures. Therefore, it is important to assess the current situation and to forecast the dynamics of N biogeochemical functioning of urban landscapes. This chapter discusses the past and current situation and touches upon the future trends of the development of Europe’s urban systems that are related to the N cycle.

12.2â•‡ Urban geography 12.2.1â•‡ Regional physiography Europe’s long historical development has led to well established trade and communication with the rest of the world. Almost nowhere in Europe is far from water, and water routes have historically facilitated contact between peoples and cultures resulting in the circulation of goods and ideas. The earliest European cities were thus built and developed

adjacent to water routes. The hundreds of miles of navigable waterways, straits and channels between many islands, peninsulas and the mainland, the accessible Mediterranean, North and Baltic Seas, all provided the routes for the exchange of merchandise. Later, the oceans became the means of longdistance spatial interaction. This historical advantage of moderate distances applies to the mainland as well. No place in Europe is very far from any other place on the continent, although nearby places are often different from each other. Short distances and large differences enable much interaction, which is typical of European geography over the past 1000 years. The climate in Europe is mild and temperate. Europe’s biomes (Bazilevich, 1979) are on average high in biomass density; for example, storage of Nr in the deciduous forest that covers large parts of central Europe can be as high as 1100 kg N/ha (for oak-dominated forest including roots). On average, the Nr content in biomass of the deciduous forest biome of Central Europe is 310 kg /ha, while in the mixed forest surrounding Moscow it is estimated at 496 kg N/ha (Bazilevich and Tytlianova, 2007). European soils are generally fertile. For example, there are many alluvial soils, formed in the river valleys and deltas, used to grow crops for centuries, while the Russian and Ukrainian black-soil (‘chernozems’) are extremely fertile. The North European Lowland topographic region, extending from Southern Britain to Western Russia, is the most densely populated physical region of Europe and a route of contact between Europeans and their neighbours to the East. The United Nations (UN) regional classification divides European countries according to their developmental stage as either Highly Industrialised (Western Europe) or Economies in Transition (Eastern Block and former USSR). The dynamics of city development and the process of urbanisation in general, differ somewhat according to region, as discussed in the next sub-section.

12.2.2â•‡ Urban demography and projections The total population of Europe is around 702 million, distributed over a land area of 9.8 million km2. Following the demographic explosion which started in the middle of the last century, European rates of population growth and of Nr produced by anthropogenic activities broadly matched the global trends, with similar increases during the period 1980 to 1990 (Erisman etâ•¯al., 2008). However, these growth rates have declined from an all time peak (2.1% per year between 1965 and 1970) to about 1.6% per year, while human influence has increased faster than population growth (Cohen, 1995). The UN forecasts that virtually all population growth from now until 2030 will be concentrated in the urban areas of the world. In Europe, the percentage of the population living in urban areas is expected to rise from 77% in 2000 to 84% in 2030, with some countries reaching 90%. However, the proportion of people living in very large urban agglomerations is still small:Â€in 2000, 48% of the population in developed countries lived in cities of less than 1 million inhabitants and by 2015 that proportion is expected to rise to 49% (UN, 2008).

251

Nitrogen flows and fate in urban landscapes Table 12.1 The largest urban areas in Europe, ranked from 200 world largest (base year 2005)

Rank (World)

Country

Urban Area

Population (106)

Land area (103 km2)

Density (103 person/km2)

16

Russia

Moscow

13.3

4.5

2.9

22

France

Paris

10.4

3.1

3.4

29

UK

London

8.3

1.6

5.1

38

Germany

Essen-Dusseldorf

7.3

2.6

2.8

52

Spain

Madrid

5.0

0.9

5.2

63

Russia

St Petersburg

4.6

1.9

3.9

68

Italy

Milan

4.2

2.4

4.6

79

Greece

Athens

3.7

0.7

5.4

111

Italy

Naples

3.0

0.8

3.9

124

Italy

Rome

2.8

9

3.2

134

Ukraine

Kiyev

2.5

0.5

4.6

143

Portugal

Lisbon

2.3

0.9

2.5

154

Germany

Frankfurt

2.7

0.7

3.4

155

UK

Birmingham

2.3

0.6

3.8

176

Netherlands

Rotterdam-Hague

2.1

0.8

2.5

181

Hungary

Budapest

2.1

0.9

2.4

183

Germany

Cologne-Bonn

2.0

0.9

2.1

186

Poland

Warsaw

2.0

0.5

3.7

Source:Â€Demographia, 2009.

Not all large urban agglomerations experience fast population growth. In fact, some of the fastest growing cities have small populations and, as population size increases, the growth rate tends to decline. Some European cities are actually shrinking, with Rome having a negative urban population growth rate of −0.04%; for example, Budapest −0.01% and St.Â€ Petersburg −0.09% (Demographia, 2009). Some East German cities are also experiencing a reduction of population due to the changing labour markets. However, in general for Europe, even cities with shrinking populations tend to be sprawling in area. In other words, even where there is little or no population pressure, a variety of factors are still driving ‘sprawl’ (i.e. increase in city area). These changes are rooted in the desire of people to realise new lifestyles in suburban areas, outside the inner city. The factors reflect micro and macro socio-economic trends, e.g. transport quality, land prices, planning policies, cultural traditions and the attractiveness of respective urban areas. Thus, four-fifths of European citizens now live in towns and cities. As cities start to increase in size, the social infrastructure grows faster than population, i.e. the surface areas of streets, electricity network length, etc., lag behind the city’s population growth, while income and certain measures of innovation outpace it (Bettencourt, etâ•¯al., 2007). According to the same author, individual human needs (housing, employment, household electrical consumption etc.) are scaled linearly for cities of different sizes. This observation provides us with the grounds for using per capita-based calculations later in this chapter.

252

On a global scale, cities with a population of 0.5 million and smaller are anticipated to grow the most rapidly in the course of the next 50 years. In Europe, already-existing cities of between 0.5 and 2 million inhabitants are projected to expand the most rapidly in the course of the next 40 years, which calls for expansion of the cities considered in the present assessment. The model of Svirejeva-Hopkins and Schellnhuber (2008) estimates that in Western Europe, the total urban area (including cities smaller than 1 million inhabitants) will increase from 1645 * 103 km2 in the year 2010 to 1661 * 103 km2 in 2030 and subsequently decrease again slightly below the 2010 value to 1641 * 103 km2, owing to the saturation in the urban population growth rate. The dynamics are different for Eastern Europe, where urban areas are projected to grow from 133 to 134 * 103 km2 by 2030, then decrease in 2040, and then increase again to 134 * 103 km2 by 2050. There is also a substantial difference in the regional relative urban areas that are currently 40% for Western Europe (of the total land) and only around 2% for Eastern Europe. This indicates the different patterns of urban developmentÂ€– sprawl versus density increase.

City sizes and types With a population of more than 13 million, Moscow is the largest European urban area today (see Table 12.1). Paris is the second largest, followed by London, Essen-Dusseldorf, Madrid and St. Petersburg. There are, in total, 58 urban areas in Europe with populations of one million and more. An urban area (urbanised area agglomeration or urban centre) is defined


as a continuously built up landmass of urban development (Demographia, 2009). It generally defines the ‘urban footprint’, or the lighted area that can be observed from an airplane at night. This chapter confines urban areas to a single metropolitan area or labour market area. What constitutes a particular metropolitan area is a matter of professional judgment. However, there is a necessity to ‘draw a line’, especially where adjacent urban areas have ‘grown together’, but remain fairly distinct labour markets. Having considered the list of all European cities, we could categorise city types as: • small cities with a population of under half a million, • medium-sized cities ranging from 0.5 to 2 million, • large cities with a population of 3 million people or more. The so-called combined urban areas (‘mega-cities’) are not characteristic of Europe; this is a typical American, Chinese, and to some extent Indian phenomenon, while there are no mega-cities in Europe to date. While Moscow and London have populations of 13 and 8 million respectively, they neither have catastrophic urban densities nor occupy vast amount of land, as typical mega-cities do. However, some twin cities are already emerging in Europe; for example Essen-Dusseldorf (Germany), Marseille–Aix-en-Provence (France), and Rotterdam-Hague (Netherlands). In this chapter we focus on the settlements of one million and more, since in terms of N fluxes, they play the role of indirect consumers (N for fertiliser, fixed to ‘feed’ them) and sources of Nr at the regional scale. Special consideration, in view of future urban trends, should be given to the type of settlements known as ‘new Â�cities’Â€ – planned communities, also called ‘new towns’, since they are still relatively small. Carefully planned from the start and typically constructed in a previously undeveloped area, they contrast with settlements that evolve in an ‘old-fashioned’ way. There are only a few of them in Europe so far, their populations are small (a few thousand) and some of them are not developing as planned. One successful example is Louvainla-Neuve, the French-speaking university town in Belgium built in 1972 with originally only 600 permanent residents in it, which has experienced rapid growth, reaching 10 477 inhabitants in 1981. This town is also an example of the ‘New Pedestrianism’ movement, e.g. where roads are in many cases directed under the city. Another representative new city is Tapiola (population of 16 000), constructed in the 1950s and 1960s by the Finnish apartment foundation and designed as a garden city. An interesting example is the city of Slobomir in Eastern Europe, which aims to become one of the major cities of post-war Bosnia–Herzegovina and Serbia, which is still under construction.

12.2.3â•‡ Urban density as an integral indicator The western part of Europe is characterised by the most uniform demographic processes. France, Europe’s second largest country, has the lowest number of large-city dwellers, at only 10.4%. By contrast, Russia has one of the highest figuresÂ€– 42%

of inhabitants reside in large towns and cities. Countries that formed part of the former Soviet Union are similar in that respect. In the Ukraine, 37% of the population live in cities with more than 150 000 residents, in Belarus 40%. The historical development of Germany and Italy led to the formation of a large number of important but smaller cities. In Germany, 26% of the population live in Großstädten (‘large cities’), while only 21% of Italians reside in cities with populations of more than 150 000 people. Poland is the country where the share of the rural population is the highest:Â€only 24% of people live in large towns and cities. With more than 8 million residents, London alone accounts for almost 12% of the UK population. According to statistics, about 51% of Britons live in towns and cities with more than 150 000 inhabitants. However, this figure could be inaccurate, since some smaller towns have been administratively merged with their surrounding rural districts. Suburban structure varies throughout Europe depending on the individual city’s location, the urban spatial typology, social status and functioning. Today, suburbs are a mosaic of mainly isolated fragments of different housing types, enriched by infrastructure facilities like retail stores and offices subÂ� divided by transport networks. Central Western Europe and especially France and Great Britain have experienced a tough social exclusion of non-privileged classes that have had to settle in compact suburbs consisting of social housing facilities. In Central and Eastern Europe, housing construction was limited to huge, industrialised mass-housing estates. Western-style suburbia never developed, which is one of the core differences between Eastern and Western urban development. Currently, this difference is rapidly disappearing, and luxurious housing estates multiply in the suburban areas of former socialist cities. The urban core is losing its inhabitants, while the suburbs grow as a whole. This is particularly true for Eastern Germany. Figure 12.1 shows the European part of Tobler’s updated world population density map (Tobler, 1995); urban densities are notably redder and large urban areas are detectable. In broad terms, this map of urban density provides an indicator of NOx emissions intensity, as can be seen by comparison with mapped tropospheric NO2 concentrations (Beirle etâ•¯al., 2004; Simpson etâ•¯al., 2011, Chapter 14, this volume). In Figure 12.2, one can clearly see that different types of cities form different clouds in the plot. For example, such major cities as Moscow, Paris and London clearly stand out; however their densities are not especially high compared to some unsustainable Eastern European and Russian cities. Samara and EkaterinburgÂ€– although occupying relatively small areas and being only medium-sizedÂ€ – have the highest population densities in Europe with 8.4 and 9 thousand inhabitants per km2 respectively. Bucharest, the capital city of Romania, is of medium size (2 million) and is also quite dense. If we look at Table 12.2 showing the time taken to travel to work in a sample of cities, we see that Bucharest is characterised by the highest time of almost 80 minutes. Since the area of city is not too large, this may indicate a high level of vehicle congestion, that the public transport networks are not efficient or that the city has developed without one core financial commercial district,

253


Figure 12.1 European population density and the location of major cities.

but with many small business areas. At the same time, it may reflect that people mainly live in dense suburbs and travel to work in a rather small downtown area. All these factors would directly influence the air quality in this city and consequently pattern of NOx emissions and atmospheric concentrations. Figure 12.3 illustrates another set of urban indicators. This shows, for example, that only 0.01% of wastewater was treated in Bucharest in 1999! Hopefully the situation has changed since then, because with 20% waste incinerated, the rest was deposited in open dumps. Obviously, the emissions of Nr into the water would be very high in this case. Based on Figure 12.2, some medium-sized German and French cities of the lower left corner of the diagram such as Marseille and Lyon appear to be more sustainably managed

254

than other cities shown. Lyon has a medium-sized population, low urban density and 100% of its wastewater is treated, while most of the solid waste is incinerated and only 4% properly land-filled, with some recycling taking place as well. While these indicators do not specify the efficiency of the water treatment and incineration plants, it is notable that Lyon also has a well developed public transport system combining buses, metro, funiculars and tram lines. The picture is of a city with a carefully organised infrastructure having the potential to reduce Nr emissions to air and water. When dealing with urban transportation, not only the average urban density and the geographical expanse of urban areas are important, but also the differences in internal population density, i.e. density gradient. The average urban density data could mask significant variations within urban areas. For

Anastasia Svirejeva-Hopkins and Stefan Reis Table 12.2 Mean travel time to work as an indicator, reflecting internal urban density gradients for selected European cities and areas (year 1999)

Mean travel time to work (in minutes)

example, London and Athens have similar population densities; however the core (central business district) densities in Athens are considerably higher than in London. The Athens suburbs, however, are among the least dense in the world. Similarly, the Essen-Dusseldorf and Milan urban areas have almost identical densities, yet core densities are considerably higher in Milan. This is because with the geographical expanse of nearly all modern, high-income urban areas, automobiles provide by far the greatest coverage, with considerably shorter travel times than public transport. For example, automobiles account for 88% of travel in the Essen-Dusseldorf urban area and somewhat more than 77% in Milan, with its steeper density gradient. These gradients also play a central role when considering wastewater treatment, which is addressed in detail in the following section. Urban density or structure has an impact on air quality and in turn on the health of urban residents. Results of one study (Ferreira etâ•¯al., 2008) indicate that although compact cities provide better air quality compared with dispersed cities, the former have greater exposures and thus a higher health risk, due to high population density. Urban density could clearly serve as some integral indicator that reflects the quality of life, including Nr pollution levels, in cities of different types. As already mentioned, many cities of lower middle ‘cloud’ in Figure 12.2 are expected to expand at a high rate in the next 50 years. Therefore, the direction they will move on a plot has important implications for the anthropogenic urban Nr emissions.

City

Country

Amsterdam

Netherlands

22

Athens

Greece

53

Copenhagen

Denmark

22

Glasgow

United Kingdom

32

Hertfordshire

United Kingdom

27

Koeln

Germany

32

Lyon

France

32

Paris

France

35

Stockholm

Sweden

35

Donetsk

Ukraine

51

Minsk

Belarus

51

Moscow

Russian Federation

62

Nizhny Novgorod

Russian Federation

35

Tbilisi

Georgia

70

Yerevan

Armenia

52

Belgrade

Serbia

35

Bucharest

Romania

78

Budapest

Hungary

40

Prague

Czech Republic

57

Riga

Latvia

27

Comparing European sub-regions

Sofia

Bulgaria

35

Warsaw

Poland

34

Zagreb

Croatia

26

One important difference highlighted by the comparison of Bucharest and Lyon, is that they are situated in Eastern and Western Europe, respectively. The UN sub-divided the world according to the economic developmental stage and, as expected, the difference between Economies in Transition (ET) and Highly Industrialised (HI) countries within Europe is clearly reflected in the urban densities. When comparing total urban areas and populations in cities of the two regions, we can see that ET exceed the values of HI region by two to six times, and that European cities have in general lower density than cities of Russia and Eastern Europe. For cities of half a million or more, Western Europe has an average urban density of 3150 inhabitants per km2; Western Europe outside the UKÂ€ – 3000; UKÂ€ – 4100; Europe except RussiaÂ€ – 4200; RussiaÂ€– 4900 (Demographia, 2009).

Source:Â€UNCHS Global Urban Indicators Database (2003).

12.3â•‡ N-fluxes and city sub-systems, including a case study of the Paris Metropolitan Area

Figure 12.2 Population size versus density for the European cities of 1Â€million people or more.

Cities show symptoms of the biogeochemical imbalances that they help to create. In urban systems, additional N inputs occur primarily via the importation of foodstuffs for humans, as well as by inadvertent ‘fertilisation’ through the production and subsequent deposition of NOx derived from the combustion of fossil fuels. Cities also experience high acid deposition.

255


100% other

90% 80%

recycling

70% 60%

open dump

50% incineration

40% 30%

land-fill

20% 10%

Nitrogen transfers in human-dominated ecosystems are inherently inefficient; there is leakage of N at each point of the food chain from fertilisation through human excretion. These leaks could lead to increased storage in soil and groundwater pools and losses to rivers. Air pollutants are transported over both short and long distances (as far as a few thousand kilometres) before being deposited on surface water, vegetation or soil (Bobbink, 1998). In this way, vegetation over a large area or in remote regions can be influenced by airborne pollutants (see Fowler etâ•¯al., 1998; Asman etâ•¯al., 1998). Elemental mass balances can frame this problem, because they identify potential excesses of inputs over outputs and likely sinks within the urban landscape (Baker etâ•¯al., 2001). Usually cities are hotspots of accumulation of N, P, and metals and, consequently, harbour a pool of material resources. By constructing mass balances at scales from the household to the city, human choice can be linked directly to biogeochemical cycling (Kaye etâ•¯al., 2005). The following sections describe the general functioning of city sub-systems and development of the urban N budget for the Paris Metropolitan Area (PAM). They provide a view of the region’s history, current status and projected impacts of N accumulation in adjacent areas, which generally has caused negative impacts. The N budget serves as a planning tool that is based on the estimation of gross N contributions from different N sources/components of the N cycle entering the system, as well as the amount of N leaving it. Such an analysis illustrates the spatial heterogeneity in both Nr creation and distribution of N from a local to a regional scale. The relatively simple N budget for Paris provides an assessment of the relative contributions of sources and the potential benefit of changes of management practices in the PAM. The subsequent sections (Sections 12.3.3 to 12.3.6) discuss urban N fluxes in a more regional context, supporting the statements of the case study for the PAM.

Zagreb (CRO)

Sofia (BG)

Warsaw (PL)

Riga (LAT)

Prague (CZR)

Budapest (HUN)

Belgrade (SRB)

Bucharest (ROM)

Tbilisi (GRG)

Yerevan (ARM)

Nizhny Novgorod (RUS)

Minsk (BLR)

Moscow (RUS)

Donetsk (UKR)

Stockholm (SE)

Lyon (F)

Paris (F)

Koeln (DE)

Glasgow (UK)

Hertfordshire (UK)

Athens (GR)

Copenhagen (DK)

Amsterdam (NL)

0%

256

Figure 12.3 Liquid and solid household waste indicators for selected European cities for the year 1999. A substantial share of wastewater is treated in the majority of the cities listed (exceptions are Zagreb, Belgrade and Bucharest) and the level of land-filling is quite diverse among Western European cities, whereas open dumps are frequently used in Eastern and South-Eastern Europe.

share of wastewater treated

12.3.1â•‡ A case study and its historic development The Paris Metropolitan Area occupies the Île-de-France, the geographic region constituting the lowland area around the city. This area, which forms the heartland of France, is drained largely by the Seine River and its major tributaries converging on Paris. The natural vegetation of the basin, broad-leafed deciduous forest biome, has been almost entirely lost to civilisation, except for a few relict forests. In order to emphasise what is happening in urban areas concerning N it is relevant to go back as far as the end of the eighteenth century. Paris, an old European capital, is a good illustration, since some major historical changes occurred before 1950, while recent decades have been characterised by an unprecedented amplification of those changes. The population of Paris has dramatically increased since the beginning of the nineteenth century (see Figure 12.4). Paris at the end of the eighteenth century and beginning of the nineteenth century already represented a substantial hub of Nr flows. The concentration of humans and animals (especially horses) in the city is estimated to have required an input of Nr of 24 g per head per day to meet the combined dietary needs. Material from cesspools and other organic matter placed in the streets came from households, animals, butchers, slaughterhouses and other industrial activities. As a result of waste infiltration through the surfaces, the content of Nr in soils and underground water was high, for example the nitrate content in water from Paris wells was up to 2.2 g/L (Boussingault, 1858). Household water supply was nearly non-existent, but the river Seine was more or less preserved from human excreta discharge. This was however not the case for small industrial rivers (like the Bièvre in Paris) where water quality was poor. Much of the N was recovered from the city waste, with urine and ‘night soil’ being collected in carriages and transferred to


2500

2000

Firewood (*1,000m3) Charcoal (*1,000 m3) Coal (*1,000 tons)

1500

1000

500

0 1855

Figure 12.4 The human population of Paris, of the Seine catchment, of the urban area of the Seine and of the whole of the Île-de-France region, 1801–1999 (millions of people).

the city refuse depots to obtain the Nr rich urine fraction and a phosphate rich fertiliser powder ‘poudrette’. The latter also contained significant amounts of Nr, though much less than the liquid fraction, especially as part of the Nr was volatilised as ammonia during its preparation (Barles and Lestel, 2007). Wood and, to lesser extent, coal combustion were responsible for emissions to the airÂ€– to these emissions of N2O (from local denitrification) would have been added, although the latter can be considered as having a much smaller scale. Overall, the main urban impact on the N cycle at that time was in the form of underground accumulation, riverine discharge and emissions to the air. The next stage of Paris’s evolution was from the mid nineteenth century to the beginning of the twentieth century. The rates of human and animal concentration in cities kept increasing with more people moving in. Food production therefore became a central issue and so did the greater needs for fertilisers. From the 1820s onwards, cities came to be recognised as sources of fertilisers and the main concern was N recovery. There were many discussions between Boussingault and Liebig about this issue. As Jean Baptiste Dumas said:Â€‘one of the most beautiful problems in agriculture lies in the art of obtaining nitrogen at low cost’ (Dumas, 1844, as translated by Barles and Lestel, 2007). In addition to the production of poudrette, from the 1830s, ammonium sulphate was manufactured in Paris using urine, and many patents involving the use of human manure and dry fertilisers were developed all around Europe. The processing of Nr rich waste was particularly well developed in Paris (Barles and Lestel, 2007). Thus by the late nineteenth century around 50% of the city’s excreta was collected and industrially processed. The excreta were settled and the eau vanne distilled industrially to produce ammonia (Vincent, 1901). Using the Margueritte process, the yield was estimated at 2.5–3 kg NHx-N per m3 of eau vanne (Vincent, 1901, p 6 ff, p 19). Based on these estimates, this would have amounted to around 800 000 tonnes of excreta processed industrially every year, from which around 2000 to 2400 tonnes of ammonium N were produced, mainly as ammonium sulphate. Combined with the processing of excretal solids to produce poudrette and other fertilisers, N recovery rates increased, but not enough to

1865

1875

1885

1895

1905

1915

1925

1935

1945

Figure 12.5 Firewood, charcoal, and coal consumption, Paris, 1855–1943.

counterbalance the effect of urban population increase in the inner city and suburbs. Water supply to households had been very much improved, but the question still remained of what to do with the water once used. Untreated wastewater provided a major source of pollution to the River Seine, with both high organic matter and ammonium content. Energy consumption (heating systems, gas production, industrial development) continued to increase. Coal progressively replaced firewood and was used either directly or turned into gas, causing the increase in related emissions (see Figure 12.5). By the start of the twentieth century a major change occurred with the rapid growth of the household water supply, the introduction of British-style flushing toilets and the development of the piped sewage system, to which 10% of the population was connected in 1895 and 70% by 1914 (Barles, 2007). Flushing toilets produced much more dilute sewage streams, which were supplied as a liquid fertiliser to surrounding agricultural land. However, industrial processing of dilute sewage was much less cost-effective, and this was therefore a major factor contributing to the obsolescence of the system of recycling the Nr containing wastes (Barles, 2007; Barles and Lestel, 2007). Thus, by 1913, the production of ammonium sulphate from sewage was already down to 600 tonnes of N (Barles, 2007), substantially less than that estimated above for the turn of the century. Between the beginning of the twentieth century and the 1970s, Paris grew as a source of N emissions to the water and air. The human population continued to increase in Paris, while the city sprawled over an even bigger area. However, the animal population decreased substantially. As horses were moved out of the city, total per capita Nr inputs decreased, yet consumption patterns and higher living standards still caused an increase in human Nr inputs, since a larger fraction of food was not eaten and contributed to waste. Table 12.3 shows these changes. New sources of industrial Nr fertilisers were discovered, such as the Haber–Bosch process (Smil, 1999; Erisman etâ•¯al., 2008). During the first half of the twentieth century, the cheaper costs associated with this process removed the immediate need to use sewage Nr as a fertiliser. In time, Nr fertiliser manufacture from the excreta of Paris became completely uneconomic, so that by the 1920s, the industry was effectively at an end;

257

Nitrogen flows and fate in urban landscapes Table 12.3 Main characteristics of dietary nitrogen balance, Paris, 1817, 1869, 1913, 1931 (Barles, 2007)

1817

1869

1913

1931

Human population

716 000

1 840 000

2 893 000

2 885 000

Horses population

16 500

50 000

55 000

10 000

Food inflows (Gg N)

6.0

17.6

23.5

19.7

Urban fertiliser produced Street sludge (Gg N)

0.5

1.3

2.1

0.7

Horse manure (Gg N)

0.6

1.8

1.8

0.4

Human manure (Gg N)

0.1

1.1

1.2

0.1

Wastewater to sewage farms (Gg N)

0

±0

4.0

4.0b

Total outflows to agriculture (Gg N)

1.2

4.2

9.1

5.2

% of food inflows

20

24

40

26

Direct discharge to Seine (Gg N)

?

?

3.1

7.0

% of food inflows

?

?

13

36

a

b

substantial fraction of the sewage was processed industrially for ammonium sulphate production A (Vincent, 1901). bâ•‡ This concerns only the dietary nitrogen. The total amount of nitrogen in wastewaters is more important. aâ•‡

Table 12.4 Sewage treatment capacity increase for the growing population (Billen etâ•¯al., 2009)

Year

Rate of connection to sewage collection system (%)

Installed domestic wastewater treatment capacity (inhabitant equivalent)

1954

9

300 000

1962

18

500 000

1971

23

2 000 000

1976

31

4 900 000

1980

38

7 200 000

1985

50

9 600 000

1991

60

9 700 000

1996

70

11 300 000

urban Nr, which had been a previously valued product, became a waste product for disposal. Wastewater treatment plants were thus constructed, focusing on removing nitrogen through denitrification, which gradually increased their treatment capacities (see Table 12.4). Surface water contamination continued to increase during this period, since the environmental issue alone (the agricultural pressure disappeared) was not important enough at that time to provoke water treatment enhancement. As heavy industry was moved out of Paris, emissions from industry (industrial N and other pollutants) became distinct from those generated by other urban processes. This, however, did not mean that industrial N emissions decreased, but rather that urban N emissions to the air became impacted by energy transitions. On the one hand, they decreased because of the increased share of electricity in the energy system (which was earlier dominated by gas produced from coal or coal itself). However, urban N

258

emissions decreased substantially only with the construction of nuclear power plants or even since coal power plants were taken outside the city. At the same time, emissions to the air increased as fuel-powered transport replaced horse transport and overall traffic increased due to urban sprawl. From the 1950s the use of Nr fertilisers increased substantially. The mean application rate was 13 kg N per ha/yr on agricultural land in France, which thus increased to 114 kg N per ha/yr in 1996. While ammonium in the River Seine was practically undetectable before the middle of the 1960s, the maximum contamination was reached during the 1970s, owing both to increased urban population (mostly in the downstream part of the sub-basins, as a result of the expansion of the Paris agglomeration) and to increased rates of sewage collection, often released into surface water without treatment. However, later on progress in wastewater treatment led to a considerable decrease in ammonium contamination during the 1990s (Barles, 2007).

12.3.2â•‡ Nitrogen budget for the Paris Metropolitan Area The urban nitrogen budget (balance) can be considered as a subset of national and regional N budgets. It incorporates imports of Nr-containing products into the city, their conversion within the city boundaries and exports outside of the urban sphere. The urban sphere incorporates the three dimensional space surrounding the urban habitat and spans all environmental media, water, air and soil. We make the first step by creating a detailed Nr mass balance for Paris and its urbanised surroundings in order to estimate the magnitude of major fluxes across the urban landscape and to see how N cycling varies among urban system components. This will help to determine which budget terms are most open to management efforts to reduce N pollution to recipient systems. The budget is shown in Figure 12.6.


Figure 12.6 Nitrogen flows quantified for the Paris Metropolitan Area for the year 2006 (PAM, numbers in Gg N per year). The quantified fluxes displayed reflect major N flows through the PAM originating from food import and fossil fuel use, as well as N2 out-flux from wastewater treatment. Notes: The calculation of fossil fuel input has been based on total gasoline consumption for France, weighted by the urban population of the Paris Metropolitan area. Rough assumptions had to be made regarding gasoline N content, fuel N conversion rate in combustion and that the average per capita consumption for France was a suitable indicator for Paris. In reality, it is likely that the urban population will rely more on public transport and this figure may be an overestimation, however, it indicates that the bulk of emissions from fossil fuel combustion stem from N2 conversion of air N content in the combustion process, whereas the fuel N contribution is comparatively small. It has to be stated as well, that this figure only covers gasoline consumption, which is assumed to be the largest contributor of fossil fuel Nr import into the city. â•… The dry and wet deposition of Nr on the soils and surface waters of the PAM were not quantified due to a lack of modelling results for this specific spatial domain. Considering a total emission to air of approx. 50 Gg N per year, dry and wet deposition may lead to a substantial contribution to nutrient input into urban soils and waters. â•… If solid waste has been thermally treated, the remaining Nr content of the waste deposited in landfills should be negligible. Because of the relatively low temperatures at which municipal waste furnaces operate, 70%–80% of NOx formed in municipal waste furnaces is associated with nitrogen in the waste and is emitted to air, alongside small amounts of NO2 and N2O (in particular from emission control equipment for flue-gas treatment of incineration plants). â•… Emissions of nitrogen containing species (mainly NH3) from untreated municipal waste in landfills due to rotting and chemical conversion processes are difficult to quantify, as they depend on the content of the waste stored, moisture and other parameters. In most cases, only CH4 emissions from landfills are monitored and used for power generation. Overall, based on Sutton etâ•¯al. (2000), volatilisation emissions of NH3 in the PAM may be of the order of 1–2 Gg Nr per year.

The estimates in this figure are derived from the mass balances of urban food consumption and nutrient flows (Faerge etâ•¯al., 2001; Magid etâ•¯al., 2006), data on N2O emissions from wastewater treatment plants (Thomsen and Lyck, 2005) with the focus on the urban sub-systems. Obviously, the urban consumption of resources produced elsewhere (notably food) gives rise to substantial leakages of Nr, and should be included in an ecological footprint analysis, as in Rees (1997) and Wackernagel and Rees (1997). Based on the indicative calculation illustrated in Figure 12.6, the Paris Metropolitan Area is a source of Nr, emitting in total the amount of 50 Gg per year to the atmosphere, the major part being attributed to the emissions from transport and energy. Although much smaller, emissions of Nr to air from the incineration of solid waste are also substantial, contributing 2 Gg per year. The amount emitted to the aquatic environment, at about 12Â€Gg N/yr, greatly depends on the type of wastewater treatment adopted. Disposal of solid wastes and incineration residues in landfills or of sewage sludge on agricultural and

non-agricultural soils (potentially leaking to the ground water over time), together amount to 17 Gg N/yr. Regarding the transformations between N2 and Nr, the largest of these occurs outside this budget, in the production of fertiliser Nr to provide food. Overall, the food Nr import of 63 Gg per year is of a similar order of magnitude to the inadvertent fixation of N2 to Nr through combustion processes. However, the fate of the Nr produced by the two processes is very different. In the case of Nr in food, most is transferred to waste waters, with over half of this being denitrified to N2 in wastewater treatment, i.e. 32 Gg, with only around 0.2 Gg per year being emitted as N2O (Tallec et al., 2007). Although NH3 volatilisation from wastewater treatment is unquantified, based on UK estimates (Sutton etâ•¯al., 2000), it is expected to be similar at around 0.2Â€Gg per year. As a result of the major loss by denitrification, this leaves only around 12 Gg per year which is returned to agricultural and non-agricultural land, with 12 Gg per year of Nr lost to the environment in receiving waters. The small fraction of the food import Nr being reused on agricultural and

259


non-agricultural land of 20%, compares with a value of nearly 40% achieved in 1913 (Table 12.3). In the case of Nr from combustion processes, effectively all of this is exported from Paris as Nr. Thus four times as much Nr is released to the environment of Paris from combustion proÂ� cesses than from the Nr originating in food imported to Paris. If this highlights the problem of Nr emissions from combustion sources for this city, it should not be forgotten that the Nr denitrified in wastewater treatment represents the loss of a valuable resource. Without commenting here on the economic viability of recycling wastewater Nr, it may simply be noted that, at an indicative value of €1 per kg fertiliser Nr, the denitrification of wastewater Nr in Paris represents an annual resource loss of €32 million per year.

12.3.3â•‡ Human food sub-system An important criterion for the assessment of a city’s functioning in relation to the N cycle could be set as the relation between urban population growth and rural productivity (nitrogencontaining food products). In other words, is there enough land surrounding a city to feed its population and how does its rate of growth compare to the growth rate of the urban population? The concept of the so-called food print has been suggested by Billen etâ•¯al. (2009) and discussed in the previous sections. However, this is not a simple relation, since the amount of land could remain the same while its productivity increases due to the introduction of new technologies based on artificial Nr fertilisation, etc. The food could also be imported. This is precisely what caused the food print of Paris to shrink during the second half of the twentieth century, while its population dramatically increased, eventually reaching 10 million inhabitants. Inputs to the human food system include imported and internally produced food, while outputs consist of discharges of unutilised food to landfills and excretion to wastewaters. The size of the area that feeds Paris remained more or less the same and corresponds to the Seine watershed being around 60 000 km2. However, if we look at the ecology of a city, we also need to be sure that water and air pollution do not impair the hinterland surrounding it, thus making the lands less productive. The average intake of protein by the population of the EU is 105 g per person per day, while for France this value is 118 g per person per day, one of the highest in the world (FAO Nutritional Studies). Based on this figure, for France the total direct human consumption of nitrogen is 420 Gg Nr per year, corresponding to about 14% of total EU human consumption of protein. Consequently, for the Paris Metropolitan Area it should be 79.9 Gg N per year. Most food protein is contained in animal products, the common protein sources for Europeans being meat from pigs, cattle and poultry and eggs (FAO Yearbook, 2005/2006). Normal well-fed adults exhibit a nitrogen balance where Nr ingested equals Nr excreted (Voegt and Voegt, 2003), therefore 100% of the outputs from the human food sub-system enter directly into the wastewater subsystem. In other words, human direct consumption should be about the same as human direct excrement to the wastewater subsystem. However, for the mass balance calculations presented

260

earlier in the text, we modified the figure to 63.3 Gg N per year based on Magid etâ•¯al. (2006), who estimated the annual production of nitrogen per person to be 6 kg, of which 0.37 kg is from detergents and the like used in washing water. The rest is related to food intake, either directly or to the food waste going to the bin or kitchen sink.

12.3.4â•‡ Sewage system:Â€N in liquid and solid fractions of urban waste The main point sources of nitrogen in discharges are human or industrial sewage treatment plants, larger agricultural units (husbandry) and, of course, untreated wastewater from urban areas (Figure 12.7). In Western Europe, population increased from 466 to 519 million inhabitants (+11%) between 1970 and 2000, which is a slower growth than in North America, for example. Human Nr in sewage increased from 4.8 to 5.7 kg per person per year in the period 1970–2000. Similarly, human P emissions increased from 0.8 to 1.0 kg per person per year, while detergent P emissions decreased from 0.3 to 0.2 kg per person per year, with a peak in the 1980s (Van Drecht etâ•¯al., 2009) . In 1970 about 64% of the population of Europe was connected to sewage systems, increasing to 79% in 2000; in the same period the amount of N removed in wastewater treatment (as denitrification to N2) increased from 10% to 50% of the Nr and P removal from 11% to 59% (Van Drecht etâ•¯al., 2009). The sum of all these changes was a slight decrease of the Nr discharge to surface water (after treatment) from 1326 to 1192 Gg/yr (Figure 12.8); P discharge to surface water decreased from 333 to 216 Gg/yr. Hence, despite the enormous investments in the construction of sewage systems and wastewater treatment facilities, the Nr flow from households to surface waters is still considerable. With higher P removal rates the flows of P are reduced more effectively. For Europe, nitrogen discharge has become an important issue, when in the mid 1970s massive recurrent blooms of gelatinous Phaeocystis flagellata (producing toxins that kill marine animals) colonies and cells were observed each spring in the coastal areas of the North Sea. Practically all investigators came to one conclusion:Â€the change of dominance from diatoms to Phaeocystis was a consequence of the nitrate enrichment of coastal waters in response to the cumulative nutrient discharges by the major North-West European Rivers, and especially of the decrease of Redfield ratio due to the abundance of phosphorus in the input flow of biogenic pollutants. A detailed description of the precise spatio-temporal interactions between human activities and the functioning of river basin ecosystems and estuaries is presented in Billen etâ•¯al., 2011 (Chapter 13, this volume). The urban sewage treatment process is primarily designed to reduce the level of pollution of watercourses by organic matter, which results in the oxidation of nutrients to inorganic forms. Technologies of nutrient removal are relatively new:Â€ ‘tertiary treatment’ normally follows the normal twostage treatment of sedimentation, followed in turn by biological treatment. These technologies allow most of the Nr and phosphorus to be removed. But the costs of removal grow very fast according to the degree of cleaning required.

Anastasia Svirejeva-Hopkins and Stefan Reis Figure 12.7 N effluent from sewage systems after wastewater treatment for 2000 for Europe (based on data from Van Drecht etâ•¯al., 2009).

Figure 12.8 Trends in human N emission by type of sanitation, sewage and N removal for 1970, 1990 and 2000 for Europe. Improved sanitation indicates N from households with connection to public sewerage, but also to other systems such as septic systems, simple pit latrines, pour-flush, and ventilated improved pit latrines. It is assumed that the N from households with no improved sanitation or with no sewage connection does not end in surface water. Therefore, the N from households and small industries that enters the surface water is the N from sewage systems with no treatment or after treatment (the red parts of the bars). Figure based on data from Van Drecht etâ•¯al. (2009).

Traditional wastewater treatment can remove 85%–95% Nr and 90%–95% phosphorus. The operational cost is around 1€ per kg nitrogen removed (i.e. denitrified to N2) and 1.5€ per kg phosphorus removed. Construction cost is at a similar level. In principle, up to 100% of N and 90% of P can be removed from wastewaters, but these are very expensive technologies (Henze etâ•¯al., 2008). The nitrogen in wastewater is, in general, in the form of ammonia, also some nitrite is present in low concentrations, but both are toxic for fish. Most technologies use a two-step process:Â€nitrification (ammonia → nitrate) and denitrification (nitrate → gaseous nitrogen). Usually, combined treatment plants are used, where nitrification and denitrification occur in different zones controlled by oxidation. For instance, the ‘Carousel’ system allows up to 50%–70% removal of the total nitrogen. Certainly, physical and chemical processes exist for ammonia removal, although they are more expensive than the microbiological ones. Chemical stripping by the addition of lime (so that the pH of the sludge rises above 11), and further passage of it through an aeration tower can raise the degree of removal up to 90% (Hammer and McKichan, 1981). In some areas septic systems (installed in allotments and suburban

261


residences) are suspected of causing an increased level of ground-water contamination. Intermittent loading or recycling of nitrified effluent are suggested as methods of improving denitrification in septic systems. This becomes especially relevant in view of the growing suburbanisation of Europe. These days, European cities usually direct all their wastewaters to treatment plants, however in the 1980s and 1990s the situation was very different, especially for the Eastern European countries (see Figure 12.3):Â€in Bucharest, for example, only 0.01% of wastewater was treated, in Belgrade 12% and in Warsaw 36%. Since then, the Urban Wastewater Treatment (UWT) Directive was issued in 1991, which regulates wastewater treatment in Europe. It requires the collection and treatment of wastewater in all agglomerations of >2000 population equivalents, p.e., (where 1 p.e. is the organic biodegradable load having a five-day biochemical oxygen demand (BOD5) of 60 g of oxygen per day); secondary treatment of all discharges from agglomerations of >2000 p.e., and more advanced treatment for agglomerations >10 000 p.e. in designated sensitive areas and their catchments. The UWT Directive requires the pre-authorisation of all discharges of urban wastewater, of discharges from the food-processing industry and of industrial discharges into urban wastewater collection systems; it moreover monitors the performance of treatment plants and receiving waters and controls the sewage sludge disposal and reuse, and treated wastewater reuse whenever it is appropriate. Sewage water reuse is one of the adaptation strategies listed in the IPPC 4th Assessment Report for the water sector (IPCC, 2007). Segregated wastewater collection enables efficiency in reuse of water and the nutrients found in wastewater. Consequently, water resources are conserved and nutrients are returned back to the soil. In this system, greywater and blackwater are collected separately from urban households. Rainwater is also harvested before it reaches wastewater collection systems. The UNDP Environment and Energy Program defines Ecological Sanitation (ECOSAN) as ‘an approach to human excreta disposal that aims at recycling nutrients back into the environment and productive systems’ (see further discussion by Oenema etâ•¯al., 2011, Chapter 23, this volume). Ideally, a community using the ECOSAN approach disposes no raw or treated wastewater into the water bodies, limiting the disposal of xenobiotics, including endocrine disrupting chemicals (EDCs), pharmaceuticals and personal care products (PPCPs) along the way. It should be noted that this approach cannot be applied to urban areas with established centralised wastewater collection and treatment systems. However, this is easily adoptable in newly developing urban settlements. Strict legislation is lacking, however, the World Health Organization (WHO, 2006) has issued ‘Guidelines for the Safe use of Wastewater, Excreta and Greywater’. Greywater is rich in terms of phosphorus but the nitrogen content is limited (Atasoy, 2007). Urine contains approximately 80% of the Nr and 55% of the phosphorus found in domestic wastewater (Leeming and Stenstrom, 2002). As was already mentioned, in conventional treatment systems, nitrogen and phosphorus are removed in tertiary treatment. Sludge, containing some of the remaining nutrients, is then disposed of most commonly either by landfill or incineration. With segregated

262

Table 12.5 Nitrogen emitted from wastewater treatment for all European cities of over 1 million, contrasting a scenario of current water treatment (80% treatment, with denitrification based approaches) with a system of latrine water recycling; based on per capita recalculations (Magid etâ•¯al., 2006).

Gg N per year New system of latrine water recycling

Receiving media

N Gg per year 80% treatment

Water (Nr)

157

26.2

Sludge (Nr)

157

52.4

Air (denitrified to N2)

418

0

0

629

732

708

Recycled (as fertiliser Nr) Total

water collection, on the other hand, water is reused and nutrients are returned back to the soil as fertiliser. One interesting historical example of the latrine sewage recycling, implemented in the middle of the nineteenth century in Copenhagen, is described in the Box 12.1. It also describes the situation in London at that time. Box 12.2 describes the beginning stage of centralised urban water management in Russia. It is indeed possible to reduce the amount of Nr entering the surface waters substantially and to entirely eliminate Nr emissions to the atmosphere from wastewater treatment plants (Magid etâ•¯al., 2006). Figure 12.14 shows the scheme suggested for the recycling of sewage waters. It is relevant to estimate the total amounts of Nr in different fluxes for all major European cities (> 1 million population) using the traditional cleaning method and the suggested utilisation. Table 12.5 shows the calculated values, which clearly highlight the advantages of the proposed utilisation scheme. Overall the production of fertiliser in recycling Nr for major European cities would have the theoretical potential to produce over 600 Gg Nr per year, equivalent to around 600 million € per year, at the same time as reducing polluting losses to the environment. Box 12.1â•‡ Urban waste management in the nineteenth century: London and Copenhagen

London In 1840 Thomas Cubitt wrote ‘… Fifty years ago nearly all LondonÂ€had every house cleaned into a large cesspool …. Now sewers having been very much improved, scarcely any person thinks of making a cesspool, but it is carried off at once into the river. The Thames is now made a great cesspool instead of each person having one of his own …’ . By then London had reached over 2Â€million inhabitants, and was the largest city in the world. Cholera outbreaks had begun some years earlier, but the cause for this was not understood. The main reason for public debate was caused by the stink of the Thames. This fired a debate on how to manage waste. At this time Justus von Liebig wrote a letter to the Prime Minister of the UK Sir Robert Peel. ..The cause of the exhaustion of the soil is sought in the customs and habits of the towns people, i.e., in the construction of water closets, which do not admit of a collection and preservation


of the liquid and solid excrement. They do not return in Britain to the fields, but are carried by the rivers into the sea. The equilibrium in the fertility of the soil is destroyed by this incessant removal of phosphates and can only be restored by an equivalent supply. …If it was possible to bring back to the fields of Scotland and England all those phosphates which have been carried to the sea in the last 50 years, the crops would increase to double the quantity of former years…’. In his book on Agricultural Chemistry (1862) von Liebig later stated that ‘The introduction of water-closets into most parts of England results in the loss annually of the materials capable of producing food for three and a half million people; the greater part of the enormous quantity of manure imported into England being regularly conveyed to the sea by the rivers …like a vampire it hangs upon the breast of Europe, and even the world; sucking its life-blood. Although von Liebig focused his argument on phosphorus, it is clear that they applied just as much to Nr. When London’s authorities decided to construct a sewage disposal rather than a recycling system suggested by Liebig, he increased his effort to find ways to replace the fertility removed by cities from farmland by artificial means. He focused in particular on developing artificial fertilisers to keep the agricultural land productive in order to feed the cities.

Copenhagen At around the same time as these developments Copenhagen was bankrupt. Similar problems with waste arose, although on a smaller scale and cholera outbreaks eventuallyÂ€ visited

Figure 12.9

Figure 12.10

Copenhagen in 1853. The future waste management system was hotly debated, but in the end the state prohibited sewers in 1858 due to insurmountable costs, and the city negotiated contracts with farmers for collection of latrine waste. Eventually this system was developed into an elaborate service industry that ensured timely collection and daily transport of latrine contents to the eastern and western outskirts of the city. Figure 12.9 shows show night soil workers empty stainless steel drums into large wooden barrels, and subsequently wash and steam rinse the drums. Furthermore they show farmers collecting night soil from the latrine wagons that were commonly known as‘The Royal Train’and‘The Chocolate Express’(see Figure 12.10). Cholera subsided during these years, and the hinterland farming community gained access to fertiliser as well as a growing market for perishable foods, resulting in better welfare. This system persisted until after the Second World War, but gradually gave way to sewers and water closets. Peri-urban farmers were strong stakeholders, protesting vociferously against the decline of the system and the resulting negative effects on their farmland productivity.

Box 12.2â•‡ The history of centralized water supply and canalisation in Moscow There were no centralised systems of water supply for cities in Russia before the end of the nineteenth century. The water was taken from the streams and wells. Cities were supplied with water in barrels (fig. 12.11).

Figure 12.11 Water, brought to the city in barrel (Miksashevsky and Korolkova, 2000).

The domestic waste discharges were dumped in the nearby water body or just on the streets. Therefore the water bodies were polluted and were the sources of infectious diseases. The canalisation systems were constructed earlier in Europe:Â€first in London, then in Paris and Berlin. The positive results were immediate, like in Berlin in the course of one year after the centralised sewage system was built, the water quality was greatly improved and the number of people who contacted cholera dropped by half, and soon the disease was entirely eliminated. With the growth of large cities and rapid increase of their inhabitants, the need for the centralised water supply developed. The late development of centralised water supply in Moscow, contrasted with the fact that smaller-scale water

263


Box 12.2 (cont.) supply systems existed earlier in Eastern than in Western Europe. For example, the archeological findings suggest their presence on the territory of Caucusus, (Russia) Great Novgorod and Ukraine (see Figure 12.12).

Figure 12.12 An example of small scale water supply: Kremlyn palace, Moscow (17th century).

The centralised large water supply system started to operate in Moscow in 1892, when the two main pumping towers were constructed (see Figure 12.13).

Figure 12.13 “Kretsletz” main water pumping towers.

In 1874, engineer M. A. Popov brought to the attention of the Russian government that sewage channels needed to be built in Moscow and suggested that sewage waters be removed from the city and purified using special irrigation fields (with later usage as fertiliser). Popov used his own funds for collection of topographic and soil data and made sewage application capacity calculations, based on fundamental population growth projections. He developed the entire project of combined sewage system and estimated the construction costs as well as costs of using sewage residue. Unfortunately, the implementation of an actual plan was delayed, due to disagreement with the external evaluator from Berlin, Gobrecht, who supported the plan at first, but then found some flaws and offered to take it over. InÂ€1890, a segregated sewage system project, developed by the engineer Kastilsky, was implemented. By 1898, 262 km of pipelines had been laid and the main pumping station was built. By August 1899, the system began to function to distribute sewage waters to agricultural irrigation fields.

12.3.5â•‡ Urban N fluxes due to the combustion of fossil fuels in stationary and mobile sources 264

The main contribution to urban air quality problems is made by the combustion of fossil fuels. Emissions come from both

stationary (residential and commercial combustion for heating and process water purposes, combined heat and power plants) and mobile sources (road and off-road transport and machinery). The general mechanisms leading to the formation of the most relevant pollutants (NO2, NOx, NH3, ozone and secondary aerosols/particular matter are illustrated by Hertel etâ•¯al. (2011, Chapter 9, this volume), which gives a detailed account of the processes leading from emissions to ambient concentrations. Here, only specific aspects of urban air quality will be discussed. Kousoulidou etâ•¯al. (2008) analysed the projections of road transport emissions until 2020 and state that while significant reductions are to be expected for relative emissions per vehicle and kilometre driven, NO2 concentrations in urban areas are not expected to fall as dramatically. This is mainly due to the change in the NO2/NOx emission ratio of new technologies, aiming for instance to reduce PM emissions from vehicle exhausts (see also Keuken etâ•¯al., 2010). This trend will most likely have implications for the attainment of ambient air quality standards for NO2 concentrations in all large European cities. Beevers and Carslaw’s (2005) earlier work concluded this for central London. In addition to the technology changes in vehicles and control equipment, an increase in annual average mileage driven in urban areas may arise from a continuing urbanisation towards the development of urban sprawls, as discussed by De Ridder etâ•¯al. (2008). Stationary sources of emissions in urban areas are residential and commercial combustion plants on the one hand (household heating and process water, open fireplaces, etc.) andÂ€– with the deregulation of the energy markets and increasing fuel pricesÂ€– decentralised small power plant units (in most cases combined heat and power, CHP, plants based on natural gas or renewable fuels, e.g. biomass) on the other hand. While solid fossil fuels have been banned for use in private stoves in some countries and regions/cities, they contribute a significant share of household heating especially in Scandinavian countries and Central and Eastern Europe. The major contributing sources obviously vary from city to city, however a few patterns can be identified by looking at the information available from individual large urban areas in Europe, such as Greater London (Table 12.6) or Berlin (TableÂ€12.7). Table 12.7 illustrates the situation in Berlin, showing a bulk of NOx emissions stemming from road transport sources. In contrast to Greater London, however, industrial emissions contribute a significantly larger share in Berlin with about 32% of facilities requiring a permit to operate, and thus being subject to regulation. Domestic fuel combustion makes up only about 11% of NOx emissions in Berlin. The above tables illustrate the relative contribution of major activities to urban air quality problems, namely high ambient concentrations of NO2, ozone and particular matter. Figure 12.15 displays the percentage of the urban population in Europe experiencing pollutant concentrations above the respective target/limit values. A clear downward trend can be observed for SO2 together with a less pronounced one for NO2. Concentrations of NO2 in general and population exposure to very high (>40μg/m3) concentrations have declined in the 10 years between 1997 and 2006 (see FigureÂ€12.16).

Anastasia Svirejeva-Hopkins and Stefan Reis Table 12.6 Share of emissions within Greater London and on a national scale for the year 1999. This table illustrates the relevance of road transport sources for urban air quality and indicates a significantly larger proportion of nitrogen (58.2%) being contributed by urban road transport. Industrial sources, in contrast, play a minor role (8.9%) and only about 33% of urban NO x emissions can be attributed to other sources

Nitrogen oxides (NOx) Fine particles (PM10) Sulphur dioxide (SO2) Carbon monoxide (CO)

Total emissions in Gg per year

percentage of emissions in Greater London

percentage of national emissions

All sources

Road transport

Road transport

Industry

Industry

68.13

58.2

8.9

44

37

2.75

67.9

22.3

20

44

3.55

38.3

39.1

1

89

173.38

93.7

1.4

69

16

Source:Â€Mayor of London’s Air Quality Strategy 2010).

8g to air

Present day distribution of Nitrogen 14 g mostly form urine and faeces delivered via waste water to sewage treatment

3.0 g to the sea

1.7 g in solid waste (organic waste from household and garden)

3g Sludge

N distribution after increased utilization or urine and faeces 12 g utilised as fertiliser for agricultural land 0 g to air

1.5 g delivered via waste water to sewage treatment

0.5 g to the sea

2.2 g in solid waste (organic waste from household and garden) to be recycled upon treatment (composting or bio-gas production)

1g Sludge

Figure 12.14 Current versus increased utilisation method of N distribution (g per capita) (Magid etâ•¯al., 2006).

At the same time as Figures 12.15 and 12.16 show the decline in exposure to high NO2 concentrations in urban areas, a more frequent occurrence of exposures to high ambient levels of ground level ozone above 120 μh/m3 (8 h mean) are observed (Figure 12.17). It is difficult to assess to what extent this increase of exposure of the urban population to high ambient levels of ground level ozone is caused by urban emissions

(resp. the reduction of urban NOx emissions and the resulting decrease of the titration effect in NOx-rich environments) and to what extent by the slowly increasing concentrations of global background ozone levels. In particular, exposure to high ambient concentrations of ozone and PM lead to adverse effects on human health, which are discussed in detail in Moldanová etâ•¯al., 2011 (Chapter 18, this volume).

265

Nitrogen flows and fate in urban landscapes Table 12.7 Emissions of NOx in Berlin according to emitting groups (Gg per year)

Data in Gg per year Nitrogen oxide

1989

1994

2000

2002

Trend 2005

Germany

2862

2226

1815

1640

1447

Berlin

70.0 (2.4%)

42.4

26.1

22.1

19.8

17.5

Emittent approved facilities

41.8

16.2

6.0

5.8

8.3 (31.9%)

6.5

Trend 2010

Domestic fuel

2.7

3.1

2.9 (10.9%)

2.9

2.7

2.6

Small trade

1.2

0.7

0.2 (0.7%)

0.2

0.2

0.1

21.4

19.0

12.4 (47.5%)

10.5

8.9

7.0

Traffic (other)

1.4

1.3

1.1 (4.3%)

1.1

1.1

1.1

Other sources

1.5

2.1

1.2 (4.6%)

1.0

1.0

0.9

Traffic (motor vehicles only)

(Source:Â€Senate Berlin, 2010).

100

PM10

NO2

O3

Figure 12.15 Percentage of urban population resident in areas where different air pollutant concentrations are higher than selected limit/target values, EEA member countries, 1997–2006. (Source:Â€EEA, 2010.)

SO2

80 60 40 20 0 1997

1998

1999

0-26 µg/m3

100

2000

2001

2002

26-32 µg/m3

2003

2004

32-40 µg/m3

2005

2006

2007

Figure 12.16 Percentage of population resident in urban areas potentially exposed to NO2 concentration levels exceeding the annual limit value in EEA member countries for the period 1997–2006. (Source:Â€EEA, 2010.)

>40 µg/m3

75

50

25

0 1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

% of urban population

12.3.6â•‡ Urban green areas and urban soils Urban open space and green areas In European cities the area occupied by open space, consisting of parks and recreation areas, is on average 30%. Organic matter production does not play a significant role in the operating mechanisms within a city. However, while the green belts play

266

purely recreational and aesthetic roles, they are very important also because they even out the air temperature fluctuations within a city, reduce noise and other pollution, and serve as a habitat for small animals and birds. It is not cost-free, however, to support their functioning, and the labour and fuel spent on irrigation, lawn management, tree planting and care, etc., increases the energy and monetary expenses of a city.


12.4â•‡ Conclusions, uncertainties and the future development of European cities

Figure 12.17 Exposure of urban population in EEA member countries to maximum ozone concentration above the 8 h-daily mean target value of 120Â€µg/m3. (Source:Â€EEA, 2010.)

Trees can nevertheless play an important role in Nr related issues in cities, such as by helping to reduce the effects of particulate matter (PM) and NOx pollution. The amounts of gaseous pollutants and particulates and the interception of aerosols are greater in woodlands than in shorter vegetation (Fowler etâ•¯al., 1998), since they have broader leafs and create turbulent mixing of air. Therefore, urban woodlands and the presence of trees in the urban environment can improve air quality quite significantly. McPherson (1998) estimated that in Chicago trees removed 234 tonnes of PM10 in 1991 and improved average hourly air quality by 0.4% (2.1% in the heavily wooded areas), while Nowak etâ•¯al. (1997) calculated that trees in Philadelphia improved air quality by 72% by removal of PM10.

Urban soils Urban soils, which play an important role as sources, sinks and transformers in the nitrogen cycle, represent an area of great concern as regards food supply and supply of sustainable drinking water, and are important in terms of aesthetics and recreation. Soils are designated as urban soils if they are located in watersheds that provide drinking water, food, waste utilisation, and natural resources to cities. Urban soils include also all soils located within cities in park areas, recreation areas, community gardens, green belts, lawns, septic absorption fields, sediment basins or other open or sealed soils inside the city. The N status and dynamic of urban soils is determined by external factors like N deposition, temperature, rainfall, groundwater N content and groundwater level and internal factors like the geogenic parent material, technogenic substrates, water and air holding capacity, dry bulk density, microbial activity, etc. Technogenic substrates (rubble, construction material, sewage sludge, refuse, and dust) play a key role in the genesis of urban soils. Furthermore they are relatively comparable among cities. Examining the N status of the most important technogenic substrates makes it possible to assess the potential behaviours of urban soils.

The present day situation in our case study, the PAM, reflects to a large extent the metropolisation processes in the area. The urban population is continuing to increase due to urban sprawl, while the density does not increase greatly, which is typical for the Western European region. The further development of the nitrification–denitrification process in wastewater treatment plants has reduced surface water pollution from cities over the last century. However, such wastewater treatment plants operate by denitrifying Nr back to N2 which can be considered as a substantial waste of an expensive resource. For Paris alone, this loss equates to a potential fertiliser value of around € 30 million per year. Urban sprawl is responsible for an increase of car use in the urban context (public transport is still not adapted to low density areas) and there is an increase of trip length. The globalisation of trade leads to an increase in transport-related N emissions. However, both the trend towards moving industries from Europe to other parts of the world and the strict regulation of industrial pollutants lead to a decrease in industrial N emissions. The renaissance of the city is a hot topic in Europe. Generally the term addresses the renaissance of the inner city and is applied to the city centre only. However, while suburbanisation is increasing, more and more European cities are expected to turn into urban regions. A partial renaissance of the inner city would probably take place, as well as partial growth of suburbia. Both will be accompanied by either partial decay of suburbia or partial decay of the inner city. There are already vast and increasing differences among cities. Since the breakdown of communism, the development is very different in different regions, but most European cities are currently exposed to drastic economic and social changes. They face tremendous new challenges such as globalisation, ageing societies, shrinking population figures, shrinking household sizes, increasing social divisions, decreasing resources of public authorities, etc. This change from a relatively stable industrial society towards a post-industrial society will shape the development of cities in Europe over the coming decades. The rejuvenation of an attractive city centre can offer the best service locations, plus it can tie a highly mobile urban middle class to a city in the long run. Creation of an efficient public transportation network connecting the suburbs with the city core is an essential aspect. The growth of European suburbia is a dynamic process, yet in most European cities, urban planning efforts concentrate on the city centre, such as in London and Berlin. If we aim to create a ‘neutral’ Nr state for cities in Europe, we have to increase recycling of food and water, minimise household waste either through reusing sewage waters or technologically improving treatment plants, and reduce Nr emissions to the atmosphere limiting travel by car as much as possible. The importance of these sources is clearly illustrated by the nitrogen budget of Paris (Figure 12.6). In particular, reducing road traffic NOx emissions has the largest single potential to decrease Nr emissions from a city such as Paris, while the use of new re-use based sewage systems, have the potential to avoid

267


the waste of Nr inherent in denitrification-based water treatment systems. Such measures could eventually turn urban areas from being a source of nitrogen to becoming nitrogen neutral. These adaptation measures have to be carefully planned and individually tailored. The following uncertainties regarding nitrogen cycling in an urban system need a better understanding:Â€the mechanisms of dry-deposition processes in urban systems with patchy vegetation; high NOx emissions and the complex patterns of air flow in densely built-up areas. The N dynamics of urban soil are very uncertain, and while soil represents a major sink of N in natural ecosystems, what happens in urban soils due to, for example, impervious surfaces (roads, etc.) has been little studied. Factors that control denitrification in urban landscapes are related to the presence of green areas within city, but those areas differ from natural landscapes. They have lower densities of biomass and altered decomposition rates. Interactions between increasing temperatures, especially in built-up areas, and photochemical smog (NO2 and ground level ozone) are complex and difficult to quantify. Yet, it can be expected that increasing global ambient temperatures may contribute to more frequent occurrences of Nr-related adverse health effects in cities. There is still some uncertainty regarding the fate of Nr in the septic tanks in low-density suburban residential areas. For example, many of these are fairly old and may not function properly, causing leaking to the groundwater. Also storm events often cause septic tanks to overflow, in which case the untreated sewage is transported directly to the surface waters. The most immediate task to bring a city to a neutral state in relation to the nitrogen cycle, is to control transport emissions in cities. There are already some examples of sustainable transport policies in cities, showing that public transport can be attractively organised for a densely built-up city, as well as for a large metropolitan area. In the city of Basel, the traffic policy aims to calm traffic and to promote the use of the bicycle. In the 1980s and the early 1990s the Basel traffic policy implemented a variety of environmentally compatible measures in different areas of transport. This multi-level policy could serve as a model for urban development in other cities. Local measures include implementation of a traffic policy with an effective combination of green transport modes; the successful testing of traffic calming measures in residential districts; the safeguarding of high standards for bicycle use; the diversification of modernised bus, tram, and rail systems; the introduction of a customer-friendly pricing policy in public transport systems; the passing of legal regulations in favour of green modes of transport. Restructuring the labour market (which is the second most important driver of urbanisation after population growth) plays an important role in creating of sustainable transportation network. The city of Copenhagen, which in 1993 introduced a Municipal Plan aiming to design a compact urban structure based on public transport, provides a good example. This required a long-term restructuring of working places according to public transport stations, enhancing and transforming the growth of the city in the harbour area,

268

strengthening the ‘green’ aspects of the city and restoring and maintaining the historical quality of specific city districts and their diversity. An increasing interest in deploying the tram-trains concept is growing across Europe in order to fight congestion whilst also cutting carbon and nitrogen emissions. This approach, using proven technologies, combines heavy rail routes with tramways to allow passengers to access key destinations in city centres from the suburbs without making a change, with the aim of attracting people who previously used cars. Germany pioneered the utilisation of combining heavy rail and street running fixed link systems but in the last few years there has also been an upsurge of interest in France and a trial is under way in the UK (connecting Sheffield, Huddersfield and Rotherham in Yorkshire). Generally speaking, the most effective N management strategies are those that are specifically tailored to individual cities and the ecosystems surrounding them. To develop such schemes will require the construction of detailed, ecosystemlevel Nr balances, to help with a deeper understanding of the interplay of inputs, geographical and climatic factors, nonspecific management practices, and deliberate Nr management practices that control the fate of Nr in urban landscapes. Nitrogen budgets can be used as a tool to provide a context for the evaluation of the extent to which human intervention in the N cycle has changed Nr distribution from local to global scales. To gain first insight into the spatial heterogeneity of Nr creation and distribution in urban landscapes, we examined an urban N budget. This is important as it illustrates the differences in Nr creation and distribution as a function of the level of urban development and geographic location.

Acknowledgements The authors gratefully acknowledge support from the Nitrogen in Europe (NinE) programme of the European Science Foundation, from the NitroEurope IP funded by the European Commission and from the COST Action 729.

References Asman, W. A. H., Sutton, M. A. and Schjørring, J. K. (1998). Ammonia:Â€emission, atmospheric transport and deposition. NewÂ€Phytologist, 139, 27–48. Atasoy, E., Murat, S., Baban A. and Tiris, M. (2007). Membrane bioreactor treatment of segregated household wastewater for reuse. Clean, 35, 465–472. Baker, L., Hoper, D., Xu, Y., Edmonds, J. and Lauver, L. (2001). Nitrogen balance for the Central Arizona–Phoenix (CAP) ecosystem. Ecosystems, 4, 582–602. Barles, S. (2007). Feeding the city:Â€food consumption and flow of nitrogen, Paris, 1801–1914. Science of the Total Environment, 375, 48–58. Barles, S. and Lestel, L. (2007). The nitrogen question:Â€urbanization, industrialization, and river quality in Paris 1830–1939. Journal of Urban History, 33, 794–812. Bazilevich, N. I. (1979). Biogeochemisty of the Earth and functional models of exchange processes in natural ecosystems. In:Â€Modern Concepts and Problem of Biogeochemistry

Anastasia Svirejeva-Hopkins and Stefan Reis Proceedings of the Biogeochemical Laboratory, vol. 17, Nauka, Moscow, pp. 56–73. Bazilevich, N. I. and Tytlanova, A. A. (2007). Biological Cycling on Five Continents:Â€Nitrogen and Elements in Natural Ecosystems. Novosibinsk, Nauka (in Russian). Beirle, S., Platt, U. and Wagner, T. (2004). Global Mean Tropospheric Nitrogen Dioxide (NO2) Vertical Column Density (VCD) for Europe: Map. Institute for Environmental Physics, University of Heidelberg. Beevers, S. D. and Carslaw, D. C. (2005). The impact of congestion charging on vehicle emissions in London. Atmospheric Environment, 39, 1–5. Bettencourt, L. M. A., Lobo, J., Helbing, Y., Kuchnert, C. and West,Â€G.Â€B. (2007). “Growth, innovations scale and the pace of life in cities”, Vol. 104, No. 17 pp. 7301–7306. Billen, G., Barles, S., Garnier, J. etâ•¯al. (2009). The food-print of Paris:Â€long-term reconstruction of the nitrogen flows imported into the city from its rural hinterland. Regional Environmental Change, 9, 1436–1498. Billen, G., Silvestre, M., Grizzetti, B. etâ•¯al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al. Cambridge University Press. Bobbink, R. (1998). Impacts of tropospheric ozone and airborne nitrogenous pollutants on natural and seminatural ecosystems:Â€a commentary. New Phytologist, 139, 161–168. Boussingault, J. B. (1858). Note relative à l’emploi industriel et domestique des eaux de puits de Paris. Gazette hebdomadaire de médecine et de chirurgie, 5, 507–508. Cohen, J. E. (1995). Population growth and Earth’s human carrying capacity. Science, 269, 341–346. Demographia (2009). World Urban Areas & Population Projections:Â€5th Comprehensive Edition (Revised April 2009). http://www.demographia.com (site accessed 30/07/2010). Dumas, J. B. (1844). Essai de statique chimique des êtres organisés, 3rd edn. Paris. De Ridder, K., Lefebre, F., Adriaensen, S. etâ•¯al. (2008). Simulating the impact of urban sprawl on air quality and population exposure in the German Ruhr area. II. Development and evaluation of an urban growth scenario. Atmospheric Environment, 42, 7070–7077. EEA (2010). http://www.eea.europa.eu/data-and-maps/figures (site accessed 30/07/2010). Erisman, J.-W., Sutton, M. A., Galloway, J. and Klimont, Z. (2008). How a century of ammonia synthesis changed the world. Nature Geoscience, 1, 636–639. Faerge, J., Magid, J. and de Vries, F. W. T. P. (2001). Urban nutrient balance for Bangkok. Ecological Modelling, 139, 63–74. FAO Nutritional Studies no.24. http://www.fao.org/ (site accessed 30/07/2010). FAO Yearbook (2005/2006). http://www.fao.org/ (site accessed 30/07/2010). Ferreira, J., Martins, H., Miranda, A. I. and Borrego, C. (2008). Population Exposure to Atmospheric Pollutants:Â€The Influence of Urban Structure. Wessex Institute of Technology Publications. Fowler, D., Flechard, C., Skiba, U., Coyle, M. and Cape, N. (1998). The atmospheric budget of oxidized nitrogen and its role in ozone formation and deposition. New Phytologist, 139, 11–23. Hammer, M. J., McKichan, K. A. (1981). Hydrology and Quality of Water Resources. John Wiley, New York.

Henze, M., van Loosdrecht, M. C. M., Ekama, G. A. and Brdjanovic,Â€D. (eds.) (2008). Biological Wastewater Treatment: Principles, Modelling and Design. Hertel, O., Reis, S. and Ambelas Skjøth, C. etâ•¯al. (2011). Nitrogen processes in the atmosphere. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman, etâ•¯al. Cambridge University Press. Hohenberg, P. M. and Hollen, L. (1985). The Making of Urban Europe, 1000–1950. Harvard University Press, Cambridge, MA. IPCC (2007). Climate Change 2007:Â€Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland. Kaye, J. P., McCulley, R. and Burke, I. C. etâ•¯al. (2005). Carbon fluxes, nitrogen cycling and soil micro-organisms in adjacent urban, native and agricultural ecosystems. Global Change Biology, 11, 575–587. Keuken, M. P., Jonkers, S., Wilmink, I. R. and Wesseling, J. (2010). Reduced NO2 and PM10 emissions on urban motorways in The Netherlands by 80 km/h speed management. Science of the Total Environment, 408, 2517–2526. Kousoulidou, J., Ntziachrios, L., Mellios, G. and Samaras, Z. (2008). Road transport emissions projections in European urban environments. Atmospheric Environment, 42, 7465–7475. Leeming, C. R. and Stenstrom, T. A. (2002). Faecal contamination of source-separated human urine based on the content of faecal sterols. Water Research, 36, 1965–1972. Magid, J., Eilersen, A. M., Wrisberg, S. and Henze, M. (2006). Possibilities and barriers for recirculation of nutrients and organic matter from urban to rural areas:Â€a technical theoretical framework applied to the medium-sized town Hillerod, Denmark. Ecological Engineering, 28, 44–54. Mayor of London (2010). http://www.london.gov.uk/mayor/ strategies/air_quality/docs/chap_2.pdf McPherson, E. G. (1998). Structure and sustainability of Sacramento’s urban forest. Journal of Arboriculture, 24, 174–190. Miksashevsky, N. V. and Korolkova, S. V. (2000). Clean Water Treatment Systems and Fitters. BHV–Petersburg, (in Russ.) Moldanová, J., Grennfelt, P., Jonsson, Å . etâ•¯al. (2011). Nitrogen as a threat to European air quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al. Cambridge University Press. Nowak, D. J. etâ•¯al. (1997). Environmental and economic benefitsÂ€of preserving forests within urban areas:Â€air and waterÂ€quality. In:Â€de Brun, Constance T. F. (ed.), The Economic Benefits of Lland Conservation. The Trust for Public Land, pp.Â€28–47. Odum, H. T. (1973). Energy, ecology and economics. Ambio, 2, 220–227. Oenema, O., Salomez, J., Branquinho, C. etâ•¯al. (2011). Integrated approaches to nitrogen management. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al. Cambridge University Press. Pataki, D. E., Alig, R. J., Fung, A. S. etâ•¯al. (2006). Urban ecosystems and the North American carbon cycle. Global Change Biology, 12, 2092–2102. Rees, W. E. (1992). Ecological footprints and appropriated carrying capacity:Â€what urban economics leaves out. Environment and Urbanisation, 4 (2), 121–130. Senate Berlin (2010). http://www.berlin.de/sen/umwelt/luftqualitaet/ de/emissionen/index.shtml Simpson, D., Aas, W., Bartnicki, J. etâ•¯al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:Â€The European Nitrogen

269

Nitrogen flows and fate in urban landscapes Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al. Cambridge University Press. Smil, V. (1999). How many billions to go? Nature, 401, 429. Sutton, M. A., Dragosits, U., Tang, Y. S. and Fowler, D. (2000). Ammonia emissions from non-agricultural sources in the UK. Atmospheric Environment, 34, 855–869. Sutton, M. A., Howard, C. M., Erisman J. W. etâ•¯al. (2011). The need to integrate nitrogen science and policies. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al. Cambridge University Press. Svirejeva-Hopkins, A. and Schellnhuber, H. J. (2008). Urban expansion and its contribution to the regional carbon emissions:Â€using the model based on the population density distribution. Ecological Modelling, 216. Tallec, G., Garnier, G., Billen, G. and Gousailles, M. (2007). Nitrous oxide emissions from denitrifying activated sludge of urban wastewater treatment plants, under anoxia and low oxygenation. Bioresource Technology, 99 2200–2209. Thomsen, M and Lyck, E. (2005). Emission of CH4 and N2O from Wastewater Treatment Plants (6B). No. 208, 1–46. 2005, Research notes from National Environmental Research Institute, Roskilde, Denmark.

270

Tobler, W., Deichmann, V., Gottsegen, J. and Maloy, K. (1995). The Global Demography Project, Technical Report TR-95–6. National Center for Geographic Information Analysis, University of Santa Barbara, CA. UNCHS (2003). Global Urban Indicators Database. http://ww2. unhabitat.org/programmes/guo/urban_indicators.asp 2003 Van Drecht, G., Bouwman, A. F., Harrison, J. and Knoop, J. M. (2009). Global nitrogen and phosphate in urban waste water for the period 1970–2050. Global Biogeochemical Cycles, 23, GB0A03, doi:10.1029/2009GB003458. Vincent, C. (1901). Ammonia and its Compounds: Their Manufacture and Uses, translated by M. J. Salter. Scott, Greenwood & Co., London. Voegt, D. and Voegt, J. G. (2003). Biochemistry, 2nd edn. John Wiley and Sons, New York. Wackernagel, M. and Rees, W. E. (1997). Perceptual and structural barriers to investing in natural capital:Â€Economics from an ecological footprint perspective. Ecological Economics, 20, 3–24. WHO (2006). Guidelines for the Safe Use of Wastewater, Excreta and Greywater, Vol. 1, Policy and Regulatory Aspects. World Health Organization, Geneva. WWF (2002). Living Planet Report. http://www.panda.org/ about_our_earth/.

Chapter

13

Nitrogen flows from European regional watersheds to coastal marine waters Lead author: Gilles Billen Contributing authors: Marie Silvestre, Bruna Grizzetti, Adrian Leip, Josette Garnier, Maren Voss, Robert Howarth, Fayçal Bouraoui, Ahti Lepistö, Pirkko Kortelainen, Penny Johnes, Chris Curtis, Christoph Humborg, Erik Smedberg, Øyvind Kaste, Raja Ganeshram, Arthur Beusen and Christiane Lancelot

Executive summary Nature of the problem • Most regional watersheds in Europe constitute managed human territories importing large amounts of new reactive nitrogen. • As a consequence, groundwater, surface freshwater and coastal seawater are undergoing severe nitrogen contamination and/or eutrophication problems.

Approaches • A comprehensive evaluation of net anthropogenic inputs of reactive nitrogen (NANI) through atmospheric deposition, crop N fixation, fertiliser use and import of food and feed has been carried out for all European watersheds. A database on N, P and Si fluxes delivered at the basin outlets has been assembled. • A number of modelling approaches based on either statistical regression analysis or mechanistic description of the processes involved in nitrogen transfer and transformations have been developed for relating N inputs to watersheds to outputs into coastal marine ecosystems.

Key findings/state of knowledge • Throughout Europe, NANI represents 3700 kgN/km²/yr (range, 0–8400 depending on the watershed), i.e. five times the background rate of natural N2 fixation. • A mean of approximately 78% of NANI does not reach the basin outlet, but instead is stored (in soils, sediments or ground water) or eliminated to the atmosphere as reactive N forms or as N2. • N delivery to the European marine coastal zone totals 810 kgN/km²/yr (range, 200–4000 depending on the watershed), about four times the natural background. In areas of limited availability of silica, these inputs cause harmful algal blooms.

Major uncertainties/challenges • The exact dimension of anthropogenic N inputs to watersheds is still imperfectly known and requires pursuing monitoring programmes and data integration at the international level. • The exact nature of ‘retention’ processes, which potentially represent a major management lever for reducing N contamination of water resources, is still poorly understood. • Coastal marine eutrophication depends to a large degree on local morphological and hydrographic conditions as well as on estuarine processes, which are also imperfectly known.

Recommendations • Better control and management of the nitrogen cascade at the watershed scale is required to reduce N contamination of ground- and surface water, as well as coastal eutrophication. • In spite of the potential of these management measures, there is no choice at the European scale but to reduce the primary inputs of reactive nitrogen to watersheds, through changes in agriculture, human diet and other N flows related to human activity.


271

Nitrogen flows from European regional watersheds

13.1â•‡ Introduction A regional territory comprises a number of natural, seminatural and artificial landscapes, themselves composed of a mosaic of interacting ecosystems. The preceding chapters in this volume have emphasised that the complexity of landscape interactions, often occurring at the interface between ecosystems, prevents a simple additive approach to the functioning of large systems and their nitrogen budget; this is particularly true for regional territories. A regional watershed can be defined as a territory structured by a drainage network. Defining the limits of a territory in accordance with the limits of a watershed simplifies budgetÂ� ing approaches, as it allows a direct estimate of export through the hydrosystem, which is one of the major output terms in the nitrogen budget. However, this simple matter of budgeting convenience is not the sole reason to focus a discussion of the nitrogen cascade on the scale of regional watersheds. Indeed, drainage networks historically played a major role in structuring the European geographical space, often determining city settlement locations, the commercial

exchanges between them and the surrounding rural areas, hence the development of agriculture. Regional watersheds are therefore pertinent spatial units for studying the interactions between humans and the environment. Moreover, the coastal marine systems located at the outlet of regional watersheds are strongly influenced by the fluxes of water and nutrients delivered by the river, so that the whole continuum of ecosystems, including the catchments’ terrestrial systems, the drainage network, the estuarine zone and the coastal sea, should all be viewed and managed as a single integrated system. This is the point of view adopted in the present chapter. The major European watersheds are shown in Figure 13.1, grouped according to the marine coastal zones where they discharge. The full database of European watersheds used for this study is available as on-line supplementary material (see Supplementary materials, Section 13.7). It includes 5872 Â�individual catchments, most of them very small rivers. The major ones, with an area larger than 10 000 km², account for 67% of the total European coastal watershed area.

Figure 13.1 Major regional watersheds in Europe and their receiving coastal marine systems.

272

Gilles Billen

Figure 13.2 Schematic representation of the flows of reactive nitrogen through a regional watershed.

In this chapter, the nitrogen cascade will be examined at the scale of the major watersheds in Europe. The fate of reactive nitrogen brought into these regional territories through atmospheric deposition, synthetic fertiliser application, crop nitrogen fixation and commercial import of food and feed will be discussed at the regional basin scale. Particular emphasis will be placed on the riverine transfer of nitrogen from the terrestrial watershed to ground, surface and marine coastal waters, and on the consequences for the health of the marine systems. The analysis will be guided by the conceptual representation of nitrogen transfers through the different components of a regional watershed illustrated in Figure 13.2. Although this figure does not show all the complex interactions between the different components of the system, it clearly indicates the major sources of reactive nitrogen, the three major types of nitrogen output (to the atmosphere in its gaseous form, to other territories as food and feed, to the coastal seas as river loading) and the major pools with a long residence time (soils and aquifers) where nitrogen can be stored (and possibly remobilised) within the system. We will first establish the reactive nitrogen input–output budget of European watersheds and discuss the difference, often improperly termed ‘retention’, between total inputs and riverine outputs to the coastal zone. We then will take stock of the various modelling approaches used at the regional scale for relating nitrogen inputs to riverine outputs. Using both model results and observed fluxes, we will then examine the longterm trends of nitrogen riverine delivery, and its relations with phosphorus and silica, which is the key to understanding their potential for coastal marine eutrophication. The role of estuaries, acting as the last filter before delivery to the sea, will be briefly examined, prior to discussing the state of eutrophication of European coastal zones.

13.2â•‡ Input–output nitrogen budgets of regional watersheds 13.2.1â•‡ Inputs to watersheds As depicted in Figure 13.2, reactive nitrogen is brought into watersheds from atmospheric deposition, crop N2 fixation and synthetic fertiliser use, as well as by net commercial import of food and feed. All these terms are estimated at a rather fine geographical resolution scale for the whole of Europe, as discussed by Leip etâ•¯al. (2011, Chapter 16, this volume). We present here only a short summary of these data. Data on atmospheric deposition of nitrogen as nitrogen oxides and as ammonium are available from the calculation of the EMEP project. Owing to the much shorter residence time of NH3 than nitrogen oxides in the atmosphere, a large part of deposited reduced nitrogen is short-distance re-deposition of emitted ammonia. Therefore, for the purposes of estimating net input of N to large watersheds, local emissions of ammonium by agricultural sources should be subtracted from deposition figures, or, as often done, only the figures for oxidised nitrogen deposition should be considered. Synthetic fertiliser application rates are calculated from the CAPRI database, which is fed by the national fertiliser application rate by crop, communicated by EFMA (European Fertilizers Manufacturers Association). Crop N2 fixation is evaluated from the data on legume crop and grassland distribution considering their respective rates of N2 fixation. As also discussed by Leip etâ•¯al. (2011, Chapter 16, this volume), net commercial input/output of nitrogen as agricultural goods can be deduced from a budget of food and feed production by agriculture (autotrophy) versus local consumption by human and domestic animals (heterotrophy), both fluxes being expressed in terms of nitrogen (Billen etâ•¯al., 2007, 2009a,

273

Nitrogen flows from European regional watersheds Figure 13.3 Distribution of the balance between autotrophy and heterotrophy of the main watersheds across Europe (EU-27) (as calculated from the CAPRI-DNDC database, Leip etâ•¯al., 2011, Chapter 16, this volume). Green watersheds have an autotrophic status, while orange or red areas represent systems with heterotrophic status; yellow watersheds are balanced.

274

15000 Somme Severn autotrophy, kgN.km–2.yr –1

2010). Urban areas, where food is consumed but not produced, are obviously of dominant heterotrophic status. Rural regions specialised in crop farming and exporting their production to distant markets have an autotrophic status, while those which orient their agricultural activities towards intensive animal farming based on the import of feed from other regions have a heterotrophic status (Figure 13.3). When applied to European watersheds, this approach shows basins such as the Scheldt or the Po which have a strongly heterotrophic status, while basins such as the Seine or the Somme are highly autotrophic systems (Figure 13.4). Based on agricultural data from the second half of the twentieth century, Figure 13.4 shows the opposite trends in the historical trajectory of two exemplary basins, the Seine and the Scheldt, during the past 50 years:Â€ the former, turning towards exclusive cereal farming, become more and more autotrophic, while the latter, specialised in intensive animal husbandry, increased its heterotrophic status. These trends are important from the perspective of the nitrogen cascade, since the dominant source of nitrogen in autotrophic watersheds consists of inorganic fertilisers, while organic forms of nitrogen dominate in the nitrogen inputs to watersheds with heterotrophic status, which modifies the subsequent cascading nitrogen pathways. Summing up all the atmospheric deposition, fertiliser application and N2 fixation data, as well as the above-calculated net commercial imports (or exports) of nitrogen as food and feed, the net anthropogenic nitrogen input (NANI, Howarth etâ•¯al., 1996) to each European watershed can be calculated. Expressed per square kilometre, it represents the intensity of anthropogenic disturbance of the N cycle by introduction of reactive nitrogen into the biosphere at the regional basin scale (Figure 13.5). Values range from a few tens of kgN/km²/yr in Scandinavian watersheds, where atmospheric deposition dominates the total, to over 10 000 kgN/km²/yr in watersheds

10000 1990

Ems

2000 Seine 5000

0

1970

Scheldt 1950 1980 2005 Po

1955 19th c

Thames

Shannon Elbe

Tevere

pre-medieval medieval Kalix 0 5000

10000 –2

heterotrophy, kgN km

yr

15000 –1

Figure 13.4 Autotrophy/heterotrophy diagram showing the historical trajectory of the Seine (● ) and Scheldt (■) watersheds (Billen etâ•¯al., 2009b). Autotrophy represents the agricultural production of food and feed, while heterotrophy represents local human and livestock consumption. The position of a number of other European basins (◯) is also shown.

bordering the North Sea, where either fertilisers or commercial imports of feed dominate, depending on their autotrophic or heterotrophic status. Net total nitrogen inputs (NTNI) to watersheds can also be defined; this differs from NANI by the natural rate of atmospheric nitrogen fixation both by lightning and by biological N2 fixation in the soils of natural ecosystems. In Europe this rate has been evaluated at 1.5–2.5 kgN/ha/yr in Scandinavian forests, 5–25 kgN/ha/yr in temperate forests and 10–35 kgN/ha/yr

Gilles Billen Figure 13.5 Basin averaged net anthropogenic nitrogen inputs to European watersheds, based on CAPRI-DNDC data (LeipÂ€etâ•¯al., 2011, Chapter 16, this volume).

in Mediterranean shrubland, based on the compilation by Cleveland etâ•¯al. (1999).

13.2.2â•‡ Observed riverine nitrogen fluxes at the catchment outlet A database of nutrient fluxes delivered from large European watersheds at their outlet into estuarine/coastal waters has been assembled as part of NinE activities. The database (available as on-line supplementary material, see Section 13.7) includes recent data on total N, P and Si fluxes. Typically, average values of annual fluxes observed between 1995 and 2005 are recorded. When only inorganic nitrogen data where available, total nitrogen was estimated using the relationship between TN and DIN discussed by Durand etâ•¯al. (2011, Chapter 6, this volume). Figure 13.6 summarises the available data. Documented watersheds in the NinE database cover 69% of the total European watershed area. Nitrogen delivery rates range from less than 200 to more than 4000Â€kgN/km²/yr. Total nitrogen fluxes exported from small watersheds may be even higher, with values over 10â•›000 kgN/km²/yr. Figure 13.7 shows that values higher than 2000 kgN/km²/yr are always associated with the presence of agriculture as a major share of land use in the catchment, although the exported nitrogen fluxes from watersheds with the same percentage of agricultural land may vary greatly, reflecting differences in agricultural practices (including the proportion of low-intensity grazing) as well as climatic and hydrological conditions.

inputs of nitrogen to watersheds is actually exported by the river to the coastal zone. Although misleading, the term ‘retention’ is used extensively in the literature to designate all the processes preventing nitrogen load (i.e. NTNI) to a watershed being transferred to the outlet of the drainage network (Dillon etâ•¯al., 1990; Howarth etâ•¯al., 1996; Windolf etâ•¯al., 1996; Arheimer, 1998; Lepistö etâ•¯al., 2001). It accounts for the net effect of various biogeochemical processes responsible for temporary or permanent N removal from the water phase (such as biological uptake and biomass production, sedimentation and denitrification) or N removal from the land phase (such as gaseous losses by denitrification and nitrification, volatilisation and N storage in permanent vegetation, soils and groundwater). Howarth etâ•¯al. (1996, 2006), Boyer etâ•¯al. (2002) and Alexander etâ•¯al. (2002) showed that the flux of nitrogen exported by North American and Western European watersheds, over a background export of 107â•›kgN/km²/yr, accounts for a mean 26% of net anthropogenic nitrogen inputs (NANI), implying that 74% of the anthropogenically introduced nitrogen is retained or eliminated in the watershed. The data gathered in the NinE database allow testing this empirical relationship for the European watersheds for which an estimate of N delivery at the outlet is available (Figure 13.8a). Apparent retention, expressed as the fraction of NANI, varies from 90% to 50%, with a mean value of 82%. The regression of N delivery versus NANI is, however, highly significant:

13.2.3â•‡ Overall ‘retention’ within regional watersheds

Nflx (kgN/km²/yr) = 0.18 * NANI (kgN/km²/yr) + 228 (r²= 0.58) (13.1)

Comparing the data in Figures 13.5 and 13.6 immediately shows that only a limited fraction of the total anthropogenic

The modelled background nitrogen export of 228 kgN/ km²/yr, although burdened with substantial uncertainty

275


Riverine TN flux, kg/km²/yr

Figure 13.6 Available observed data on N delivery by European watersheds. Data from Humborg etâ•¯al. (2003, 2006, 2008); Radach and Pätsch (2007); Lancelot etâ•¯al. (1991); Billen etâ•¯al. (2009a, b); Neal and Davies (2003); Meybeck etâ•¯al. (1988); Ludwig etâ•¯al. (2009); Cociasu etâ•¯al.(1996); Johnes and Butterfield (2002); OSPAR (2002); REGINE (2010). See online supplementary material (Section 13.7) for original data.

14000 12000 10000 8000 6000 4000 2000 0 0

20

40

60

80

100

% Agriculture

Figure 13.7 Exported total nitrogen fluxes at the outlet of small to mediumsized watersheds from different areas of Europe as a function of the share of agricultural land (arable land and managed grassland) in total land cover. (data from Baltic countries (○), Germany and Czech Rep (□), United Kingdom (■), France (● ) and the Netherlands (◆)).

(± 100 kgN/km²/yr), agrees well with the experimental data from the boreal zone (Mattsson etâ•¯al., 2003; Kortelainen etâ•¯al., 2006; see below). Looking further to the variability of the retention factor, Howarth etâ•¯al. (2006) found that it could be correlated with a simple climate variable such as precipitation or discharge (Q), with retention decreasing from 95% to 40% of total NANI when the specific discharge increases from 100 to 800 mm/yr in North American watersheds. The even more pronounced effect of specific discharge on N retention was also underlined by Behrendt and Opitz (2000) and Billen etâ•¯al. (2009b). A sigmoid relationship of runoff has been proposed by Billen etâ•¯al. (2010) for world watersheds. From the data presented in Figure 13.8a, no clear relationship of retention with specific runoff emerges, which could explain the variability observed around relation (13.1), although a

276

trend around a sigmoid relationship of discharge is observed when the data are grouped into Scandinavian, temperate or Mediterranean river systems (Figure 13.8b). Other factors such as variability in temperature- and soil moisture-induced biogeochemical processes in watershed soils are likely to play a role as well. The presence of lakes and their location in the watershed with respect to the outlet are also important factors contributing to overall retention (Arheimer and Brandt, 1998; Lepistö etâ•¯al., 2006). On the basis of relationship (13.1), it is possible to use the distributed data on NANI (Figure 13.5) to calculate the most likely value of riverine-specific nitrogen delivery by undocumented European watersheds, thus interpolating the observed data of Figure 13.6. The overall nitrogen riverine delivery and retention for the watersheds of the major costal areas of Europe calculated on this basis are summarised in Table 13.1. According to this analysis, the total flux of nitrogen discharged by rivers into European coastal waters is 4760 ktonN/yr, accounting for 22% of the total amount of new nitrogen brought by anthropogenic processes to the corresponding watersheds (21 550 kton/yr). Although sizeable uncertainties affect these estimations, they clearly show the extreme perturbation of the N cycle in most European watersheds. They also stress that water resources contamination is only one of many important pathways of the cascade followed by anthropogenic nitrogen introduced into regional watersheds. Understanding and predicting the relation between N-related human activities in a watershed and the amount of N transferred by the hydrosystem is therefore a key scientific question. It is also a major management issue, as many measures can potentially act on retention processes.

Gilles Billen 5000 50% ret y = 0.18x + 228 r 2 = 0.54

FlxN, kgN/km2/yr

4000

3000

2000

90%ret

1000

0 0

5000

10000

15000

20000

2

NANI, kgN/km /yr

a.

1.25 nordic

fraction of NTNI exported

1.00

temperate

0.75 southern

0.50

13.3.2â•‡ A typology of models for regional watershed N transfers

0.25

0.00 0 b.

aquatic processes of nitrogen retention and makes no difference between the pathways through which nitrogen inputs are introduced to the hydrosystem, either as diffuse processes on the terrestrial watershed or as discrete point injection directly into the drainage network (Figure 13.9). This distinction between diffuse and point sources of nitrogen is important because different retention processes act on each of them:Â€landscape processes including storage in soil organic matter or biomass pools, soil denitrification or ammonia volatilisation, storage in deep aquifers, etc., act on diffuse sources of nitrogen. In-stream processes, including river bed denitrification or sediment storage, act on point sources and on diffuse sources after the action of landscape processes. Identifying and quantifying the various pathways of nitrogen contamination of and transformation in hydrosystems is therefore of prime importance for understanding the nitrogen cascade at the regional scale. Various models have been developed for this purpose and will be briefly described in the following section. To implement them, detailed data on point and non-point sources of nutrients to the hydrosystem are required. As far as point sources of nitrogen from urban wastewater are concerned, a detailed inventory is available at the European scale, thanks to the efforts of the EC within the implementation of the Water Framework Directive (Bouraoui etâ•¯al., 2009). The geographical pattern shown (Figure 13.10) closely follows the distribution of large cities across Europe, with the transversal dorsal of rich cities extending from Birmingham to Milan (‘The Blue Banana’, Brunet, 2002). The definition of diffuse sources depends considerably on the particular model used, as will be shown below.

500

1000

1500

2000

runoff, mm/yr

Figure 13.8 (a) Specific N flux delivered at the outlet of European watersheds as a function of the net anthropogenic nitrogen inputs (NANI). The heavy line represents the regression across all points. The lighter lines represent N retention of 50% and 90% of NANI. (b) Fraction of NTNI delivered at the outlet as a function of runoff for Scandinavian (◊), temperate (■) and Mediterranean (▲) watersheds. The line represents the best fit of the sigmoid relationship proposed by Billen etâ•¯al. (2010):Â€fraction NTNI exported = exp(−(Q−Qm)²)/Qs²), where Q is the mean specific runoff (mm/yr) of each watershed and Qm and Qs are climate-specific parameters.

13.3â•‡ Modelling N fluxes through watersheds 13.3.1â•‡ Point and diffuse sources of nitrogen to the hydrosystem The above NANI approach is based on a pure black-box input– output budget of the watershed as a whole. It quantifies the overall retention without identifying the processes responsible for it. In particular, it does not distinguish between terrestrial and

A large number of models have been used for quantifying nutrient transport and retention at the regional river-basin scale, which relies on a wide range of different assumptions and different methods for the description of nutrient sources, catchment characteristics and the physical and biogeochemical processes involved. Table 13.2 lists a number of such models, with their general basic equations and principles, their required input data and a list of watersheds where they have been applied. All models have been validated in diverse catchment areas and all are capable of satisfactorily predicting nutrient export from land use and point inputs, but they differ in the system boundaries, their spatial resolution, the complexity of their representation of the processes and their temporal resolution. A first difference lies in the definition of the system modeled:Â€some models work with the drainage network only; others encompass the entire watershed, including part or all of the soil and groundwater landscape components as well. This difference implies different ways of defining the diffuse sources of nitrogen and of dealing with the nitrogen dynamics in the soil–plant system above the root zone (plant growth, mineralisation, immobilisation, denitrification, etc.). For instance, many models, such as N-exret and RivR-N, are typically

277

Nitrogen flows from European regional watersheds Table 13.1 Overall nitrogen input, riverine delivery and percentage retention for the watershed of the major European coastal areas. (The documented area represents the percentage total watershed area for which observed data of nitrogen delivery are available; relationship (13.1) with NANI is used for the undocumented areas, then the overall specific delivery is calculated for the whole coastal zone watershed area.)

Basin area km²

Documented %

Specific delivery (weighted average) kgN/km²/yr

Net input (NANI) ktonN/yr

Total delivery ktonN/yr

Arctic Ocean

Retention of NANI % —

Norway & Finland Arctic coast

122 929

28

192

10

24

—

NW Norway coast

132 905

38

225

11

30

—

White Sea

281 927

0

244

—

69

—

83 974

66

309

116

26

78

Bothian Sea

499 488

88

211

181

105

42

Gulf of Finland

421 306

90

271

872

114

87

Gulf of Riga

137 846

90

634

712

87

88

S Baltic and Belt area

506 045

86

699

2720

354

87

47 135

27

339

—

16

—

Skagerak & Kattegat

196 115

80

509

302

100

67

Wadden Seas & W Danish coast

447 931

83

2017

3392

904

73

Seine BightÂ€– Belgian coast fringe

129 219

78

2024

888

262

70

E English & Scottish coast

115 361

48

1502

582

173

70

34 356

99

2667

289

92

68

Biscay Bay

267 545

92

1175

1217

314

74

Portuguese & W Spanish coasts

370 252

61

623

1202

231

81

W Scottish, Irish & Welsh coasts

Baltic Sea SE Sweden coast

North Sea W Norway coast

Atlantic Ocean Brittany

134 597

58

1386

473

186

61

S Welsh & English Coast

18 569

14

1374

103

26

75

Irish Sea

42 638

24

1758

303

75

75

S Spanish coast

42 440

0

988

179

42

76

305 052

74

811

1135

247

78

Tyrrhenian Sea

74 650

24

1122

359

84

77

Central Mediterranean Sea

13 433

0

995

57

13

77

Mediterranean Sea Lyon Gulf

Ionian Sea

61 775

9

767

193

47

75

Adriatic Sea

237 711

40

1191

1148

283

75

Aegean Sea

155 260

65

530

789

82

90

Black Sea W Black Sea coast Total

991 235

81

782

4317

775

82

5 868 102

69

811

21 550

4761

78

drainage network models which define diffuse inputs of nutrients on the basis of an export coefficient approach, i.e. by associating an empirically determined mean annual nitrogen flux of nutrients to the watershed’s different land use classes, without taking into account specific internal soil processes. The Seneque/Riverstrahler model uses a similar approach, except that the mean annual concentrations of surface and base flow

278

runoff are associated with land use/lithologic classes, these concentrations being either empirically defined or from offline runs of plant–soil models at the plot or landscape scales, as described by Cellier etâ•¯al. (2011, Chapter 11, this volume). Full watershed models, such as Swat, Inca, Green or EveNFlow, fully integrate a description of the processes occurring in the top soil-plant system and define diffuse sources as the inputs

Gilles Billen Figure 13.9 Comparison of the input– output view of the watershed behind the black-box NANI approach (a), with a view of the watershed distinguishing between point sources from urban wastewater and diffuse sources from agricultural soils, and considering landscape and in-stream retention processes separately (b). Refer to Figure 13.2 for a more detailed view.

to soil as atmospheric deposition, biological nitrogen fixation, inorganic fertiliser and manure application. The complexity in the description of soil nutrient dynamics varies considerably, however, between these models, from simple regression relationships (e.g. Green) to detailed process modelling (e.g. Swat). Other models, such as Sparrow, Polflow and Moneris, define diffuse sources as the soil nitrogen surplus, i.e. the difference between total inputs to soil and outputs as crop yield, which is assumed to be directly transferred to the hydrosystem. The spatial resolution also differs widely between the models. Lumped approaches do not consider any spatial distribution of sources and sinks within the watershed; fully distributed models, on the other hand, account for the spatial variability of processes, nutrient inputs and watershed characteristics. The latter obviously require high-resolution spatial referencing of all constraint data, which might be difficult to obtain for large regional applications. Semi-distributed models illustrate the

intermediate case where sub-basins are divided into uniform units (e.g. in terms of land use or vegetation zones) as in the Green or Swat models, or a drainage network into a regular scheme of confluence of tributaries with mean characteristics by stream-order as in the Riverstrahler approach. Most often, depending on data availability and spatial resolution, models may compromise between fully distributed and totally lumped methods, also adapting to the level of process-description details. Indeed, another major difference between models lies in the complexity of their representation of the processes affecting the nitrogen transfer and transformations, ranging from a very simplified to an extremely complex description. Statistical regression models, such as Green, Sparrow and Polflow, consist in simple correlations of stream monitoring data with watershed sources and landscape properties and provide empirical estimates of nutrient stream export, based on a few explanatory

279

Nitrogen flows from European regional watersheds Figure 13.10 N point source emissions (urban communities and industries) on a 1â•›km² grid over Europe (from Bouraoui et al., 2009).

variables or predictors. The low data and time requirements of such methods explain their popularity for modelling nutrient fate in large river basins. Nevertheless, as soon as forecasting, i.e. explaining and describing the evolution over time of nutrient export and its dependence on the several controlling factors, is needed, mathematical models should be used for mechanistically describing the physical and biogeochemical underlying processes, as in the Inca, Swat and Seneque/ Riverstrahler approaches. Owing to their very detailed deterministic description of the processes, these models reduce site specificity to generically explain nutrient transfers with no or minimal need for calibration, hence providing a true understanding of the mechanisms involved. However, the intensive data requirement of such approaches leads to hybrid methods based on empirical relations for quantifying processes whose interactions are mechanistically expressed. While most regression models only predict mean annual nutrient fluxes, mechanistic models such as Swat, Inca and Seneque/Riverstrahler necessarily take into account seasonal patterns and provide inter-annually and seasonally variable flux and concentration results, which might be of prime importance for assessing the effect on receiving marine ecosystems. Finally, the models may differ in terms of the variables described, either total N or the different chemical (organic and mineral, dissolved or particulate) forms, and possibly other nutrients such as P and Si.

13.3.3â•‡ Comparison of different modelling approaches (EUROHARP) As illustrated above, a large range of modelling approaches of the nitrogen cascade in large watersheds are available, each of them corresponding to a specific objective and perspective.

280

The EUROHARP project (Kronvang etâ•¯al., 2009) aimed at providing a broad range of end-users with unbiased guidance for an appropriate choice of model to satisfy existing European requirements on harmonisation, reliability and transparency for quantifying diffuse nutrient losses. It focused on diffuse nutrient losses, nitrogen and phosphorus in particular, from agricultural land to surface freshwater systems and coastal waters, to help end-users implement the Water Framework Directive and the Nitrate Directive. Nine different models were applied to 17 catchments in Europe covering a broad range of climatic, pedologic and farming practices gradients. The models were selected by each participant in the project as one of the official models being used in assessing nutrient losses to surface waters. The models selected included Moneris (Behrendt etâ•¯al., 2002), Swat (Neitsch etâ•¯al., 2001) and TRK/ HBV-NP (Brandt and Ejhed, 2002) (see Table 13.2), among others. The models are fully described in Schouman etâ•¯al. (2003) and vary from simple loading functions to complex fully distributed mechanistic models. All models were applied to three core catchments, located in the UK, Italy and Norway, in order to fully investigate the similarities and differences in the various approaches not only in estimating the losses, but also in assessing the contribution of different pathways of losses, nutrient turnover, etc. At least four of the models were applied to each of the 14 remaining catchments in order to test their applicability. Overall it was concluded that no single model appeared consistently superior in terms of its performance across all three core catchments. Indeed, according to the output variable considered, depending on the goodness of fit of the test used, the models ranked differently on the three core catchments. The largest variations between model predictions (largest standard deviations) were found for the three Mediterranean catchments mostly due to the limited data availability when

281

Distributed drainage network model (1×1â•›km grid cells)

Lumped full watershed models

Semi-distributed (lumped by subbasins)

Semi-distributed drainage network (HBV-NP) or full watershed (TRK) model (sub-basins divided into elevation zones)

Lumped watershed model

Global-NEWS models Dumont etâ•¯al., 2005 Seitzinger etâ•¯al., 2009 Mayorga etâ•¯al., 2010

Green Grizzetti et al., 2005, 2008 Bouraoui etâ•¯al., 2009

(TRK)/HBV-NP Arheimer, 1998 Arheimer etâ•¯al., 1998 Bergström etâ•¯al., 1987 Brandt, 1990 Petersson et al., 2001

Moneris Behrendt, 2002

Geographical resolution

N-Exper is over Lepistö et al., 2001, 2006

Model and authors

Land use including tile drained areas Runoff divided into several pathways Point emissions (PS) Diffuse sources (DF) defined as soil N surplus

Daily meteorological data, point inputs (P), atmospheric deposition, land-use specific soil-leaching concentrations, potentially produced by a soil model (TRK)

Land use, rainfall, drainage network morphology (length, lake area), point emissions (PS), diffuse sources (DS) (fertiliser and manure application, atm. deposition, biological N2 fixation)

Land use, runoff, lake and reservoirs (dam) area, point emissions (PS) diffuse sources (DS) (fertiliser and manure application, atm. deposition, biological N2 fixation, minus N in crops and grass removed from land)

Point emissions, distributed land use, drainage network morphology,

Required input data

Nutrient outlet load = Σ (for different pathways) α.f(residence time).[DS.+ PS] where αs are a priori calibrated retention parameters

Mechanistic model of N soil dynamics (TRK) Simple calibrated first order, temperature-dependent, in-stream retention parameter (HBV-NP)

Nutrient outlet load = α.f(L,Area).[DS.β.f(R) + (PS+UL)] where α and β are calibrated in-stream and watershed retention parameters L is the total river length and Area the lake area in the watershed, R is rainfall, UL is the upstream load.

Nutrient outlet load = α.f(WA, RA, IR).[DS.β.f(Roff ) + PS)] where α and β are calibrated in-stream and watershed retention parameters WA is the watershed area RA is the reservoir (dammed) area IR is the water removed for irrigation Roff is runoff

N export coefficient by land use class; in-stream and riparian retention parameterised

Basic equation(s)/principle(s) for nutrient transfer representation

Table 13.2 A summarised description of a sample of models for nutrient transport and retention at the scale of regional watersheds

Mean annual

Daily

Annual mean

Annual mean

Annual mean

temporal resolution

Total N total P

DIN, orgN, totP

Total N total P

DIN, DON, PON

Total N

variables

German river basins

Entire Baltic Sea drainage basin

All European basins

All world watersheds

All Finnish watersheds

Watersheds

282

Distributed full watershed model (regular grid cells)

Semi-distributed full watershed model (sub-basin structure based on monitored reaches)

Distributed full watershed model

Semi-distributed full watershed model

Semi-distributed (Riverstrahler) or fully distributed (Sénèque) drainage network model

Sparrow Smith etâ•¯al., 1997; Preston and Brakebill, 1999; Alexander etâ•¯al., 2000, 2001

Inca Whitehead etâ•¯al., 1998 Wade etâ•¯al., 2002a, b

Swat Arnold etâ•¯al., 1998, 1999 Neitsch etâ•¯al., 2001, 2005

Riverstrahler/Sénèque Billen etâ•¯al., 1994 Garnier etâ•¯al., 1995, 2002; Ruelland etâ•¯al., 2007; Thieu etâ•¯al., 2009

Geographical resolution

Polflow De Wit, 2001

Model and authors

Table 13.2 (cont.)

Meteorological data, drainage network morphology, point emissions, land use, nutrient concentrations of superficial and base flow to streams for each land use class.

Meteorological data, topographic slope, soils, land use, nutrient emissions, agricultural management strategies

Daily meteorological and hydrological series, basin characteristics, point inputs, land use, growing seasons of crops, diffuse emissions (included fertilisers and livestock)

Basin characteristics (air temperature, precipitation, land-surface slope, soil permeability, stream density, and wetland area) and drainage network characteristics (discharge, time of travel), discharge data Point sources (PS) Diffuse sources (DF) defined as fertilisers and manure application, nonagricultural runoff and atm. deposition

Runoff and aquifer residence times, basin topography, soil & aquifer types, Point sources (PS) Diffuse sources (DF) defined as soil N surplus for each grid cell

Required input data

Kinetic formulation for each process describing the in-stream dynamics of nutrients, phytoplankton, zooplankton, bacteria (Rive model)

Mechanistic description of water, nutrient and pesticide routing and transformation in the watershed; mixing equations and simple parametric relationships for drainage network processes

Detailed mechanistic approach: * differential equations for describing losses in plant/ soil system and instream processes (nitrification, denitrification, sediment dynamics, biological uptake); * reaction rates are calibrated.

Nutrient load in each reach x = αf(cz, tt). [DS(x).β + (PS(x)+ Σ UL(x))] where α and β are calibrated in-stream and watershed retention parameters In-stream N retention depends on channel size (cz) and time of travel (tt) (first-order kinetics) Watershed retention depends on basin characteristics. UL is the upstream load.

Nutrient load at grid cell x= αf(slope).[DS(x).βf(s,r) + (PS(x)+UL(x))] where α and β are calibrated in-stream and watershed retention parameters In-stream N retention depends on slope and runoff Watershed retention depends on soil type and residence time in aquifers Denitrification in groundwater:Â€regression on residence times and infiltration. UL is the upstream load.

Basic equation(s)/principle(s) for nutrient transfer representation

10-Day periods

Daily

Daily

Mean annual

Mean annual

temporal resolution variables

NO3, NH4, diss&partorg N, P, Si, orgC, phyto/ zooplankton, bacteria

NO3, NH4, diss&partorgN

NO3, NH4, total P

Total N total P

Total N total P

Watersheds

Seine, Somme, Scheldt, Mosel, Danube, Kalix, Lule, Red rivers

Many watersheds in Europe and America

A wide range of catchments across Europe

Major US watersheds

Rhine, Elbe, Norrström

Gilles Billen Table 13.3 Budget of nitrogen to the basins of a number of rivers as calculated by the MONERIS Model for the 2001–2005 period (Behrendt etâ•¯al., unpublished data)

kgN/km²/yr (%)

Danube

Rhine

Weser

Elbe

Odra

Input to land (before landscape retention)

2080

4400

5390

3810

2840

â•… from fertilisers and manure

930

2660

3500

2300

1430

â•… from atmospheric deposition

1150

1740

1890

1510

1410

Landscape retention

1490

3010

4030

2740

2100

â•… diffuse sources (after landscape retention)

590 (70)

1390 (76)

1360 (87)

1070 (78)

740 (80)

â•… background

60

80

70

50

40

â•…â•… from fertilisers and manure

210

670

720

530

360

â•…â•… from atmospheric deposition

330

740

570

490

340

Point sources

250 (30)

450 (24)

210 (13)

310 (22)

190 (20)

Total inputs to hydrosystem

840 (100)

1840 (100)

1570 (100)

1380 (100)

930 (100)

Delivery at outlet

560 (67)

1490 (81)

1300 (83)

920 (67)

400 (43)

In-stream retention

280 (33)

350 (19)

270 (17)

460 (33)

530 (57)

Table 13.4 . Nitrogen budget for three large European watersheds under wet and dry hydrological conditions, calculated by the Riverstrahler model (Trifu-Raducu, 2002; Thieu etâ•¯al., 2009)

Danube

Seine

Scheldt

kgN/km²/yr (%)

1993 (dry)

1996 (wet)

1996 (dry)

2001 (wet)

1996 (dry)

2001 (wet)

Diffuse sources

966

1079

2012

3905

1227

2837

Point sources

281

281

553

553

1013

1013

Total inputs

1247 (100)

1360 (100)

2566 (100)

4459 (100)

2240 (100)

3850 (100)

Delivery at outlet

474 (38)

667 (49)

1378 (54)

2311 (52)

1213 (54)

2084 (54)

Groundwater storage

—

—

356 (14)

707 (16)

193 (9)

392 (10)

Riparian retention

273 (22)

312 (23)

524 (20)

1161 (26)

607 (27)

1167 (30)

In-stream retention

492 (39)

367 (27)

185 (7)

132 (3)

225 (10)

207 (5)

Reservoir retention

8 (0.6)

14 (1)

13 (0.5)

40 (1)

—

—

compared to the other catchments. Another critical factor affecting model results in these catchments was the model formulation, since in general most models were not developed to cover the Mediterranean regions typically characterised by nonpermanent flow, high rainfall intensity, etc. A similar limitation was found in the Norwegian catchments where none of the models considered frozen soils. The EUROHARP project highlighted that one of the major sources of discrepancy between the models is the quantification of the retention process. As most models are based on a mass balance approach, in order to accurately quantify the export of nitrogen at the catchment outlet, the models tended to adjust river retention accordingly, resulting in differences in retention estimates larger than one order of magnitude. It is important to note that even if most models did reproduce water and nutrient losses at the outlet reasonably well, the pathways of losses differed considerably between the models. To increase the reliability of the prediction of diffuse losses, it is suggested to scrutinise the internal processes and pathways simulated by the models whenever possible. The overall conclusion was that

the selection of the best model for N loss estimation should be made on a case-by-case basis depending on the catchment type, the purpose of the application, data availability, model limitations, expertise, etc. The parallel use of several models should always be recommended.

13.3.4â•‡ Environmental controls of N retention processes As explained in the above discussion, different models provide different visions of the nature and quantitative importance of retention processes. It is possible, however, to draw a number of general conclusions from the results of various models. Of particular interest in this respect are the results of those models which provide a detailed estimation of different pathways of nitrogen transfer through the watershed and the drainage network, and the corresponding retention. Examples of such results are presented in Tables 13.3 and 13.4, respectively from the Moneris model (a lumped, annual, calibrated model) and the Sénèque model (a distributed, seasonal, mechanistic model).

283

Nitrogen flows from European regional watersheds 80

The effect of nitrogen delivery on the coastal zone is highly dependent on the accompanying fluxes of the other nutrients required for the development of marine phytoplankton, particularly phosphorus and silica. For this reason, P and Si delivery rates were also gathered in the database established in the scope of the ENA (Figure 13.12). Table 13.5 summarises the data grouped according to the main coastal marine receiving areas. These are inter-annual average values, and it must be stated again that annual fluxes at the outlet of a regional basin can vary within a factor of two between a dry and a wet hydrological year. Moreover, since not all basins in each coastal zone watershed are documented, missing information has been obtained by extrapolation from nearby documented areas. The figures in Table 13.5 should therefore be considered rough estimates.

70 60

% N retention

50 40 30 20 10 0 –10 0

5

10

15

20

% lake area in the watershed Figure 13.11 In-stream N retention (%) as a function of the percentage lake area in the watershed for 30 Finnish river systems (data from Lepistö etâ•¯al., 2006).

These models show the significance of processes occurring in the watershed’s upper soil layers, in the unsaturated zone and in the riparian wetlands for eliminating or storing nitrogen surplus from agricultural soil. Conversely, tile drainage, which affects agricultural soils in large areas in Europe (including Great Britain, France, the Netherlands, Denmark, Norway, the Baltic countries, Poland and Germany), accelerates nitrogen transfer to surface water. In-stream retention largely depends on the residence time of water masses through the drainage network and is therefore dependent on both the specific runoff and the presence of lakes and ponds (Figure 13.11).

13.4â•‡ N, P and Si delivery from watersheds 13.4.1â•‡ The present situation The level of nitrogen surface water contamination as revealed by the N delivery at the outlet of the major regional watersheds of Europe is depicted in Figure 13.6 above. Delivery rates are at least twice the background value in most of Europe except in the Scandinavian areas, and rates more than ten times the background are not unusual. This reflects a severe level of surface and groundwater contamination, which is described and discussed in Grizzetti etâ•¯al., 2011 (Chapter 17 this volume). The total flux of nitrogen delivered to the sea along the EU27 coasts can be estimated at 4.8 TgN/yr of which 4.3 TgN/ yr comes from EU27 (4 171 851 km² watershed) and 0.5 TgN/ yr from outside EU27 (449 085 km²). This rate of N delivery is nearly five times the estimated natural background (0.98Â€TgN/yr).

284

13.4.2â•‡ The potential for coastal eutrophication (ICEP) It is now well recognised that the basic cause of coastal eutrophication is related not only to the general nutrient enrichment of the marine system, but also to the imbalance in the delivery of nitrogen (and phosphorus) with respect to silica. Indeed, many authors (Officer and Ryther, 1980; Conley etâ•¯al., 1993; Conley, 1999; Turner and Rabalais, 1994; Justic etâ•¯al., 1995; Billen and Garnier, 1997, 2007; Turner etâ•¯al., 1998; Cugier etâ•¯al., 2005) have shown that coastal eutrophication is the consequence of excess nitrogen and phosphorus delivery with respect to silica, in relation to the requirements of diatom growth. They underlined that coastal enrichment with nutrients brought in proportion of the Redfield ratios (Redfield etâ•¯al., 1963), characterising the requirement of diatom growth, seldom causes problems, but, on the contrary, stimulates a healthy and productive food web, as is the case in upwelling areas where new planktonic primary production is mostly ensured by diatoms, while non-siliceous algae are restricted to regenerated production. By contrast, coastal eutrophication problems are the manifestation of new production of non-siliceous algae sustained by external inputs of nitrogen and phosphorus brought in excess over silica, thus in conditions where diatom growth is limited. Based on this view of coastal eutrophication, Billen and Garnier (2007) developed an indicator of coastal eutrophication potential (ICEP) of riverine nutrient inputs. This represents the carbon biomass potentially produced in the receiving coastal water body through new production sustained by the flux of nitrogen or phosphorus (depending on which one is limiting with respect to the other) delivered in excess over silica. For the purposes of a river-to-river comparison, it is expressed by unit of watershed area, in kgCkm2/day. It can be calculated by the following relationships (based on the Redfield molar C:N:P:Si ratios 106:16:1:20): N-ICEP = [NFlx / (14*16)Â€– SiFlx / (28*20) ] * 106 * 12 if N/P < 16 (N limiting) P-ICEP = [PFlx / 31Â€– SiFlx / (28*20) ] * 106 * 12 if N/P > 16 (P limiting)

Gilles Billen Figure 13.12 Available observed data on total P (a) and Si (b) delivery by European watersheds (see Figure 13.6 and supplementary material for references).

where PFlx, NFlx and SiFlx are, respectively, the mean specific fluxes of total phosphorus, total nitrogen and dissolved silica delivered at the outlet of the river basin, expressed in kgP/km²/ day, in kgN/km²/day and in kgSi/km²/day. A negative ICEP value indicates that silica is present in excess over the limiting nutrient (among nitrogen and phosphorus) and thus characterises the absence of eutrophication problems. Positive values indicate an excess of nitrogen or phosphorus over the potential for diatom growth, thus a condition for harmful non-siliceous algae development. As defined,

the ICEP does not take into account the particular conditions determining the response of the coastal zone into which the river is discharging, but simply represents the potential impact of the riverine fluxes. According to the N/P ratio of nutrient loading, N or P is the potential limiting nutrient. The ICEP should theoretically be calculated with respect to this nutrient. However, even in the case where P is limiting, a large excess of nitrogen with respect to silica is probably a risk for coastal eutrophication. This is because P is rapidly recycled in the marine environment, so

285

Nitrogen flows from European regional watersheds Table 13.5 Average specific fluxes of N, P and Si delivered by rivers into the different European coastal areas (P and Si flux values in italics and in brackets are educated guesses for undocumented areas). Corresponding Indicator of Coastal Eutrophication Potential (Billen and Garnier, 2007), calculated as the CÂ€equivalent of either N (N-ICEP) or P (P-ICEP) brought in excess of Si with respect to the requirements of diatom growth

Weighted average river loading kgN/km²/yr

kgP/km²/yr

ICEP (N)

kgSi/km²/yr

ICEP (P)

Limitation

mgC/km²/day

Arctic Ocean Norway & Finland Arctic coast

192

3.1

865

−2.4

−5.0

P

NW Norway coast

225

6.0

992

−2.7

−5.5

P

White Sea

244

(4.5)

(900)

−1.8

−5.0

P

309

7.5

268

3.1

−0.8

P

−4.2

Baltic Sea SE Sweden Coast Bothian Sea

210

10

860

−2.1

Gulf of Finland

271

12

145

3.3

Gulf of Riga

634

15

434

7.2

−1.0

P

S Baltic and Belt area

698

33

497

7.8

0.6

P

W Norway coast

339

17

1427

−3.6

−7.0

P

Skagerak & Kattegat

509

644

3.9

−3.2

P

8.1

P

5.0

P

0.45

P P

North Sea

Wadden seas & W Danish coast

2017

Seine BightÂ€– Belgian coast fringe

2024

E English & Scottish coast

1502

Brittany

2667

Biscay Bay

7.4 136

1158

24

93

881

26

265

1074

17

23

N

21

858

36

−2.9

P

Atlantic ocean 1175

62

1429

9.4

−1.9

P

Portuguese & W Spanish coast

623

32

485

6.7

0.6

P

W Scottish, Irish & Welsh coast

1386

73

(1250)

14

0.5

P

S Welsh and English coast

1374

148

(1250)

14

8.9

P

Irish Sea

1758

216

1054

21

988

(30)

(485)

12

811

44

815

S Spanish coast

18

P

0.35

P

−0.11

P

Mediterranean Sea Lyon Gulf Tyrrhenian Sea

7.5

1122

30

800

12

−1.6

P

Central Mediterranean Sea

995

(25)

(800)

10.5

−2.1

P

Ionian Sea

767

21

800

6.9

−-2.6

P

Adriatic Sea

1191

92

2130

5.3

−2.9

P

Aegean Sea

530

246

1360

−0.2

19

N

782

31

267

11.5

Black Sea W Black Sea coast

that P limitation might not be effective as long as high nitrogen concentrations are available. Moreover, there is evidence that toxin production by non-siliceous as well as siliceous algae is enhanced in high nitrogen concentrations (Murata etâ•¯al., 2006). An additional reason for considering the N-ICEP even in situations where the N/P ratio is above the Redfield ratio is that excess N not used in a coastal zone is likely to be exported to adjacent areas where it might cause eutrophication problems.

286

1.8

P

For the European rivers for which N, P and Si loading are documented the (N- and P-) ICEPs have been calculated (FigureÂ€13.13). The mean values extrapolated to all European coastal areas are summarised in Table 13.5. Figure 13.13 clearly shows that excess N or P delivery with respect to silica is widespread in Europe, with the exception of northern Scandinavia. In most of Europe, nitrogen excess is much more pronounced than phosphorus excess; the reverse is true only in the southern part of the Balkan peninsula,

Gilles Billen Figure 13.13 Calculated values of N-ICEP (upper panel) and P-ICEP (lower panel) at the outlet of European watersheds.

where specific phosphorus delivery is still very high (see Figure 13.12).

13.4.3â•‡ Historical trends The land- and waterscape of Europe is the heritage of millennia of a complex human history which modified the land cover as well as the river morphology and hydrology. For a number of watersheds, retrospective studies have reconstructed past trends of nutrient inputs, transfer and delivery to the

coastal sea in response to changes in the constraints imposed by human society, using both historical records and modelling approaches (Andersson and Arheimer, 2003, for Swedish rivers; Behrendt etâ•¯al., 2002, for the Odra and Danube; Billen etâ•¯al., 2005, for the Scheldt basin; Billen etâ•¯al., 2007, for the Seine basin; Stalnacke etâ•¯al., 2003, for Latvian rivers). Such historical studies allow assessing the present degree of perturbation of European watersheds with respect to either pristine or historical situations. They are also particularly useful

287


to examine the time lag involved in the response of compartments of the system with a very long life time, such as large aquifers and urban structures. We present here a summary of the general findings of these studies at the European scale.

Reconstituted pristine situations The pristine level of nitrogen inputs to river systems corresponds to background nitrogen concentration in runoff water from unperturbed forested areas, plus the input of litter from riparian trees. For a hypothetical pristine, entirely forested Seine watershed, Billen etâ•¯al. (2007) estimated this to be 120– 300 kgN/km²/yr. The corresponding delivery at the outlet of the drainage network was in the range 60–150 kgN/km²/yr according to hydrological conditions. Similarly, Thieu etâ•¯al. (2010) calculated values in the range 50–250 kgN/km2/yr for the pristine state of the Seine, Somme and Scheldt rivers. These figures are consistent with the value of 228 kgN/km²/yr found above for the y intercept of delivery vs NANI (Figure 13.8, relation (13.1)). They are also close to the values reported for present delivery rates of Swedish and Finnish rivers (see Table 13.1). Spatially representative long-term databases from 42 unmanaged headwater catchments covered by peatlands and forests showed average long-term N export around 130 kgN/km2/yr (site specific range, 29–230â•›kgN/km2/yr; Kortelainen etâ•¯al., 2006) and 140 kgN/km2/yr (site specific range, 77–230â•›kgN/ km2/yr; Mattsson etâ•¯al., 2003). Corresponding modelled pristine figures for phosphorus and silica delivery from the Seine, Somme and Scheldt basins are in the range of 8–30 kgP/km²/yr and 350–1500 kgSi/km²/yr (Thieu etâ•¯al., 2010a, b). Observed phosphorus delivery in boreal Finnish rivers is 2–5 kgP/km²/yr. The average long-term P export from unmanaged Finnish catchments was 5.0 kgP/km²/yr (range, 1.7–15 kgP/km²/yr; Kortelainen etâ•¯al., 2006) and 5.4Â€kgP/km²/yr (range, 2.1–18 kgP/km²/yr; Mattsson etâ•¯al., 2003).

Preindustrial agricultural systems Traditional agricultural practices involved rotation alternating a fallow period and one or two cereal crops, and using manure fertilisation. Estimated nitrogen delivery from landscapes characterised by such agrarian systems varies between 300 and 800Â€kgN/Â�km²/yr in the Seine basin based on a few available measurements dating back to the nineteenth century (Billen etâ•¯al., 2007). In an attempt to evaluate the nitrogen delivery from the Seine, Somme and Scheldt basins under a hypothetical scenario with generalised organic farming over their whole agricultural areas, Thieu etâ•¯al. (2010a, b) obtained figures ranging from 430 to 950 kgN/km²/yr depending on the basin and the hydrology. Phosphorus release from agricultural soils increased significantly with respect to pristine levels because of higher erosion losses. Direct release of phosphorus from point sources from even small cities also leads to increased phosphorus contamination of surface water. For the Seine watershed, the estimated delivery in the periods preceding the twentieth century was in the range of 15–50 kgP/km²/yr. The question of the role played by agriculture in increasing silica delivery is still under debate. It was generally assumed

288

that dissolved silica concentration in runoff water, because it originates from rock weathering, only depends on the watershed’s lithology. However, some authors stressed the role of vegetation in a terrestrial silica cycle involving active uptake of silica from soil by plants and release of biogenic silica under the form of phytoliths, the dissolution or erosion of which contributes to the inputs of silica from soils to the surface water. Agriculture could therefore have influenced the diffuse sources of silica to river systems (Conley, 2002; Humborg etâ•¯al., 2004). Rantakari and Kortelainen (2008) demonstrated that in a randomly selected Finnish lake database, SiO2 had highest correlation coefficient with TIC and CO2 in lakes surrounded by peatlands, the relation between SiO2 and inorganic carbon was less close in lakes surrounded by forests or agricultural land, supporting the important role played by biogenic Si cycling. The decomposition of organic matter produces organic acids and carbon dioxide, both of which enhance weathering and thus SiO2 concentrations.

From 1950 to 1985 In most regions of Europe, the second half of the twentieth century was characterised by both increased urbanisation, often with few wastewater treatment infrastructures, and generalisation of modern agricultural practices with increased use of synthetic fertilisers. In the Seine basin, N delivery peaked in the 1980s at 1500–3000â•›kgN/km²/yr (according to hydrological conditions). In the Odra river (where low specific discharge is responsible for high retention), Behrendt etâ•¯al. (2005a) calculated an increase from 270 kgN/km²/yr in 1960 to 595 kgN/km²/yr in 1980. As far as phosphorus is concerned, this period is also characterised by the substitution of traditional soap products by P-containing washing powders, which led to a fourfold increase in the per capita P release rate in urban wastewater. In the Seine basin, delivery rates as high as 350 kgP/km²/yr were reached at the end of 1980. In the Odra River the increase during the 1960–1980 period was from 20 to 50Â€kgP/km²/yr. Silica release from domestic wastewater, although not insignificant (Sferratore etâ•¯al., 2006), is relatively low with respect to the N and P content of urban wastewater. Moreover, eutrophication of surface water, owing to N and P contamination, often resulted in increased retention of dissolved silica related to a more intense diatom development in rivers. Impoundments of large reservoirs also led to increased silica retention either by algae development and trapping of dissolved silica or biogenic particulate silica produced upstream, or by reducing rock weathering in flooded, former wetlands (Humborg etâ•¯al., 2004, 2006). As a result, the period of industrialisation and urbanisation of the second half of the twentieth century was characterised by a significant decrease in silica delivery, while N and P fluxes increased tremendously.

The economic transition of Eastern countries The period following the collapse of the former USSR was characterised by major changes in agricultural and industrial activity in all countries of Eastern Europe, resulting in a considerable decrease in fertiliser application and industrial wastewater discharge.

Gilles Billen 10

per capita N flux, kgN/cap/yr

In the Baltic countries (Estonia, Latvia, and Lithuania), formerly specialised in cattle farming, import of mineral fertilisers and feedstuff decreased by a factor of 15 between 1987 and 1996, and the livestock was reduced fourfold, decreasing the use of manure. Yet, this dramatic reduction of the intensity of agriculture led to only a slow and limited response in Latvian rivers’ N load, due to the inertia of the soils and aquifer compartments (Stalnacke etâ•¯al., 2003), while the response in terms of P load was more visible. Similar observations are reported by Behrendt etâ•¯al. (2005a) for the Odra River. For the Danube basin, although some disagreement exists on its amplitude, a clear decrease of N and P delivery to the Black Sea was reported in the early 1990s (by 9–23% for N, by 25–35% for P), as a result of decreased diffuse sources and point sources (Behrendt etâ•¯al., 2005b).

13.5â•‡ The estuarine filter 13.5.1â•‡ Typology of European estuaries Before reaching the sea, the flux of nutrients delivered at river outlets has to cross their estuarine zones, which are often biogeochemically very active systems. The different coastal systems of Europe, however, offer a wide range of estuarine types, differing in their filtering effect for riverine nutrients. Meybeck and Dürr (2009) have proposed a typology of estuaries, based on coastal morphology, tidal influence and freshwater discharge, distinguishing (i) fjords and fjärds (deep glacial valleys filled with marine water, but where snow melt leads to rapid transit of surface freshwater to the coastal zone), (ii) rias (drowned river valleys, dominated by seawater dynamics), (iii) macrotidal estuaries (where the tidal circulation gives rise to the development of a turbidity maximum zone), (iv) deltas (prograding wedges of sediment at the river mouth with restricted entrance of seawater) and (v) lagoons (littoral shallow brackish

6

Seine Oder Danube Scheldt

4

2

0 1900

The recent trends

2 per capita P flux, kgP/cap/yr

During the past 10–15 years, considerable efforts have been devoted to improving surface water quality in most European countries. This resulted in a spectacular decrease of point inputs of nutrients through wastewater discharge. The effect is particularly striking for phosphorus, as improved wasteÂ� water treatment was accompanied by the substitution of polyphosphate as a sequestering agent in washing powders, which resulted in a three- to four-fold reduction of the per capita rate of phosphorus release in domestic wastewater. Tertiary treatment of nitrogen in wastewater purification plants is also in progress, particularly in northern European countries. Diffuse inputs of nutrients by agriculture, however, are still at a high level, in spite of the agro-environmental measures advocated by most European Water Authorities. The inertia of soil and aquifer reservoirs, mentioned above, is here added to the conservatism of many components of the socio-economic agricultural sphere. As a result, phosphorus delivery is rapidly decreasing at the outlet of most European rivers, while nitrogen delivery is still increasing or at best levelling off (Figure 13.14).

8

1920

1940

1960

1980

2000

1960

1980

2000

Seine Oder Danube Scheldt

1

0 1900

1920

1940

Figure 13.14 Trends of N and P delivery for different European rivers during the twentieth century, normalised to the total population of the watershed. Data are a combination of observations and a model reconstruction from Billen etâ•¯al., 2005, 2007 (the Scheldt and the Seine), Behrendt etâ•¯al., 2002, 2005a,b (the Oder and the Danube). For the Seine and Scheldt, the error bars show the values for wet and dry hydrological conditions, while the other values refer to mean hydrological conditions.

ponds with permanent or temporary sea water exchange). Karstic areas are characterised by direct inputs of groundwater to the sea. Figure 13.15 shows the dominant types of estuaries along the coasts of Europe.

13.5.2â•‡ Estuarine nutrient retention Estuarine nutrient processing is highly varied and too few studies are available to make any generalising quantitative statement on the filtering effect of estuaries on riverine nutrient fluxes. Figure 13.16 summarises a number of European case studies where the retention of the land-based nitrogen loading during the transit through the estuarine zones has been evaluated. These studies highlight the effect of residence time on overall retention. Nixon etâ•¯al. (1996) proposed a relationship similar to that proposed by Kelly etâ•¯al. (1987) for lakes, relating N retention during estuarine transit to depth and residence time; this relationship fits generally well with the data assembled for European estuaries (Figure 13.16).

289

Nitrogen flows from European regional watersheds Figure 13.15 Dominant types of estuaries along Europe’s coastlines (Meybeck and Dürr, 2009).

100

13.6â•‡ Nitrogen delivery and coastal eutrophication

% N retention

80 Arcachon lagoon

60

13.6.1â•‡ Coastal eutrophication in European coastal waters

Scheldt

40

Oder

20 Seine 0 1

10

Norsmind fjord Tweed 100

Danube Pô Tyne 1000

10000

depth/residence time, m/yr Figure 13.16 Observed N retention during transit through some European estuaries, plotted versus the depth/residence time ratio. The line represents the relationship found by Nixon etâ•¯al. (1996) for a number of North Atlantic American estuaries. European case studies include deltas [the Danube (TrifuRaducu, 2002); the Po (De Wit and Bendoricchio, 2001), the Rhone (Pettine etâ•¯al., 1998; El-Habr and Golterman, 1987), Oder (Pastuszak etâ•¯al., 2005)], macrotidal estuaries (Seine (Garnier etâ•¯al., 2010), the Scheldt (Billen etâ•¯al., 1985), the Tyne and Tweed (Ahad etâ•¯al., 2006)), a fjord (Norsminde fjord (Nielsen etâ•¯al., 1995) and a lagoon Arcachon lagoon, DeWit etâ•¯al., 2005).

Those estuarine systems where substantial nitrogen processing is occurring, including nitrification and denitrification, are often characterised by N2O concentrations far above saturation, indicating that they act as a source for this greenhouse gas. This has been observed in the Tamar (Law etâ•¯al., 1992), the Humber (Barnes and Owens, 1998), the Scheldt (de Wilde and de Bie, 2000), and the Seine (Garnier etâ•¯al., 2006) estuaries, where emission rates ranged from 0.4 to 5 gN-N2O/m²/yr. In the case of the Tyne estuary where high ammonium release occurs in the rapidly flushed estuarine zone, an area of high N2O emission is observed in the adjacent coastal zone (Ahad etâ•¯al., 2006).

290

Riverine delivery considerably affects (positively or negatively) the ecological functioning of coastal marine ecosystems, as it most often represents the major source of new nutrients for primary production. Satellite determination of coastal marine algal biomass have been available since the early 2000s, based on the radiometric observation of changes of seawater colour from blue to green as the chlorophyll concentration increases. A composite image of chlorophyll distribution in European coastal zones is shown as an example in Figure 13.17. The conversion of the optical signal to in situ pigment concentration relies on the calibration of algorithms which are highly dependent on the presence of various organic and inorganic constituents of seawater and can lead to severe overestimation of actual biomass (Darecki and Stramski, 2004). Qualitatively, however, Figure 13.17 clearly shows the effect of riverine nutrient discharge on algal biomass distribution in European coastal zones. In and of itself, this enhancement of primary producer biomass would not be a problem, if it were not often accompanied by profound changes in the structure of the food webs and a decline of zooplankton grazing and commercial fish production (Vasas etâ•¯al., 2007), as well as by diverse harmful manifestations such as organic matter accumulation, toxin production, anoxia, etc. A detailed account of the problems related to coastal eutrophication is provided by Voss etâ•¯al. (2011, ChapterÂ€8, this volume).

Gilles Billen

Figure 13.17 Composite satellite image of mean chlorophyll concentration along the European coasts in 2007 (from MODIS-Aqua satellite data, source:Â€JRC, http://marine.jrc.ec.europa.eu/). Annual mean and maximum values in brackets of direct measurements at selected stations are also shown to provide an absolute reference (Lancelot etâ•¯al., 2005; Solidoro etâ•¯al., 2009; M. Voss, personal communication).

13.6.2â•‡ Comparing indicators and observation of coastal eutrophication Ignoring the role played by the estuarine filter on riverine nutrient delivery, the data gathered in Table 13.5 can be compared with the available observations of coastal eutrophication along European coastlines (Figure 13.18). On the basis of the N/P ratio of nutrient river loading calculated in Table 13.5, phosphorus presently appears as the potentially limiting nutrient in most coastal areas, except in the Aegean Sea, where phosphorus loading is still extremely high. This situation is recent and contrasts with that of the 1980s when, owing to much higher P loading, nitrogen was likely to be the limiting nutrient in most European coastal areas (see Section 13.4.3 and Figure 13.14). Note that the Gulf of Finland, as well as many other areas of the Baltic Sea, are still regarded as N-limited for most of the growth period (Graneli etâ•¯al., 1990; Tamminen and Andersen, 2007). P limitation increases towards the north in the Gulf of Bothnia (Tamminen and Andersen, 2007). Admittedly,

the N/P calculation based on riverine deliveries does not take into account the effect of the biogeochemical processes in receiving coastal waters (sedimentation, denitrification, sediment release). Thus, in the Baltic Sea these processes tend to shift the ratios from estuarine P towards N limitation in the open sea (Pitkänen and Tamminen, 1995; Tamminen and Andersen, 2007). The ICEP shows negative values (whether it is estimated on the basis of N or P) in the Arctic coastal zones as well as in the northern Baltic. Positive ICEP values are reached in the southern Baltic, as well as in the North Sea and along most Atlantic coasts. In the Mediterranean, positive ICEP values are observed in rivers flowing into the Adriatic and Tyrrhenian seas. The western Black Sea coast is also characterised by high ICEP values. The distribution of European coastal areas designated as subject to the risk of eutrophication in the discussion above fits well with the observations of eutrophication problems, although the manifestations of eutrophication might be quite different depending on the local physiographical and hydrological

291


Figure 13.18 Calculated Indicator of Coastal Eutrophication Potential (ICEP) by European coastal region, based on the data from Table 13.5. Identification of the major coastal areas where eutrophication problems are recorded.

conditions:Â€blooms of toxic algae, as in the Seine Bight (Cugier etâ•¯al., 2005) and in the Baltic (Hansson, 2008; HELCOM, 2009), massive development of mucilaginous, unpalatable, algal species in the North Sea (Lancelot etâ•¯al., 1987, 2005, 2007), the Black Sea (Cociasu etâ•¯al., 1996) and the Adriatic Sea (Marchetti, 1991), deposition of increasing amounts of organic material resulting in anoxic bottom waters as in the northern Adriatic (Justic, 1991), Danish coastal waters (Babenerd, 1990) and Baltic coastal zones (HELCOM, 2009). In Brittany and on the other Atlantic coasts, the very rapid dilution of fresh water masses due to tidal currents often prevents the development of dense planktonic blooms. Eutrophication is mainly apparent from the development of benthic macro algae close to the coast, although development of toxic dinoflagellate blooms might also be a problem during summer when the water column is stratified. The continental coastal zone of the English Channel and the North Sea, from Normandy to the Danish coast, is one of the more severely eutrophicated areas in the world, with the occurrence of heavy blooms of Phaeocystis globosa colonies every spring, responsible for the accumulation of mucus foam on the beaches (Lancelot, 1995). Marine ecological model simulations constrained by river nutrient load simulations suggest that the maximum biomass reached by Phaeocystis increased threefold from 1950 to 1990 and has now decreased by about 20% (Lancelot etâ•¯al., 2007). The Baltic Sea is a nearly enclosed brackish-water area, with seawater renewal occurring through the narrow Danish Straits and Sound areas linking the Baltic to the North Sea. Major inflows of seawater have only occurred rarely in recent decades, leaving the water in the deeper basins without a renewal of oxygen. Salinity stratification, small water volume and long residence time are the main physical reasons for the

292

sensitivity of the Baltic Sea to eutrophication (Leppäranta and Myrberg, 2009). The sea is heavily impacted by nutrient loading and anoxic conditions promoting release of inorganic phosphorus from the sediments (Pitkänen et al, 2001). The impacts of eutrophication are manifested as various symptoms such as increased nutrient concentrations and phytoplankton biomass, oxygen deficiency and elimination of benthic fauna, as well as frequent blooms of filamentous cyanobacteria (Lundberg, 2005). The northwestern Adriatic Sea, subject to the inputs of the Po River, also suffers from the development of non-siliceous algae leading to the production of mucilaginous substances. Detection of organic-walled dinoflagellate on sediment cores revealed a clear shift to eutrophication conditions from 1930 onwards, reaching a peak in the 1960–1980 period. Subsequently, eutrophication levels decreased, although dinocyst diversity suggests that the ecosystem has not completely recovered (Sangiorgi and Donders, 2004). The western coast of the Black Sea has been experiencing a severe process of degradation since the early 1960s. From a diverse ecosystem with rich ecological resources, it evolved into a low biodiversity zone where jellyfish and ctenophores replaced zooplankton–fish food chains. An almost total collapse of fisheries occurred in the late 1980s (Mee, 1992; Lancelot etâ•¯al., 2002). The considerably reduced Danube nutrient discharge over the past 15 years, following the collapse of industry and agriculture in the former Soviet countries of the Danube catchment area, however, induced a trend towards restoration of the marine ecosystem. The species diversity of macrozooÂ� benthos has increased since 1996 in front of the Danube delta (Horstmann etâ•¯al., 2003), but ctenophores and medusa still dominate the zooplankton, preventing the full regeneration of fish populations.

Gilles Billen

13.7â•‡ Conclusions Urbanisation and the spread of industrial fertilisation techniques in agriculture in most European territories have led to an unprecedented opening of the nitrogen cycle which resulted in increased inputs of reactive nitrogen to watersheds. Compared with the background pristine inputs of N through natural biological fixation and atmospheric deposition (460– 1800 ktonN/yr, based on the figures compiled by Cleveland, 1999), net anthropogenic inputs of reactive nitrogen to EU27 (21 540 ktonN/yr) are two to ten times higher. A fraction of only about 20% of these inputs ultimately reaches the outlet of the hydrographic network of large river systems, while both landscape and aquatic processes contribute to retention of the remaining 80% of anthropogenic inputs. Landscape processes include storage of nitrogen in the soil organic matter pool and in the groundwater. This is temporary storage, which simply confers a great inertia of the response of riverine delivery to changes in diffuse inputs:Â€ depending on the residence time of nitrogen in these reservoirs, the reaction to any change in land use and agricultural practices in terms of nitrogen flux at the basin outlet can be delayed by several decades. Soil and riparian zone denitrification are other processes contributing substantially to landscape retention; the elimination of nitrate by this pathway unfortunately is accompanied by harmful emissions of N2O. In-stream nitrogen retention processes are dominated by benthic denitrification both in the river bed and in small water storage structures such as ponds and shallow reservoirs. This process also leads to N2O emissions, however. The relative role played by lakes in terms of N retention within watersheds is important:Â€ two major processes involved are denitrification and sedimentation. Wastewater treatment must be considered a retention process when it involves specific processes for N elimination, most often through denitrification, accompanied, once again, by N2O emissions. In spite of these effective retention mechanisms, many of which can still be improved by suitable management, nitrogen delivery to coastal systems at the European scale, now totalling 4750 ktonN/yr, increased more than fourfold with respect to the pristine state, and approximately threefold with respect to the pre-1950 situation. At the same time, phosphorus delivery increased, but is now decreasing again close to preindustrial levels, owing to effective P abatement measures in urban wastewater purification implemented in most European countries. Silica delivery, on the other hand, is decreasing due to both reduced rock weathering and enhanced retention in watersheds, mostly linked to dam construction (Humborg etâ•¯al., 2008). The consequence of these changes is that the riverine input of nutrients to the coastal zones, which used to be a major factor contributing to the richness of these areas providing most of the fish catch, is now largely imbalanced, resulting in severe eutrophication problems. Particularly affected are the south-eastern continental coast of the North Sea, the Baltic Sea (except the Gulf of Bothnia), the coasts of Brittany, the Adriatic Sea and the western Black Sea coastal area.

Better knowledge and understanding of the processes leading to retention and elimination of reactive nitrogen once introduced within watersheds would certainly allow better management of land- and waterscapes with the objective of reducing the N fluxes transferred to the sea and to the atmosphere as reactive species. However, whatever the potential of such management measures, there will be no other choice for durably improving the situation than reducing the anthropogenic nitrogen load, through changes in agriculture, human diet and other nitrogen flows related to modern human activity.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission) and the COST Action 729. It grew from the discussions held in two workshops held in Paris in January 2007 (with the support of both NinE and LOICZ) and in Dourdan in November 2008 (with the support of NinE and COST729). It also benefited from the participation in other collaborative networks including TIMOTHY Interuniversity Attracting Pole of the Belgian Science Policy and the AWARE EC-FP7 programme. We acknowledge Amelie Danacq for establishing Table 13.2. We also acknowledge Hast Behrendt for his active participation in this workshops mentioned above; he has since passed away. We dedicate this chapter to his memory.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press:Â€ www.cambridge.org/ena and the Nitrogen in Europe website:Â€www.nine-esf.org/ena.

References Ahad, J. M. E., Ganeshram R. S., Spencer, R. G. M. etâ•¯al. (2006). Evaluating the sources and fate of anthropogenic dissolved inorganic nitrogen (DIN) in two contrasting North Sea estuaries. Science of the Total Environment, 372, 317–333. Alexander, R. B., Smith, R. A. and Swartz, G. E. (2000). Effect of stream channel size on the delivery of nitrogen to the Gulf of Mexico. Nature 403, 758–761. Alexander, R. B., Smith, R. A., Schwarz, G. E. etâ•¯al. (2001). Atmospheric nitrogen flux from the watersheds of major estuaries of the United States:Â€an application of the SPARROW watershed model. In: Nitrogen Loading in Coastal Water Bodies:Â€An Atmospheric Perspective, ed. R.Â€Valigura,R . Alexander,M . Castro, etâ•¯al. American Geophysical Union, Madison, WI, pp.Â€119–170. Alexander, R. B., Johnes, P. J., Boyer, E. W. and Smith, R. A. (2002). A comparison of models for estimating the riverine export of nitrogen from large watersheds. Biogeochemistry, 2002, 295–339. Andersson, L. and Arheimer, B. (2003). Modelling of human and climatic impact on nitrogen load in a Swedish river 1885–1994. Hydrobiologia, 497, 63–77. Arheimer, B. (1998). Riverine nitrogenÂ€– analysis and modelling under nordic Conditions. Ph.D. Thesis. Linköping University, Sweden. Arheimer, B. and Brandt, M. (1998). Modelling nitrogen transport and retention in the catchments of southern Sweden. Ambio, 27, 471–480.

293

Nitrogen flows from European regional watersheds Arnold, J. G., Srinivasan, R., Muttiah, R. S. and Williams, J. R. (1998). Large area hydrologic modeling and assessment. Part I:Â€Model development. Journal of the American Water Resources Association, 34, 73–89. Arnold, J. G., Srinivasan, R., Muttiah, R. S., Allen, P. M. and Walker,Â€C. (1999). Continental scale simulation of the hydrologic balance. Journal of the American Water Resources Association, 35, 1037–1052. Babenerd, B. (1990). Eutrophication and algal blooms. Water Pollution Research Reports, 12,Â€215–223. Barnes, J. and Owens, N. J. P. (1998). Denitrification and nitrous oxide concentrations in the Humber Estuary, UK, and adjacent coastal zones. Marine Pollution Bulletin, 37, 247–260. Behrendt, H. (2002) The Model System MONERIS. EUROCAT WD. Behrendt, H. and Opitz, D. (2000). Retention of nutrients in river systems:Â€dependence on specific runoff and hydraulic load. Hydrobiologia, 410, 111–112. Behrendt, H., Huber, P., Kornmilch, M. et al. (2002). Estimation of the nutrient inputs into river basins: experiences from German rivers, Regional Environmental Changes, 3, 107–117. Behrendt, H., Opitz, D., Korol, R. and Stronszka, M. (2005a). Changes of the nutrient loads of the Odra River during the last century: their causes and consequences. In: Proceedings of the ICID 21st European Regional Conference ‘Integrated Land and Water Management:Â€Towards Sustainable Rural Development’, Frankfurt/ Oder, Germany, April 2005. Behrendt, H., van Gils, J., Schreiber, H. and Zessner, M. (2005b). Point and diffuse nutrient emissions and loads in the transboundary Danube River Basin. Part II: Long term changes. Archives of Hydrobiology (Suppl.), 158, 221–247. Bergström, S., Brandt, M. and Gustafson, A. (1987). Simulation of runoff and nitrogen leaching from two fields in southern Sweden. Hydrological Science Journal, 32, 191–205. Billen, G . and Garnier, J. (1997). The Phison River Plume:Â€coastal eutrophication in response to changes in land use and water management in the watershed. Aquatic Microbiology and Ecology, 13, 3–17. Billen, G. and Garnier, J. (2007). River basin nutrient delivery to the coastal sea:Â€assessing its potential to sustain new production of non siliceous algae. Marine Chemistry, 106, 148–160. Billen, G., Somville, M., De Becker, E. and Servais, P. (1985). A nitrogen budget of the Scheldt hydrographical basin. Netherlands Journal of Sea Research, 19, 223–230. Billen, G., Garnier, J. and Hanset, P. (1994). Modelling phytoplankton development in whole drainage networks:Â€the RIVERSTRAHLER model applied to the Seine river system. Hydrobiologia, 289, 119–137. Billen, G., Garnier, J. and Rousseau, V. (2005). Nutrient fluxes and water quality in the drainage network of the Scheldt basin over the last 50 years. Hydrobiologia, 540, 47–67. Billen, G., Garnier, J., Némery, J. etâ•¯al. (2007). Nutrient transfers through the Seine river continuum:Â€mechanisms and long term trends. Science of the Total Environment, 375, 80–97. Billen, G., Barles, S., Garnier, J., Rouillard, J. and Benoit, P. (2009a). The food-print of Paris:Â€long term reconstruction of the nitrogen flows imported to the city from its rural hinterland. Regional Environmental Change, 9, 13–24. Billen, G., Thieu, V., Garnier, J. and Silvestre, M. (2009b). Modelling the N cascade in regional watersheds:Â€the case study of the Seine, Somme and Scheldt rivers. Agriculture, Ecosystems and Environment, 133, 234–246. Billen, G., Beusen, A., Bouwman, L. and Garnier J. (2010). Anthropogenic nitrogen autotrophy and heterotrophy of the

294

world’s watersheds:Â€past, present, and future trends. Global Biogeochemical Cycles, 24, GB0A11, doi:10.1029/2009GB003702. Boyer, E. W., Goodale, C. L., Jaworski, N. A. and Howarth, R. W. (2002). Anthropogenic nitrogen sources and relationships to riverine nitrogen export in the northeastern USA. Biogeochemistry, 57/58, 137–169. Brandt, M. (1990). Simulation of runoff and nitrogen transport from mixed basins in Sweden. Nordic Hydrology, 21, 13–34. Brandt,M. and Ejhed, H. (2002). TRK, Transport-RetentionKällfördelning, Belastning Pahavet, Swedish Enviroment Protection Agency, Report 5247, Stockholm, Sweden, 2002, (in Swedish). Brunet, R. (2002). Lignes de force de l’espace Européen. Mappemonde, 66,Â€14–19. Cellier, P., Durand, P., Hutchings, N. etâ•¯al. (2011). Nitrogen flows and fate in rural landscapes. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Cleveland, C. C., Townsend, A. R., Schimel, D. S. etâ•¯al. (1999). Global patterns of terrestrial biological nitrogen fixation in natural ecosystems. Global Biogeochemical Cycles, 13, 623–645. Cociasu, A., Dorogan, L., Humborg, C. and Popa, L. (1996). Longterm ecological changes in Romanian coastal waters of the Black Sea. Marine Pollution Bulletin, 32, 32–38. Conley, D. J. (1999). Biogeochemical nutrient cycles and nutrient management strategies. Hydrobiologia, 289, 87–96. Conley, D. J. (2002). Terrestrial ecosystems and the global biogeochemical silica cycle. Global Biogeochemical Cycles, 16, 1121, doi: 10.1029/2002GB001894. Conley, D. J., Schelske, C. L. and Stoermer, E. F. (1993). Modification of the biogeochemical cycle of silica with eutrophication. Marine Ecology Progress Series, 81, 121–128. Cugier, P., Billen, G., Guillaud, J. F., Garnier, J. and Ménesguen,Â€A. (2005). Modelling eutrophication of the Seine Bight under present, historical and future Seine river nutrient loads. Journal of Hydrology, 304, 381–396. Darecki, M. and Stramski, D. (2004). An evaluation of MODIS and SeaWiFS bio-optical algorithms in the Baltic Sea. Remote Sensing of the Environment, 89, 326–350. de Wilde, H. P. J. and de Bie, M. J. M. (2000). Nitrous oxide in the Schelde estuary:Â€production by nitrification and emission to the atmosphere. Marine Chemistry, 69, 203–216. De Wit, M. (2001). Nutrient fluxes at the river basins scale. Part I:Â€The PolFlow model. Hydrological Processes, 15, 743–759. De Wit, M. and Bendoricchio, G. (2001). Nutrient fluxes in the Po basin. Science of the Total Environment, 273, 147–161. De Wit, R., Leibreich, J., Vernier F. etâ•¯al. (2005). Relationship between land-use in the agro-forestry system of les Landes, nitrogen loading to and risk of macro-algal blooming in the Bassin d’Arcachon coastal lagoon (SW France). Estuarine, Coastal and Shelf Science, 62, 453–465. Dillon, P. J., Evans, R. D. and Molot, L. A. (1990). Retention and resuspension of phosphorus, nitrogen, and iron in a central Ontario lake. Canadian Journal of Aquatic Science, 47, 1269–1274. Dumont, E., Harrison, A., Kroeze, C., Bakker, E. J. and Seitzinger, S.Â€P. (2005). Global distribution and sources of dissolved inorganic nitrogen export to the coastal zone:Â€result from a spatially explicit, global model. Global Biogeochemical Cycles 19, GB4S02, doi:10.1029/2005 GB002488 Durand, P., Breuer, L. and Johnes, P. J. (2011). Nitrogen turnover processes and effects in aquatic ecosystems. In:Â€The European Nitrogen Assessment, ed. M., Sutton, C., Howard, J.W. Erisman, etâ•¯al. Cambridge University Press.

Gilles Billen El-Habr, H. and Golterman, H. L. (1987). Input of nutrients and suspended matter into the Golfe du Lion by the river Rhone? Chemosphere, 16, 417–418. Garnier, J., Billen, G. and Coste, M. (1995). Seasonal succession of diatoms and chlorophyecae in the drainage network of the River Seine: observations and modelling. Limnology and Oceanography, 40,Â€750–765. Garnier, J., Billen, G., Hannon, E. etâ•¯al. (2002). Modeling transfer and retention of nutrients in the drainage network of the Danube River. Estuarine and Coastal Shelf Science, 54, 285–308. Garnier, J., Cébron, A., Tallec, G. etâ•¯al. (2006). Nitrogen behaviour and nitrous oxide emission in the tidal Seine River estuary (France) as influenced by human activities in the upstream watershed. Biogeochemistry, 77, 305–326. Garnier, J., Billen, G., Némery, J. and Sebilo, M. (2010). Transformations of nutrients (N, P, Si) in the turbidity maximum zone of the Seine estuary and export to the sea. Estuarine and Coastal Shelf Science (in press). Graneli, E., Wallström, K., Larsson, U., Graneli, W. and Elmgren, R. (1990). Nutrient limitation of primary production in the Baltic Sea area. Ambio, 19, 142–151. Grizzetti, B. and Bouraoui, F. (2006). Assessment of Nitrogen and Phosphorus Environmental Pressure at European Scale. European Commission, Joint Research Centre, Luxembourg. Grizzetti, B., Bouraoui, F., de Marsily, G. and Bidoglio, G. (2005). A statistical approach to estimate nitrogen sectorial contribution to total load. Water Science and Technology, 51, 83–90. Grizzetti, B., Bouraoui F. and De Marsily, G. (2008). Assessing nitrogen pressures on European surface water, Global Biogeochemical Cycles, doi:10.1029/2007GB003085. Grizzetti, B., Bouraoui, F., Billen, G. etâ•¯al. (2011). Nitrogen as a threat to European water quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Hansson, M. (2008). Cyanobacterial blooms in the Baltic Sea. HELCOM Indicator Fact Sheets 2008. Available online:Â€http:// www.helcom.fi/environment2/ifs/ifs2008/ en_GB/ cyanobacteriaBlooms/ [Viewed 24 March 2009] HELCOM (2009). Eutrophication in the Baltic Sea: An Integrated Thematic Assessment of the Effects of Nutrient Enrichment in the Baltic Sea Region. HELCOM, Helsinki. Horstmann, U., Davidov, A., Cociasu, A. and Velikova, V. (2003). Der Einfluss verringerter Nährstofffrachten der Donau auf das Schwarze Meer. Österreichische Wasser- und Abfallwirtschaft, 55, 205–212. Howarth, R. W., Billen, G., Swaney, D. etâ•¯al. (1996). Regional nitrogen budgets and riverine N and P fluxes for the drainages to the North Atlantic ocean:Â€natural and human influences. Biogeochemistry, 35, 75–139. Howarth, R. W., Boyer, E. W., Marino, R. etâ•¯al. (2006). The influence of climate on average nitrogen export from large watersheds in the northeastern United States. Biogeochemistry, 79, 163–186. Humborg, C., Danielsson, A., Sjoberg, B. and Green, M. (2003). Nutrient land-sea fluxes in oligothrophic and pristine estuaries of the Gulf of Bothnia, Baltic Sea. Estuarine, Coastal and Shelf Science, 56, 781–793. Humborg, C., Smedberg, E., Blomqvist, S. etâ•¯al. (2004). Nutrient variations in boreal and subarctic Swedish rivers:Â€landscape control of land-sea fluxes. Limnology and Oceanography, 49, 1871–1883. Humborg, C., Pastuszak, M., Aigars, J. etâ•¯al. (2006). Decreased silica land-sea fluxes through damming in the Baltic Sea catchment: significance of particle trapping and hydrological alterations. Biogeochemistry, 77, 265–281.

Humborg, C., Smedberg, E., Rodriguez Medina, M., Mörth, C.-M. (2008). Changes in dissolved silicate loads to the Baltic SeaÂ€– The effects of lakes and reservoirs Journal Marine Systems. 73:Â€223–235 Johnes, P. J. and Butterfield, D. (2002). Landscape, regional and global estimates of nitrogen flux from land to sea:Â€errors and uncertainties. Biogeochemistry, 57/58,Â€429–476. Justic, D. (1991). Hypoxic conditions in the northern Adriatic sea: historical development and ecological significance. Geological Society of London, Special Publications, 58, 95–105 Justic, D., Rabalais, N. N., Turner, R. E. and Dortch, Q. (1995). Changes in nutrient structure of river-dominated coastal waters, stoichiometric nutrient balance and its consequences. Estuarine and Coastal Shelf Science, 40, 339–345. Kelly, C. A., Rudd, J. M, Hesslein, R. H. etâ•¯al. (1987). Prediction of biological acid neutralization in acid-sensitive lakes. Biogeochemistry, 3, 129–140. Kortelainen, P., Mattsson, T., Finer, L. etâ•¯al. (2006). Controls on the export of C, N, P and Fe from undisturbed boreal catchments, Finland. Aquatic Sciences, 68, 453–468. Kronvang, B., Borgvang, S. A. and Barkved, L. J. (2009). Towards European harmonised procedures for quantification of nutrient losses from diffuse sources:Â€the EUROHARP project. Journal of Environmental Monitoring, 11, 503–505. Lancelot, C. (1995). The mucilage phenomenon in the continental coastal waters of the North Sea. Science of the Total Environment, 165, 83–102. Lancelot, C., Billen, G., Sournia, A. etâ•¯al. (1987). Phaeocystis blooms and nutrient enrichment in the continental coastal zones of the North Sea. Ambio, 16, 38–46. Lancelot, C., Billen, G. and Barth, H. (1991). The Dynamics of Phaeocystis Blooms in Nutrient Enriched Coastal Zones, Water Pollution Research Reports Series No. 23, Environmental RD Programme , CEC, Brussels. Lancelot, C., Martin, J. M., Panin, N. and Zaitsev, Y. (2002). The North-Western Black-Sea:Â€a pilot site to understand the complex interaction between human activities and the coastal environment. Estuarine, Coastal and Shelf Science, 54, 279–283. Lancelot, C., Spitz, Y. and Gypens, N. etâ•¯al. (2005). Modelling diatomPhaeocystis blooms and nutrient cycles in the Southern Bight of the North Sea:Â€the MIRO model. Marine Ecology Progress Series, 289, 63–78. Lancelot, C., Gypens, N., Billen, G., Garnier, J. and Roubeix, V. (2007). Testing an integrated river–ocean mathematical tool for linking marine eutrophication to land use:Â€The Phaeocystis-dominated Belgian coastal zone (Southern North Sea) over the past 50 years. Journal of Marine Systems, 64, 216–228. Law, C. S. Rees, A. P. and Owens N. J. P. (1992). Nitrous oxide:Â€estuarine sources and atmospheric flux. Estuarine, Coastal and Shelf Science, 35, 301–314. Leip, A. Achermann, B., Biller, G. etâ•¯al. (2011). Integrating nitrogen fluxes at the European scale. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman et al.,Â€Cambridge University Press. Lepistö, A., Kenttämies, K. and Rekolainen, S. (2001) Modeling combined effects of forestry, agriculture and deposition on nitrogen export in a northern river basin in Finland. Ambio, 30, 338–348. Lepistö, A., Granlund, K., Kortelainen, P. and Räike, A. (2006). Nitrogen in river basins:Â€Sources, retention in the surface waters and peatlands, and fluxes to estuaries in Finland. Science of theTotal Environment, 365, 238–259. Leppäranta, M. and Myrberg, K. (2009). Physical Oceanography of the Baltic Sea. Springer, New York.

295

Nitrogen flows from European regional watersheds Ludwig, W., Dumont, E., Meybeck, M. and Heussner, S. (2009). River discharges of water and nutrients to the Mediterranean and Black Sea:Â€major drivers for ecosystem changes during past and future decades? Progress in Oceanography, 80, 199–217. Lundberg, C. (2005). Conceptualizing the Baltic Sea ecosystem an interdisciplinary tool for environmental decision making. Ambio, 34, 433–439. Marchetti, R. (1990). Algal blooms and gel production in the Adriatic sea. In: Barth, H., Fegan, L. (eds.) Eutrophication Related Phenomena in the Adriatic Sea and in other Mediterranean Coastal Zones, Water Pollution Research Reports Series No. 16, Enviromental RD Programme, CEC, Brussels, pp. 21–42. Mattsson, T., Finer, L., Kortelainen, P. and Sallantaus, T. (2003). Brook water quality and background leaching from unmanaged forested catchments in Finland. Water, Air and Soil Pollution, 147,Â€275–297. Mayorga, E., Seitzinger, S. P., Harrison, J. A. et al. (2010). Global Nutrient Export from WaterSheds 2 (NEWS 2):Â€model development and implementation. Environmental Modeling and Software, 25, 837–853. Mee, L. D. (1992). The Black Sea in crisis:Â€the need for concerted international action. Ambio, 21, 278–286. Meybeck, M., Cauwet, G., Dessery, S. etâ•¯al. (1988). Nutrients (organic C, P, N, Si) in the eutrophic River Loire (France) and its estuary. Estuarine Coastal and Shelf Science, 27, 595–624. Meybeck, M. and Dürr, H. H. (2009). Cascading filters of river material from headwaters to regional seas:Â€the European example. In SCOPE 70, Chapter 7. Murata, A., Sandric Chee Yew Leonga, Yuji Nagashima, Y. and Taguchi, S. (2006). Nitrogen: phosphorus supply ratio may control the protein and total toxin of dinoflagellate Alexandrium tamarense. Toxicon, 48, 683–689. Neal, C. and Davies, H. (2003). Water quality fluxes for eastern UK rivers entering the North Sea:Â€a summary of information from the Land Ocean Interaction Study (LOIS). Science of the Total Environment, 314/316, 821–882. Neitsch, S. L., Arnold, J. G., Kiniry, J. R. and Williams, J. R. (2001). Soil and Water Assessment Tool-use Manual Version 2000. Blackland Research Center, Agricultural Research Service, Texas, USA. Neitsch, S. L., Arnold, J. G., Kiniry, J. R. and Williams, J. R. (2005). SWAT Theoretical Documentation. www:bcr.tamus.edu/swat/ Nielsen, K., Nielsen, L. P. and Rasmussen, P. (1995). Estuarine nitrogen retention independently estimated by the denitrification and mass balance methods:Â€a study of Norsminde Fjord, Denmark. Marine Ecology Progress Series, 119, 275–283. Nixon, S. W., Ammerman, J. and Atkinson, L. (1996). The fate of nitrogen and phosphorus at the land–sea margin of the North Atlantic Ocean. Biogeochemistry, 35, 141–180. Officer, C. B. and Ryther, J. H. (1980). The possible importance of silicon in marine eutrophication. Marine Ecology Progress. Series., 3, 383–391. OSPAR (2002). Data Report on the Comprehensive Study of Riverine Inputs and Direct Discharges (RID) in 2000. OSPAR Commission, London. Pastuszak, M., Witek, Z., Nagel, K., Wielgat, M. and Grelowski, A. (2005). Role of the Oder estuary (southern Baltic) in transformation of the riverine nutrient loads. Journal of Marine Systems, 57, 30–54. Peterson, B. J., Wollheim, W. M., Mulholland, P. J. etâ•¯al. (2001). Control of nitrogen export from watersheds by headwater streams. Science, 292, 86–90. Pettine, M., Patrolecco, L., Camusso, P. and Crescenzio, S. (1998). Transport of carbon and nitrogen to the northern Adriatic Sea by the Po River. Estuarine, Coastal and Shelf Science, 46, 127–142.

296

Pitkänen, H. and Tamminen, T. (1995). Nitrogen and phosphorus as production limiting factors in the estuarine waters of the eastern Gulf of Finland. Marine Ecology Progress Series, 129, 283–294. Pitkänen, H., Lehtoranta, J. and Räike, A. (2001). Internal nutrient fluxes counteract decreases in external load:Â€the case of the estuarial eastern Gulf of Finland, Baltic Sea. Ambio, 30, 195–201. Preston, D. P. and Brakebill, J. W. (1999). Application of Spatially Referenced Regression Modeling for the Evaluation of Total Nitrogen Loading in the Chesapeake Bay Watershed, (WRI Report 99–4054. USGS Water Resources Division MD-DE-DC, Maryland). Radach, G. and Pätsch, J. (2007). Variability of continental riverine freshwater and nutrient inputs into the North Sea for the years 1977–2000 and its consequences for the assessment of eutrophication. Estuaries and Coasts, 30, 66–81. Rantakari, M. and Kortelainen, P. (2008). Controls of organic and inorganic carbon in randomly selected Boreal lakes in varied catchments. Biogeochemistry, 91, 151–162. Redfield, A. C., Ketchum, B. H., Richards, F. A. (1963). The influence of organisms on the composition of sea-water. In: The Sea, ed. M.Â€N.Â€Hill, John Wiley & Sons, New York, pp. 12–37. REGINE (2010). www.nve.no Ruelland, D., Billen, G., Brunstein, D. and Garnier, J. (2007). SENEQUE 3:Â€a GIS interface to the RIVERSTRAHLER model of the biogeochemical functioning of river systems. Science of the Total Environment, 375, 257–273. Sangiorgi, F. and Donders, T. H. (2004). Reconstructing 150 years of eutrophication in the north-western Adriatic Sea (Italy) using dinoflagellate cysts, pollen and spores. Estuarine, Coastal and Shelf Science, 60, 69–79. Schouwman, O. F., Silgram, M. P., Groenendijk, P. etâ•¯al.(2003). Description of nine nutrient loss models:Â€capabilities and suitability based on their characteristics. Journal of Environmental Monitoring, 11, 506–514. Seitzinger, S. P., Mayorga, E., Kroeze, C. etâ•¯al. (2009). Global river nutrient export trajectories 1970–2050:Â€a millennium ecosystem assessment scenario analysis. Global Biogeochemical Cycles, doi:10.1029/2009GB003587. Sferratore, A., Garnier, J., Billen, G., Conley, D. and Pinault, S. (2006). Silica diffuse and point sources in the Seine watershed. Environmental Science and Technology, 40, 6630–6635. Smith, R. A., Schwarz, G. E. and Alexander, R. B. (1997). Regional interpretation of water-quality monitoring data. Water Resources Research, 33, 2781–2798. Solidoro, C., Bastianini, M., Bandelj, V. etâ•¯al.(2009). Current state, scales of variability, and trends of biogeochemical properties in the northern Adriatic Sea. Journal of Geophysical Research, 114, doi:10.1029/2008JC004838. Stalnacke, P., Grimvall, A., Libiseller, C., Laznik, M. and Kokorite, I. (2003). Trends in nutrient concentrations in Latvian rivers and the response to the dramatic change in agriculture. Journal of Hydrology, 283, 184–205. Tamminen, T. and Andersen, T. (2007). Seasonal phytoplankton nutrient limitation patterns as revealed by bioassays over Baltic Sea gradients of salinity and eutrophication. Marine Ecology Progress Series, 340, 121–138. Thieu, V., Billen, G. and Garnier, J. (2009). Nutrient transfer in 3 contrasting N-W European watersheds:Â€the Seine, Somme and Scheldt rivers. A comparative application of the Seneque/ Riverstrahler model. Water Resources Research, 43, 1740–1748. Thieu, V., Garnier, J. and Billen, G. (2010a). Assessing the effect of nutrient mitigation measures in the watersheds of the Southern Bight of the North Sea. Science of the Total Environment (inÂ€press).

Gilles Billen Thieu, V., Billen, G., Garnier, J. and Benoît, M. (2010b). Nutrient cycling in a hypothetical scenario of generalized organic agriculture in the Seine, Somme and Scheldt watersheds. Regional Environmental Changes (in press). Trifu-Raducu, M.-C. (2002) Transfert des nutriments dans le bassin du Danube et apports à la Mer Noire:Â€modélisation et bilans. Thèse, Université P & M Curie. Paris Turner, R. E. and Rabalais, N. N. (1994). Evidence for coastal eutrophication near the Mississippi River Delta. Nature, 368, 619–621. Turner, R. E., Qureshi, N. A., Rabalais, N. N. etâ•¯al. (1998). Fluctuating silicate:nitrate ratios and coastal plankton food webs. Proceedings of the National Academy of Sciences of the USA, 95, 13048–13051. Vasas, V., Lancelot, C., Rousseau, V. and Jordan, F. (2007). Eutrophication and overfishing in temperate nearshore pelagic food webs:Â€a network perspective. Marine Ecology Progress Series, 336, 1–14. Voss, M., Baker, A. and Bange, H. W. (2011). Nitrogen processes in coastal and marine systems. In:Â€The European Nitrogen Assessment,

ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Wade, A. J., Durand, P., Beaujouan, V. etâ•¯al. (2002a). A nitrogen model for European catchments:Â€INCA, new model structure and equations. Hydrology and Earth System Sciences, 6, 559–682. Wade, A. J., Whitehead, P. G. and Butterfield, D. (2002b). The Integrated Catchments model of phosphorus dynamics (INCA-P), a new approach for multiple source assessment in heterogeneous river systems:Â€model structure and equations. Hydrology and Earth System Sciences, 6, 583–606. Whitehead, P. G., Wilson, E. J. and Butterfield, D. (1998). A semidistributed Nitrogen Model for Multiple Source Assessments in Catchments (INCA):Â€Part 1Â€– Model Structure and Process Equations. Science of the Total Environment, 210/211, 547–558. Windolf, J., Jeppesen, E., Jenseen, J. P. and Kristensen, P. (1996). Modelling of seasonal variation in nitrogen retention and in-lake concentration:Â€a four-year mass balance study in 16 shallow Danish lakes. Biogeochemistry, 33, 25–44.

297

Chapter

14

Atmospheric transport and deposition of reactive nitrogen in Europe Lead author: David Simpson Contributing authors: Wenche Aas, Jerzy Bartnicki, Haldis Berge, Albert Bleeker, Kees Cuvelier, Frank Dentener, Tony Dore, Jan Willem Erisman, Hilde Fagerli, Chris Flechard, Ole Hertel, Hans van Jaarsveld, Mike Jenkin, Martijn Schaap, Valiyaveetil Shamsudheen Semeena, Philippe Thunis, Robert Vautard and Massimo Vieno

Executive summary Nature of the problem • Observations of atmospheric reactive nitrogen (Nr) deposition are severely restricted in spatial extent and type. The chain of processes leading to atmospheric deposition emissions, atmospheric dispersion, chemical transformation and eventual loss from the atmosphere is extremely complex and therefore currently, observations can only address part of this chain.

Approaches • Modelling provides a way of estimating atmospheric transport and deposition of Nr at the European scale. A description of the different model types is provided. • Current deposition estimates from models are compared with observations from European air chemistry monitoring networks. • The main focus of the chapter is at the European scale; however, both local variability and and intercontinental Nr transfers are also addressed.

Key findings/state of knowledge • Atmospheric deposition is a major input of Nr for European terrestrial and freshwater ecosystems as well as coastal sea areas. • Models are key tools to integrate our understanding of atmospheric chemistry and transport, and are essential for estimating the spatial distribution of deposition, and to support the formulation of air pollution control strategies. • Our knowledge of the reliability of models for deposition estimates is, however, limited, since we have so few observational constraints on many key parameters. • Total Nr deposition estimates cannot be directly assessed because of a lack of measurements, especially of the Nr dry deposition component. Differences among European regional models can be significant, however, e.g. 30% in some areas, and substantially more than this for specific locations.

Major uncertainties/challenges • There are very few measurements of many of the key compounds (e.g. gaseous HNO3, coarse-nitrate, NH3), which are needed to enable comprehensive model evaluation. Data on all compounds should be available at the same site if the mass-balance of Nr is to be assessed, pointing to the need for integrated site measurements in air monitoring networks. • The main needs for oxidised Nr compounds are to evaluate how well the models capture the partitioning between gaseous HNO3 and either fine or coarse nitrate aerosol. For reduced Nr compounds, better estimates of NH3 emissions are needed, and how these are affected by meteorological factors as well as agricultural practices, coupled with an understanding of biosphere–atmosphere exchange. • Dry deposition of particles, sub-grid fluxes of NHx compounds, and effects of topography on wet deposition are especially difficult to parameterise properly.

Recommendations • There is a significant need for studies to constrain uncertain model parameters. This includes measurements of both the gas and particle phases of Nr compounds, and of atmosphere–biosphere fluxes of Nr compounds over sensitive ecosystems. • A balanced programme of observations and models is needed and is critical to future understanding of atmospheric transport and deposition of Nr containing pollutants at local to global scales. The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

298

David Simpson

14.1â•‡ Introduction This chapter attempts to answer the overriding question:Â€what are the atmospheric inputs of reactive nitrogen (Nr) in Europe, and how well can we estimate these? This issue is of particular concern for semi-natural ecosystems and sea areas, where the atmospheric supply of nitrogen can form an appreciable part of the total nitrogen load. As discussed elsewhere in this report, nitrogen measurements are of course essential for understanding the state of the atmosphere, and hence to help answer the first part of this question. However, as outlined by Hertel etÂ€al., 2011 (Chapter 9, this volume), the chain of processes linking emissions, atmospheric dispersion, chemical transformation and loss from the atmosphere of Nr compounds is extremely complex. Observations can typically address only a small portion of this chain. In particular for this chapter, observations of atmospheric deposition are severely restricted in spatial extent and type. Typically only the ‘wet’ deposition of atmospheric nitrogen can be observed, and even this issue is fraught with uncertainty when we try to measure deposition to canopies. Nitrogen in the form of both ammonium and nitrate, together with most other plant nutrients, is strongly affected by canopy exchange (mainly uptake on the surface of the foliage), which affects throughfall composition. Unlike for S-species, N-species can be retained by the forest canopy, and throughfall is not a reliable indicator of total deposition. Another emerging and difficult field is that of organic nitrogen and its contribution to especially wet deposition (Cape etÂ€ al., 2001; Gonzalez Benitez etÂ€ al., 2009, 2010). (This issue is discussed by Hertel etÂ€al., 2011 (ChapterÂ€9, this volume), but not in this chapter as the sources of much of the measured organic nitrogen are still unclear. Further, the models presented in this chapter only consider oxidised organics, such as PAN, rather than reduced compounds.) The situation for dry deposition is even worse, with no routine method of measuring dry deposition. Some flux data are available from a limited number of sites employing microÂ�meteorological methods (Fowler etÂ€al., 2001, 2009), but estimates of particle deposition rates are still very uncertain (Pryor etÂ€al., 2008a,b,c). Measurements of the dry deposition of gaseous nitrogen species usually rely on the measurement of concentrations and estimates of deposition velocity (Zhang etÂ€al., 2009). Given the lack of an observed deposition field over Europe, models are thus an essential tool for our understanding of the nitrogen cycle. A wide variety of models is available, but most aim to provide some or all of the following benefits. • To allow for spatially comprehensive estimates of pollutant concentrations, and for mapping of deposition patterns over large areas. • To integrate our understanding of atmospheric chemistry and transport. Models address emissions, dispersion and transport over multiple scales, chemical transformation, and dry and wet removal of pollutants. • To allow an exploration of the relative importance of different physical/chemical processes, in order to test hypotheses, and to focus attention on the most important mechanisms.

• To predict future pollution levels, including ‘what-if ’ scenarios in which different policy options are explored. • A comparison of model predictions against observed values is essential if scientists and policy makers are to have confidence that we understand the nitrogen cycle. Models are of course necessarily approximations to the real world, and they have to be evaluated thoroughly against Â�measurements if we are to have any confidence in their Â�abilities. For these reasons this chapter will focus mostly on models (in particular chemical transport models, CTMs) and their results, although with strong coupling to measurements. Section 14.2 will briefly present the types of models Â�typically used to assess atmospheric deposition, Section 14.3 will present deposition estimates from global to local scales. Model evaluation will be discussed in Section 14.4, and Section 14.5 will discuss the remaining uncertainty and challenges. The main focus of the chapter is the European regional scale, and deposition issues, but we will also present results covering scales from global to local (~1 km scale) in order to place the results in context.

14.2â•‡ Types of models A bewildering variety of models is available for air pollution studies, with applications ranging from near-source dispersion or process studies to global scale. For example, the European Topic Centre on Air and Climate Change model documentation system lists 123 different models, developed around the world (EIONET, 2010). The United States Environmental Protection Agency maintains a similar list (EPA, 2010). Recent reviews of different types of models and their applications can be found in Bleeker etÂ€al. (2009), Hertel etÂ€al. (2006), Holmes and Morawska (2006), Seinfeld and Pandis (1998), Sportisse (2007) and van Pul etÂ€al. (2009), for example. The number of models partly reflects the difficulties of the task at hand, with models limited by basic theoretical principles as well as by practical problems. Difficulties arise from our limited understanding of many biological, meteorological and chemical processes, the difficulty of specifying many of the important inputs for modelling nitrogen exchange (e.g. NH+4 levels in vegetation, soil water, atmospheric emissions, surface properties), and the still-real problems of computer processing power. Thus, all models are compromises in which all aspects of the problem are simplified to some extent. The goal, and art, of modelling is to capture the most important processes for the problem at hand, so that the model is useful for its purpose, and can be relied upon to a reasonable extent. In this chapter we will discuss the main types of models typically used for problems related to oxidised or reduced nitrogen in relation to air quality issues. The main focus is Europe, but we also discuss applications from local scale to global scale. This section discusses the main types of models which are typically used to calculate nitrogen inputs to ecosystems or water surfaces, namely plume, Lagrangian or Eulerian models. For a discussion of other types of models (e.g.

299

Atmospheric transport and deposition of reactive nitrogen

computational fluid dynamics models, canyon models) see Hertel etÂ€al. (2006).

Plume models Plume models are widely used in operational local-scale modelling of releases from industry and power plants. Examples of such models are the American AERMOD (Cimorelli etÂ€al., 2005), the UKADMS (Carruthers etÂ€al., 1994), the Dutch OPS (Duyzer etÂ€al., 2001; van Pul etÂ€al., 2004) and the Danish OML (Olesen etÂ€ al., 1992). An inter-comparison study of plume models has shown a reasonably good agreement for most conditions (Olesen, 1995) and this type of model is in general suited for application in local-scale air pollution regulation, or for emissions verification. An example of a plume model applied specifically for NH3 deposition modelling is OMLDEP (Hertel etÂ€ al., 2006). Further discussion of these and other models can be found in Hertel etÂ€al. (2006) and Holmes and Morawska (2006).

Lagrangian models In Lagrangian models, an air parcel is tracked along a trajectory computed from wind speed and wind direction. Lagrangian models may use just one, or many, vertical layers. Where more than one layer is used, the approximation is usually made that all layers are transported with the same velocity, e.g. for ACDEP (Hertel etÂ€ al., 1995) or FRAME (Singles etÂ€ al., 1998; Fournier etÂ€ al., 2005). Although global-scale models usually use the Eulerian framework, the UK STOCHEM model uses a Lagrangian formulation in which very many independent air parcels are followed and allowed to exchange material with each other (Collins etÂ€al., 1997). Lagrangian models are typically computationally fast since they are usually applied to a restricted number of receptor points or air parcels, and with simplified treatment of meteorology and dispersion. In some models, this allows for a more advanced treatment of other aspects, e.g. of Â�chemistryÂ€ – the UK Photochemical Trajectory Model makes use of the Master Chemical Mechanism (MCM), with MCM v3–1 treating about 13 500 reactions between 5900 species (Johnson etÂ€ al., 2006; Jenkin etÂ€al., 2003), or of detailed aerosol dynamics, e.g. UHMA (Korhonen etÂ€al., 2004). Figure 14.1 provides a relevant example of the type of detailed atmospheric processing which can be analysed with MCM.

Eulerian models In Eulerian models, calculations are performed simultaneously for a grid of cells. For each of these grid-cells, advection, turÂ� bulent exchange, chemistry, and dry and wet deposition are computed. Examples of such models are the EMEP model (Berge and Jakobsen, 1998; Simpson etÂ€al., 2003), CHIMERE (Bessagnet etÂ€al., 2004), LOTOS (Schaap etÂ€al., 2008), MATCH (Robertson etÂ€al., 1999), RADM (Chang etÂ€al., 1987), CMAQ (Binkowski and Roselle, 2003), STEM (Carmichael etÂ€al., 1991), and DEHM-REGINA (Frohn etÂ€al., 2001). Eulerian models are generally more computer-resource demanding than plume and Lagrangian models, especially when a high geographical resolution is desired. However, such

300

Figure 14.1 Example of the use of a Lagrangian box model for chemical simulation. These calculations used a pseudo-Lagrangian boundary layer box model to represent initial passage of an air mass over an urban area (three hours duration), during which time the box received enhanced emissions. Subsequently, the box received background emissions, which were based on the UK average. Discussed in more detail in Jenkin etÂ€al. (2006), these calculations demonstrated particularly important contributions from organic nitrates and PANs, a conclusion that is in broad agreement with observations.

models are generally recognised to provide the most comprehensive framework for chemical transport models (Sportisse, 2007; Seinfeld and Pandis, 1998). Unlike plume or Lagrangian models, Eulerian modelling involves calculations for the full spatial domain, and the structure allows for straightforward inclusion of complex meteorology and multiple, chemically interacting sources. Such models have been developed with simple one-way nesting (Kessler etÂ€al., 2001; Vieno etÂ€al., 2009, 2010) and more accurate and advanced models with two-way nesting also exist (Frohn etÂ€al., 2001).

Inferential models The last type of model mentioned here, inferential modelling, is very different, in that no chemical transport modelling is done at all. Where flux measurements are not available (i.e. at the vast majority of sites), an estimate of dry deposition fluxes may still be made by combining measured concentration data with relevant (micro)meteorological data and estimated deposition velocities. Inferential modelling is important for its ability to provide a deposition estimate which is heavily observation-based, and as a framework for evaluating differences between model formulations for deposition velocity. This method has been applied within e.g. the US-CASTNET network (Clarke etÂ€ al., 1997) and the EU-NitroEurope network (Sutton etÂ€al., 2007; Flechard etÂ€al., 2010). See also Table 14.1.

14.3â•‡ Atmospheric deposition of reactive nitrogen This section gives an overview of modelling results concernÂ�ing the deposition of reactive nitrogen to land and sea areas. Scales ranging from global to local are covered, but most emphasis is given to the European scale. Unfortunately, observations cannot provide maps of total Nr deposition, as typically only wet deposition can be measured, and then at spatially heterogeneous

David Simpson Table 14.1 Advantages and disadvantages of common types of chemical transport models (adapted from Hertel etÂ€al., 2006)

Advantages

Disadvantages

Scale T

Highly simplified formulation. Difficulties with complex meteorology, chemistry. Cannot account for interactions between sources.

0–20 km

Short-falls in the description of transport and dispersion. The uncertainty increases with distance along the trajectory. Forward trajectory models can only handle simplified chemistry. Computationally demanding for a large number of receptor points.

1–500 km

Generally computationally demandingÂ€– especially for three dimensional models with high resolution, e.g. including nesting techniques. Difficulties in handling plumes.

10 km–global

Plume Fast, analytic solutions, easy to apply.

Lagrangian Fast for carrying out multiple model runs that concern a limited number of receptor points. Generally easy to apply for most purposes. Eulerian Allows comprehensive description of combined transport, dispersion and chemical modelling. Enables high-resolution three-dimensional simulations, treatment of complex terrain, and one or two-way nesting.

Notes:Â€There are examples of all model types at essentially all scales, but we give here the main domain of application of the different types.

networks, so this section focuses on model results. However, Section 14.4 will present further data on observed wet deposition in the context of model evaluation studies, and discuss some of the uncertainties surrounding these estimates.

14.3.1â•‡ Atmospheric deposition:Â€global scale Global emissions of NO, NH3 and SO2 may have increased by more than a factor of three since the pre-industrial era. Regionally, these increases have been even more substantial, and emissions from large portions of North America, Europe and Asia increased by more than a factor of ten during the past century (van Aardenne etÂ€al., 2001). Recent studies (Galloway etÂ€al., 2004) indicate substantial further increases of emissions and deposition toward 2050. Other scenario studies suggest that increasing air pollution control will stabilise or reduce emissions by 2030 (Cofala etÂ€ al., 2007). The need to understand and predict such changes has led to a flurry of activity on global-scale modelling in recent years, further promoted by the establishment of the UNECE Task Force on Hemispheric Transport of Air Pollution (HTAP, 2010). An extensive recent study of global N-deposition is that of Dentener etÂ€al. (2006). This study focused on global and regional deposition fluxes of both oxidised and reduced nitrogen compounds for the present day and near future (2030), using an ensemble of 23 models. This study showed reasonable agreement with observations in Europe and North America, where 60%–70% of the model-calculated wet deposition rates agree to within ±50% with quality-controlled measurements (Dentener etÂ€al., 2006). The same models systematically overestimate NHx deposition in South Asia, and underestimate NOy deposition in East Asia (Figure 14.2). These questions were addressed in several multi-model studies of nitrogen deposition. Recently the UNECE Task Force on Hemispheric Transport of Air Pollution evaluated the hemispheric transport of ozone, aerosol and precursors between

four world regions. As discussed in Sanderson etÂ€al. (2008) (see also Erisman etÂ€al., 2011, Chapter 2, Figure 2.10, this volume), Europe substantially impacts parts of Asia and North America, and, vice versa, Europe is mostly influenced by emissions from North America. A few percent of NOy emissions from North America reach Europe. The TF HTAP interim report states that on average 75% of the NOx emissions in Europe are deposited within Europe, with small fractions falling on North AmericaÂ€ (1%), South Asia (2%), East Asia (2.5%), and the remainder deposited in the oceans, and Russia.

14.3.2â•‡ Atmospheric deposition over Europe In this section we focus on modelling results for the so-called regional scale models, those which are designed to run over large areas of Europe, with grid sizes of typically 30–50 km. One important model in this context is the EMEP model (Berge and Jakobsen, 1998; Simpson etÂ€al., 2003, 2006a), as it is widely employed within the European air pollution abatement strategy and legislation work (Sliggers and Kakebeeke, 2004). This model is typically run with a 50 × 50 km2 grid size, although first results for 10 × 10 km2 are now available (Fagerli etÂ€al., 2008). Other models typically applied at this scale include CHIMERE (Bessagnet etÂ€ al., 2004), LOTOS (Schaap etÂ€ al., 2008), MATCH (Robertson etÂ€al., 1999), and DEHM-REGINA (Frohn etÂ€ al., 2001). Intercomparison of some or all of these models was presented in van Loon etÂ€al. (2004, 2007), Vautard etÂ€al. (2007) and Stern etÂ€al. (2008). We will begin by illustrating the results from a so-called ensemble of chemistry transport models, which includes most of those mentioned above. Using an ensemble of models rather than a single model to simulate air quality for assessment or emission scenario evaluation purposes provides two new pieces of information. Firstly, the average (or the median) over this ensemble is a new result by itself, which is expected to have a smaller error because individual model errors cancel each

301


Figure 14.2 Annual wet deposition of (A) NO –3 (HNO3 and aerosol nitrate), (B) NHx (NH3 and aerosol ammonium), for current-year (~Â€yearÂ€2000) simulation along with measurements grouped in 5° latitude and 10° longitude. The numbers within the circles indicate the number of stations in this latitude/longitude band. Units mg(N) m−2 (100Â€mg(N)Â€m−2Â€= 1 kg(N) ha−2). From Dentener etÂ€al. (2006).

other to a certain extent. Secondly, the spread of the ensemble can be a measure of the uncertainty in model simulations. In the EURODELTA study (van Loon etÂ€al., 2007; Schaap etÂ€al., 2010; Vautard etÂ€al., 2008) seven modelling teams simulated the air quality over the European domain for the full year of 2001 using a harmonised emission database. Figure 14.3 Â�illustrates the total Nitrogen deposition obtained from the ensemble-mean, along with the standard deviation of these results. Firstly, these results demonstrate the strong spatial variation in nitrogen deposition, with clear maxima over the Benelux area and Po Valley region of Northern Italy. The standard deviation of model results is large, however, e.g. representing about 30% of the mean value over the Netherlands. Maps of just the wet-deposition component of one model, EMEP, will be presented and compared with observations in Section 14.4.3.

Deposition to ecosystems Figure 14.3 presented total N deposition to model grids, but for assessing the vulnerability of ecosystems to deposition one

302

needs to know the deposition rates to each type of land-cover within the grids. Importantly, deposition loads to forests are typically greater than to other ecosystems, enhanced by their greater aerodynamic roughness, and their ability to capture fine-particles (Ruijgrok etÂ€al., 1997; Pryor etÂ€al., 2008b). Here we present some examples calculated with the EMEP model, as this model utilises a so-called mosaic approach, in which deposition rates are calculated for up to 18 different landcover types per grid (Simpson etÂ€al., 2001, 2003). Calculated deposition to two important ecosystems are illustrated in Figure 14.4, with deposition given per unit area of ecosystem. This figure clearly illustrates the large gradients of N-deposition across Europe, and that areas in north-west Europe receive the highest loadings of N-deposition. Deposition to forests is significantly higher than to semi-Â�natural areas. Similar calculations for croplands show even lower deposition rates than to semi-natural, partly due to the fact that crop lands are only vegetated for part of the year. In order to further illustrate the sources of this nitrogen deposition, Figure 14.5 shows the calculated relative

David Simpson

Figure 14.3 Calculated nitrogen deposition from an ensemble of seven models for 2001, together with the standard deviation (right) of the model estimates. Units:Â€mg(N)m−2 (100mg(N)m−2 = 1â•›kg(N)ha−2) (EURODELTA study, see text). (a)

Sum CF

(b)

100

Sum CF

100 3000

3000 2000

80

2000

80

1500

1500 1100

60

1100

60

900

900 700

40

700

40

500

500 300

20

300

20

100

100

40

60

80

100

120

40

60

80

100

120

Figure 14.4 Calculated N-deposition densities (mg(N) per m2 of ecosystem per year) to different ecosystems (year 2000):Â€(a) coniferous forest, (b) semi-natural. Source:Â€EMEP MSC-W.

contributions of dry and wet deposition, for oxidised and reduced nitrogen, to the total reactive nitrogen deposition to forests. In the Nordic countries, dry and wet deposition of oxidised nitrogen dominate, although wet deposition of reduced nitrogen accounts for around 20%–30% of the total in forest ecosystems, and somewhat more for non-forest ecosystems. Dry deposition of reduced-N is, however, the most significant contributor in many areas of central Europe, including parts of France, UK, Ireland and the Netherlands. Over southern Europe dry deposition tends to dominate over wet, as should be expected given the lower precipitation rates.

Deposition to European seas The atmospheric input of Nr to sea is significant. It has been estimated that approximately one quarter of the total nitrogen

input to the Baltic Sea comes from airborne nitrogen deposited directly into the sea (HELCOM, 2005) and around 30% for the North Sea (Rendell et al., 1993). A number of modelling studies have examined deposition to sea areas, with the Baltic and North Sea receiving most attention (Bartnicki and Fagerli, 2008; Hertel etÂ€al., 2002, 2003; Langner etÂ€ al., 2009; de Leeuw etÂ€ al., 2001, 2003; Schlunzen and Meyer, 2007). For example, Hertel etÂ€al. (2002) estimated around 40% of the nitrogen deposition over the North Sea to originate from agriculture activities and around 60% from emissions from combustion sources. As seen in Figure 14.6, wet deposition dominates over the dry deposition of nitrogen for three of the four sea areas. The dominance of wet deposition was also found by de Leeuw etÂ€al. (2003) and Hertel etÂ€al. (2002) for the North Sea (more than 80%). This dominance is expected

303

Atmospheric transport and deposition of reactive nitrogen (a)

(b)

100

50

30

80

100

50

30

80

20 60

15

10

40

20 60

15

10

40

5 20

5 20

80

100

120

140

80

(c)

100

120

140

(d)

100

50 30

80

100

50 30

80

20

20 60

15 10

40

60 15 10

40

5 20

5 20

80

100

120

140

80

100

120

140

Figure 14.5 Calculated percentage contributions to total nitrogen deposition over coniferous forest:Â€(a) dry deposition of oxidised N; (b) dry deposition of reduced N; (c) wet deposition of oxidised N; (d) wet deposition of reduced N. Calculations for the year 2000, from Simpson etÂ€al. (2006a).

since compounds such as NO2 and PAN have low deposition rates to water surfaces. Further, the sea surface is usually aerodynamically smooth compared to land (especially forest), and so dry deposition of even soluble compounds is relatively less important over sea than land. In general, nitrogen deposition originating from emissions on land have a strong gradient towards the sea. Ammonia is efficiently dry deposited close to the source areas and most of the reduced nitrogen that reaches the open sea comes in the form of ammonium particles which are efficiently wet deposited. NOx deposition has a somewhat weaker gradient, reflecting a longer residence time in the atmosphere (NO and NO2 do not deposit efficiently, but are transformed to HNO3 which is efficiently dry deposited or forms nitrate aerosols.) Furthermore,

304

slower deposition processes of aerosols over water surfaces are assumed in the model. The Mediterranean Sea is also different with respect to the share of agriculture related nitrogen deposition and deposition originating from emissions from combustion sources. Whilst the other seas have similar contributions of oxidised and reduced nitrogen deposition, the share of oxidised nitrogen deposition is more than 70% for the Mediterranean Sea, owing largely to the large contribution from ship traffic emissions. In fact, for three of the four European seas discussed here, ship traffic emissions are among the most important contributor to oxidised nitrogen deposition to the sea area (Table 14.2). The exception is the Black Sea where emissions from Russia,

David Simpson (a)

Table 14.2 Contribution from international ship traffic emissions to oxidised nitrogen deposition in European seas (%). From Bartnicki and Fagerli (2008)

Receptor

Ship contribution

Baltic Sea

22%

North Sea

17%

Mediterrenean Sea

34%

Black Sea

(b)

7%

Turkey and Ukraine contribute around 20% each, with contribution from international ship traffic of around 7%. The main contributors to reduced nitrogen deposition are in general countries along the coast lines.

14.3.3â•‡ The local scale and scaling issues

(c)

(d)

Figure 14.6 Time series of annual atmospheric load (Gg N/yr) of nitrogen to the European seas in the period 1995–2005. Oxidised and reduced dry and wet deposition and total nitrogen deposition are shown. From Bartnicki and Fagerli (2008).

As noted above, the grid resolution of models varies from typically less than 1 km in the most detailed local modelling to around 1° (c. 100 km) or larger for global scale models. Grid resolution affects not only the detail of model outputs, but has profound effects on the treatment of non-linear processes. Affected processes include for example the rate of oxidation of NOx in plumes, and subsequent partitioning of NOy into either rapidly depositing HNO3 or longer-lived aerosol nitrate particles, or the bi-directional exchange of both oxidised and reduced nitrogen, where a mosaic of regions with high and low concentrations may well have a different net exchange to that found in a calculation where all concentrations are smeared out over a grid square. Many of the scaling problems associated with especially NH3 modelling have been addressed in a series of recent reviews and so are not covered in detail hereÂ€ – the reader is referred to Bleeker etÂ€al. (2009), Hertel etÂ€al. (2006), Holmes and Morawska (2006), Loubet etÂ€al. (2009) and van Pul etÂ€al. (2009). Here we will concentrate on the comparability of regional and local scale modelsÂ€– and on the issues associated with bridging these scales. It was noted in Loubet etÂ€al. (2009) (and refs cited therein) that the combination of hot-spot sources and effective deposition processes lead to sources and sinks of NHx being spatially heterogeneous at a scale of a square kilometer or less. Direct measurement of NHx deposition near hot spots is challenging due to intense local advection, and indirect estimates using mass balance, 15N labelling, SF6 to NH3 ratio methods, as well as modelling studies, have estimated that the fraction recaptured within 2 km downwind from the source of NH3 emitted ranges between 2% and 60% (Asman, 1998; Loubet and Cellier, 2001; Sommer and Jensen, 1991; Theobald etÂ€al., 2001; Loubet etÂ€al., 2006). As another example, field studies in the Netherlands (Asman etÂ€ al., 1988) and the UK (Fowler etÂ€ al., 1998) show that individual sources lead to a large downwind gradient in concentration and deposition. Figure 14.7 shows an example of a farm scale emission and deposition gradient from 28 to 2â•›g m−3 and 40 to 5 kgNâ•›ha−1â•›yr−1, respectively, within a distance of about 300 m.

305


4800 kg NH3-N y –1

Distance in m

Poultry Unit

15 50 42 16 28.9 9.5

76 12

126 8

270

5

Deposition kg ha–1

6.5 3.9 1.6

Concentration NH3 µgm–3 Sum of deposition within 270m of farm woodland is: 155 kg N y –1 (3.2% of emissions)

Figure 14.7 Farm scale NH3 emission and deposition, illustrating the rapid fall-off in deposition levels with distance from source (adapted from Fowler etÂ€al., 1998).

Figure 14.8 Annual mean cumulative deposition of NHx species as a function of downwind distance, calculated with the OPS model.

Figure 14.8 illustrates the cumulative deposition of ammonia and ammonium plotted against the distance downwind of a source (e.g. an animal house) as calculated by the OPS model (van Jaarsveld, 2004). Owing to the high dry deposition velocity of ammonia and the relatively low release height the loss

of material is substantial in the first kilometres. Almost 20% is already deposited after 1 km transport and 50% after 50 km. Indeed, an important aspect of ammonia is that local deposition is almost fully determined by dry deposition of NH3. After approx. 50 km wet deposition of NH+4 becomes the dominant deposition form. Modelling the transport and deposition of ammonia, therefore, requires relatively high resolutions, both in the horizontal and vertical dimension. As of today, no single model is capable of reproducing a sufficiently wide range of length and time scales. Practical solutions include the (dynamical) nesting of small scale models into large-scale models or the use of output from large scale models to provide the boundary conditions for small scale models. The development of Eulerian models with flexible Â�resolution allows a systematic assessment of the effects of scale on model predictions. For example, Figure 14.9 illustrates the effect of increasing resolution on modelled deposition of reduced nitrogen using the same model (EMEP) at both 50 km resolution and 5 km resolution (see Vieno etÂ€ al., 2009). The increased resolution affects both the detail of the simulation, but also the Â�location of the deposition. Deposition over hillsides more closely reflects the patterns of precipitation in the United Kingdom, and thus becomes more comparable to the results obtained by the UK CBED methodology (Smith etÂ€al., 2000). This improvement partly reflects improved modelling of dispersion, but also partly improved meteorological modelling.

14.4â•‡ Comparison with observations Although models are essential for mapping deposition, their trustworthiness can only be assessed by comparison with measurements. Unfortunately, as noted in Section 14.1, observations are lacking for many important aspects of the deposition process, so comprehensive evaluations are impossible. However, even routine measurements of parameters such as air concentrations, or concentrations in deposition, give valuable information.

Figure 14.9 Calculated dry deposition of reduced nitrogen for UK, calculated from EMEP model runs at two resolutions:Â€50 km (left) and 5 km (right). Units mg(N) m−2 (100 mg(N) m−2 = lkg(N) ha−1). Source:Â€EMEP4UK model (Vieno etÂ€al., 2009a,b).

306

David Simpson

It should also be noted that global and regional scale models cannot be expected to reproduce small-scale variations in deposition regimes, caused by such factors as local emissions (especially important for NH3 close to agricultural sources, see Sutton etÂ€al., 1998), topography (which has strong effects on rainfall amount and deposition, see Dore etÂ€ al., 1992; Fowler etÂ€al., 1988; Hertel etÂ€al., 2011, Chapter 9, this Â�volume), or where processes not included in the model (e.g. occult deposition) are important. These problems are difficult to address, but by comparison with measurements we can make an assessment of the degree of agreement between the model and observed values. Here we focus on the evaluation of European-scale deposition estimates, but start with a brief introduction to the EMEP measurement network (Section 14.4.1) and an evaluation of the air concentrations (Section 14.4.2). Reliable modelling of gas and aerosol air concentrations is a necessary (but not sufficient) prerequisite for reliable modelling of atmospheric inputs to ecosystems and seas.

(a)

(b)

NO3

TNH4

14.4.1â•‡ The EMEP network The main measurement network providing European-scale data on reactive nitrogen concentrations and deposition is the EMEP network (EMEP, 2010). As discussed in detail in Fagerli and Aas (2008), 24 EMEP sites have reported nitrate and ammonium in precipitation from around 1980, with a good coverage of North and Central Europe and partly Eastern Europe. The measurements of these compounds in air did not start until the end of the 1980s. Nineteen sites reported long-term data series, but the majority of these sites were located in Nordic countries. In general, few long-term measurements are available from the south-east of Europe. Nearly all of the air measurements conducted within the EMEP network are made using the filter pack method. It is well known that this method is biased for separate gas and particulate nitrogen compounds (EMEP, 1996). Ammonium nitrate on the aerosol filter may dissociate into gaseous nitric acid and ammonia that will be captured by the impregnated filters in the filter pack sampler. This causes a negative interference on the particle filter and positive interferences on the impregnated filters. The opposite may happen if ammonia or nitric acid is captured on the front aerosol filter. An artifact free separation of these gases and particles can be achieved using denuders, but only two EMEP sites had used this method at the time of the Fagerli and Aas (2008) study. EMEP has an extensive quality control of the data that are included in the database. Laboratories that fail badly in fieldand lab inter-comparisons (Aas and Hjellbrekke, 2005) are flagged. The data sets are graded according to their quality, and for model evaluation it is clearly best to use the data with the best quality (Fagerli and Aas, 2008).

14.4.2â•‡ Air concentrations In this section we illustrate the performance of chemistry transport models for reactive nitrogen using the ensemble of models introduced in Section 14.3.2. Figure 14.10 compares the

Figure 14.10 Modelled and measured seasonal variation of particulate nitrate and ammonium for 2001. The data represent the average monthly mean values for Ns stations over Europe. Number of stations (Ns) is indicated in Table 14.3. Units:Â€μg m−3. From Schaap etÂ€al. (2010).

mean seasonal variation of the ensemble mean model and its members to the observed variation for particulate nitrate and ammonium from the EMEP network. The spread of the models is a measure of gaps and uncertainty in our knowledge. For example, Figure 14.10 indicates a higher uncertainty for nitrate than for ammonium. These EURODELTA results indicate that in general, the models are able to capture the seasonal variation of the single components, but with significant uncertainty. Table 14.3 shows the relative root mean square error (RRMSE) and mean correlation of concentrations found from this comparison. RRMSE values are higher for the nitrogen components than for sulphate, with largest values seen for total nitrate (TNO3). This latter finding is partly an artefact though, related to the number and distribution of the measurement sites over Europe for different compounds. Nitrate (only six sites here) is measured in north-western and central Europe, at sites characterised by flat-terrain, continental meteorology and high pollution levels. By contrast, total nitrate (21 sites here) is mostly measured in less polluted areas near the sea and/or in areas with complex terrain. Low pollution levels and complex terrain are generally associated with lower model skills. All models show this characteristic for all species. Indeed, in north western and central Europe RMSE values for TNO3 tend to be lower than for NO3, reflecting the sensitivity of the nitrate partitioning to ambient conditions and precursor gas concentrations.

307

Atmospheric transport and deposition of reactive nitrogen Table 14.3 Comparison of modelled and observed inorganic species for seven CTMs and their ensemble mean. Tables gives RRMSE value (%) and (in parentheses) temporal correlation coefficients between daily modelled and measured values, averaged over a number of stations (Ns) in Europe

Model

SO4

SO2

NO3

TNO3

NH4

NHx

Ns

36

27

6

21

â•‡ 8

19

EMEPv3.1

54 (0.61)

101 (0.52)

69 (0.59)

LOTOS-EUROS

51 (0.54)

85 (0.51)

62 (0.55)

106 (0.50)

50 (0.61)

62 (0.49)

87 (0.44)

52 (0.53)

56 (0.35)

MATCH

62 (0.62)

99 (0.57)

52 (0.56)

88 (0.51)

44 (0.61)

50 (0.57)

CHIMERE

59 (0.45)

139 (0.52)

62 (0.53)

84 (0.37)

46 (0.54)

62 (0.41)

RCG DEHM

57 (0.55)

107 (0.43)

53 (0.62)

75 (0.43)

48 (0.55)

74 (0.38)

68 (0.55)

83 (0.51)

71 (0.38)

160 (0.43)

49 (0.54)

54 (0.49)

TM5

59 (0.50)

n.a

97 (0.59)

143 (0.50)

62 (0.57)

67 (0.43)

Ensemble

44 (0.68)

91 (0.58)

46 (0.66)

â•‡ 92 (0.56)

40 (0.66)

50 (0.54)

Notes:Â€RRMSE is the relative root-mean square error, i.e. RMSE divided by the observations and in %; Ns is the number of stations; SO4 is particulate sulphate; NO3 and NH4 are particulate NO–3 and NH+4â†œ; TNO3 is the sum of HNO3+NO–3; NHx is the sum of NH3+ particulate NH+4.

The lower RRMSE values for NH3 and NHx compared to the oxidised compounds reflect the fact that the majority of the ammonium is bound to sulphate. The skill of the ensemble average is generally higher than the skill of the individual models. The better skill of the ensemble average or median has been shown earlier in several recent studies for air quality (Delle Monache and Stull, 2003; Pagowski etÂ€al., 2005; McKeen etÂ€al., 2007; van Loon etÂ€al., 2007; Vautard etÂ€al., 2008; Schaap etÂ€al., 2010) as well as for transport of passive tracers (Galmarini etÂ€al., 2004; Riccio etÂ€al., 2007). In climate assessments (IPCC, 2007) for example, model ensembles has become essential to evaluate the state of the knowledge of the scientific community and the spread of its uncertainty.

14.4.3â•‡ Wet deposition Comparison of model results for wet deposition or concentrations in precipitation is in many ways trickier than comparing gas concentrations. Hertel etÂ€al., 2011 (Chapter 9, this volume) discusses the physical/chemical processes controlling wet-deposition of Nr compounds, and the important role that topography can play in enhancing deposition rates. As noted in van Loon etÂ€al. (2004), the most important issue concerning the wet removal of species in CTM models is probably the meteorological input; model performance for wet deposition fluxes or concentrations in precipitation is strongly limited by the quality of the NWP models providing meteorological data. For example, models generally have problems with sub-grid precipitation, simulating precipitation more often, but in lower amounts, than reality. As precipitation scavenging is a complex and non-linear process (Barrie, 1992), such issues will cause errors in modelled wet deposition that are difficult to evaluate. There are also many uncertainties inherent in the deposition monitoring methods themselves (Draaijers and Erisman, 1993; Erisman etÂ€al., 2005). The precipitation amount may vary quite a lot over short distances, especially in mountain areas, and the sites are not always representative for the average gridded precipitation amount. For the EMEP network, the agreement between

308

precipitation measured at EMEP sites and the EMEP model is within 30% at almost all sites. Some of this discrepancy is of course due to uncertainties with the NWP model, but some is also due to precipitation sampling problems. In an early intercomparison of six different CTMs used in Europe, van Loon etÂ€al. (2004) found very poor model performance for the wet deposited components, despite fair to good performance for airborne components. A clear result of this study was that no model achieved good correlation coefficients (the best was just r = 0.35) for wet components, and bias and RMSE values could be very substantial (up to 60%–70% for wet deposition fluxes) relative to observed values. These results were much worse than equivalent results for concentrations in air. The models used in this study have been improved to some extent since this intercomparison, but it seems likely that a study using today’s models would still show discrepancies of up to 50%. The relatively poor agreement between modelled and observed wet deposition fluxes is not a specific feature of this inter-comparison or these models. Large differences between models were also found in the global models participating in the COSAM study, in which the wet deposition efficiency ranged over a factor of 4 (Roelofs etÂ€al., 2001). A similar spread was also found for global models by Dentener etÂ€al. (2006) and Textor etÂ€al. (2006). The EMEP model seems to have been subject to most Â�evaluation against observed wet deposition estimates. Standard scatter plots showing the performance of the model against observed concentrations of NO–3 and NH+4 can be found in the yearly EMEP status reports; see Fagerli and Hjellbrekke (2008) and Berge and Hjellbrekke (2010). The model has also been compared to observed wet deposition for nitrogen from the ICP-forest network (Simpson etÂ€al., 2006b). Differences in mean values between modelled and observed (ICP-forest) SO2â•›–4, NO–3 and NH+4 total and wet deposition were within 20% in 1997 and 30% in 2000, with the EMEP model showing slightly lower values than the observations (Simpson etÂ€al., 2006b). Modelled and observed concentrations of SO2â•›–4, NO–3 and NH+4 in precipitation were very similar on average (differences of 0%–14%),

David Simpson Figure 14.11 Comparison of modelled and observed annual wet deposition of (a) NHx and (b) NO –3 (HNO3 and aerosol nitrate). Data are for 2001 in the EMEP model with observations. The bullets depict observations with the same colour bar as the modelled field. Measured annual deposition is calculated by using the measured precipitation amount and the nitrate and ammonium concentration in precipitation.

(a) 100 90 80 70 60

1100 1000

50

900 800 700

40

600

30

500

20

400

10

200

300 100 20

30

40

50

60

70

80

90

100

110

120

130

(b) 100 90 80 70 60

1100 1000

50

900 800 700

40

600

30

500

20

400

10

200

300 100 20

30

40

50

60

70

80

90

100

110

120

and the correlation between modelled and observed data is rather high for this type of comparison (between r2 = 0.4–0.8 for most components and years). Figure 14.11 compares measured wet deposition of oxidised and reduced nitrogen against results from the EMEP model. In these plots the measured deposition is calculated using the measured precipitation amount and the nitrate and ammonium concentration in precipitation. For reduced nitrogen, Figure 14.11 a reveals good agreement between modelled and

130

measured values, across almost all of Europe. The high modelled values near northern Italy are reflected in the measurements. Unfortunately, other regions with high predicted wet deposition have only a limited number of measurement sites (e.g. Netherlands, Belgium), and so it is difficult to evaluate model performance here. The EMEP model has a tendency to under-predict wet deposition in Nordic sites. For oxidised nitrogen (Figure 14.11b), five sites stand out with much higher measured wet deposition than modelled.

309

Atmospheric transport and deposition of reactive nitrogen Table 14.4 Comparison of observed and modelled (EMEP) contributions (%) of dry and wet deposition of oxidised (OXN) and reduced nitrogen (RDN) to total N-deposition (OXN+RDN) at Speulderbos forest, Netherlands, 1995. From Simpson etÂ€al. (2006a)

Observed Dry+Wet

Modelled Dry+Wet

Dry

Wet

Dry+Wet

Dry

Wet

OXN

18

11

29

22

â•‡ 9

31

RDN

47

24

71

54

15

69

The reason for this seems to be that the observed precipitation at the sites far exceeds the modelled precipitation (e.g. by a factor of two for the Norwegian site). However, there is a very good agreement between model results and measurements at almost all other sites, which gives some confidence that the modelled budget of wet-deposition is within the uncertainty of the measured value.

14.4.4â•‡ Dry deposition Although wet deposition represents an important fraction of N-deposition over Europe, dry deposition is also important. Hertel etÂ€al. (2011), (Chapter 9 this volume) discusses the physical/chemical processes controlling dry-deposition of Nr compounds. Unfortunately, the EMEP network has no specific measurements of dry deposition, so we cannot present maps of modelled versus observed dry deposition. However, dry deposition monitoring has been performed over many years at Speulderbos forest in the Netherlands (Erisman etÂ€ al., 1997, 2001), the site with by far the highest deposition loads within the EU NOFRETETE project (Pilegaard etÂ€ al., 2005). Erisman etÂ€ al. (2001) presented estimates of wet and dry deposition of oxidised and reduced nitrogen for Speulderbos, over the period 1995–1998, and Simpson etÂ€al. (2006a) compared EMEP model estimates against these. The modelled total deposition for the Speulderbos grid square in 1995 was 5200 mg(N) m−2, within 10% of that found in the measurements (4798 mg(N)/m−2). Table 14.4 illustrates the percentage breakdowns of this total deposition between wet/ dry/OXN/RDN components. These relative contributions are remarkably similar, with reduced nitrogen accounting for about 2/3 of total deposition, and dry deposition dominating both the oxidised and reduced-N contributions. Ongoing studies within NitroEurope (Sutton etÂ€al., 2007) suggest problems with dry-deposition estimates, however. As part of the EU Nitro-Europe project, inferential modelling is being conducted with deposition codes from three European dry deposition models at selected sites of the NitroEurope (NEU) inferential network (Flechard etÂ€ al., 2010). The Â�deposition modules are from the UK-CBED model (Smith etÂ€ al., 2000), the Dutch IDEM model (Bleeker etÂ€ al., 2004) and the EMEP scheme (Simpson etÂ€ al., 2001, 2003). This study has suggested that NH3 is the single highest atmospheric Nr dry input in many parts of Europe. At sub-urban sites of the NEU network, HNO3 and particulate NO3− and NH+ could also contribute significant fractions of total dry deposition. There were, however, substantial discrepancies between models, with annual deposition rates varying as

310

Dry+Wet

much as two-fold between models at given monitoring sites. This highlights the variability in model parameterisations, stemming from the variability in measured deposition rates and canopy resistances. For NH3, the stomatal compensation point and the external leaf-surface (or non-stomatal) resistance are the largest sources of divergence between models. The effective annual mean deposition velocity (Vd) predicted by the CBED model is negative for the cropland and grassland sites, as a result of a non-zero compensation point for these land-use classes, but otherwise the lowest Vd for NH3 is always that predicted by the EMEP scheme. The discrepancies can be ascribed to different parameterisations for the non-stomatal resistances. Model estimates of aerosol Vd differ greatly among the Â�various modelling approaches and parameterisations (see Ruijgrok etÂ€ al., 1997, for a review), but it is in the size range 0.1–1.0 μm that the variability and uncertainty are Â�greatest. Whereas mechanistic models predict very low deposition velocities for fine aerosols, typically of the order of 0.1â•›mmâ•›s−1, field measurements suggest that Vd is 1–3 orders of magnitude higher (Gallagher etÂ€ al., 2002; Zhang etÂ€ al., 2001). Still, such field measurements are also subject to great uncertainty (Pryor etÂ€al., 2008b,c; Rannik etÂ€al., 2003). This is especially relevant for reactive nitrogen in the aerosol phase, as NH+4 and NO−3 are mostly (>90%) present as sub-micron particles.

14.4.5â•‡ Evaluation of emissions Emissions are the most important input to all CTM models, essential to both good model performance against observations and to the reliability of any emission control scenarios. Sources of reactive nitrogen to the atmosphere have been discussed for instance in Hertel etÂ€al. (2011) (Chapter 9 this volume), and uncertainties in inventories will be discussed in Section 14.5. Satellite-borne instruments, e.g. GOME (Burrows etÂ€ al., 1999), SCIAMACHY (Bovensmann etÂ€ al., 1999), aboard the ENVISAT satellite, or OMI (Boersma etÂ€ al., 2007), represent an interesting possibility to assess emissions, or at least CTM results which can relate to emissions. Such satellites Â�provide global coverage of some compounds at a spatial resolution of a few tens of kilometres. One of the first outstanding pictures provided by the use of such data was the decreasing NO2 Â�column trends over North America and Europe, and the increasing trend over China (Richter etÂ€al., 2005; van der A etÂ€al., 2006). However, the extent to which tropospheric Â�columns can be used for characterising air quality, namely surface concentrations, is not

David Simpson

obvious, due to measurements uncertainty and column vs. surface representativeness. Blond etÂ€al. (2007) showed in particular that spatial variability of surface concentrations in and near European cities are not captured by satellite measurements. Satellite measurements have also been used to constrain (Martin etÂ€ al., 2003) or estimate NOx emissions (Leue etÂ€ al., 2001). Konovalov etÂ€ al. (2005) and Blond etÂ€ al. (2007) show that the spatial distribution of tropospheric NO2 is generally well simulated by chemistry-transport models. However, sub-Â�regional model underestimations, especially in Southern Europe, are present, which suggests an underestimate of NOx emissions. Other models have been tested against satellite measurements, especially over China (Ma etÂ€al., 2006). Konovalov etÂ€al. (2006) attempted to invert measurements in order to obtain emissions at regional scale over Europe, using relations fitted to a chemistry-transport model. This method was also applied to the estimation of NOx emission decadal trends (Konovalov etÂ€al., 2008). It showed marked differences between trends in ‘bottom-up’ and satellite derived NOx emission trend estimates in Southern European regions, while trends are consistent in Northern areas. Satellites show also some potential for the evaluation of modelled fields of NH3 and hence of their emissions (Beer etÂ€al., 2008; Clarisse etÂ€al., 2009). Current retrieval methods require improvement, however, before this potential can be realised.

14.5â•‡ Uncertainties and challenges Estimation of the atmospheric inputs to Nr deposition is challenging because of uncertainties in the whole chain of proÂ� cessesÂ€ – emissions, dispersion, chemistry and deposition. Monks etÂ€ al. (2009) have discussed many of the issues with regard to oxidised nitrogen, and reviews such as those by Bleeker etÂ€ al. (2009) or van Pul etÂ€ al. (2009) provide much information on reduced nitrogen. Emissions and processes have been discussed in Hertel etÂ€al., 2011 (Chapter 9 this volume). Here we highlight those issues specific to modelling and measurements rather than processes. As noted in Hertel etÂ€al. (2006), there is a substantial discrepancy in the relative importance of various physical and chemical processes that need to be taken into account in local and regional scale models for N deposition, due to the differences in time scale. On the local scale the dispersion of pollutants is the most important process with regards to the concentration levels, whereas beyond about 5Â€km it is increasingly necessary to have good descriptions of wet and dry deposition processes, and atmospheric chemistry.

Measurements Generally, there are a variety of issues related to the measurements of N compounds in the atmosphere, due to their large number, low concentrations, reactivity and gas/aerosol interactions (Laj etÂ€ al., 2009). Highly reactive gases such as NH3 or HNO3 are very challenging to detect because of their Â�interaction with parts of the instruments, resulting in slow sensor response times. Their very short lifetime in the atmosphere means that they are highly variable, spatially and temporally. HONO is notoriously difficult to measure, usually with

positive bias caused by photolysis (it is thought) of nitrate on sampling lines and inlets. The measurement of aerosol compounds such as ammonium and nitrate also requires sophisticated instrumentation. Indeed, one of the critical limitations for model evaluation is the lack of good measurement data on the partitioning of oxidised nitrogen between HNO3, and fine and coarse particulate nitrate. Of the EMEP sites discussed by Fagerli and Hjellbrekke (2008), only a few sites reported results for the gaseous compounds, 15 for HNO3 and 11 for NH3, and of these only two use denuders whilst the others use filter-pack methodsÂ€– results from the latter are very uncertain. Organic N in the atmosphere is in general not measured as gas/particles with the exception of occasional measurements of PAN, amines and organic nitrates (but except for PAN not routinely monitored). The recent paper by Gonzalez Benitez etÂ€al. (2010) shows the potential scale of the problem. Measurement technology is available to measure organic N compounds such as amines and PANs, but the technology is still quite expensive, especially when it comes to continuous measurements. Another basic problem is that the surface measurements which are typically available give only a partial picture of some important chemical components. For example, gaseous NO3 is often discussed as a potentially important loss mechanism for hydrocarbons, providing a night-time alternative to OH radicals in driving chemistry (see Wayne etÂ€ al., 1991; Brown etÂ€ al., 2006). Unfortunately, this compound is extremely difficult to measure, and has very large vertical gradientsÂ€– surface concentrations are both modelled (Fish etÂ€al., 1999) and measured to be very small, even when boundary layer values are significant. Over sea areas, evaluation of models is further complicated as validation is usually against measurements at coastal sites; observations in the open sea rarely exists. Furthermore, for the Black Sea and the Mediterranean Sea, very few observations exist for any location.

Meteorology Uncertainties in the meteorological data used by models are often difficult to quantify, because many of the parameters that are critical for air pollution modelling are not measured, or only available through a series of assumptions. As well as precipitation (Section 14.4.3), an important example is the height of the boundary layer, or mixing height (Hmix), which controls the dispersion of all pollutants in the boundary layer but which is difficult to define even when radiosonde data are available (Seibert etÂ€al., 2000; Stern etÂ€al., 2008). Other important parameters include friction velocity and stability, both of which are crucial for deposition estimates. It can be noted that the uncertainties of meteorological inputs to CTMs receives relatively little attention, and these uncertainties are probably significant.

Emissions As noted above, emissions are the most important input to all CTM models, but they often receive little attention despite being subject to substantial uncertainties (Reis etÂ€al., 2009). As an example, looking at two recent emission inventories EDGAR

311


v.4 (EDGAR, 2010), and the UK National Emission Inventory (UKNAEI, 2010) for the year 2005, emission figures for NOx and NH3 for the UK both differ by about 30%. Such uncertainties are likely ubiquitous in European inventories, and indeed greater in countries with few emission measurement activities and where emission inventory development receives little funding.

Deposition modelling There are very many uncertainties regarding the magnitude (and even direction) of surface–atmosphere exchange, especially concerning dry deposition, occult deposition and emissions from soils. Particle deposition rates are one obvious source of uncertainty (Pryor etÂ€al., 2008a, c). As another example, the non-stomatal resistance term for NH3 is not only a function of ambient NH3 but also of the concentrations of acid gases (HNO3, SO2, HCl) which neutralise NH3 in water films on vegetation (Flechard etÂ€al., 1999; Fowler etÂ€al., 2009). To quantify this effect mechanistically requires dynamic chemical modelling with very short time steps, which precludes the implementation of such schemes in regional models, and some models (EMEP, IDEM) use the NH3/SO2 ratio as a proxy in empirical parameterisations. Flux networks such as Nitro-Europe will hopefully help reduce uncertainties in some parts of the Nr deposition budget, but there is a clear need for both improved instrumentation and analysis techniques before reliable estimates of Nr deposition can be made.

14.6â•‡ Conclusions This chapter has attempted to answer the overriding question:Â€what are the atmospheric inputs of reactive nitrogen (Nr) in Europe, and how well can we estimate these? We have focused mainly on presenting results from models, partly because of the limitations of measurements, but also because models allow for spatially comprehensive estimates of pollutant concentrations, and for mapping of deposition patterns over large areas. Models are also key tools to integrate our understanding of atmospheric chemistry and transport. Models address emissions, dispersion and transport over multiple scales, chemical transformation, and dry and wet removal of pollutants. This chapter, along with Hertel etÂ€al., 2011 (Chapter 9 this volume), has also discussed many of the uncertainties associated with deposition estimates of Nr. For specific locations, and especially at fine scales, such estimates can be very uncertain, with considerable variations on spatial scales of less than a kilometre. Differences between modelled and observed wet deposition of more than a factor of two are not uncommon for specific sites, especially in regions of complex topography. However, it should be remembered that mass considerations provide a strong constraint on the uncertainties in Nr deposition. Globally, all emissions of Nr will deposit somewhere, so that uncertainties in the deposition are equal to uncertainties in the emissions. Over Europe this equivalency still holds to a large extent, as the lifetime of emitted Nr is usually less than a few days. Such considerations probably explain why the current generation of chemical transport models perform quite

312

well (within 30%–50% say) when compared to the available (albeit very limited) observational data for wet deposition. Estimation of deposition at fine scales remains however a formidable task, and this poses challenges for estimating exceedances of critical levels for sensitive ecosystems. Improvements in models, emissions, measurements and understanding of physical/chemical processes will be needed before we can map fine-scale Nr deposition with confidence. Ideally, model evaluation and improvement of deposition estimates should be guided by direct measurements of fluxes of Nr, but such data are extremely expensive. In any case, observations of airborne components can also play a strong role in improving models, as there are many aspects of atmospheric chemistry which are still not properly evaluated. The main needs for oxidised compounds are probably to evaluate how well the models capture the partitioning of Nr between gaseous HNO3 and either fine or coarse nitrate. For reduced compounds, better estimates of emissions are needed, and how these are affected by meteorological factors as well as agricultural practices, coupled with an understanding of biosphere-atmosphere exchange. Such work should benefit from detailed studies from research networks such as NitroEurope, EUCAARI and EUSAAR, and from field campaigns (Laj etÂ€al., 2009; Kulmala etÂ€al., 2009; Sutton etÂ€al., 2007; Tang etÂ€al., 2009). Long-term monitoring, and a balanced hierarchy of a limited number of so-called super-sites (level 3 in EMEP terminology) and larger numbers of simpler level 1 and 2 sites (UNECE, 2009) would still be a crucial requirement, however, in order to assess (among other things) emission inventories, atmospheric processes, and long-term model performance.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), COST Action 729 and EMEP under the LRTAP Convention.

References Aas, W. and Hjellbrekke, A.-G. (2005). Data Quality 2003: Quality Assurance and Field Comparisons. EMEP/CCC Report 6/2005, The Norwegian Institute for Air Research (NILU), Kjeller, Norway. Asman, W. (1998). Factors influencing local dry deposition of gases with special reference to ammonia. Atmos. Environ., 32, 415–421. Asman, W., Drukker, B. and Janssen, A. (1998). Modeled historical concentrations and depositions of ammonia and ammonium in Europe. Atmos. Environ., 22, 725–735. Barrie, L. (1992). Scavenging ratios:Â€black magic or a useful scientific tool? In:Â€Proc. of 5th Int. Conf. on Precipitation Scavenging and Atmosphere–Surface Exchange Processes, ed. S. Schwartz and W. Slinn, vol. 1, the Georgii volume; Precipitation scavenging processes, pp. 403–420, Hemisphere Publ. Corp., Washington, DC. Bartnicki, J. and Fagerli, H. (2008). Airborne load of nitrogen to European seas. Ecol. Chem. Eng. S, 15, 297–313. Beer, R., Shephard, M. W., Kulawik, S. S. etÂ€al. (2008). First satellite observations of lower tropospheric ammonia and methanol. Geophys. Res. Lett., 35.

David Simpson Berge, E. and Jakobsen, H. A. (1998). A regional scale multi-layer model for the calculation of long-term transport and deposition of air pollution in Europe. Tellus, 50, 205–223. Berge, H. and Hjellbrekke, A.-G. (2010). Acidifying and Eutrophying Components:Â€Validation and Combined Maps, supplementary material to emep status report 1/2010, available online at www. emep. int, The Norwegian Meteorological Institute, Oslo, Norway. Bessagnet, B., Hodzic, A., Vautard, R. etÂ€al. (2004). Aerosol modeling with CHIMEREÂ€– preliminary evaluation at the continental scale. Atmos. Environ., 38, 2803–2817. Binkowski, F. and Roselle, S. (2003). Models-3 community multiscale air quality (CMAQ) model aerosol componentÂ€– 1. Model description. J. Geophys. Res., 108. Bleeker, A., Reinds, G., Vermeulen, A., de Vries, W. and Erisman,Â€J. (2004). Critical Loads and Present Deposition Thresholds of Nitrogen and Acidity and Their Exceedances at the Level II and Level I Monitoring Plots in Europe, ECN Report ECN-C-04–117, Energy Research Centre of the Netherlands (ECN). Bleeker, A., Sutton, M., Acherman, B. etÂ€al. (2009). Linking ammonia emission trends to measured concentrations and deposition of reduced nitrogen at different scales. In: Atmospheric Ammonia. Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S. M. Baker, pp. 123–180, Springer, New York. Blond, N., Boersma, K. F., Eskes, H. J. etÂ€al. (2007). Intercomparison of SCIAMACHY nitrogen dioxide observations, in situ measurements and air quality modeling results over Western Europe. J. Geophys. Res., 112, D10 311. Boersma, K., Eskes, H., Veefkind, J. etÂ€al. (2007). Near-real time retrieval of tropospheric NO2 from OMI. Atmos. Chem. Phys. 7, 2103–2118. Bovensmann, H., Burrows, J. P., Buchwitz, M. etÂ€al. (1999). SCIAMACHYÂ€– mission objectives and measurement modes. J.Â€Atmos. Sci., 56, 127–150. Brown, S., Ryerson, T., Wollny, A. etÂ€al. (2006). Variability in nocturnal nitrogen oxide processing and its role in regional air quality. Science, 311, 67–70. Burrows, J. P., Weber, M., Buchwitz, M. etÂ€al. (1999). The Global Ozone Monitoring Experiment (GOME)Â€–Â€mission concept and first scientific results, J. Atmos. Sci., 56, 151–175. Cape, J., Kirika, A., Rowland, A. etÂ€al. (2001). Organic nitrogen in precipitationÂ€–Â€real problem or sampling artefact? Scientif. World, 1, 230–237. Carmichael, G., Peters, L. and Saylor, R. (1991). The STEM-II regional scale acid deposition and photochemical oxidant model. 1. Overview of model development and applications. J. Geophys. Res., 25, 2077–2090. Carruthers, D., Holroy, D. etÂ€al. (1994). UK-ADMSÂ€– a new approach to modeling dispersion in the Earth’s atmospheric boundary layer. J. Wind Eng. Industr. Aerodynam., 52, 139–153. Chang, J., Brost, A., Isaksen, I. etÂ€al. (1987). A three-dimensional Eulerian acid deposition modelÂ€–Â€physical concepts and formulation. J. Geophys. Res., 92, 14681–14700. Cimorelli, A. J., Perry, S. G., Venkatram, A. etÂ€al. (2005). AERMOD:Â€a dispersion model for industrial source applications. 1:Â€General model formulation and boundary layer characterization. J. Appl. Meteorol., 44, 682–693. Clarisse, L., Clerbaux, C., Dentener, F., Hurtmans, D. and Coheur, P.-F. (2009). Global ammonia distribution derived from infrared satellite observations. Nature Geosci., 2, 479–483. Clarke, J., Edgerton, E. and Martin, B. (1997). Dry deposition calculations for the clean air status and trends network. Atmos. Environ., 31, 3667–3678.

Cofala, J., Amann, M., Kupiainen, K. and Hoglund-Isaksson, L. (2007). Scenarios of global anthropogenic emissions of air pollutants and methane until 2030. Atmos. Env., 41, 8486–8499. Collins, W., Stevenson, D., Johnson, C . and Derwent, R. (1997). Tropospheric ozone in a global-scale three-dimensional Lagrangian model and its response to NOx emission controls. J.Â€Atmos. Chem., 26, 223–274. de Leeuw, G., Cohen, L. Frohn, L. etÂ€al. (2001). Atmospheric input of nitrogen into the North SeaÂ€–Â€ANICE project overview. Cont. Shelf Res., 21, 2073–2094. de Leeuw, G., Skjoth, C, Hertel, O. etÂ€al. (2003). Deposition of nitrogen into the North Sea. Atmos. Environ., 37, S145–S165. Delle Monache, L. and Stull, R. (2003). An ensemble air-quality forecast over western Europe during an ozone episode. Atmos. Environ., 37, 3469–3474. Dentener, F., Drevet, J., Lamarque, J. F. etÂ€al. (2006). Nitrogen and sulfur deposition on regional and global scalesÂ€–Â€a multimodel evaluation. Glob. Biogeochem. Cycles, 20. Dore, A., Choularton, T. and Fowler, D. (1992). An improved wet deposition map of the United Kingdom incorporating the seederfeeder effect over mountanous terrain. Atmos. Environ., 26A, 1375–1381. Draaijers, G. and Erisman, J. (1993). Atmospheric sulfur deposition to forest standsÂ€– Through-fall estimates compared to estimates from inference. Atmos. Environ., 27A, 43–55. Duyzer, J., Nijenhuis, B. and Weststrate, H. (2001). Monitoring and modelling of ammonia consequences deposition in agricultural areas of the Netherlands. Water, Air and Soil Pollut.Â€Focus, 1, 131–144. EDGAR (2010). v.4, http :Â€//edgar. j re.ec.europa.eu EIONET (2010). http://air-climate.eionet.europa.eu/databases/MDS/ index_html EMEP (1996). EMEP Manual for Sampling and Chemical Analysis, EMEP/CCC-Report 1995, last revision in 2001. Norwegian Institute for Air Research, Kjeller, Norway. Online at http ://www. nilu.no/proj ects/ccc/manual/index.html EMEP http ://ebas.nilu.no (2010). www. emep.int EPA (US Environmental Protection Agency) (2010). www.epa.gov/ scram001/aqmindex. htm Erisman, J., Draaijers, G., Duyzer, J. etÂ€al. (1997). Particle deposition to forestsÂ€– summary of results and application. Atmos. Environ., 31, 321–322. Erisman, J. W., Hensen, A., Fowler, D. etÂ€al. (2001). Dry Deposition monitoring in Europe, Water, Air Soil Pollut.Â€Focus, 1, 17–27. Erisman, J., Vermeulen, A., Hensen, A. etÂ€al. (2005). Monitoring and modelling of biosphere/atmosphere exchange of gases and aerosols in Europe. Environ. Pollut., 403–413. Erisman, J., van Grinsven, H., Grizzetti, B. etÂ€al. (2011). The European nitrogen problem in a global perspective. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Fagerli, H. and Aas, W. (2008). Trends of nitrogen in air and precipitation:Â€model results and observations at EMEP sites in Europe, 1980–2003. Environ. Pollut., 154, 448–461. Fagerli, H. and Hjellbrekke, A. (2008). Acidification and eutrophication. In: Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe in 2006, EMEP Status Report 1/2008, pp. 41–56, The Norwegian Meteorological Institute, Oslo, Norway. Fagerli, H., Wind, P., Gauss, M. etÂ€al. (2008). Improved resolution in EMEP models. In:Â€Transboundary Acidification, Eutrophication

313

Atmospheric transport and deposition of reactive nitrogen and Ground Level Ozone in Europe in 2006, EMEP Status Report 1/2008, pp. 89–104, The Norwegian Meteorological Institute, Oslo, Norway. Fish, D., Shallcross, D. and Jones, R. (1999). The vertical distribution of NO3 in the atmospheric boundary layer. Atmos. Environ., 33, 687–691. Flechard, C. R., Fowler, D., Sutton, M. A. and Cape, J. N. (1999). A dynamic chemical model of bi-directional ammonia exchange between semi-natural vegetation and the atmosphere. Q. J. R. Meteorol. Soc, 125, 2611–2641. Fournier, N., Tang, Y., Dragosits, U., De Kluizenaar, Y. and Sutton, M. (2005). Regional atmospheric budgets of reduced nitrogen over the British Isles assessed using a multi-layer atmospheric transport model. Water, Air Soil Pollut., 162, 331–351. Fowler, D., Cape, J., Leith, I. etÂ€al. (1988). The influence of altitude on rainfall composition at Great Dun Fell. Atmos. Environ., 22, 1355–1362. Fowler, D., Pitcairn, C. E. R., Sutton, M. A. etÂ€al. (1998). The mass budget of atmospheric ammonia in woodland within 1 km of livestock buildings. Environ. Pollut., 102, 343–348. Fowler, D., Coyle, M., Flechard, C. etÂ€al. (2001). Advances in micrometeorological methods for the measurement and interpretation of gas and particle nitrogen fluxes. Plant and Soil, 228, 117–129. Fowler, D., Pilegaard, K., Sutton, M. etÂ€al. (2009). Atmospheric composition changeÂ€–Â€ecosystems–atmosphere interactions. Atmos. Environ., 43, 5193–5267. Frohn, L. M., Christensen, J. H., Brandt, J. and Hertel, O. (2001). Development of a high resolution integrated nested model for studying air pollution in Denmark. Phys. Chem. Earth, B, 26, 769–774. Gallagher, M. W., Nemitz, E., Dorsey, J. R. etÂ€al. (2002). Measurements and parameterizations of small aerosol deposition velocities to grassland, arable crops, and forestÂ€–Â€influence of surface roughness length on deposition, J. Geophys. Res., 107. Galloway, J., Dentener, F., Capone, D. etÂ€al. (2004). Nitrogen cycles:Â€past, present, and future. Biogeochemistry, 70, 153–226. Galmarini, S., Bianconi, R., Addis, R. etÂ€al. (2004). Ensemble dispersion forecasting.Â€2.Â€Application and evaluation. Atmos. Environ., 38, 4619–4632. Gonzalez Benitez, J. M., Cape, J. N., Heal, M. R., van Dijk, N. and Diez, A. V. (2009). Atmospheric nitrogen deposition in south-east ScotlandÂ€–Â€quantification of the organic nitrogen fraction in wet, dry and bulk deposition. Atmos. Environ., 43, 4087–4094. Gonzalez Benitez, J. M., Cape, J. N. and Heal, M. R. (2010). Gaseous and particulate water-soluble organic and inorganic nitrogen in rural air in southern Scotland. Atmos. Environ., 44, 1506–1514. HELCOM (2005). Airborne Nitrogen Loads to the Baltic Sea, Helsinki Commission. Hertel, O., Christensen, J., Runge, E. etÂ€al. (1995). Development and testing of a new variable scale air-pollution modelÂ€– ACDEP. Atmos. Environ., 29, 1267–1290. Hertel, O., Ambelas Skjoth, C., Frohn, L. M. etÂ€al. (2002). Assessment of the atmospheric nitrogen and sulphur inputs into the North Sea using a Lagrangian model. Phys. Chem. Earth, B, 27, 1507–1515. Hertel, O., Skjoth, C., Brandt, J. etÂ€al. (2003). Operational mapping of atmospheric nitrogen deposition to the Baltic Sea. Atmos. Chem. Phys., 3, 2083–2099.

314

Hertel, O., Skjoth, C. A., Lofstrom, P. etÂ€al. (2006). Modelling nitrogen deposition on a local scaleÂ€– a review of the current state of the art. Environ. Chem., 3, 317–337. Hertel, O., Reis, S., Ambelas Skjoth, C. etÂ€al. (2011). Nitrogen processes in the atmosphere. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Holmes, N. and Morawska, L. (2006). A review of dispersion modelling and its application to the dispersion of particlesÂ€–Â€an overview of different dispersion models available. Atmos. Environ., 40, 5902–5928. HTAP (UNECE Hemispheric Task Force on Atmospheric Pollutants) (2010). www. htap.org IPCC (2007). Climate Change 2007. The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the International Panel on Climate Change, www.ipcc.ch. Jenkin, M. E., Saunders, S. M., Wagner, V. and Pilling, M. J. (2003). Protocol for the development of the Master Chemical Mechanism, MCM v3 (Part B):Â€tropospheric degradation of aromatic volatile organic compounds. Atmos. Chem. Phys., 3, 181–193. Jenkin, M., Utembe, S. and Johnson, D. (2006). The simulated speciation and distribution of oxidized organic nitrogen in the atmospheric boundary layer. In:Â€Understanding and Quantifying the Atmospheric Nitrogen Cycle, ed. R. Cox, D. Fowler, P. Monks and P. Borrell, pp. 130–136, ACCENT, Keele University, UK. Johnson, D., Utembe, S. R. and Jenkin, M. E. (2006). Simulating the detailed chemical composition of secondary organic aerosol formed on a regional scale during the TORCH 2003 campaign in the southern UK. Atmos. Chem. Phys., 6, 419–431. Kessler, C., Brucher, W., Memmesheimer, M., Kerschgens, M. and Ebel, A. (2001). Simulation of air pollution with nested models in North Rhine-Westphalia. Atmos. Environ., 35, S3–S12. Konovalov, I., Beekmann, M., Vautard, R. etÂ€al. (2005). Comparison and evaluation of modelled and GOME measurement derived tropospheric NO2 columns over Western and Eastern Europe. Atmos. Chem. Phys., 5, 169–190. Konovalov, I. B., Beekmann, M., Richter, A. and Burrows, J. P. (2006). Inverse modelling of the spatial distribution of NOx emissions on a continental scale using satellite data. Atmos. Chem. and Phys., 6, 1747–1770. Konovalov, I. B., Beekmann, M., Burrows, J. P. and Richter, A. (2008). Satellite measurement based estimates of decadal changes in European nitrogen oxides emissions. Atmos. Chem. Phys., 8, 2623–2641. Korhonen, H., Lehtinen, K. E. J. and Kulmala, M. (2004). Multicomponent aerosol dynamics model UHMA:Â€model development and validation. Atmos. Chem. Phys., 4, 757–771. Laj, P., Klausen, J., Bilde, M. etÂ€al. (2009). Measuring atmospheric composition change. Atmos. Environ., 43, 5351–5414. Langner, J., Andersson, C. and Engardt, M. (2009). Atmospheric input of nitrogen to the Baltic sea basinÂ€–Â€present situation, variability due to meteorology and impact of climate change. Boreal Environ. Res., 14, 226–237. Leue, C., Wenig, M., Wagner, T. etÂ€al. (2001). Quantitative analysis of NOx emissions from GOME satellite image sequences. J. Geophys. Res., 106, 5493–5505. Loubet, B. and Cellier, P. (2001). Experimental assessment of atmospheric ammonia dispersion and short range dry deposition in a maize canopy. Water, Air and Soil Pollution: Focus, 1, 157–166. Loubet, B., Cellier, P., Milford, C. and Sutton, M. A. (2006). A coupled dispersion and exchange model for short-range dry

David Simpson deposition of atmospheric ammonia. Quarterly Journal of the Royal Meteorological Society, 132, 1733–1763. Loubet, B., Asman, W. A., Theobald, M. R. etÂ€al. (2009). Ammonia deposition near hot spots:Â€processes, models and monitoring methods. In:Â€Atmospheric Ammonia. Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S.Â€M.Â€Baker, pp. 205–267, Springer, New York. Ma, J., Richter, A., Burrows, J. P., Nuss, H. and van Aardenne, J. A. (2006). Comparison of model-simulated tropospheric NO2 over China with GOME-satellite data. Atmos. Environ., 40, 593–604. Martin, R V., Jacob, D. J., Chance, K. V. etÂ€al. (2003). Global inventory of nitrogen dioxide emissions constrained by space-based observations of NO2 columns. J. Geophys. Res., 108, 4537. McKeen, S., Chung, S. H., Wilczak, J. etÂ€al. (2007). Evaluation of several PM2.5 forecast models using data collected during the ICARTT/NEAQS 2004 field study. J. Geophys. Res., 112. Monks, P., Granier, C., Fuzzi, S. etÂ€al. (2009). Atmospheric composition changeÂ€– global and regional air quality. Atmos. Environ., 43, 5268–5350. Olesen, H., Lofstrem, P., Berkowicz, R. and Jensen, A. (1992). An improved dispersion model for regulatory useÂ€– the OML model. In: Air Pollution Modelling and its Application IX, ed. H. van Dop and G. Kallos, Plenum Press, New York. Olesen, H. R. (1995). The model validation exercise at M01:Â€overview of results. Int. J. Environ. Pollut. 5, 761. Pagowski, M., Grell, G., McKeen, S. etÂ€al. (2005). A simple method to improve ensemble-based ozone forecasts. Geophys. Res. Lett., 32. Pilegaard, K., Skiba, U., Ambus, P. etÂ€al. (2005). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 651–661. Pryor, S. C., Barthelmie, R. J., Sorensen, L. L. etÂ€al. (2008a). Upward fluxes of particles over forestsÂ€–Â€when, where, why? Tellus, 60, 372–380. Pryor, S. C., Gallagher, M., Sievering, H. etÂ€al. (2008b). A review of measurement and modelling results of particle atmosphere– surface exchange. Tellus, 60, 42–75. Pryor, S. C., Larsen, S. E., Sorensen, L. L. and Barthelmie, R. J. (2008c). Particle fluxes above forests:Â€Observations, methodological considerations and method comparisons. Environ. Pollut., 152, 667–678. Rannik, U., Aalto, P., Keronen, P., Vesala, T. and Kulmala, M. (2003). Interpretation of aerosol particle fluxes over pine forestÂ€–Â€dry deposition and random errors. J. Geophys. Res., 108, 4544. Reis, S., Pinder, R. W., Zhang, M., Lijie, G. and Sutton, M. A. (2009). Reactive nitrogen in atmospheric emission inventories. Atmos. Chem. Phys., 9, 7657–7677. Rendell, A. R., Ottley, C. J., Jickells, T. D. and Harrison, R. M. (1993). The atmospheric input of nitrogen species to the North Sea. TellusÂ€B, 45, 53–63. Riccio, A., Giunta, G. and Galmarini, S. (2007). Seeking for the rational basis of the Median Model:Â€the optimal combination of multimodel ensemble results. Atmos. Chem. Phys., 7, 6085–6098. Richter, A., Burrows, J. P., Nuß, H., Granier, C . and Niemeier, U. (2005). Increase in tropospheric nitrogen dioxide over China observed from space. Nature, 437, 129–132. Robertson, L., Langner, J. and Engardt, M. (1999). An Eulerian limited-area atmospheric transport model. J. Appl. Meteorol., 38, 190–210. Roelofs, G., Kasibhatla, P., Barrie, L. etÂ€al. (2001). Analysis of regional budgets of sulfur species modeled for the COSAM exercise. Tellus, 53B, 673–694.

Ruijgrok, W., Tieben, H. and Eisinga, P. (1997). The dry deposition of particles to forest canopyÂ€–Â€a comparison of model and experimental results. Atmos. Environ., 31, 399–415. Sanderson, M., Dentener, R, Fiore, A. etÂ€al. (2008). A multimodel study of the hemispheric transport and deposition of oxidised nitrogen. Geophys. Res. Lett., 35, L17 815, doi:10.1029/2008GL035389. Schaap, M., Timmermans, R. M. A., Roemer, M. etÂ€al. (2008). The LOTOS-EUROS model:Â€description, validation and latest developments. Int. J. Environ. Pollut., 32, 270–290. Schaap, M., Vautard, R., Bergstrom, R. etÂ€al. (2010). Evaluation of long-term aerosol simulations from seven air quality models and their ensemble in the EURODELTA study (submitted). Schlunzen, K. H. and Meyer, E. M. I. (2007). Impacts of meteorological situations and chemical reactions on daily dry deposition of nitrogen into the southern North Sea. Atmos. Environ., 41, 289–302. Seibert, P., Beyrich, F., Gryning, S.-E. etÂ€al. (2000). Review and intercomparison of operational methods for the determination of the mixing height. Atmos. Environ., 34, 1001–1027. Seinfeld, J. and Pandis, S. (1998). Atmospheric Chemistry and Physics. From Air Pollution to Climate Change, John Wiley and Sons, New York. Simpson, D., Tuovinen, J.-P., Emberson, L. and Ashmore, M. (2001). Characteristics of an ozone deposition module. Water, Air Soil Pollut.Â€Focus, 1, 253–262. Simpson, D., Fagerli, H., Jonson, J. etÂ€al. (2003). The EMEP Unified Eulerian Model – Model Description, EMEP MSC-W Report 1/2003, The Norwegian Meteorological Institute, Oslo, Norway. Simpson, D., Butterbach-Bahl, K., Fagerli, H. etÂ€al. (2006a). Deposition and emissions of reactive nitrogen over European forests:Â€a modelling study. Atmos. Environ., 40, 5712–5726. Simpson, D., Fagerli, H., Hellsten, S., Knulst, J., and Westling, O. (2006b). Comparison of modelled and monitored deposition fluxes of sulphur and nitrogen to ICP-forest sites in Europe. Bio-geosciences, 3, 337–355. Singles, R., Sutton, M. and Weston, K. (1998). A multi-layer model to describe the atmospheric transport and deposition of ammonia in Great Britain. Atmos. Environ., 32, 393–399. Sliggers, S. and Kakebeeke, W. (eds.) (2004). Clearing the Air: 25 Years of the Convention on Long-Range Transboundary Air Pollution, United Nations, Economic Commission for Europe, Geneva, http:Â€//www.unece .org/env/lrtap Smith, R., Fowler, D., Sutton, M. A., Flechard, C. and Coyle, M. (2000). Regional estimation of pollutant gas dry deposition in the UKÂ€–Â€model description, sensitivity analyses and outputs. Atmos. Environ., 34, 3757–3777. Sommer, S. G., and Jensen, E. S. (1991). Foliar absorption of atmospheric ammonia by ryegrass in the field. Journal of Environmental Quality, 20, 153–156. Sportisse, B. (2007). A review of current issues in air pollution modeling and simulation. Comp. Geosci., 11, 159–181. Stern, R., Builtjes, P., Schaap, M. etÂ€al. (2008). A model intercomparison study focussing on episodes with elevated PM10 concentrations. Atmos. Environ., 42, 4567–4588. Sutton, M., Nemitz, E., Erisman, J. etÂ€al. (2007). Challenges in quantifying biosphere–atmosphere exchange of nitrogen species. Environ. Poll., 150, 125–139. Sutton, M. A., Milford, C., Dragosits, U. etÂ€al. (1998). Dispersion, deposition and impacts of atmospheric ammoniaÂ€–Â€quantifying local budgets and spatial variability. Environ. Pollut., 102, 349–361.

315

Atmospheric transport and deposition of reactive nitrogen Tang, Y., Simmons, I., van Dijk, N. etÂ€al. (2009). European scale application of atmospheric reactive nitrogen measurements in a low-cost approach to infer dry deposition fluxes. Agricult., Ecosyst. Environ. 133, 183–195. Theobald, M. R., Millford, C., Hargreaves, K. J. etÂ€al. (2001). Potential for ammonia recapture by farm woodlands: design and application of a new experimental facility. 1(S2), 791–801. Textor, C., Schulz, M., Guibert, S. etÂ€al. (2006). Analysis and quantification of the diversities of aerosol life cycles within AeroCom. Atmos. Chem. Phys., 6, 1777–1813. UNECE (2009). EMEP Draft revised monitoring strategy, http:// www.unece.org/env/ documents/2009/EB/ge1/ece.eb.air. ge.1.2009.15.e.pdf UKNAEI (UK National Emission inventory) (2010). www. uknaei. org. uk van Aardenne, J. A., Dentener, F. J., Olivier, J. G. J., Goldewijk, C. G. M. K. and Lelieveld, J. (2001). A 1 degrees × 1 degrees resolution data set of historical anthropogenic trace gas emissions for the period 1890–1990. Glob. Biogeochem. Cycles, 15, 909–928. van der A, R. J. M., Peters, D. H. M. U., Eskes, H. etÂ€al. (2006). Detection of the trend and seasonal variation in tropospheric NO2 over China J. Geophys. Res., 111, D12317. van Jaarsveld, J. A. (2004). The Operational Priority Substances Model – Description and Validation of OPS-Pro 4.1, RIVM Report 500045001/2004, RIVM, Bilthoven, The Netherlands. van Loon, M., Roemer, M. and Builtjes, P. (2004). Model Intercomparison in the Framework of the Review of the Unified EMEP Model, TNO-report R2004/282, TNO, TNO Environment and Geosciences, Utrecht, The Netherlands, http://www.tno.nl/ content.cfm? context=markten&content=publicatie&laagl=186&l aag2=l&item_ id=737&Taal=2 van Loon, M., Vautard, R., Schaap, M. etÂ€al. (2007). Evaluation of longterm ozone simulations from seven regional air quality models and their ensemble. Atmos. Environ., 41, 2083–2097.

316

van Pul, A., Jaarsveld, H. V., Meulen, T. v. d. and Velders, G. (2004). Ammonia concentrations in the NetherlandsÂ€–Â€spatially detailed measurements and model calculations. Atmos. Environ., 38, 4045–4055. van Pul, A., Hertel, O., Geels, C. etÂ€al. (2009). Modelling of the atmospheric transport and deposition of ammonia at a national and regional scale. In:Â€Atmospheric Ammonia: Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S. M. Baker, pp. 301–358, Springer, New York. Vautard, R., Builtjes, P., Thunis, P. etÂ€al. (2007). Evaluation and intercomparison of ozone and PM10 simulations by several chemistry transport models over four European cities within the CityDelta project. Atmos. Environ., 41, 173–188. Vautard, R., Schaap, M., Bergstrom, R. etÂ€al. (2008). Skill and uncertainty of a regional air quality model ensemble. Atmos. Environ. 43, 4822–4832. Vieno, M., Dore, A. J., Wind, P. etÂ€al. (2009). Application of the EMEP Unified Model to the UK with a horizontal resolution of 5 × 5 km2. In: Atmospheric Ammonia. Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S. M. Baker, pp. 367–372, Springer, New York. Vieno, M., Dore, A. J., Stevenson, D. S. etÂ€al.Â€2010 Modelling surface ozone during the 2003 heat wave in the UK. Atmospheric Chemistry and Physics, 110, 7963–7978. Wayne, R., Barnes, I., Biggs, P. etÂ€al. (1991). The nitrate radical:Â€physics, chemistry, and the atmosphere. Atmos. Environ. 25A, 1–203. Zhang, L., Gong, S., Padro, J. and Barrie, L. (2001). A size-segregated particle dry deposition scheme for an atmospheric aerosol module. Atmos. Environ., 35, 549–560. Zhang, L., Vet, R., O’Brien, J. M. etÂ€al. (2009). Dry deposition of individual nitrogen species at eight Canadian rural sites. J.Â€Geophys. Res., 114, D02301.

Chapter

15

Geographical variation in terrestrial nitrogen budgets across Europe Lead author: Wim de Vries Contributing authors: Adrian Leip, Gert Jan Reinds, Johannes Kros, Jan Peter Lesschen, Alexander F. Bouwman, Bruna Grizzetti, Fayçal Bouraoui, Klaus Butterbach-Bahl, Peter Bergamaschi, Wilfried Winiwarter

Executive summary Nature of the problem • Nitrogen (N) budgets of agricultural systems give important information for assessing the impact of N inputs on the environment, and identify levers for action.

Approaches • N budgets of agro-ecosystems in the 27 EU countries are established for the year 2000, considering N inputs by fertiliser application, manure excretion, atmospheric deposition and crop fixation, and N outputs by plant uptake, gaseous emissions, mineralisation, leaching and runoff. • Country N budgets for agro-ecosystems are based on the models INTEGRATOR, IDEAg, MITERRA and IMAGE. Fine geographic distribution is depicted with the former two models, which have higher spatial resolution. INTEGRATOR is the only available model for calculating non-agricultural terrestrial N budgets systems.

Key findings/state of knowledge • For EU-27, the models estimate a comparable total N input in European agriculture, i.e. 23.3–25.7 Mton N yr−1, but N uptake varies largely from 11.3–15.4 Mton N yr−1, leading to total N surpluses varying from 10.4–13.2 Mton N yr−1. Despite this variation, the overall difference at EU-27 is small for the emissions of NH3 (2.8–3.1 Mton N yr−1) and N2O (0.33–0.43 Mton N yr−1) but estimates vary largely at a regional scale. The estimated sum of N leaching and runoff at EU-27 is roughly equal to the sum of NH3, N2O and NOx emissions to the atmosphere, but estimates vary by a factor two, from 2.7 to 6.3 Mton N yr−1. • Trends in N fluxes in agro-ecosystems since 1970 show an increase in N inputs by fertilisers and manure up to 1985, followed by a decrease since 1985 in response to a decrease in crop production and in animal numbers. Actually, livestock decreased since 1970, but in the period 1970–1985 the N input by manure excretion still increased due to an increase in N excretion rates. • In non-agricultural system (forests and semi-natural vegetation), the estimated total N input is near 3.2 Mton N yr−1, while the net N uptake is near 1.1 Mton N yr−1, leading to a surplus near 2.1 Mton N yr−1. Compared to agricultural systems, the estimated N fluxes in non-agricultural systems are about five times lower for N2O emissions and 10 times lower for NOx and NH3 emissions and for the sum of N leaching and runoff.

Major uncertainties/challenges • The largest uncertainties in flux values, as estimated from inter-model comparison, concerns N leaching and runoff, followed by N2O emissions, from agricultural ecosystems.

Recommendations • Future research should focus on reducing the fluxes with the most uncertainty (N leaching and runoff, followed by N2O emissions, from agricultural ecosystems), including studies on denitrification. • To improve model assessments and enable model validation, databases should be set up of:Â€(i) N contents in major crops/vegetation in various regions (to improve estimates of N uptake and N surplus), (ii) NH3 and N2O emissions based on inverse modelling approaches


317

Geographical variation in terrestrial nitrogen budgets

(to validate N emission calculations) and N concentrations in ground water and surface water (to validate N leaching and N runoff assessments). • The number of countries with estimated NH3-N emissions in 2000 exceeding the National emission ceilings for 2010 depends on the model approach and varies between 7 and 18. Exceedance of critical N concentrations in surface water is highly model-dependent. It is relevant that data use, both on activity data and emission or leaching factors is harmonised for models predicting air emissions and N loss to waters for consistent environmental decision-making relevant to air quality, ecosystem deposition and water quality.

15.1â•‡ Introduction The major share of new reactive nitrogen (Nr) is introduced into the environment with the purpose of producing agricultural commodities. Excess N input, however, causes a number of ecological and human health effects, like acidification, eutrophication, elevated N saturation of forest soils, climate change and biodiversity impacts (see also Grizzetti etâ•¯al., 2011; Moldanová etâ•¯al., 2011; Butterbach-Bahl etâ•¯al., 2011, Dise etâ•¯al., 2011; Velthof etâ•¯al., 2011, Chapters 17–21, this volume). An indication of the potential impact of N inputs in agriculture can be derived by an overview of all N inputs and N outputs, here referred to as an N budget. N budgets of agro-ecosystems are generally constructed (i) to increase the understanding of nutrient cycling, (ii) for use as performance indicator and to raise awareness in nutrient management and environmental policy, and (iii) as regulating policy instrument to monitor and enforce a certain nutrient management policy in practice (Oenema etâ•¯al., 2003). Sometimes, the term N balance is also used, but this term is consistently used in this chapter to denote the N surplus, defined as the sum of all N inputs minus N removal by feed and food, in line with its use by the Organisation for Economic Co-operation and Development (OECD) (OECD, 2001, 2007). We use the word N budget for a complete N flux assessment. In this chapter, we present N budgets of agro-ecosystems and non-agricultural terrestrial ecosystems in Europe as performance indicator, illustrating the N use efficiency of agro-ecosystems and the loss of excess N to the environment (air and water). We summarise the present knowledge on European N budgets for terrestrial ecosystems by using a range of different modelling and input data assessment approaches. This way we implicitly assess uncertainties. As a part of the budget approach, the chapter includes key N fluxes, including N inputs by manure, fertiliser, deposition and fixation, N uptake, emissions of ammonia (NH3), nitrous oxide (N2O) nitrogen oxides (NOx) and di-nitrogen (N2), and the sum of N leaching and runoff, to provide an overall picture of the N status of Europe. The assessment concentrates at discussing data at the country level with the EU-27 as geographical scope, even though the calculations are performed in many models at much higher resolution in order to cover the nonlinearity of the soil proÂ� cesses. Most data are available around the year 2000 and so most of the data presented are reflecting the situation around this year. However, we include also a discussion of the past trends of important elements in the N-budgets since 1970 onwards. In Section 15.2, we first describe the modelling approaches and input data that are available to assess terrestrial N fluxes at

318

the European scale. We then present results in terms of farm and land N budgets for agricultural systems, including trends in N budgets in the period 1970–2000 (Section 15.3) followed by land N budgets for non-agricultural terrestrial systems (Section 15.4). An overall evaluation of the results is given in Section 15.5. This includes an evaluation of the validity of the presented model approaches by comparison of model results with independent datasets, whenever available. Furthermore, the relevance of N budgets and their trends with respect to effects on ecosystems and the reliability of N budgets at various geographic scales are discussed. For a complete overview of aggregated N fluxes across media and sectors for countries throughout Europe, we refer to Leip etâ•¯al., 2011a (Chapter 16, this volume). Details on N sources in deposition are given in Simpson etâ•¯al., 2011 (Chapter 14, this volume).

15.2â•‡ Methodological approaches and input data to assess terrestrial nitrogen budgets at the European scale 15.2.1â•‡ Approaches to assess nitrogen budgets at regional scale While we are interested to obtain N budgets for agriculture on a regional, country or European level, we need to differentiate different budgeting approaches by the respective system boundaries used. We distinguish three basic approaches in regional N budget studies, using the farm, land or soil as the gate at which the N inputs and outputs are quantified (see Table 15.1). (1) Farm nitrogen budget (called farm-gate budget by Oenema etâ•¯al., 2003); it records the amounts of N in all kinds of products that enter and leave the farm via the farm-gate. Throughputs, as for example uptake of grass by animals, or the application of manure, are not part of the farm N budget. The surplus/deficit, i.e. the difference between inputs and outputs, is a measure of total N losses, adjusted for possible changes in the storage of nutrients in the farming system. Examples of this approach are the now abolished MINAS (Mineral Accounting System) regulatory nutrient book-keeping system in the Netherlands (Oenema etâ•¯al., 1998; Neeteson, 2000), and the OSPARCOM method (Oslo and Paris Conventions for the Prevention of Marine Pollution) focusing on N and P discharges into the North Sea and Baltic Sea from the surrounding countries (OSPARCOM, 1994). In the simple farm N budget, the N surplus is not further specified, whereas N (NH3, N2O, NOx and N2) losses from

Wim de Vries Table 15.1 Definition of N inputs, N outputs and N surpluses in regional farm, land and soil nitrogen budgets for agricultural systems

Budget

System boundary

Simple

Detailed

N Inputs

N Outputs

N Surplusa

Farm

Farm N budget

Agricultural system N budget

Fertiliser, feed (concentrates), external organic N sources, N fixation and N deposition, net N manure import, and withdrawals

Sold animal (meat, milk, etc.) and crop products

N (NH3, N2O, NOx and N2) emissions and N leaching/runoff from housing and manure storage systems and soil; soil N stock changes

Land

Gross N budget (OECD approach)

Land system N budget

Fertiliser, manure excretion, external organic sources, crop residues returned on soils, N fixation, N deposition, net N manure import/ export, and withdrawals

Harvest of crop products (in arable land) or above ground removal of grass, crop residues

N (NH3, N2O, NOx and N2) emissions and N leaching/runoff from housing and manure storage systems and soil; soil N stock changes

Soil

Soil N budget

Soil system N budget

Fertiliser, manure application, grazing inputs, external organic sources, crop residues returned on soils, N fixation and N deposition

Removal of crop products (in arable land) or above ground removal of grass; crop residues, soil N stock changes

N (NH3, N2O, NOx and N2) emissions and N leaching/runoff from soil

a

N surplus is specified in the detailed N-budgets

the housing and manure storage systems and from soil to the air and to aquatic systems are specified in a detailed agricultural system budget, as illustrated in Figure 15.1. An example of this approach is the CAPRI-DNDC model (Leip etâ•¯al., 2009). (2) Land nitrogen budgets (called gross N balances by the OECD). It records all N that enters a farm land (including housing and manure storage systems) and leaves the farmland by crop products. Nitrogen inputs include fertiliser, animal manure production/excretion, biological N fixation and N deposition. This approach is used for example by the OECD as environmental performance indicator for agriculture (OECD, 2001, 2007). In the simple approach, called gross N budget (gross N balance by the OECD), the N surplus is not further specified, whereas N losses from the housing and manure storage systems and from soil to the air and to aquatic systems are specified in a detailed land system budget. This approach is used in this chapter. (3) Soil nitrogen budget (called soil surface budget by Oenema etâ•¯al., 2003). It records all N that enters the soil and that leaves the soil via crop uptake, including nutrient gains and losses within the soil. Nitrogen inputs via animal manure are adjusted for losses of N emissions in housing and manure management systems; all other N inputs are the same as for the land N budget. Nitrogen output (defined here as output of ‘useful product’) is corrected by the changes of N storage; accumulation of N in organic matter is regarded as useful because it improves soil quality and can potentially contribute to crop growth in following

years. Soil N surplus (see Table 15.1) is then a measure for the total N loss from the soil to either the atmosphere (NH3, N2O, NOx and N2 emissions) or the hydrosphere (NÂ€leaching to ground water and N runoff to surface water). In the soil N budget, this N surplus is not further specified, whereas in the soil system budget all N inputs and outputs, including N gains and losses within and from the soil are specified. It should be noted, that in the literature the soil N budget mostly differs from our definition, as the NH3 emission from soils is often already corrected for while the soil N changes are included in the calculation of the surplus (Oenema etâ•¯al., 2003). The N surplus gross N budget includes the sum of all nutrient emissions from agriculture into soil, water, and air (OECD, 2007) and is thus often used as the indicator of agricultural pressure on water quality (EEA, 2005), as it allows identifying areas with high risk of N leaching. Detailed budgets are able to resolve the individual pathways of N as presented in Table 15.1. It is important to remember that different accounting methods cover different N flows. Animal housing and manure management systems are not included in the soil budgets, while they are accounted for in farm and land budgets. In the land N budgets, the N excreted in the manure is considered, while in the soil N budget only the N in applied manure, corrected for losses in housing and manure management systems, is accounted for. Manure used for other purposes (e.g. burning) is not considered in both approaches. With respect to ‘mineral N fertiliser’, the farm N budget considers fertiliser purchases, while mineral fertiliser applications are relevant for the land and soil N budgets.

319


While for soil budgets the system boundaries are usually the top soil layer (surface to rooting depth), and covers thus only land-based agricultural production, farm and land budgets include also the livestock sector. As for the farm (and agricultural systems) budget, the boundary is the farm, they don’t consider manure and animal intake of N in fodder produced in the farm as input or output. However, if data are available, they are often quantified as N throughput. The difference in farm, land and soil budgets is illustrated further in Leip etâ•¯al. (2010).

15.2.2â•‡ Modelling approaches There are several operative activities that estimate N budgets for the European Union and for Europe at various spatial resolutions. Table 15.2 gives an overview of main model approaches that have been used for assessing total agricultural emissions of different forms of reactive N for various parts of Europe (from EU15 to whole Europe), at various geographic resolutions (from grid to country) and for different time periods. The approaches included in Table 15.2 are:Â€(i) complete land system N budget models for agriculture, using yearly time steps (INTEGRATOR, CAPRI, IDEAg, MITERRA, IMAGE), (ii) emission factor approaches for both agricultural and total annual NH3, N2O and NOx emissions to the atmosphere (GAINS, EMEP, EDGAR, UNFCCC-IPCC) and (iii) N loss models to either surface water (GREEN) or ground water (EPIC). In the supplementary information to this chapter (SuppÂ� lementary material, Chapter 15 & 16), a description of the various models mentioned above and the meaning of their abbreviations is given. In short, the complete land system N budget models are able to calculate all N fluxes to and from a land system, as defined in Table 15.1. First of all, these models are able to assess the N surplus or gross soil N budget according to (see Table 15.1):Â€N surplus = input (mineral fertilisers + livestock manure excretion corrected for transport + other organic sources + left crop residues + biological fixation + atmospheric deposition)Â€ – total crop removalÂ€ – total forage uptake. The models are also all able to simulate the fate of the N surplus in terms of NH3, N2O, NOx and N2 emissions from housing and manure storage systems, N accumulation in or release from the soil (not in all models) and N losses by leaching and runoff. The emission factor approach models are limited to atmospheric emissions, but unlike the land system N budget models they include all sectors, including traffic and industry. Similarly N loss models are limited to estimates of N losses to surface water and/or ground water, but they generally include all N sources, including human sewage and direct deposition inputs to surface water. In this chapter, we focus on complete N budgets for agriculture, as derived with INTEGRATOR, IDEAg (CAPRI based model), MITERRA and IMAGE. More details on these models is given in the supplementary materials at Chapter 15 and 16 and in De Vries etâ•¯al. (2010b). We also include a comparison of results of NH3, N2O and NOx emissions with the emission factor approaches (GAINS, EMEP, EDGAR, UNFCCC-IPCC), while results of the model GREEN are shown to illustrate the impact of diffuse sources versus point sources.

320

There are also detailed ecosystem models available that provide process-level descriptions for either daily NH3, N2O and NOx emissions, such as the DNDC model (Li etâ•¯al., 2000) or N leaching, such as the EPIC model (Bouraoui Aloe, 2007; Van der Velde etâ•¯al., 2009) that have been applied to derive N fluxes at regional scale in Europe. The DNDC model has for example been used to assess N2O and NOx emissions for both forests (Kesik etâ•¯al., 2005) and agricultural land (Butterbach-Bahl etâ•¯al., 2009) at a fixed 10 km × 10 km grid, while the EPIC model that has been applied to study the effect of agricultural practices and biofuel cultivation on N leaching (Bouraoui and Aloe, 2007; Van der Velde etâ•¯al., 2009). However, these models do not include emissions from housing systems and in case of EPIC also not explicitly from soils, and are therefore not included in the model comparison presented in this paper. Some results are, however, shown in the Supplementary material (Chapter 15 and 16).

15.2.3â•‡ Data sets to estimate nitrogen inputs and outputs In order to understand the operation of models, an overview of internationally coherent datasets used by the models is given. In addition to these international datasets, often national information also exists, but in general this cannot be assessed by activities operating on a European scale. Inputs of N to agricultural systems include N fertiliser, N manure due to application and grazing, N deposition and N fixation. Data sets that are relevant for the assessment of N uptake are crop yields and element contents in crops, while N and C pools are relevant for the assessment of N emission fluxes. The assessment of N fluxes to the air (emissions of NH3, N2O, NOx, and N2) and water (N leaching to ground water, N surface runoff and subsurface flow to surface water) requires data on emission and leaching parameters in the various models to make such predictions. An overview of the data used by all the four complete N budget models is given in De Vries etâ•¯al. (2011). More information on the datasets that are used to calculate the amount of fertiliser and manure N applied to soil is given in Supplementary material ChapterÂ€15 and 16. In biogeochemistry models, soil C and N contents often strongly determine the N2O flux. Maps of present concentrations and pools of C and N in the soil and C/N ratios in the soil distinguishing between agricultural soils and non-agricultural soils can be based on various databases, i.e. WISE/SOTER, European Soil Data Base (ESDB2) and ICP forests database. More information on approaches and results is given in the Supplementary material (Chapter 15 and 16).

15.3â•‡ Farm and land nitrogen budgets for agricultural systems In the following sections, data on farm and soil N budgets are presented focusing on two recently developed model systems, i.e. IDEAg and INTEGRATOR. IDEAg consists of three

Wim de Vries Table 15.2 Overview of available models approaches for assessing emissions of different forms of Nr for various parts of Europe at various geographic resolutions and for various time periods

Model approach

Element flux considered

Method

Sectors considered

Area involved

Geographic resolution

Time

Complete land N budget models INTEGRATOR (De Vries etâ•¯al., 2010)

N2O, NOx and NH3 emission, N leaching, N runoff

Adapted MITERRA approach for agricultural systems. Statistical model for terrestrial systems

Agriculture, terrestrial systems

EU-27+3

NCUa

1970–2000

MITERRA (Velthof etâ•¯al., 2007, 2009)

N2O, NOx and NH3 emission, N leaching, N runoff

Emission and leaching factor approach for agricultural systems

Agriculture

EU-27

NUTS2

2000

CAPRI (Britz, 2005; Britz etâ•¯al., 2005; CAPRI 2010)

NH3, N2O, NÂ€surplus

Mass-budget model using an emission-factor approach

Agriculture

EU-27

NUTS2

Base year currently 2002 projections up to 2012

IDEAg, (Leip etâ•¯al., 2008)

N2O, NOx and NH3 emission, N leaching

Economic model for agriculture, linked to mechanistic model to simulate soil NÂ€budget

Agriculture

EU-27

HSMUa

2000

IMAGE (Alcamo, 1994; Leemans etâ•¯al., 1998; MNP, 2006; IMAGE, 2010)

N2O, NOx and NH3 emission, N leaching, NÂ€runoff

Extended emission factor approach with consideration of mitigation technologies

All sectors

Europe Global

Country

Present, projections

N emission models to atmosphere GAINS (Höglund-Isaksson and Mechler, 2005; Winiwarter, 2005) http://gains. iiasa.ac.at/ gains/EU/index. login?logout=1

N2O, NOx and NH3 emission


All sectors

Europe Global

Country

Present, projections

EDGAR (Van Aardenne, 2002) http://edgar.jrc.it

NH3, N2O and NOx emission


All sectors

Global

1 × 1 degree. The latest version (released 11/2008) is 0.1â•›× 0.1 degree

Past and present

EMEP (Simpson etâ•¯al., 2003, 2006; EMEP, 2010a)

NOx and NH3 emission N deposition

Emissions (disaggregated from official national inventories) and Atmospheric dispersion model

All sectors

Europe

50 km × 50Â€km; 5 × 5 km possible (e.g. Vieno etâ•¯al., 2009)

Past, present and projections up to 2030

321

Geographical variation in terrestrial nitrogen budgets Table 15.2 (cont.)

Model approach UNFCC/IPCC (IPCC, 2006; UNFCCC, 2010)

Element flux considered N2O (and NOx) emission

Sectors considered

Area involved

Geographic resolution

Emission factor approach on activity data

All sectors

Europe and other ‘Annex-I’ countries (industrialised)

Country

1990– present

Method

Time

N loss models to hydrosphere GREEN (Grizzetti etâ•¯al., 2005, 2008; Bouraoui etâ•¯al., 2009)

Total N diffuse emissions to waters and total N runoff

Geospatial empirical regression model

Agriculture and Point Sources

Europe

Sub-catchments (average size 180 km2)

1985–2005

EPIC (Bouraoui and Aloe, 2007; Van der Velde etâ•¯al., 2009)

NO3, NH4, total N, soluble and particulate N runoff, NÂ€leaching

Detailed mechanistic model

Agriculture, terrestrial systems

EU-27 + Swiss

10 km × 10 km grid (including multiple crops)

1985–2005

a

HSMUâ•›=â•›Homogeneous Spatial Mapping Units; NCU = NitroEurope Calculation Units. Units refer to clusters of 1 km2 grid cells that are characterised by similar environmental and/or agronomic conditions

elements:Â€(i) the CAPRI-SPAT downscaling model (Leip etâ•¯al., 2008); (ii) the DNDC-CAPRI meta-model (Britz and Leip, 2009b); and (iii) an interface combining results of the DNDCCAPRI meta-model with elements of CAPRI-SPAT, yielding a database with environmental indicators that are inherently consistent and operating at the level of individual crops. These models use the most detailed geographically explicit input data currently available, thus allowing the best way to map the various N fluxes included in the N budget. In particular, the DNDC-CAPRI meta-model is based on detailed spatial information, partly based on biophysical model simulations. A special feature of INTEGRATOR is that it includes historical data up to 1960, thus allowing the assessment of trends in N budgets. Despite the high spatial resolution of the data available in these model systems, results presented in this chapter are mainly restricted to model comparisons at the Europewide scale (tables of complete N budgets) and at the national scale (scatter plots of N fluxes). Detailed maps are limited to N input by manure and fertiliser and to NH3 and N2O, emissions from the agricultural system (both housing systems and soil) as derived by IDEAg and INTEGRATOR. Detailed maps of total N emissions divided in various sectors are further presented in Leip etâ•¯al., 2011a (Chapter 16 this volume).

15.3.1â•‡ Farm nitrogen budget The IDEAg model system can be used to provide an updated picture of a farm N-budget for Europe. In IDEAg, a combination of the farm budget (animal and crop production in relation with the EU and global market) and soil N budget has been implemented (see Figure 15.1). As explained above, the farm N budget comprises as inputs feed intake and as output animal products, both driven by the economic situation of the farm (i.e. region). The N surplus is exported to manure management systems and finally applied to crops or excreted on

322

grassland by grazing animals (other uses of manure are not significant in Europe and are not considered in IDEAg). IDEAg also calculates the fate of animal and crop products and distinguishes human consumption, processing by the industry to generate feed concentrates, biofuels or other products and, inand export for each commodity considered. Also, losses at the market (and at the farm) are estimated. As a result, the IDEAg system is able to depict a detailed picture of N-flows of the agriculture sector at the European scale.

15.3.2â•‡ Land nitrogen budgets Detailed land nitrogen budgets at European level An overview of a detailed European (EU27) field scale (land) N budget is presented in Table 15.3. The table compares results derived with INTEGRATOR (De Vries etâ•¯al., 2010) with information from IDEAg (Britz and Leip, 2009a), MITERRA (Velthof etâ•¯al., 2007, 2009) and IMAGE (De Vries etâ•¯al., 2009). Furthermore, the sum of the officially submitted data to the UNFCCC secretariat by the 27 EU countries, as reported in the Annual European Community greenhouse gas inventory, are presented (EEA, 2008). Results include N (NH3, N2O, NOx and N2) emissions from housing systems to give complete emission estimates from the agricultural system. Consequently, we include manure excretion instead of manure application as input to the system. For EU27, the four models estimate a total N input in European agriculture of 23.3–25.7 Mton N yr−1, which is mainly due to fertiliser and animal manure inputs and to a lesser extent caused by atmospheric deposition and N fixation. The N uptake varies from 11.3–15.4 Mton N yr−1 leading to total N surpluses (N input not used by the plants) varying from 10.4 to 13.2 Mton N yr−1 at EU27 level. The lowest surplus is calculated by INTEGRATOR, as it assesses the highest uptake. The various models give in general very similar results

Wim de Vries Figure 15.1 N budget for the agricultural sector in EU27 for the year 2002 as calculated by the IDEAg model.

for the emissions of NH3 (2.8–3.1 Mton N yr−1). Comparable estimates are also derived for the direct N2O emissions Â�(0.33–0.43 Mton N yr–1), but NOx emissions vary by a factor 10 (0.02–0.22 Mton N yr–1). The sum of N leaching and N runoff also varies largely. The estimates by IDEAg and IMAGE are nearly twice as large as the estimates by INTEGRATOR and MITERRA, causing a much lower estimated N2 emission by IDEAg and IMAGE as compared to INTEGRATOR and MITERRA (Table 15.3). An important difference in this context is also that both INTEGRATOR and IDEAg include mineralisation estimates, whereas this input term is neglected in MITERRA and IMAGE. In INTEGRATOR, the net release is mainly determined by the N mineralisation in drained peat soils. In IDEAg, mineralisation of all soils is obtained from the DNDC meta-model and then scaled in two steps (the second jointly with N2 flux estimates) to close the N budget. More details on the N emission sources calculated by the various models are given in Table 15.4. Results show that the difference in NH3 emissions between IDEAg versus the other three models is the result of the higher emissions from housing and manure storage systems. Another notable difference is the much higher N2O and NOx emission from grazing by IMAGE as compared to the other models (Table 15.4). Reasons for the various similarities and differences can be summarised as follows. • All model give similar results for the N inputs by fertiliser as they use the same FAO data regarding fertiliser rates. • Deviations between inputs by manure application are larger due to different sources for animal numbers, but specifically due to deviating N excretion rates. • Differences in biological N fixation mainly follow from the different values used to derive N fixation of pulses/legumes as a fraction of the harvested N amount, as summarised in

the supplementary material chapters 15 & 16 (see also De Vries etâ•¯al. (2010b). • The NH3 emissions by INTEGRATOR, IDEAg and MITERRA are comparable as they are based on the same GAINS dataset. There are however differences in N manure and fertiliser distribution and this affects the N leaching that is affected by soil type, land use, etc. • The difference in N2O emissions is limited on a European wide scale, considering the differences in N2O emission factors used. In INTEGRATOR, these emissions are determined as a function of soil type, land use, manure type, etc. In IDEAg, results are based on the DNDC-CAPRI meta-model, whereas MITERRA uses standard emission fractions based on IPCC. These differences do, however, affect the spatial variation in N2O emissions (see Section 15.3.3). • The higher estimated sum of total N leaching and runoff by IDEAg and IMAGE are mainly due to higher leaching and runoff fractions. In IDEAg, N leaching is based on the DNDC meta-model whereas N leaching by the other models depends on various environmental factors as described in detail in De Vries etâ•¯al. (2011). Apparently, the difference in parameterization of the factors and in geographic resolution leads to strongly different results.

Land nitrogen inputs and nitrogen surplus at country level Land N budgets at country-scale for agriculture for the year 2000 calculated by INTEGRATOR, IDEAg, MITERRA and IMAGE for the various EU countries are presented in the Supplementary materials (Chapter 15 and 16). A scatter diagram of the N inputs as calculated with the INTEGRATOR model compared to IDEAg, MITERRA and IMAGE is given in Figure 15.2. The four approaches generally agree for fertiliser input and N inputs by manure, which is logical as it has the same

323

Geographical variation in terrestrial nitrogen budgets Table 15.3 Annual N budgets of agricultural land in Europe in 2000, including N (NH3, N2O, NOx and N2) emissions from housing systems and from soil. Output terms in italic are summations of more detailed N fluxes and should not be added in the calculation of the total N output

N budget (Mton N yr−1) Source

INTEGRATOR EU 27–2000

MITERRAa EU 27–2000

IMAGEa EU 27–2000

1.0

0.8

1.4

IDEAg EU25–2002

UNFCCb EU27–2002

Input to land Biological fixation

1.3

Manure excretion

10.3

8.8

10.4

9.8

9.1

Synthetic fertiliser

11.5

11.4

11.3

11.3

10.6

2.7

2.1

2.0

2.8

—

25.7

23.3

24.5

25.3

20.8

Atmospheric deposition Total

1.1

Output from land Plant removal

15.4c

12.5

11.3

13.5

—

N accumulation

−3.3

−3.5

—

—

—

2.9

3.1

2.9

2.8

3.1

0.40

0.43

0.33

0.43

0.4

0.21

0.11

0.02

0.22

—

7.0

4.5

7.2

2.5

nd

7.6

5.1

7.8

3.1

—

Emissions of NH3 N2O

d

NO and NO2 N2 Total (De)nitrification N leaching

2.8

5.7

2.0

—

—

0.35

0.4

0.75

—

-

3.1

6.1

2.7

5.9

6.6

Total surplus

10.4

10.8

13.2

11.8

—

Total

25.7

23.3

24.5

25.3

—

N surface runoff Total leaching/runoff

Details of the comparison between MITERRA and IMAGE are described in De Vries etâ•¯al. (2009). Source:Â€EEA (2008). c Uptake includes the removal from grassland, rough grazing areas and the net crop removal from arable land. d N2O emission refers to direct N2O-N emission only that is calculated by all models. a b

basis although the IDEAg N manure inputs are consistently lower (see also Table 15.4). There are relatively large differences for the other N inputs (deposition and fixation) at country level, but this hardly affects the total N inputs by the four models, which are comparable for all countries. Total N uptake is quite different between the various approaches. As with the results at European scale (see Table 15.4), INTEGRATOR results are consistently higher than the other models. The uptake mostly decreases according to INTEGRATOR > IMAGE > IDEAg > MITERRA. Furthermore, there is quite some scatter at country level. This is reflected in an even larger scatter for the N surplus per country, indicating an uncertainty near 50% for country estimates of the N surplus.

Nitrogen emissions to air and water at country level Instead of quantifying just the gross N surplus, the N excess input can be further defined in terms of N (NH3, N2O, NOx and N2) emissions to the atmosphere, N leaching and N runoff. The N budget models described before can derive such detailed agricultural N budgets not only at European level (see Section 15.2.1), but also at country level. An example of

324

such an output calculation using INTEGRATOR is given in Table 15.5. To gain insight in the comparability of the results obtained, a comparison is given of agricultural emissions of NH3-N, N2O-N and NOx-N and N leaching for 27 EU countries for the year 2000 as derived with INTEGRATOR with those obtained by the complete N budget models (IDEAg, MITERRA and IMAGE). Furthermore, results for the N emissions were compared with standard activity data-emission factors approaches (UNFCC/IPCC, 2010; GAINS, 2010; OECD, 2010; EDGAR, 2010; and EMEP 2010b). Data used for the results of the various models for NH3-N, N2O-N and NOx-N are found in the Supplementary data for Chapter 15. A comparison of country emissions for NH3-N, N2O-N and NOx-N and of N leaching plus runoff (kton N yr−1) within EU 27 as derived with INTEGRATOR with the various other approaches is given in Figure 15.3. Results show comparable estimates for NH3 emissions, which is due to the use of comparable databases for the estimation. Both INTEGRATOR and MITERRA use the N excretion and NH3 emission constants derived by GAINS and consequently, the differences should be

Wim de Vries Table 15.4 Annual N emissions from agriculture in Europe for the year 2000

N emissions in 2000 (kton N yr−1) N source

Emission source

INTEGRATOR EU 27–2000

NH3

Housing and storage

1189

Fertiliser application

1413

b

Manure application

N2O

MITERRAa EU 27–2000

IMAGEa EU 27–2000

1428

1279

1048

678

540

798

759

823

683

Grazing

271

201

231

319

Total agriculture

2873

3066

2873

2848

Housing and storage

55

48

54

52

242

316

208

289

Grazing

124

67

66

92

Indirect emissions

43

80

51

Total agriculture

401 (444)

431 (531)

328 (379)

N application

NO and NO2

IDEAg EU 25–2002

c

d

76 d

434 (510)d

Housing and storage

20

32

36

0

N applicationc

123

16

25

23

Grazing

63

59

32

196

Total agriculture

207

108

93

219

Details of the comparison between MITERRA and IMAGE are described in De Vries etâ•¯al. (2009). b Includes emissions through soil inputs by fertiliser and manure application. c Includes emissions through soil inputs by fertiliser and manure application, deposition, mineralisation, fixation and crop residues. d The value in brackets are the total N2O emissions calculated by INTEGRATOR, IDEAg, MITERRA and IMAGE including also indirect N2O emissions due to N leaching and NH3 and NOx emissions. a

small and are mainly due to the use of different statistics for animal numbers. Furthermore, all models use comparable statistics for N fertiliser use and NH3 emissions from manure. The differences in different N2O emissions, however, are much larger, reflecting the larger variation in model approaches, specifically the use of N2O emission factors. For example, a comparison of INTEGRATOR results with the N2O emissions reported by the EU countries to the UNFCCCIPCC shows quite a disagreement. For MITERRA, there is a good agreement with estimated N2O emission from manure management, and direct soil N2O emission (Velthof etâ•¯al., 2009), since both methods are based on the same N2O emission fractions as a function of N inputs. Deviations between UNFCCC figures and MITERRA are thus only due to differences in activity data and the use of specific emission factors by some countries. By contrast, INTEGRATOR uses emission factors that depend on N source and environmental conditions. In both INTEGRATOR and MITERRA, the estimated indirect N2O emission (not shown here) are much smaller than those reported to the UNFCCC, owing to both a lower N2O emission factor and a lower N leaching fraction. Firstly, the revised IPCC emission factor for N leaching (IPCC, 2006) was used in both INTEGRATOR and MITERRA-EUROPE (i.e. 0.0075 kg N2O-N for each kg N that leaches), whereas the values of the UNFCCC for most countries were obtained using the former emissions factor of 0.025 kg N2O-N per kg N leached (IPCC, 1997). Secondly, IPCC uses a simple method to calculate leaching, i.e. 30% of the total N input via fertiliser, manure, grazing

and other sources leaches to ground water and surface water (Mosier etâ•¯al., 1998). INTEGRATOR and MITERRA use a different approach to calculate N leaching which resulted in leaching losses of 11% of the total N input in EU-27. The NOx emissions appear to be very uncertain (see FigureÂ€15.3). This is in line with results obtained by ButterbachBahl etâ•¯al. (2009), who applied the approach used in IMAGE and three other empirical emission models, using the same input data for all models. More information on that approach and related results is given in the Supporting material in Chapters 15 and 16. The sum of N leaching plus runoff also varies largely within EU 27 and is systematically higher for IDEAg and IMAGE as compared to INTEGRATOR and MITERRA, in line with the results at European level. This implies that the used N leaching factors are highly uncertain and need further refinement.

15.3.3â•‡ Mapping the European agricultural nitrogen fluxes The national N inputs and N outputs presented in Section 15.3.2 do not show the regional differences in N fluxes. In this section we provide maps showing such differences, focusing on presentations with IDEAg and INTEGRATOR for agricultural ecosystems in EU-27 for the year 2000. These two models were used to illustrate the geographic variation in model results, because of their highly disaggregated model input data. With respect to the emission of greenhouse gases, such as N2O, it is

325


Figure 15.2 A comparison of country N inputs by fertiliser, manure, other inputs (deposition and fixation), total N inputs, total net N uptake and N surplus within EU27 as derived with INTEGRATOR, IDEAg, MITERRA and IMAGE for the year 2000 (IDEAg is 2002).

crucial to know whether total emissions for the area considered are correct, whereas accurate information on the spatial distribution of the emissions is less relevant. The latter aspect is, however, crucial when assessing the risk of elevated NH3 emissions, and related N deposition, and of N leaching and

326

N runoff in view of eutrophication impacts on terrestrial and aquatic ecosystems. Here, aggregation of input data for large areas may cause accurate average N deposition and N leaching levels, but a strong deviation in the area exceeding critical N deposition loads or critical N concentrations in ground water

Wim de Vries Table 15.5 N emissions to air and water calculated at country level with INTEGRATOR for the year 2000

N output fluxes (kg N ha−1 yr−1) Countrya

Area (Mha)

Emission NH3

Emission N2O

Emission NOx

Emission N2

Leaching + runoff

Austria

3.336

13.2

2.1

0.9

23.1

11.1

Belgium

1.779

41.6

5.6

2.2

70.3

32.6

Bulgaria

6.816

4.7

0.9

0.4

16.3

4.4

Czech. Rep

4.776

11.1

2.5

1.0

38.7

18.4

Denmark

3.273

19.9

1.8

0.9

27.5

25.4

Estonia

1.846

3.8

1.1

0.5

25.5

5.4

Finland

6.914

2.0

0.4

0.1

17.6

4.6

France

35.346

14.8

2.6

1.2

30.8

13.7

Germany

21.566

20.2

2.4

1.1

42.4

19.5

Greece

8.404

6.2

1.2

0.7

20.9

10.4

Hungary

6.739

8.6

1.5

0.7

38.1

10.1

Ireland

5.043

15.5

5.4

2.8

37.3

11.5

18.434

17.6

1.9

0.9

33.3

19.1

Latvia

3.343

2.7

0.6

0.3

15.9

6.0

Lithuania

4.246

5.4

1.2

0.5

32.5

17.9

Luxembourg

0.144

20.8

6.9

0.0

34.6

13.9

Netherlands

2.491

52.6

4.8

2.4

94.3

45.0

Italy

Poland

20.265

10.6

1.3

0.4

32.0

15.6

Portugal

5.411

8.1

1.1

0.6

30.1

14.4

Romania

14.517

7.9

1.1

0.5

23.8

8.1

Slovakia

2.664

9.4

1.5

0.8

21.4

13.5

Slovenia Spain Sweden UK

0.779

20.5

2.6

1.3

28.2

12.8

35.027

7.2

0.8

0.4

16.2

7.6

7.914

4.2

0.6

0.3

12.9

5.8

16.237

15.2

4.1

1.8

39.1

15.3

EU-27

237.310

EU-27b

237.310

a b

12.1 2873

1.9 444

0.9 207

29.3 6965

13.2 3136

Data for Cyprus and Malta are not included. Data given in kton N yr-1.

and surface water (De Vries etâ•¯al., 2009). For this reason, it is relevant to make use of models with the highest level of spatial detail with respect to inputs and outputs, such as IDEAg and INTEGRATOR. The datasets mentioned in the Supplementary materials in ChaptersÂ€15 and 16 in combination with various downscaling techniques have been used to ‘regionalise’ the agricultural N inputs from statistical data at national or subnational level to the NCU or HSMU level.

Nitrogen inputs Inputs by manure and fertiliser Input of mineral N fertiliser and manure N as derived by IDEAg and INTEGRATOR are shown in Figure 15.4. The legend of 170 kg N is chosen as this is the maximum allowed manure N input in the EC, with

the exception of a derogation (accepted after 2000) of 250 kg N for the Netherlands and 230 kg N for Denmark, Germany and Austria. High manure N application rates occur in areas of high livestock density in Europe and include parts of Denmark, the Netherlands, Belgium, Wales, Ireland, Catalonia and Galicia in Spain, and the north of Italy. Regions of high N fertiliser input can be identified in most intensive agricultural areas in Europe, again including Denmark, Belgium, the Netherlands, UK and Ireland, Brittany (France) and the Po Valley (Italy). Results show that an exceedance of the N manure input of 170 kg N occurs mainly in various dense livestock population areas, such as the Netherlands, where even the derogation of 250 kg N is often exceeded in the year 2000. There is a

327


(a) NH3-N emission agriculture (kton.yr –1) 800

IDEAg Miterra

600

IMAGE GAINS

400

EDGAR 200

0

200

0

600

400

800

NH3-N emission agriculture INTEGRATOR

(b) N2O-N emission agriculture (kton.yr –1) 100

IDEAg Miterra

80

IMAGE

60

GAINS

40

EMEP

20

OECD

0

EDGAR UNFCC/ IPCC 0

20

40

60

80

100

N2O-N emission agriculture INTEGRATOR (kton.yr –1)

(kton.yr –1)

(c) NOx-N emission agriculture (kton.yr –1)

(d) N leaching (kton.yr –1) 1500

70 IDEAg

60 50

Miterra

40

IMAGE

30

EDGAR

20

IDEAg

1000

Miterra IMAGE

500

EMEP

10

0

0 0

10

20

30

40

50

60

70

NOx-N emission agriculture INTEGRATOR (kton.yr –1)

0

500

1000

N leaching INTEGRATOR

1500

(kton.yr –1)

Figure 15.3 A comparison of country emissions for NH3-N, N2O-N and NO-N and of the sum of N leaching and runoff for the year 2000 within EU 27 as derived with INTEGRATOR and with various other model approaches (IDEAg, MITERRA, IMAGE, GAINS, EDGAR, EMEP and UNFCC/IPCC).

clear difference between IDEAg and INTEGRATOR in western France, where the latter model calculates much higher N manure inputs. The reason for this difference is seemingly a different disaggregation of animal numbers. In general N application by mineral fertiliser is higher in IDEAg, specifically in Western Europe, but also in the Nordic countries where it is possibly an artefact due to division of N inputs by very small areas of agricultural land (Figure 15.4a, b). Inversely, N application by animal manure, including grazing, is generally higher in INTEGRATOR, except for parts of the Netherlands and Denmark. INTEGRATOR shows hot-spots, e.g. in parts of France and Eastern Europe that are not resulting from IDEAg (Figure 15.4c, d). A comparable picture for the estimated N inputs by mineral fertilisers and animal manure for the year 2000 in EU25 is given by Grizzetti etâ•¯al. (2007), using a 10 km × 10â•›km resolution. Details on the approach, combining agricultural statistics on administrative basis and geographic land cover information, are given in Grizzetti etâ•¯al. (2007).

328

NH3 and N2O emissions Calculations by both INTEGRATOR and IDEAg show that the regional variation in total NH3 and N2O emissions is large (Figure 15.5). Hot spots are located in areas with intensive animal husbandry in the Eastern and central part of Ireland, in England and Wales, in the Netherlands, Belgium, Denmark, in north-western and southern Germany, in the north of Italy and in the Catalonia region in Spain. In general, NH3 emissions calculated by IDEAg are higher than by INTEGRATOR in Western and Central Europe, but the reverse is true for the Nordic countries (Figure 15.5a, b). Inversely, N2O emissions calculated by IDEAg are higher everywhere, specifically in the Nordic countries, where the high emissions might be an artefact of the extremely high N fertiliser input but lower in the UK and Ireland (Figure 15.5c, d). The variation in NH3 and N2O emissions is in general comparable with the geographic variation in N surpluses, which in turn are strongly related to the variation in manure N inputs. The high correlation between

Wim de Vries

(a) N fertiliser (kg N/ha/yr) 0 - 25 25 - 50 50 - 100 100 - 170 170 - 250 >250

(c) N manure (kg N/ha/yr) 0 - 25 25 - 50 50 - 100 100 - 170 170 - 250 >250

(b) N fertiliser (kg N/ha/yr) 0 - 25 25 - 50 50 - 100 100 - 170 170 - 250 >250

(d) N manure (kg N/ha/yr) 0 - 25 25 - 50 50 - 100 100 - 170 170 - 250 >250

Figure 15.4 Nitrogen application from mineral fertiliser (a, b) and manure, including grazing (c, d) in the year 2000 in EU-27. Calculation with IDEAg on the geographic resolution of HSMUs (left) and with INTEGRATOR on the geographic resolution of NCUs (right). Grey shading in the EU-27 denotes non-agricultural areas. Countries outside EU-27 are also included by a grey shade.

329


(a) NH3 emission (kg N/ha/yr) 0-5 5 - 10 10 - 15 15 - 20 20 - 40 >40

(c) N2O emission (kg N/ha/yr) 0-1 1-2 2-4 4-6 6-8 >8

(b) NH3 emission (kg N/ha/yr) 0-5 5 - 10 10 - 15 15 - 20 20 - 40 >40

(d) N2O emission (kg N/ha/yr) 0-1 1-2 2-4 4-6 6-8 >8

Figure 15.5 Total NH3 emissions (a, b) and N2O emissions (c, d) from agriculture in the year 2000 in EU-27. Calculation with IDEAg on the geographic resolution of HSMUs (left) and with INTEGRATOR on the geographic resolution of NCUs (right). Grey shading in the EU-27 denotes non-agricultural areas. Countries outside EU-27 are also included by a grey shade.

330

Wim de Vries

N surplus, being a main driver for N emissions and manure application is illustrated in detail by Leip etâ•¯al. (2011b).

Nitrogen losses to ground water and surface water Nitrogen losses to either ground water or surface water can be achieved using models, which include the major N inputs and the main processes of N transport and transformation, including surface runoff (overland flow) and runoff (interflow) to surface water and leaching to ground water. Various models have been developed and applied to address the issue of N fate in the river basin, and they vary for process description, scale of study and data requirement (http://euroharp. org). On a European wide scale, both detailed (EPIC) and simple process based models (INTEGRATOR, IDEAg) and statistical models (GREEN) are available (see Table 15.2). Here, we show results derived with both INTEGRATOR and IDEAg and with GREEN. The estimated regional variation N losses from soil to both ground water and surface water in 2000 as derived with IDEAg and INTEGRATOR is given in Figure 15.6. It should be emphasised that INTEGRATOR estimates are only slightly influenced by meteorological data, since the model uses N leaching fractions that depend on soil type, land use, soil organic content, precipitation surplus, temperature and rooting depth (Velthof etâ•¯al., 2009). In IDEAg, however, N leaching from soils is based on the DNDC-CAPRI meta-model (Britz and Leip, 2009a), which in turn is derived from CAPRI-DNDC model simulations using meteorological

data to asses water fluxes and related N leaching fluxes. In this context, use is made of the JRC-MARS database, being a spatial interpolation of more than 1500 weather stations across Europe onto a 50 km × 50 km grid (Orlandi and Van der Goot, 2003). In line with Table 15.4, results obtained by IDEAg show a much higher N leaching rate all over Europe, as compared to INTEGRATOR. Most likely, the N leaching by IDEAg is an overestimation, since there is a reasonable comparison between measured NO3 concentrations in ground water and those estimated by the MITERRA model, being the agricultural module in INTEGRATOR in an adapted form (see Section 15.5.1 on model evaluation). Figure 15.7 (left) shows an estimate of N diffuse losses to surface water for the year 2000 for Europe (Grizzetti etâ•¯al., 2008; Bouraoui etâ•¯al., 2009), based on the GREEN model taking into account N sources, river network and climate conditions. According to these estimates, the regions affected by higher N losses to surface waters include Belgium, the Netherlands, the Po Valley (Italy), the Brittany region (France), which are already totally or partially designated as Nitrates Vulnerable Zones (Nitrate Directive). Figure 15.7 (right) shows the estimated N source apportionment per sub-catchment for Europe for the year 2000. This map provides a picture of the relative contribution of diffuse sources (mainly agriculture) and point sources (mainly urban settlements) to the water N pollution. According to these estimates, agriculture is the main

Figure 15.6 Regional pattern of N leaching plus runoff in the year 2000 in EU-27 based on calculations with IDEAg on the geographic resolution of HSMUs (left) and with INTEGRATOR on the geographic resolution of NCUs (right). Grey shading in the EU-27 denotes non-agricultural areas. Countries outside EU-27 are also included by a grey shade.

331


N Arctic Ocean

N Arctic Ocean

N Diffuse emissions from agriculture (kg N/ha) 0–2

0 – 25

3 – 10

26 – 50 51 – 75

11 – 20 21 – 30

Atlantic Ocean

76 – 85

Atlantic Ocean

86 – 100

> 30 North Sea

North Sea

English Channel

English Channel

Black Sea

Bay Of Biscay

250

500

1,000 Kilometers

Black Sea

Bay Of Biscay

Mediterranean Sea

0

N Diffuse sources (%) (per sub-catchment)

Mediterranean Sea

0

250

500

1,000 Kilometers

Figure 15.7 Regional pattern of N loads to surface water as diffuse emissions (left) and N source apportionment (right) for Europe in year 2000, based on calculations with the GREEN model on a geographic resolution of sub-catchments (average size 180 km2).

contributor of N for surface waters in most of the river basins, while in Mediterranean catchments point sources have a relative higher contribution, which is probably due to a less effective implementation of waste water treatments and the lower precipitation and thus N losses to surface waters.

15.3.4â•‡ Trends in nitrogen fluxes since 1970 Trends in N fluxes since 1970 up to the year 2000 are derived on the basis of INTEGRATOR using the following. • Data on N fertiliser use, animal numbers and crop yields from the FAO database. • Scaled N excretion rates to those used for 2000 on the basis of RAINS/GAINS data. The scaling is based on a simple N excretion model described by Witzke and Oenema (2007), using the milk production as a scaling factor for dairy cattle and the meat production as a scaling factor for other cattle, pigs and poultry. Data on the milk and meat production per country in the period 1970–2000 were taken from the FAO database. • N deposition history based on historical NOx emissions by EMEP and NH3 emissions by INTEGRATOR, while adding non-agricultural sources from IMAGE and using an emission-deposition matrix based on the EMEP model (EMEP, 2009). • Constant N fixation rates for the grassland and arable land, but using FAO data on trends in the area of dry pulses and soy beans, mainly affecting N fixation. Information on trends in data for alfalfa and clovers, affecting the estimate for biological fixation by grasslands are missing and consequently we assumed no trends in N fixation rates by grassland. • Scaled N contents in crops, based on a change in N availability (this is automatically calculated in INTEGRATOR).

332

• Trends in NH3 emission factors in view of changes in housing systems and manure application techniques. For the year 2000, GAINS data are used for the fraction of housing systems and manure application techniques with high, medium and low emissions per country. For the period 1970–1980, we assumed that all emission fractions were high and in the period 1980–2000, we assumed a linear interpolation from high emissions to the present emission percentage. Note that the available data on both crop yields and N fertiliser use in the FAO databases include trends in N use efficiency, which is mostly defined as the crop yield divided by the N input by fertiliser (Bouwman etâ•¯al., 2005). Results derived by INTEGRATOR for the trends in all N inputs, N surplus and N outputs, in terms of N emissions to the atmosphere and N leaching to ground water and surface water, for the period 1970–2000 are given in Figure 15.8. The results show a steady increase of N inputs by fertilisers in the period 1970–1985, followed by a decrease since then, mainly in response to the increased or decreased crop production in those periods (or vice versa). Despite a slight decrease in cattle, the N input by manure excretion has increased up to 1985 due to an increase in N excretion rates, related to an increase in milk production, followed by a slight decrease in response to the decrease in livestock and the relatively constant excretion rates. The trend is also influenced by the increase in pigs anol poultry between 1970–2000 (see Oenema etâ•¯al., 2007), but the dominant effect is that of changes in N excretion rates by dairy cattle. There is a more clear increase in the average N input in agricultural systems than in the total N input, due to a decrease in agricultural area. This holds also for the trends in the total N uptake and the related N surplus for the period 1970–2000. Results show a slightly declining trend in NH3 emission in response to a decline in livestock since 1990, but the trends in N2O and NOx emissions and N leaching are almost constant.

Wim de Vries

(a)

(b)

Balance (kg N.ha–1.yr–1)

(Input (kg N.ha–1.yr –1)

150

150 Animal manure Fertiliser 100

100

Deposition

Total input

Fixation 50

Uptake

Mineralisation

50

Surplus

Total 0 1970 1975 1980 1985 1990 1995 2000

0 1970

1975

1980

1985

Year

1990

1995

2000 Year

(c) Emission and leaching (kg N.ha–1.yr–1) 40 NH3

30

N2O NOx

20

N2 10

0 1970

Leaching

1975

1980

1985

1990

1995

2000 Year

Figure 15.8 Trends in average N inputs (a), and average N surplus (b) and average N outputs (c) at EU-27 level for the period 1970–2000 as estimated by INTEGRATOR. NB:Â€leaching stands for leaching plus runoff.

Trends in N2 emissions, being most uncertain, are clearly increasing up to 1985 and declining afterwards (Figure 15.8).

15.4â•‡ Land nitrogen budgets for non-agricultural systems 15.4.1â•‡ Detailed land nitrogen budgets at European level An overview of the land N budget for all terrestrial nonÂ�agricultural systems (forests and semi-natural vegetations) at the European scale (EU-27) as calculated with INTEGRATOR is given in Table 15.6. For non-agricultural systems, there is no differentiation between land and soil N budgets as all fluxes are related to the soil system. N deposition is derived with an emission deposition matrix, using NOx and non-agricultural NH3 emissions from EMEP and NH3 emission estimates from agriculture by INTEGRATOR as inputs. The N manure input to semi-natural vegetations is mainly due to rough grazing, but it also includes some manure application being calculated in the MITERRA sub-model of INTEGRATOR. For forests, rough

grazing is assumed to be negligible. Net N immobilisation (accumulation) in both forests and semi-natural vegetations is calculated as a fraction of the net N input, which is dependent on the C/N ratio of the soil, using an approach described in De Vries etâ•¯al. (2006). NH3 emissions in forests are background emissions due to wild animals derived from Simpson etâ•¯al. (1999), whereas the NH3 emission from short vegetations is calculated as a fraction of the N manure input by grazing animals. In forests, the estimated N2O, NO and N2 emissions by INTEGRATOR are derived with a statistical relationship with environmental factors based on results of a European wide application of the process oriented biogeochemical model Forest-DNDC (Li etâ•¯al., 2000) by Kesik etâ•¯al. (2005). Apart from this meta-model of Forest-DNDC, INTEGRATOR includes an empirical relationship with various environmental factors, based on hundreds of measurements assessed in the literature (Bloemerts and de Vries, 2009). In short vegetations, the N2O and NO emissions are calculated as a fraction of the N input, using emission factors that are a function of N source, soil type, pH, precipitation and temperature (see Supplementary materials Chapter 15 and 16). Finally, N leaching is assessed by multiplying the net N input by an N leaching factor and N2 emissions

333


to the size of the country. There is also a large uncertainty in the N flax, specifically in the N2O and NOx emissions, as discussed below by comparing results of various model approaches.

Table 15.6 The annual N budget of forest soils and semi-natural vegetation (EU 27) in Europe in 2000, as derived with INTEGRATOR.

N budget (kton N yr−1) Semi-natural vegetations

Total nature

1003

1003

1367

345

1712

271

214

485

1638

1562

3200

Net uptake

302

779

1081

N accumulation

729

−26

703

Source

Forests

Inputs Manure input (grazing) Deposition Biological N fixation Total

—

Outputs

0

Emissions of

0

NH3

21

221

242

N2O

45

37

82

NOx

13

18

31

256

431

687

N2 N leaching + runoff Total

272

113

385

1638

1572

3210

are then calculated as N input minus all N output terms. In forests, N2 emission is already calculated and N leaching is calculated as all NÂ€input minus all N output terms. The results show that while the total N input is comparable in forests and semi-natural vegetations, N deposition dominates the N input in forests, whereas manure input by grazing animals dominates the N input in semi-natural vegetations. This high manure input also causes a much larger NH3 emission in semi-natural vegetations as compared to forests. Compared to semi-natural vegetations, net N uptake and N2 emissions are lower in forests, whereas N accumulation (net N immobilisation) and N leaching are higher. In semi-natural vegetation, net N growth uptake is set equal to N excretion by grazing animals, since these animals continually remove the vegetation, but also excrete nearly the same amount on the field. In percentage of the N surplus (N input minus N uptake), the N leaching and runoff is approximately 20% from forests and 8% from seminatural vegetations, being (much) lower than the default IPCC factor of 30%.

15.4.2â•‡ Nitrogen budgets at country level and regional level N budgets calculated at country level An overview of the N budget for forests for the EU-27 countries, based on INTEGRATOR results, is given in Table 15.7. In this table, removal refers to the net N removal due to wood harvesting and accumulation stands for the N pool change in the soil. Results show large variations in all N fluxes, related partly

334

N2O emissions and NO emissions at country level and regional level A comparison of the results per country by the original Forest-DNDC model with those obtained by the meta-model in INTEGRATOR is presented in Figure 15.9. For regionalisation purposes, Forest-DNDC was coupled to GIS with a resolution of 50â•›km by 50 km holding all relevant information for initialising (soil and forest stand properties) and driving the model (atmospheric input, daily meteorological data). Before application of Forest-DNDC on a regional scale, the model was evaluated for its suitability by applying it to different field sites of the NOFRETE project, which were well distributed across Europe. For further details, we refer to Kesik etâ•¯al. (2005). Results of INTEGRATOR are based on the application of meta-models for N2O and NO from DNDC at NCU level, while making checks on the N balance. We checked whether the N input by deposition and fixation, minus the net N uptake by trees, minus the calculated total N emission and N immobilisation is above a minimum N leaching rate (near zero kg N). If this is not the case, both N emission and N immobilisation are reduced, assuming that these terms are more uncertain than the estimated N deposition and N uptake. Only in cases where zero N emission and N immobiliÂ� sation still leads to a leaching rate below the minimum value, the N fixation is increased. The rationale behind this check is that in low N input systems, where trees take all the N to maintain growth, there is not enough NÂ€available for N emissions, unless there is net N mineralisation (e.g. drained forest on peat soils). The results with the meta-model for N2O are quite comparable with the original DNDC model (Figure 15.9), except for two countries (Sweden and Finland), where the original DNDC model predicted an N2O emission of 11.9 and 10.5 kton N yr−1, whereas the meta-model predicted an N2O emission of 0.7 and 2.3 kton N yr−1, respectively. This large difference is due to the check on the N balance. In these Nordic countries with low N inputs, N is simply not available for large N2O emissions. The results with the meta-model for NOx are generally lower than the original model and this holds again specifically for Sweden and Finland but also for other countries such as Germany and France. Apart from the N balance checks, the differences are also due to the large dependence of the NOx emissions on soil properties, such as pH, being differently used in the INTEGRATOR meta-model application that in the original DNDC model. Regional patterns of the N2O and NO emissions for forests calculated with INTEGRATOR are presented in Figure 15.10. Regional patterns obtained with Forest-DNDC are presented in the supporting material in Chapters 15 and 16, showing higher N2O and NO emissions calculated by Forest-DNDC, as compared to INTEGRATOR, in the Nordic countries for Â�reasons given above.

335

−1

Data given in kton N yr .

135.342

a

EU-27a

1.773

135.342

EU-27

United Kingdom

2.583

Romania

1.895

9.117

Portugal

Sweden

0.306

Poland

1.115

2.658

Netherlands

Spain

0.091

Luxembourg

6.757

1.804

Lithuania

25.247

8.195

Latvia

Slovenia

0.300

Italy

Slovakia

1.717

11.243

France

Ireland

2.036

Finland

3.369

0.358

Hungary

10.261

Denmark

Estonia

19.606

2.549

Czech. Rep.

14.636

3.418

Bulgaria

Germany

0.611

Greece

3.698

Belgium

Area (Mha)

Austria

Country

1367

10.1

10.5

69.8

66.8

0.8

4.1

28.0

1.6

424.3

3.5

20.5

9.9

2.7

433.6

1.4

6.0

1.7

11.7

15.2

37.3

41.8

0.5

18.0

9.4

21.5

16.5

Deposition

271

2.0

2.0

26.7

20.2

0.1

0.6

5.2

0.6

59.6

0.2

2.0

2.0

0.6

54.7

0.3

1.0

0.5

1.0

2.6

19.3

11.4

0.1

2.0

2.0

2.0

2.0

Fixation

1638

12.1

12.5

96.4

87.0

0.8

4.7

33.2

2.2

483.8

3.7

22.5

11.9

3.4

488.3

1.8

7.0

2.2

12.7

17.8

56.5

53.1

0.5

20.0

11.4

23.5

18.5

Total

N input fluxes (kg N ha−1 yr−1)

302

2.2

2.2

20.9

26.9

0.1

0.8

9.7

0.0

40.8

0.4

5.1

1.6

0.8

121.7

0.4

1.2

0.1

1.6

3.4

13.5

13.4

0.1

3.2

0.4

4.1

4.3

Removal

Table 15.7 N budgets calculated at country level for forests with INTEGRATOR for the year 2000

729

5.4

3.2

39.9

23.6

0.4

2.3

10.4

1.3

248.5

2.1

9.9

5.1

1.2

251.2

0.2

2.8

1.2

6.7

7.5

22.4

9.0

0.3

10.9

5.4

11.5

8.1

Accumulation

21

0.15

0.35

0.93

1.12

0.00

0.02

0.36

0.07

5.97

0.05

0.13

0.00

0.00

7.05

0.01

0.11

0.02

0.24

0.19

0.54

0.21

0.03

0.20

0.26

0.37

0.12

Emission NH3

45

0.33

0.13

5.38

4.40

0.00

0.03

1.13

0.23

12.61

0.08

0.39

0.54

0.14

6.79

0.03

0.18

0.06

0.13

0.54

1.91

1.48

0.01

0.02

0.18

0.31

0.03

Emission N2O

13

0.10

0.01

1.09

0.25

0.00

0.00

0.02

0.21

10.57

0.14

0.09

0.27

0.06

0.47

0.00

0.01

0.00

0.14

0.05

0.19

0.21

0.00

0.01

0.00

0.23

0.01

Emission NOx

N output fluxes (kg N ha−1 yr−1)

256

1.9

3.0

26.2

27.4

0.1

0.5

7.1

0.2

36.2

0.0

2.6

1.4

0.4

68.8

0.2

1.3

0.6

0.8

3.0

8.3

8.4

0.0

0.9

2.8

1.9

2.5

Emission N2

272

2.0

3.6

2.0

3.2

0.2

1.0

4.5

0.2

129.2

1.0

4.2

3.0

0.8

32.3

0.9

1.4

0.2

3.1

3.2

9.6

20.4

0.1

4.7

2.3

5.1

3.5

Leaching + Runoff


Figure 15.9 A comparison of country emissions for N2O emissions (left) and NOx emissions (right) from forests in EU 27 for the year 2000, calculated by DNDC, as estimated by Kesik etâ•¯al. (2005), and calculated with INTEGRATOR.

Figure 15.10 Regional pattern for the emissions of N2O (left) and NO (right) from forest soils in EU 27 in the year 2000 as derived with INTEGRATOR. Grey shading in the EU27 denotes non-agricultural areas. Countries outside EU27 are also included by a grey shade.

Nitrogen losses to ground water and surface water The geographic variation in estimated NO3-N leaching and runoff from forest soils and short vegetations (with rough grazing) in 2000, as derived with INTEGRATOR, is shown in Figure 15.11. In line with the high N deposition inputs, N leaching below forests is high in the Netherlands and Germany and low in the Nordic countries and in Spain. In the

336

Nordic countries, N leaching does not reflect the N deposition pattern, mainly due to impacts of temperature. In the north, growth is very limited owing to low temperatures, this leading to extremely low N uptake rates. N leaching from seminatural vegetations reflects the high N manure input regions due to rough grazing, mainly occurring in western UK and central Europe.

Wim de Vries

Figure 15.11 Regional pattern of N leaching and surface runoff to ground water and surface water (left) and semi-natural vegetation (right) from forest soils in EU-27 in the year 2000 as estimated by INTEGRATOR. Grey shading in the EU-27 denotes non-agricultural areas. Countries outside EU-27 are also included by a grey shade.

15.5â•‡ Discussion and conclusions 15.5.1â•‡ Model evaluation Comparability of model results In general, results of various N budget models (INTEGRATOR, IDEAg, MITERRA and IMAGE) in terms of annual N inputs and N fluxes on a European (EU27) wide scale are reasonably comparable for the year 2000. This holds specifically for N fertiliser inputs that are all based on the same source and to a lesser extent for N manure input where livestock sources are mostly comparable, but where N excretion rates differ. Despite the overall comparability, the estimated geographic variation in N inputs differs considerably between models. A comparison of agricultural emissions of NH3-N, N2O-N and NOx-N for all the 27 EU countries as derived with the four complete N budget models and standard activity data-emission factors approaches (UNFCC/IPCC, GAINS, OECD, EDGAR and EMEP) also shows comparable estimates for NH3. The differences in N2O emissions, however, are much larger, while NOx emissions are most uncertain. This holds both on a European wide scale and with respect to the geographic variation in the emissions. Very uncertain are also the N leaching and runoff estimates, which show a very large deviation between models. This holds both for the European wide estimates and for the geographic variation. Most uncertain are also N2 emissions that are often calculated as a rest term from all other N inputs and outputs in

a budget approach. It is important to mention that this seemingly simple compound is almost not measurable and model results are quite speculative as they cannot be validated. The N2 release can be derived from radioactive labelling and there are only a handful of studies focusing on N2 measurements. In view of a complete N budget, it would be worthwhile to put more emphasis on the measurement of N2.

Comparison of results with inverse modelling results for nitrous oxide emissions Inverse modelling is an important tool for regional emission estimates and independent verification of international agreements on emission reductions, such as the United Nations Framework Convention on Climate Change (UNFCCC) and the Kyoto Protocol (IPCC, 2001; Bergamaschi, 2007). Atmospheric measurements combined with inverse atmospheric models can provide independent ‘top-down’ emission estimates of atmospheric trace gases. Inverse modelling has been widely applied for CO2 and CH4 (IPCC, 2007), while only relatively few studies are available for N2O. The first inverse analysis of the global N2O cycle was presented by Prinn etâ•¯al. (1990) based on a 9-box model and atmospheric observations from the ALE-GAGE network for 1978–1988. They concluded that beside the use of fertiliser and fossil fuel combustion in mid latitudes, tropical sources (probably from tropical land use change) are likely to play an important role for the global budget and the observed N2O increase (32%–39% for 1978–1988). The more recent studies of Hirsch etâ•¯al. (2006)

337

Geographical variation in terrestrial nitrogen budgets Table 15.8 N2O emissions for the year 2000 for Ireland and UK and for Western Europe as derived with INTEGRATOR and based on results by the 222Rn tracer method and the inverse model NAME (after Messager etâ•¯al., 2008)

N2O emissions (kg N2O-N ha−1 yr−1)

Area Ireland + UK Western Europe

a

INTEGRATOR

Rn Tracer method

Inverse model NAME

Agriculture

Totalb

8.3–9.8

9.0–11.1

6.8

11.2

6.6–8.9

7.5–10.2

4.7

7.7

222

T he sum of emissions from France, Germany, the Netherlands, Belgium, Luxembourg and United Kingdom. b Total emissions by INTEGRATOR were derived by multiplying the agricultural N2O emissions with the ratio of total/agricultural N2O emissions based on GAINS. a

and Huang etâ•¯al. (2008), based on 3D global inverse models suggest an even larger contribution of the tropical sources between 0 and 30oN. First inverse modelling estimates of European N2O emissions were provided by Ryall etâ•¯al. (2001) and Manning etâ•¯al. (2003), using N2O observations from Mace Head and the NAME Lagrangian particle model. Their estimates for North West European countries showed an agreement within ~30% or better with emissions reported to UNFCCC. Another example is downscaled emissions for parts of Europe based on the NAME model and a model-independent approach using the 222Rn tracer method, presented by Messager etâ•¯al. (2008). A comparison of N2O emissions derived by INTEGRATOR with those estimates is given in Table 15.8. Results show that the comparison is reasonable. It needs to be emphasised, however, that top-down approaches generally estimate total emissions, while emission reported to UNFCCC cover only anthropogenic emissions. Hence, for quantitative comparisons good bottom-up estimates of the natural sources are needed. While the above European top-down emission estimates are based on one single station only (Mace Head), improved emission estimates require the use of further atmospheric measurements, to provide a better coverage of the European domain. Additional continuous N2O measurements are now available from the European RTD project CHIOTTO (‘Continuous HIgh-precisiOn Tall Tower Observations of greenhouse gases’) for 2006, which has set up a European network of tall towers for GHG measurements. The measurements from the CHIOTTO towers and further monitoring stations are currently used in the NitroEurope project to provide European N2O emission estimates using five independent inverse models. A particular challenge constitutes the fact that measurements from different stations / networks may have small calibration offsets, hence requiring sophisticated bias correction procedures in the inverse modelling systems. Results from the NitroEurope inverse modelling will be available early 2011. There are also great opportunities for constraining NH3 or NOx emissions by independent datasets based on wet concentration measurements and satellite measurement (Gilliland etâ•¯al., 2003; Konovalov etâ•¯al., 2010). Whenever such datasets come available, they will be used for independent model validation.

338

Comparison of results with measurements for nitrate concentrations in ground water and N concentrations in surface water Use was made of data on NO3 concentration measurements in groundwater in the period 2000–2003 (EC, 2007) to validate the results of the MITERRA-Europe model. The measurements of NO3 concentration showed that 17% of EU-27 monitoring stations had NO3 concentrations above 50 mg NO3 l−1, 22% were in the range of 25 to 50 mg NO3 l−1 and 61% of the groundwater stations had a concentration below 25 mg NO3 l−1 (EC, 2007). A preliminary validation of the MITERRA model on these NO3 concentration measurements (Velthof etâ•¯al., 2009) showed that the distribution of calculated mean NO3 concentrations in NUTS2 regions of EU-27 according to MITERRAEUROPE agrees very well with the distribution of the means of measured NO3 concentrations in the EU-27. For the year 2000, MITERRA estimates that 16% of the NUTS2 regions had NO3 concentrations above 50 mg NO3 l−1, 20% were in the range of 25 to 50 mg NO3 l−1, and 65% had a concentration below 25 mg NO3 l−1. The calculated NO3 concentrations were also in the same range of the means of measured NO3 concentrations in groundwater bodies. For Belgium, Czech Republic, Denmark, the Netherlands and Poland, the calculated NO3 concentrations appear somewhat higher than the measured NO3 concentrations. Possible reasons for these apparent differences are that monitoring stations measure NO3 concentrations at various depths, while MITERRA-EUROPE estimates NO3 concentration in the soil water at uniform depth (below rooting zone). Moreover, monitoring stations may include forests and natural land, whereas MITERRA-EUROPE only calculates NO3 concentration for agricultural land. Finally, it has to be realised that the model results refer to the NO3 concentration in leachate to ground water and not to the concentrations in ground water as measured in the ground water stations.

15.5.2â•‡ Nitrogen budgets and effects on ecosystems There is an increasing demand by policy makers for easy to interpret and understand indicators that assess the environmental performance and ‘sustainability’ of agriculture. Results presented before thus need to be interpreted in view of possible

Wim de Vries Table 15.9 Variation in number of countries with estimated NH3-N emissions in 2000 exceeding the National emission ceilings for 2010, depending on the model approach

Model

Number of countries exceeded

Percentage countries exceededa

Total exceedance kton NH3-N yr−1

INTEGRATOR

7

28

103

IDEAg

7

28

264

MITERRA

9

36

75

IMAGE

7

29

167

GAINS

10

40

109

EDGAR

18

72

1269

EMEP

14

56

261

OECD

12

63

245

a

T he countries included in the calculation were 25 (EU27 minus Cyprus and Malta) for INTEGRATOR, MITERRA, GAINS, EDGAR and EMEP, 24 for IDEAg and IMAGE and 19 for OECD. The percentage equals the number exceeded divided by these totals.

effects to be of use in policy making. Below, we discuss various options for performance indicators, based on either gross or detailed N budget approaches, including the exceedance of the following. • Maximum N manure inputs and NH3 emission ceilings. Note that these are policy criteria based on impacts but not critical levels related to actual impacts. • Critical NH3 concentrations and critical N loads in view of biodiversity impacts and in view of elevated N saturation of forest soils, associated with damage by plagues and diseases. • Critical NO3 concentrations in ground water in view of health effects and critical N concentrations in surface waters in view of eutrophication of terrestrial ecosystems. The assessment is focused on the year 2000. Trends in the changes of risks can be derived from the trends in N fluxes since 1970, as presented earlier.

Nitrogen surpluses and manure nitrogen inputs as performance indicators In the Pan European initiative, SEBI2010, which stands for Streamlining European 2010 Biodiversity Indicators, the agricultural N balance (implying the N surplus) is one of the 26 indicators that are developed to monitor progress towards the European target to halt the loss of biodiversity by 2010 (see: http://biodiversity-chm.eea.europa.eu/information/indicator/ F1090245995). The N surplus is, however, a typical pressure indicator and not an effect indicator, since agro-ecosystems and environment both have a strong impact on the actual N emissions to the atmosphere and the N (NH4 and NO3) concentrations in leaching and runoff water, being relevant for the effects that may occur. For example, ammonia losses from agriculture are associated predominantly with animal production systems. Nitrate concentrations in the leachate to groundwater depend not only on N balance (N surplus) but also on climate (excess rainfall which dilutes the concentration), and soil type, affecting denitrification. As a result, the relationship between N surplus and N fluxes to the air and to water is diffuse.

Because of this complexity and variability, there are very few common and accepted reference levels against which to evaluate nutrient surpluses. In the Netherlands, the regulatory policy instrument MINAS has been used in the past in which reference values for N surpluses have been set tentatively at 60 and 100 kg per ha for arable land on sandy soils and clayey soils, respectively, and at 140 and 180 kg per ha for grassland on sandy soils and clayey soils, respectively. At present, N surplus is not used as a performance indicator in policy making. Instead, use is made of a maximum N application rate by animal manure of 170 kg N with exceptions (so-called derogations) of 250 kg N for the Netherlands and 230 kg N for Denmark, Germany and Austria (after the year 2000). Maps of the N input by animal manure for the year 2000 (Figure 15.4) indicate that there still exist a number of areas in Europe where this limit is exceeded.

Ammonia emission and related ammonia concentrations and nitrogen deposition as performance indicators The variation in NH3 emissions will affect the N deposition on terrestrial ecosystems. Plant species diversity of terrestrial ecosystems is affected largely by N deposition and in this context empirical and model based critical N loads have been derived. Specifically in intensive livestock areas with high NH3 emissions, the resulting N deposition may lead to an exceedance of critical N loads. In this context, national emission ceilings (NEC) have been set. A comparison of NECs for 2010 (EEA, 2010) and results of total NH3 emissions by the various models described in this chapter is given in Table 15.9. For INTEGRATOR, IDEAg, MITERRA, and IMAGE, the estimated agricultural NH3 emissions per country were multiplied by a factor 1.07, since approximately 7% of the NH3 emissions come from non-agricultural sources. The number of countries with estimated NH3-N emissions in 2000 exceeding the National emission ceilings for 2010 depends on the model approach and varied between 7 and 18, while the total exceedance varied between 75 and 1269 kton NH3-N yr−1. The large exceedances derived by EDGAR are clearly deviating from all other model approaches. The lowest emission exceedances

339


diversity, despite the limited emission reductions in NH3 (see also Figure 15.8 lower graph for the period 1980–2000). This effect is specifically due to NOx emission reductions in that period.

Nitrogen leaching and nitrogen runoff as performance indicators

Figure 15.12 A comparison of the estimated national emissions and national emission ceilings for NH3, derived with INTEGRATOR and various other model approaches (IDEAg, MITERRA, IMAGE, GAINS, EDGAR, EMEP and OECD/IPCC) for the year 2000.

are estimated by INTEGRATOR, MITERRA and GAINS, all being based on the same animal numbers and NH3 emission factors. The variation in NH3-N emission exceedances, limited to those countries where all models calculate an exceedance is illustrated in Figure 15.12. For most countries, the exceedance is comparable, but for some countries the variation is considerable up to a fourfold variation. Insight in the actual risk of elevated NH3 emissions on terrestrial ecosystems can amongst others be derived by comparing either the actual NH3 concentration with a critical NH3 concentration in view of plant species diversity impacts. Recently updated critical levels are 1 µg.m−3 for lichens and bryophytes and 3 µg.m−3 for herbaceous plants (Cape etâ•¯al., 2009). A comparison of EMEP model predicted NH3 concentrations with these critical levels during the last 15 years show that NH3 concentrations violate the limit for lichens and bryophytes except for Fennoscandia and Scotland, as presented in Moldanová etâ•¯al., 2011 (Chapter 18, this volume). The limit for herbaceous plants is also exceeded in parts of Western Europe and Northern Italy. Indirectly, insight in the actual risk of elevated NH3 emissions on terrestrial ecosystems can also be derived by comparing present N depositions, which are largely determined by NH3 emissions together with NOx emissions, with the critical N deposition at the European scale. The critical N deposition is related to impacts on plant species diversity and is either derived from empirical field data or by model assessments, as discussed in Dise etâ•¯al., 2011 (Chapter 20 this volume). The exceedance of critical N loads in view of impacts on plant species diversity is one of the 26 performance indicators in SEBI 2010. A comparison of exceedances of critical N loads in 1980 and in 2010 is given in Dise etâ•¯al., 2011 (Chapter 20 this volume), showing that the N emission reductions in the past three decades has led to a significant reduction in the risk of N affecting plant species

340

Critical NO3 concentrations in ground water in view of health effects and critical N concentrations in surface waters in view of eutrophication of aquatic ecosystems are also important targets to evaluate the N leaching and N runoff fluxes on a European wide scale. A critical NO3 concentration in view of health impacts is set at 50 mg NO3â•›l−1. Eutrophication is the result of nutrient (both N and P) enrichment in the aquatic system, but the severity of the phenomenon largely depends on the specific regional characteristics, climate, morphology, water residence time, nutrients ratio, tropic web status, and generally on the ecosystem resilience. Therefore, similar nutrient loads may produce different effects in reason of the regional sensitivities. Similarly, the impacts are related not only to N loads, but rather to its specific synergies with the availability of other elements, such as carbon, phosphorus and silica (see also Billen etâ•¯al., 2011; Grizzetti etâ•¯al., 2011, Chapters 13 and 17 this volume). Nevertheless, N concentrations in surface waters, being a major driving force of the problems, are used as a proxy to evaluate the risk for water eutrophication. A critical limit of 0.5–1.0 mg N l−1 has been proposed by Camargo and Alonso (2006) based on an extensive study on the ecological and toxicological effects of inorganic N pollution in aquatic ecosystems. At present, N concentrations are generally exceeding those limits (see also Grizzetti etâ•¯al., 2011, Chapter 17 this volume).

15.5.3â•‡ Conclusions and recommendations Key findings regarding the temporal and geographic variation in N budgets in agricultural and other terrestrial ecosystems over Europe are as follows. • Trends in N fluxes in agro-ecosystems since 1970 show an increase in N inputs by fertilisers and manure up to 1985, followed by a decrease since 1985 in response to a change in crop production and in animal numbers. Actually, livestock decreased since 1970, but in the period 1970– 1985 the N input by manure excretion still increased due to an increase in milk production and related N excretion rates. • For EU-27, the models estimates a total N input in European agriculture for the year 2000 of 23.3–25.7 Mton N yr−1 which is mainly due to fertiliser and animal manure inputs and to a lesser extent by atmospheric deposition and N fixation. Total N inputs at EU-27 level are comparable for all models, since they all use comparable basic data on fertiliser use and animal numbers. There exist a number of areas in Europe where a maximum N application rate by animal manure of 170 kg N is exceeded. The N uptake varies from 11.3–15.4 Mton N yr−1 leading to total N surpluses varying from 10.4–13.2 Mton N yr−1 at EU-27 level.

Wim de Vries

• The four complete N budget models for agro-ecosystems give in general very similar results for the emissions of NH3 (2.8–3.1 Mton N yr−1) and N2O (0.33–0.43 Mton N yr−1) but vary largely for NOx (0.02–0.23 Mton N yr−1). Similar results and differences are found when including standard activity data-emission factors approaches (UNFCC/IPCC, GAINS, OECD, EDGAR and EMEP). • Even though NOx emissions are more uncertain, the uncertainty in the NH3 emissions is more important for the overall uncertainty in the reactive N budget, since NOx contribute little to the overall N budget. The contribution of agriculture to total NOx emissions is less than 5%, while the contribution of agricultural NH3 emissions is more than 90%, making the variation in NH3 emissions more important. The uncertainty is illustrated by the number of countries with estimated NH3-N emissions in 2000 exceeding the National emission ceilings for 2010. Depending on the model approach, this number varies between 7 and 18, while the total exceedance varied between 75 and 1269 kton NH3-N yr−1. • The estimated sum of N leaching and runoff at EU 27 is roughly equal to the sum of NH3, N2O and NOx emissions to the atmosphere, but estimates vary by a factor two, from 2.7–6.3 Mton N yr−1. This strongly affects the area with N concentrations exceeding critical N concentrations in surface water. • In non-agricultural system (forests and semi-natural vegetation), the estimated total input is near 3.2 Mton N yr−1, while the net N uptake is near 1.1 Mton N yr−1, leading to a surplus near 2.1 Mton N yr−1. Compared to agricultural systems, the estimated N fluxes in non-agricultural systems are about 5 times lower for N2O emissions and 10 times lower for NOx and NH3 emissions and for the sum of N leaching and runoff. • The regional variation in N fluxes is mainly determined by N inputs, being highest in areas with high livestock density and intensive agricultural crop production areas, while land/soil characteristics and climate are secondary factors influencing the magnitude of N fluxes. Recommendations that can be made based on this assessment are as follows. • Future research priorities should focus on major uncertainties, in particular N2O emissions and N leaching and runoff from agricultural ecosystems. Furthermore, studies on denitrification are needed to reduce the large uncertainty in this process at the European scale. • A database should be set up of N contents in various plants and in various regions to improve estimates of N uptake and N surplus at the European scale. • Information on NH3 concentrations in air should be used in inverse modelling approaches to derive independent datasets to validate the various NH3 emission calculations. • A European-wide monitoring network of ground- and surface water, using standardised methods and covering a range of habitats, should be initiated to provide consistent

and reliable information on the long-term effects of air pollution on water quality, to be used for validation of N budget models. • It is relevant that data use is harmonised for models predicting air emissions and N loss to waters for consistent environmental decision-making relevant to air quality, ecosystem deposition and water quality.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope IP (funded by the European Commission), the COST Action 729. We also thankfully acknowledge Suvi Monni (JRC) for sending updated EDGAR emission data.

Supplementary materials Supplementary materials (as referenced in the chapter) are available online through both Cambridge University Press: www.cambridge.org/ena and the Nitrogen in Europe website: www.nine-esf.org/ena.

References Alcamo, J. (ed.) (1994). IMAGE 2.0. Integrated modeling of global climate change. Water, Air and Soil Pollution, 76, 1–318. Bergamaschi, P. (ed.) (2007). Atmospheric Monitoring and Inverse Modelling for Verification of National and EU Bottom-up GHG InventoriesÂ€– report of the workshop ‘Atmospheric Monitoring and Inverse Modelling for Verification of National and EU Bottom-up GHG Inventories’ under the mandate of Climate Change Committee Working Group I, Casa Don Guanella, Ispra, Italy (08–09 March 2007). European Commission Joint Research Centre, Institute for Environment and Sustainability. Billen, G., Silvestre, M., Grizzetti, B. etâ•¯al. (2011). Nitrogen flows from European regional water sheds to coastal marine waters. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.Â€W. Erisman etâ•¯al., Cambridge University Press. Bloemerts, M. and de Vries, W. (2009). Relationships between Nitrous Oxide Emissions from Natural Ecosystems and Environmental Factors. Alterra Wageningen UR, Wageningen, The Netherlands. Bouraoui, F. and Aloe, A. (2007). European Agrochemicals Geospatial Loss Estimator:Â€Model Development and Application, EUR Report 22690 EN. Office for Official Publications of the European Communities, Luxembourg. Bouraoui, F., Grizzetti, B. and Aloe, A. (2009). Nutrient Discharge from River to Seas for Year 2000. Office for Official Publications of the European Communities, Luxembourg. Bouwman, A. F., van Drecht, G. and van der Hoek, K. W. (2005). Nitrogen surface balances in intensive agricultural production systems in different world regions for the period 1970–2030. Pedosphere, 15, 137–155. Britz, W. (ed.) (2005). CAPRI Modelling System Documentation. Common Agricultural Policy Regional Impact Analysis: “Development of a Regionalised EU-wide Operational Model to Assess the Impact of Current Common Agricultural Policy on Farming Sustainability”. J05/30/2004Â€– Deliverable 1, Bonn. Britz, W., Heckelei, T. and Kempen, M. (2005). Description of the CAPRI Modeling System, final report of the CAPRI-DynaSpat

341

Geographical variation in terrestrial nitrogen budgets Project. Institute for Food and Resource Econommics, University of Bonn, Bonn, Germany. Britz, W. and Leip, A. (2009a). Development of marginal emission factors for N losses from agricultural soils with the DNDC-CAPRI meta-model. Agriculture, Ecosystems and Environment, 133, 267–279. Britz, W. and Leip, A. (2009b). Development of marginal emission factors for N losses from agricultural soils with the DNDC-CAPRI meta-model. Agriculture Ecosystems and Environment, 133, 267–279. Butterbach-Bahl, K., Kahl, M., Mykhayliv, L. etâ•¯al. (2009). A European-wide inventory of soil NO emissions using the biogeochemical models DNDC/Forest-DNDC. Atmospheric Environment, 43, 1392–1402. Butterbach-Bahl, K., Nemitz, E., Zaehle, S. etâ•¯al. (2011). Nitrogen as a threat to the European greenhouse balance. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W.Â€Erisman etâ•¯al., Cambridge University Press. Camargo, J. A. and Alonso, A. (2006). Ecological and toxicological effects of inorganic nitrogen pollution in aquatic ecosystems:Â€a global assessment. Environment International, 32, 831–849. Cape, J. N., Van der Eerden, L. J., Sheppard, L. J., Leith, I. D. and Sutton, M. A. (2009). Evidence for changing the critical level for ammonia. Environmental Pollution, 157, 1033–1037. CAPRI (2010). http://www.ilr1.uni-bonn.de/agpo/rsrch/capri/ capri_e.htm De Vries, W., Reinds, G. J., Gundersen, P. and Sterba, H. (2006). The impact of nitrogen deposition on carbon sequestration in European forests and forest soils. Global Change Biology, 12, 1151–1173. De Vries, W., Kros, J., Voogd, J. C. etâ•¯al. (2009). Comparing Predictions of Nitrogen and Greenhouse Gas Fluxes in Response to Changes in Live stock, Land Cover and Land Management Using Models at a National, European and Global Scale. Alterra Wageningen UR, Wageningen, The Netherlands. De Vries, W., Kros, J., Reinds, G. J., Wieggers, R., Velthof, G., Oudendag, D., Oenema, O., Nabuurs, G. J., Schelhaas, M. J., Perez Soba, M., Rienks, W., de Winter, W., van den Akker, J., Bakker, M., Verburg, P., Eickhout, B. and Bouman, L. (2010). INTEGRATOR:Â€A Modelling Tool for European-wide Assessments of Nitrogen and Greenhouse Gas Fluxes in Response to Changes in Land Cover, Land Management and Climate – Calculation Procedures, Application Methodology and Examples of Scenario Results. Alterra, Wageningen. Alterra Report (in preparation). De Vries, W., Leip, A., Reinds, G. J. etâ•¯al. (2011). Comparison of land nitrogen budgets for European agriculture by various modeling approaches. Environmental Pollution (in press). Dise, N. B., Ashmore, M., Belyazid, S. etâ•¯al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. EC (2007). Report from the Commission to the Council and the European Parliament on Implementation of Council Directive 91/676/EEC Concerning the Protection of Waters against Pollution Caused by Nitrates from Agricultural Sources for the Period 2000– 2003. Commission of the European Communities, Brussels. EDGAR (2010). http://www.mnp.nl/edgar/model/v32ft2000edgar/ EEA (2005). Agriculture and Environment in EU15: The IRENA Indicator Report. European Environment Agency, Copenhagen. EEA (2008). Annual European Community Greenhouse Gas inventory 1990–2006 and Inventory Report 2008, Submission to the UNFCCC Secretariat. European Environment Agency, Copenhagen.

342

EEA (2010). http://www.eea.europa.eu/highlights/more-eu-memberstates-to-miss-2010-air-pollutant-limits/nec-status-preliminaryresults-2008-data.pdf EMEP (2009). Status Report 1/09 Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe in 2007. Joint MSC-W & CCC and CEIP Report, Norwegian Meteorological Institute, Oslo, Norway. (available through www.emep.int). EMEP (2010a). http://www.emep.int EMEP (2010b). http://www.ceip.at/ GAINS (2010) http://www.iiasa.ac.at/web-apps/apd/gains/EU/index. login?logout=1 Gilliland, A. B., Dennis, R. L., Roselle, S. J. and Pierce, T. E. (2003). Seasonal NH3 emission estimates for the eastern united states based on ammonium wet concentrations and an inverse modeling method. Journal of Geophysical Research, 108, 4477. Grizzetti, B., Bouraoui, F., de Marsily, G. and Bidoglio, G. (2005). A statistical method for source apportionment of riverine nitrogen loads. Journal of Hydrology, 304, 302–315. Grizzetti, B., Bouraoui, F. and Aloe, A. (2007). Spatialised European Nutrient Balance. Report EUR 22692 EN. Grizzetti, B., Bouraoui, F. and De Marsily, G. (2008). Assessing nitrogen pressures on European surface water. Global Biogeochemical Cycles, 22, GB4023, doi:10.1029/2007GB003085. Grizzetti, B., Bouraoui, F., Billen, G. etâ•¯al. (2011). Nitrogen as a threat to European water quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Hirsch, A. I., Michalak, A. M., Bruhwiler, L. M. etâ•¯al. (2006). Inverse modeling estimates of the global nitrous oxide surface flux from 1998–2001. Global Biogeochemical Cycles, 20, GB1008, doi:Â€10.1029/2004GB002443. Höglund-Isaksson, L. and Mechler, R. (2005). The GAINS Model for Greenhouse Gases, Version 1.0:Â€Methane (CH4), Interim Report IR-05–54. International Institute for Applied Systems Analysis. Huang, J., Golombek, A., Prinn, R. etâ•¯al. (2008). Estimation of regional emissions of nitrous oxide from 1997 to 2005 using multinetwork measurements, a chemical transport model, and an inverse method. Journal of Geophysical Research 113, doi:10.1029/2007JD009381. IMAGE (2010), http://www.mnp.nl/en/themasites/image/ index.html IPCC (1997). Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories. UK Meteorological Office, Bracknell, UK. IPCC (2001). Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories. Published for the IPCC by the Institute for Global Environmental Strategies, Japan. IPCC (2006). 2006 IPCC Guidelines for National Greenhouse Gas Inventories, prepared by the National Greenhouse Gas Inventories Programme, Eggleston, ed. H. Simon etâ•¯al. Institute for Global Environmental Strategies, Japan. IPCC (2007). Climate Change 2007:Â€The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press. Kesik, M., Ambus, P., Baritz, R. etâ•¯al. (2005). Inventories of N2O and NO emissions from European forest soils. Biogeosciences, 2, 353–375.

Wim de Vries Konovalov, I. B., Beekmann, M., Richter, A., Burrows, J. P. and Hilboll,Â€A. (2010). Multi-annual changes of NOx emissions in megacity regions:Â€nonlinear trend analysis of satellite measurement based estimates. Atmospheric Chemistry and Physics Discussions, 10, 10925–10968. Leemans, R., Kreileman, E., Zuidema, G. etâ•¯al. (1998). The IMAGE User Support System:Â€Global Change Scenarios from IMAGE 2.1. National Institute for Public Health and the Environment, Bilthoven The Netherlands. Leip, A., Marchi, G., Koeble, R. etâ•¯al. (2008). Linking an economic model for European agriculture with a mechanistic model to estimate nitrogen and carbon losses from arable soils in Europe. Biogeosciences, 5, 73–94. Leip, A., Weiss, F. and Britz, W. (2009). Agri-environmental nitrogen indicators for EU27. In: Proceedings of the Conference on Integrated Assessment of Agriculture and Sustainable Development:Â€Setting the Agenda for Science and Policy (AgSAP 2009), Egmond aan Zee, The Netherlands, 10–12 March 2009, ed. M. K. Van Ittersum, J. Wolf and H. H. Van Laar. Wageningen University and Research Centre, Wageningen, The Netherlands, pp. 184–185. Leip, A., Achermann, B., Billen, G. etâ•¯al. (2011a). Integrating nitrogen fluxes at the European scale. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Leip, A., Britz, W., Weiss, F. and de Vries, W. (2011b). Farm, land, and soil nitrogen budgets for agriculture in Europe. Environmental Pollution (in press). Li, C., Aber, J., Stange, F., Butterbach-Bahl, K. and Papen, H. (2000). A process oriented model of N2O and NO emissions from forest soils:Â€I. Model development. Journal of Geophysical Research, 105, 4369–4384. Manning, A. J., Ryall, D. B., Derwent, R. G., Simmonds, P. G. and O’Doherty, S. (2003). Estimating European emissions of ozonedepleting and greenhouse gases using observations and a modeling back-attribution technique. Journal of Geophysical Research, 108, 4405. Messager, C., Schmidt, M., Ramonet, M. etâ•¯al. (2008). Ten years of CO2, CH4, CO and N2O fluxes over Western Europe inferred from atmospheric measurements at Mace Head, Ireland. Atmospheric Chemistry and Physics Discussions, 8, 1191–1237. MNP (2006). Integrated Modelling of Global Environmental Change: An Overview of IMAGE 2.4. Netherlands Environmental Assessment Agency (MNP), Bilthoven, The Netherlands. Moldanová, J., Grennfelt, P., Jonsson, Å. etâ•¯al. (2011). Nitrogen as a threat to European air quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Mosier, A., Kroeze, C., Nevison, C. etâ•¯al. (1998). Closing the global N2O budget:Â€nitrous oxide emissions through the agricultural nitrogen cycle (OECD/IPCC/IEA phase II development of IPCC guidelines for national greenhouse gas inventory methodology). Nutrient Cycling in Agroecosystems, 52, 225–248. Neeteson, J. J. (2000). Nitrogen and phosphorus management on Dutch dairy farms:Â€legislation and strategies to meet the regulations. Biology and Fertility of Soils, 30, 566–572. OECD (2001). Environmental Indicators for Agriculture, Volume 3, Methods and Results. Organisation for Economic Co-operation and Development, Paris, France. OECD (2007). Environmental Indicators for Agriculture, Volume 4, Organisation for Economic Co-operation and Development, Paris, France.

OECD (2010). http://stats.oecd.org/wbos/Index.aspx/datasetcode= ENVPERFINDIC_TAD_2008 and http://www.oecd.org/document /56/0,3343,en_2649_33793_40374392_1_1_1_1,00.html Oenema, O., Boers, P. C. M., van Eerdt, M. M. etâ•¯al. (1998). Leaching of nitrate from agriculture to groundwater:Â€the effect of policies and measures in the Netherlands. Environmental Pollution, 102, 471–478. Oenema, O., Kros, H. and de Vries, W. (2003). Approaches and uncertainties in nutrient budgets:Â€implications for nutrient management and environmental policies. European Journal of Agronomy, 20, 3–16. Oenema, O., Oudendag, D. and Velthof, G. L. (2007). Nutrient losses from manure management in the European Union. Livestock Science, 112, 261–272. Orlandi, S. and Van der Goot, E. (2003). Technical Description of Interpolation and Processing of Meteorological Data in CGMS, European Commission, DG JRC, Agrifish. Unit, http://mars.jrc.it/ mars/content/download/640/4574/file/GridWeather.doc OSPARCOM (1994). OSPARCOM Guidelines for Calculating Mineral Balances. Working Group on Nutrients, NUT 94/8/1-E, Bern, Switzerland. Prinn, R., Cunnold, D., Rasmussen, R. etâ•¯al. (1990). Atmospheric emissions and trends of nitrous oxide deduced from 10 years of ALEGAGE data. Journal of Geophysical Research, 95, 18369–18385. Ryall, D. B., Derwent, R. G., Manning, A. J., Simmonds, P. G. and O’Doherty, S. (2001). Estimating source regions of European emissions of trace gases from observations at Mace Head. Atmospheric Environment, 35, 2507–2523. Simpson, D., Winiwarter, W., Börjesson, G. etâ•¯al. (1999). Inventorying emissions from nature in Europe. Journal of Geophysical Research, 104, 8113–8152. Simpson, D., Fagerli, H., Jonson, J. E. etâ•¯al. (2003). Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe PART I. Unified EMEP Model Description. Norwegian Meteorological Institute, Oslo, Norway. Simpson, D., Butterbach-Bahl, K., Fagerli, H. etâ•¯al. (2006). Deposition and emissions of reactive nitrogen over European forests:Â€a modelling study. Atmospheric Environment, 40, 5712–5726. Simpson, D., Aas, W., Bartnicki, J. etâ•¯al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. UNFCCC (2010). www.unfccc.int UNFCC/IPCC (2010). http://unfccc.int/ghg_data/ghg_data_unfccc/ items/4146.php Van Aardenne, J. A. (2002). Uncertainties in emission inventories. PhD thesis, Wageningen University, Wageningen. Van der Velde, M., Bouraoui, F. and Aloe, A. (2009). Pan-European regional-scale modelling of water and N efficiencies of rapeseed cultivation for biodiesel production. Global Change Biology, 15, 24–37. Velthof, G., Oudendag, D. and Oenema, O. (2007). Development and Application of the Integrated Nitrogen Model MITERRA-EUROPE: Alterra report 1663.1. Alterra, Wageningen, The Netherlands. Velthof, G. L., Oudendag, D. A. and Witzke, H. P. (2009). Integrated assessment of nitrogen emission losses from agriculture in EU-27 using MITERRA-EUROPE. Journal of Environmental Quality, 38, 1–16. Velthof, G., Barot, S., Bloem, J. etâ•¯al. (2011). Nitrogen as a threat to European soil quality. In:Â€The European Nitrogen Assessment,

343

Geographical variation in terrestrial nitrogen budgets ed. M. A. Sutton, C. M. Howard, J. W. Erisman etâ•¯al., Cambridge University Press. Vieno, M., Dore, A. J., Wind, P. etâ•¯al. (2009). Application of the EMEP Unified Model to the UK with a horizontal resolution of 5 × 5 km2 Atmospheric Ammonia. In: Detecting Emissions Changes and Environmental Impacts, ed. M. A. Sutton, S. Reis and S.Â€M.Â€Baker, Springer, New York, pp. 367–372.

344

Winiwarter, W. (2005). The GAINS Model for Greenhouse Gases – Version 1.0:Â€Nitrous Oxide (N2O). IIASA Interim Report IR-05–55, International Institute for Applied Systems Analysis. Witzke, H. P. and Oenema, O. (2007). Assessment of Most Promising Measures Task 3 Service Contract “Integrated Measures in Agriculture to Reduce Ammonia Emissions”. Alterra, Wageningen, The Netherlands.

Chapter

16

Integrating nitrogen fluxes at the European scale Lead author: Adrian Leip Contributing authors: Beat Achermann, Gilles Billen, Albert Bleeker, Alexander F. Bouwman, Wim de Vries, Ulli Dragosits, Ulrike Döring, Dave Fernall, Markus Geupel, Jürg Herolstab, Penny Johnes, Anne Christine Le Gall, Suvi Monni, Rostislav Nevečeřal, Lorenzo Orlandini, Michel Prud’homme, Hannes I. Reuter, David Simpson, Guenther Seufert, Till Spranger, Mark A. Sutton, John van Aardenne, Maren Voß and Wilfried Winiwarter

Executive summary Nature of the problem • Environmental problems related to nitrogen concern all economic sectors and impact all media:Â€atmosphere, pedosphere, hydrosphere and anthroposphere. • Therefore, the integration of fluxes allows an overall coverage of problems related to reactive nitrogen (Nr) in the environment, which is not accessible from sectoral approaches or by focusing on specific media.

Approaches • This chapter presents a set of high resolution maps showing key elements of the N flux budget across Europe, including N2 and Nr fluxes. • Comparative nitrogen budgets are also presented for a range of European countries, highlighting the most efficient strategies for mitigating Nr problems at a national scale. A new European Nitrogen Budget (EU-27) is presented on the basis of state-of-the-art Europe-wide models and databases focusing on different segments of Europe’s society.

Key findings • From c. 18 Tg Nr yr−1 input to agriculture in the EU-27, only about 7 Tg Nr yr−1 find their way to the consumer or are further processed by industry. • Some 3.7 Tg Nr yr−1 is released by the burning of fossil fuels in the EU-27, whereby the contribution of the industry and energy sectors is equal to that of the transport sector. More than 8 Tg Nr yr−1 are disposed of to the hydrosphere, while the EU-27 is a net exporter of reactive nitrogen through atmospheric transport of c. 2.3 Tg Nr yr−1. • The largest single sink for Nr appears to be denitrification to N2 in European coastal shelf regions (potentially as large as the input of mineral fertilizer, about 11 Tg N yr–1 for the EU-27); however, this sink is also the most uncertain, because of the uncertainty of Nr import from the open ocean.

Major uncertainties • National nitrogen budgets are difficult to compile using a large range of data sources and are currently available only for a limited number of countries. • Modelling approaches have been used to fill in the data gaps in some of these budgets, but it became obvious during this study that further research is needed in order to collect necessary data and make national nitrogen budgets inter-comparable across Europe. • In some countries, due to inconsistent or contradictory information coming from different data sources, closure of the nitrogen budget was not possible.

Recommendations • The large variety of problems associated with the excess of Nr in the European environment, including adverse impacts, requires an integrated nitrogen management approach that would allow for creation and closure of N budgets within European environments. • Development of nitrogen budgets nationwide, their assessment and management could become an effective tool to prioritize measures and prevent unwanted side effects.


345

Integrating nitrogen fluxes at the European scale

16.1â•‡ Introduction The concept of the nitrogen cascade was introduced to describe the ‘[…] multiple linkages among the ecological and human effects of reactive nitrogen molecules as they move from one environmental system to another’ (Galloway etÂ€al., 2003). The quantification of the nitrogen cascade requires accurate estimation of the fluxes across the sectoral and media boundaries for a large geographic entity, from regional to national, continental and global scale. Such a complete nitrogen (N)-budget was first presented for Europe by van Egmond etÂ€al. (2002). In this chapter, an update is given of the information on complete N-budgets, including all major N2 and reactive nitrogen (Nr) fluxes, both at country and at continental level. The European Nitrogen Assessment (ENA) provides an overview of the processes and pathways associated with the cascade of Nr through the environment, and also the order of magnitude of the associated problems, based on recent scientific literature and latest available model results. Each of the chapters focuses on one specific sector (for example de Vries etÂ€al., Chapter 15, on the N fluxes from agriculture and natural ecosystems and Svirejeva-Hopkins etÂ€al., Chapter 12, on the effect of urbanization), on one specific medium (such as for example Simpson etÂ€ al., Chapter 14, on the transport of Nr in the atmosphere and Billen etÂ€al., Chapter 13, looking at nitrogen from the perspective of watersheds or on one specific aspect in the nitrogen cascade (for example the transformation processes in soils in Butterbach-Bahl etÂ€ al., Chapter 6, or chemical reactions occurring in the atmosphere in Hertel etÂ€al., Chapter 9). The inter-connections between these specific assessments are manifold and reflect the interactions that nitrogen undergoes in the environment across the borders of scales, sectors and media. The present chapter stands at the interface between the sections describing nitrogen issues and those explaining nitrogen problems and suggesting nitrogen solutions. Two objectives are identified. (i) To give an overview of the most important N-fluxes in Europe in a gridded representation, i.e. compiling a number of ‘key maps’ that help the understanding of regional differences of the main N-indicators. (ii) To show aggregated fluxes of nitrogen across media and sectors, i.e. integrated national N-budgets for those countries where they have already been established and a new ‘European Nitrogen Budget’ based on the evidence compiled and the filling in of gaps according to the latest scientific knowledge. The second goal, in particular, is a challenging one, as most research is done in individual disciplines and only a few ‘integrated’ models exist today that are able to give a comprehensive overview of the nitrogen budget at a national scale. However, while some decades ago the focus of research was on individual fluxes (e.g. nitrate concentrations in rivers), and specific tools (e.g. models of N2O fluxes from agricultural soils), in recent

346

years progress has been made in developing tools and databases which cross sector- and media-boundaries and are able to consider effects such as ‘pollution swapping’ and to evaluate tradeoffs. The need to mitigate environmental problems related to nitrogen in an integrated way has led to the development of ‘national nitrogen budgets’ aimed at helping to find the most efficient and cost-effective solutions to abate these problems. Still, the establishment of a nitrogen budget requires (i)Â€the co-operation of experts from different disciplines and/or (ii)Â€ the integration of various dedicated models. Nitrogen budgeting at the national scale often relies on the first solution, as the density of experimental observations at the national scale might be sufficient to come up with good estimates of nitrogen fluxes between sub-systems, and model-results can (if needed) be used to fill gaps. Such national or (in the case of Europe) supra-national nitrogen budgets are increasingly recognized to be a very useful tool for visualizing the complexity of nitrogen issues, also in relation to society as a whole. These national nitrogen budgets help to support prioritization of policies and provide a first assessment of the impact a policy might have at various points in the nitrogen cascade. In order to understand the fluxes that are included in the establishment of European and national nitrogen budgets, we employ a number of models covering partial aspects of an overall budget. Each model has its strengths and weaknesses. Therefore combining the best features of these models provides a means for the construction of a cross-sector, crossmedia European Nitrogen Budget. The following sections give an overview of the main data sources and models used for the key-maps presented in Section 16.3 as well as for the European Nitrogen Budget (for the EU-27) presented in Section 16.4, where we also present national N-budgets developed by country experts on the basis of national data sources.

16.2â•‡ Data sources For the assessment of key nitrogen fluxes at the European scale and the development of a European Nitrogen Budget, one cannot rely on statistical or observational data as they do not exist for most of the fluxes that need to be considered. Observational data are scarce and unevenly distributed over the European area, so that a statistical up-scaling is often not possible. Instead, models are needed that extrapolate nitrogen (and other) fluxes at large scales on the basis of existing environmental or statistical information. Here, we make use of the results of such models. The models were selected on the basis of the following criteria:Â€(i) applicability at the European scale; (ii) a high spatial data resolution, the European scale notwithstanding; and (iii) a focus in the parameterization of nitrogen fluxes in the compartments considered (see also de Vries etÂ€al., 2011, ChapterÂ€15 this volume). The CAPRI-DNDC-based Integrated Database for European Agriculture (IDEAg) gives currently the most complete information on the flow of nitrogen into and through the agricultural sector in Europe, calculating also reactive nitrogen and greenhouse gas fluxes (Leip etÂ€al., 2008; Leip etÂ€al., 2010a). The Indicator Database for European Agriculture builds mainly

Adrian Leip

on the results from the economic model for agriculture CAPRI (Britz and Witzke, 2008) and the biophysical model for soil nutrient turnover DNDC (Li, 2000) through a meta-modelling approach (Britz and Leip, 2009). IDEAg covers all nitrogen fluxes related to agricultural activities in Europe. It has recently been extended to cover also Nr emissions from sewerage systems in accordance with the methodology developed for the IMAGE model (Bouwman etÂ€al., 2006). The INTEGRATOR model is an integrated model specifically designed to help developing integrated policies. It has been developed to assess responses of nitrogen and greenhouse gas (GHG) emissions to European-scale changes in land use, land management and climate. INTEGRATOR links modules calculating N and GHG emissions from housing and manure storage systems, agricultural and non-agricultural soils and surface waters, while accounting for the interaction between different sources through an emission–deposition model for NH3 and NOx. It uses relatively simple and transparent model calculations based on the use and adaptation of available model approaches, including empirical model approaches and statistical relations between model outputs and environmental variables. The model focuses on the derivation of high resolution spatially explicit data (De Vries etÂ€al., 2009). The Emission Database for Global Atmospheric Research (EDGAR) calculates emissions of air pollutants and greenhouse gases on a grid for use in atmospheric circulation models covering all relevant anthropogenic emission sectors (Olivier etÂ€al., 2005; Van Aardenne etÂ€al., 2001). EDGAR was used as the standard database for deriving emission estimates as it provides a consistent emission calculation of emissions for the whole territory considered and a sophisticated downscaling procedure to map emissions at high spatial resolution, including ship and aviation emissions and detailed sub-sector disaggregation. Two datasets have been applied in this report. N2O emissions have been taken from EDGARv4.0 (JRC/PBL, 2009) and the NOx and NH3 emissions are taken from the EDGAR-CIRCE dataset (Van Aardenne etÂ€al., 2009). The Unified EMEP model is used to estimate atmospheric transport and deposition as calculated by the Europeanscale EMEP MSC-W Chemical transport model (European Monitoring and Evaluation Programme, Meteorological Synthesizing CentreÂ€ – West). The EMEP models have been instrumental to the development of air quality policies in Europe since the late seventies, mainly through their support to the strategy work under the Convention on Long-range Transboundary Air Pollution, and became the reference atmospheric dispersion model for use in the Integrated Assessment Models supporting the development of air quality polices under the EU Commission. The Unified EMEP model is designed to calculate air concentrations and deposition fields for major acidifying and eutrophying pollutants, photo-oxidants and particulate matter (Simpson etÂ€al., 2006). Additional information on these models can be found in the supplementary material (see supplementary material Â�ChapterÂ€15 and 16) including also a comparison of total atmospheric Nr emissions fluxes by various data sets. Details on the data sets used by these models to estimate Nr fluxes is given in

Table 16.1 Overview table of main models used in this chapter to generate the key maps and the European Nitrogen Budget (ENB)

Model

Nitrogen fluxes estimated for the ENB

IDEAg

emissions and nitrogen leaching and run-off from agriculture exchange of nitrogen between the soil and the livestock sectors application of mineral fertilizer to agricultural soils feed and food trade land productivity, consumption of nitrogen nitrogen input to and emissions from sewage treatment systems

INTEGRATOR

emissions and nitrogen leaching from forests and rough grazing

EDGAR

emissions from stationary combustion (energy sector, industry, residential sector) and industrial processes emissions from transport nitrogen input to solid waste management emissions from solid waste management

EMEP

atmospheric deposition

de Vries etÂ€al. (2011, Chapter 15, this volume). A summary of the models used and the main data obtained from each of these models is given in Table 16.1. Each model focuses on different sectors and the models are thus complementary. Information for agriculture is available also from INTEGRATOR and EDGAR; because the IDEAg is the most complete source of information for agriculture the data in this chapter is taken from this model. This avoids most inconsistencies between the data presented. However, inconsistencies cannot be completely excluded and are mainly due to:Â€ (i) different atmospheric deposition data used in the IDEAg and INTEGRATOR model and the EMEP deposition data used in this chapter; and (ii) Nr fluxes from coastal areas which are not included in any of the Europe-wide models. Covering complex processes on a continental scale, the models are bound to rely on simplifying assumptions regarding input data sets, and the parameterization of the processes and the results presented here are consequently associated with large uncertainties. A proper assessment of these uncertainties, however, is very difficult as independent data that can be used to quantify the uncertainties are missing. An attempt to quantify the uncertainty of these (and other) models is currently being done within the European integrated project NitroEurope-IP (Sutton etÂ€al., 2007; NitroEurope, 2010). So far, the best approximation at an uncertainty assessment is done by comparing the in- and outputs of a wide range of models, as done by de Vries etÂ€al. (2011, Chapter 15 this volume).

16.3â•‡ Key maps of nitrogen fluxes in Europe The purpose of this section is to present the spatial distribution of various types of key nitrogen fluxes over Europe that are

347


Figure 16.1 Nitrogen input to agricultural soils in EU27 for the year 2002. The map shows total reactive N input to agricultural soils (cropland and grassland) yrÂ€–1 for a grid at 1 km × 1 km, the values are in kg N per total pixel area [kg N km−2 yr −1 total area]. The pie diagram at the right side gives the split of N input [Gg N rounded to 10Â€Gg N yr –1] for EU27:Â€mineral fertilizer, manure (intentionally applied manure and manure deposited by grazing animals), atmospheric deposition, biological nitrogen fixation and crop residues returned to the soil. The histogram shows the split of N input [Gg N yearâ•›−1, rounded to 10 Gg N yearâ•›−1] by country. Basis:Â€Indicator Database for European Agriculture (IDEAg) V1, 2009. Method:Â€Mineral fertilizer data are obtained from FAO at the national level and are distributed to crops and regions by the CAPRI model using information from IFA/FAO. Distribution to the grid is done on the basis of estimated crop N requirements using information of the potential and water-limited yield for the soil-climate conditions and N supply from biological fixation, atmospheric deposition, and manure nitrogen supply. Manure N supply is estimated from manure availability on the basis of a livestock density map, crop demand and typical share of nitrogen supply by organic nitrogen. The data are net of nitrogen losses occurring before the application of manure to the soil. All data are estimated in consistence with regional values using the highest posterior density approach (Heckelei et al., 2005). Nitrogen deposition data are from EMEP (2008). Biological nitrogen fixation is estimated as a crop-dependent fraction of above-ground nitrogen. Crop residues are estimated from crop-specific fraction and N-content of crop residues. Details on the distribution algorithm can be found in Leip et al. (2008) and Britz and Leip (2009).

responsible for environmental problems. Maps are derived on the basis of fine scale resolution data (1 km × 1 km). The only way to obtain data on such a high resolution was to apply models and combine their results with measurements where available. Most of the models and data sources used have already been presented and explained in detail in earlier chapters. All together, 11 such key maps are selected. They can be grouped into three categories. (i) Drivers for and pressures of nitrogen in terrestrial ecosystems, including both agricultural and nonagricultural systems. Here the total load of nitrogen on agricultural soils (Figure 16.1) and the gross soil nitrogen budget for agricultural (Figure 16.2) and non-agricultural soils (Figure 16.3) can be regarded as key-indicators. Input of nitrogen through atmospheric deposition (Figure 16.4) is a major pressure on (semi-) natural ecosystems.

348

(ii) Emissions of reactive nitrogen to the atmosphere and to the hydrosphere. This is a very important pressure for environmental problems related to nitrogen. We show separate maps on NH3 (Figure 16.5), NOx (Figure 16.6), and N2O (Figure 16.7) emissions across Europe. Each of these compounds is dominated by different source categories (energy/transport for NOx, the livestock sector for NH3, soils for N2O), so that the distinction between these compounds gives also an idea of the spatial distribution of the main driving forces for reactive nitrogen generation. Emissions of nitrogen towards aquatic systems are presented in Figure 16.8. (iii) Secondary nitrogen indicators. Three indicators have been selected:Â€the total productivity of agricultural land (Figure 16.9), and the total consumption of reactive nitrogen by humans (Figure 16.10) and by animals

Adrian Leip

Figure 16.2 Soil system nitrogen surplus for agricultural soils in EU27 for the year 2002. The map shows reactive N surplus for a grid of 1 km × 1 km, the values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of surplus [Gg N yearâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU27 into the loss pathways:Â€NH3 emissions from soils, NOx emissions from soils N2O emissions from soils N2 emissions from soils, N leaching and runoff. The histogram shows the split of the N surplus [Gg N yr−1] into different loss pathways by country. Basis:Â€Indicator Database for European Agriculture V1, 2009. Method:Â€Nitrogen input to agricultural soils is estimated as given in Figure 16.1. Removal of nitrogen by crops and harvested or grazed grass is estimated from regional and national Eurostat statistics, downscaled to the grid on the basis of potential yield from (Genovese et al., 2007). Total N surplus at the grid scale is split into individual fluxes on the basis of the MITERRA approach (NH3 emissions and run-off) as implemented in CAPRI (Britz and Witzke, 2008; Velthof et al., 2009) and the DNDC-CAPRI meta-model (Britz and Leip, 2009) for N2, N2O, NOx and N-leaching. The spatial distribution is done on the basis of the nitrogen input data. Changes in soil-nitrogen stocks are also estimated with the DNDC-CAPRI meta-model; according to the soil-system approach they are nitrogen output as thus not included in the split of the N-surplus. A closed nitrogen budget is obtained according to the method described by Leip et al. (2009a).

(FigureÂ€16.11). While the first indicator shows the potential of the land to feed its population in Europe, the other two indicators give a good idea of the ‘life style’ of citizens. Taken together, these indicators give information on the sustainability of land use and are the basis for the watershed assessment discussed in Chapter 13 (Billen etÂ€al., 2011, Chapter 13 this volume).

16.3.1â•‡ Drivers for and pressures of nitrogen in terrestrial ecosystems We select two indicators describing drivers for the environmental load of Nr and two indicators for the pressure of Nr on the environment. The total Nr input to agricultural soils includes intentionally applied (organic or mineral) fertilizer and manure from grazing livestock as well as biological nitrogen-fixation and atmospheric deposition. Also crop residues returned to the soil are included in total Nr-inputs. Reactive nitrogen additions

are required to fulfil the needs of plants without compromising the productivity of the soil. At the same time, however, excessive Nr additions to agricultural soils lead to high pressures on the environment. Atmospheric deposition is the main source of Nr for natural land and forests. Atmospheric deposition is fuelled mainly by the emissions of NOx from energy-related sources and NH3 lost from agriculture. There is limited capacity in natural ecosystem to absorb Nr. While initial Nr addÂ� itions to forests can lead to a stimulation of plant growth and the build-up of soil organic matter, additions to Nr-saturated systems have adverse effects on the system’s functioning and most Nr is leached (Aber, 1992; Butterbach Bahl etÂ€ al., 2011, Chapter 6 this volume). We define nitrogen surplus as the difference between total Nr inputs to a system and the useful outputs following the definition of the soil-system approach (de Vries etÂ€al., 2011, Chapter 15 this volume). Inputs are total Nr-input as defined above, while useful Nr outputs are harvested crops, including crop residues, and grazed grass. Changes of

349


Figure 16.3 Soil system nitrogen surplus for forest soils (forests, scrublands, heather) in EU-27 for the year 2000. The map shows total reactive N surplus for a grid of 1 km × 1 km, the values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of surplus [Gg N yearâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU27 into the loss pathways:Â€emissions of NH3, NOx, N2O and N2 and N leaching. The histogram shows the split of the N surplus [Gg N yrâ•›−1] into the loss pathways by country. Basis:Â€INTEGRATOR, 2009. Method:Â€Nitrogen surplus is estimated as the sum of gaseous nitrogen fluxes and nitrogen leaching. Nitrogen leaching is calculated in INTEGRATOR from the difference of total N input, via N-deposition, biological nitrogen fixation and manure input where relevant, and the previously estimated nitrogen losses, via uptake by plant growth, NH3 losses, nitrification/denitrification gas losses, and net nitrogen immobilization. However, a minimum nitrogen leaching rate is postulated which is obtained from the water flux and a concentration of 0.02 mg N l−1 (Stoddard, 1994). A check is made if the minimum N-leaching rate is achieved; otherwise this is obtained following pre-defined rules as described by de Vries et al. (2009).

soil Nr stocks can occur in both a positive direction (filling-up the nitrogen pool) and negative direction (depletion of the Nr pool). This is a transient process and can be reversed by changÂ� ing farm management. According to the soil-system approach, changes in soil Nr stocks adjust the accountable quantity of useful outputs. They have an equal impact on the Nr surplus, however, not being part of a detailed split of the fate of N-surplus (for a detailed discussion see Leip etÂ€al., 2010a). Nitrogen surplus on forest soils is defined by analogy:Â€Nr-inputs are fertilizer application, atmospheric nitrogen deposition and biological nitrogen fixation, while Nr-outputs are nitrogen uptake by the plants and Nr immobilization in the soil. The input of Nr to agricultural soils is dominated by the input of mineral fertilizer. Worldwide, the production of mineral fertilizer is the most important source (about 65%) of the net increase of Nr in the environment. While already for the global nitrogen cycle, human influence is larger than the natural dimensions of the nitrogen cycle (Galloway and Cowling, 2002).

350

In Europe the anthropogenic effect is even stronger. Application rates of mineral fertilizer and manure per hectare of utilized agricultural land is shown in Figure 15.7 (de Vries etÂ€al., 2011, Chapter 15 this volume). The pattern in the map in Figure 16.1, which shows the input of nitrogen for total surface area, is different from the maps presented in Chapter 15 as it gives an idea of the share of utilized agricultural area (UAA) across Europe. High shares of UAA up to more than 90% are found in intensive farming areas, such as the Po Valley in Italy, Central Spain, Western France and Romania (Leip etÂ€ al., 2008). For other regions, such as Finland, the Baltic countries or mountainous regions, low shares of agricultural land of generally below 10% yield low N input data even though the application rates per hectare of cultivated land can reach high values, as is the case in certain Finnish regions.The range of manure input is very large, and covers values from 7 kg N per hectare UAA in Romania, to over 230 kg N (ha UAA)−1 in the Netherlands. For some countries, extensive rearing of ruminant animals predominates and

Adrian Leip

Figure 16.4 Atmospheric nitrogen deposition to in EU-27 for the year 2001. The map shows total reactive N deposition for a grid at 1 kmÂ€× 1 km, the values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of N deposition to the different ecosystems [Gg N yrâ•›−1, rounded to 10 Gg N yrâ•›−1] for EU27:Â€coniferous forests, deciduous forests, cropland, seminatural land, and inland water surfaces as well as deposition to the coastal shelf and the deep ocean, which are not shown in the map. The histogram shows the split of N deposition [Gg N yrâ•›−1] by country. Basis:Â€EMEP MSC-W model, rv3_3, 2009. Method:Â€Atmospheric N-deposition is calculated with the European-scale EMEP MSC-W Chemical transport model (European Monitoring and Evaluation Programme, Meteorological Synthesizing CentreÂ€– West). The EMEP model was designed primarily for the calculation of acidifying substances, ozone and particles over Europe (Simpson et al., 2003, see also www.emep.int; Simpson et al., 2011, Chapter 14). The chemical scheme uses about 140 reactions between 70 species (see Andersson-Sköld et al., 1999, and references therein), and makes use of the Equilibrium Simplified Aerosol Module (EQSAM) of Metzger et al. (2002) to describe equilibria between the inorganic aerosol components. Routine N-deposition fields from the EMEP model are available at ww.emep.int. The model uses a sub-grid calculation procedure (so-called ‘mosaic’ approach) to calculate deposition separately to 19 different land-cover categories, taking into account vegetation cover, phenology and surface-characteristics. Calculations of forest-specific deposition estimates, also exploring the role of forest soil-NO emissions from Kesik et al. (2005), were presented in Simpson et al. (2006).

the input of organic nitrogen occurs mainly through deposition of manure by grazing animals, e.g. 86% for Ireland and Greece according to CAPRI model estimates while this is only 20% and less in countries such as Poland, Slovenia and Denmark. Atmospheric deposition and biological nitrogen fixation account together for only 12% of total Nr-input to agricultural soils. Generally, biological N-fixation decreases with increasing Nr input due to increasing competitiveness of non-leguminous crops (Weigelt etÂ€al., 2009). However, as CAPRI estimates biological N-fixation to be a constant fraction of above-ground crop Nr uptake (75% for leguminous crops and 5% for grass), this effect is not considered and leads to a likely over-Â�estimation of biological N-fixation in intensive regions such as North France and the Netherlands in comparison to extensive grassland areas such as in Poland and Romania. The size of the N surplus in the agricultural sector (FigureÂ€16.2) is a measure of the sustainability of the agricultural production

process, since a surplus will eventually lead to shifting the environmental problems to other places outside of the agricultural sector or abroad, including a possible time lag. The contribution of nitrogen leaching to the fate of total N-surplus varies from 26% to 73%. N-leaching is mainly a function of soil texture:Â€heavy clay soils in Central and South Europe offer larger opportunities for denitrification than soils with high organic carbon content and sandy soils, which are less resistant to nitrogen losses to the water. Consequently, losses of N2 are negatively correlated to N-leaching. Fluxes of N2 are very difficult to measure and therefore treated in most models as residual loss-pathway. In the IDEAg, N2 flux estimates are based on results of the DNDC model, but constrained by estimates of NH3 fluxes calculated as in the MITERRA model (Velthof etÂ€al., 2009, see also Chapter 15 this volume) and estimates of N-leaching, N2O and NOx fluxes from the same DNDC metamodel. Nevertheless, the ratio of N2/N2O has a range between

351


Figure 16.5 Total NH3 emissions in EU27 around the year 2000. The map shows the sum of NH3 emissions from terrestrial ecosystems, industry and waste management for a grid of 1 km × 1 km, the values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of total NH3 emissions [Gg N yrâ•›−1, values for agriculture are rounded to 10 Gg N yrâ•›−1] for EU27:Â€agricultural soils including manure application, manure in housing systems and manure management systems excluding manure application, forest soils, emissions from waste, mainly composting of solid waste, energy, and the chemical industry. The histogram shows the split of NH3 emissions [Gg N yrâ•›−1] by country. Basis:Â€Agriculture:Â€Indicator Database for European Agriculture V1, 2009; forest soils:Â€INTEGRATOR, 2009; industrial processes and waste management:Â€EDGARCIRCE (Van Aardenne et al., 2009). Method:Â€(i) Agriculture:Â€emissions are estimated for manure and mineral fertilizer as described by Weiss (2010). Manure emissions are estimated for animal housing and manure management systems and following application on the basis of a mass-conserving approach. NH3 loss factors are taken from the GAINS model for liquid and solid manure and the emission is reduced according to an assumed implementation level for NH3 emission reduction measures using again default GAINS data (Klimont and Brink, 2004; Velthof et al., 2009). Emissions from mineral fertilizer nitrogen are calculated separately for urea and non-urea fertilizers. (ii)Â€Forest soils:Â€emissions are calculated using a constant natural background flux (after Simpson et al., 1999). (iii) Industrial processes:Â€production data are from US geological survey statistics, UN industrial commodity statistics and data from SRI Consulting (2005). The emission factors are EMEP/EEA (2009). Emissions from industrial processes are allocated spatially based on point source maps for the most important source categories, and using population density for the remaining categories. (iv) Waste management:Â€the amount of solid waste composted is estimated based on national reports to the UNFCCC (2008) and on data from European Compost Network (ECN, 2008). The emission factor is from EMEP/EEA (2009). Emissions from the waste sector are spatially distributed based on human population density.

5 and 30 (EU27-average 11.6) in-line with current understanding of the nitrogen cycle (Butterbach-Bahl etÂ€al., 2011, Chapter 6 this volume; Seitzinger etÂ€al., 2006). Fluxes of NH3 contribute between 6% and 17% to the total soil N-surplus and depend on the type of manure or mineral fertilizer nitrogen applied, which is also country-specific. Countries with a high share of urea applied and/or a high livestock density have high losses of nitrogen to the atmosphere as NH3. Unlike agricultural soils, the most important nitrogen change for forest soils and soils under semi-natural land (see Figure 16.3) is Nr accumulation in the soil (about 50% for forest soils and 30% for semi-natural land). The most important Nr loss-pathways for forest soils are N2 and nitrogen leaching. As the map shows the N-surplus calculated per square kilometre total area, it shows the spatial

352

variation of two important factors:Â€the N surplus per hectare of forest area, which is mainly affected by the water balance and soil properties, and the forest area itself, which is particularly high in mountains and in Northern Europe. Therefore, the spatial pattern of nitrogen surplus in forests mimics to a certain degree the topography of Europe, and is as such in contrast to agricultural N surplus. NH3 emissions are not a significant loss pathway of Nr from forest soils, as the input of mineral fertilizer and manure to forest soils is negligible in many countries, but it accounts for one third of the Nr losses from rough grazing land. NH3 emissions include volatilization from urine, livestock manures (slurry and solid manure) and background emissions (see Simpson etÂ€al., (1999)).

Adrian Leip

Figure 16.6 Total NOx emissions in EU-27 around the year 2000. The map shows the sum of NOx emissions from agriculture (both agricultural soils and manure in housing and manure management systems), forest soils, industrial processes, combustion (stationary and mobile) sources, and waste management (incineration) for a grid at of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram on the right side gives the split of total NOx emissions [Gg N yrâ•›−1, rounded to 10 Gg N yrâ•›−1] for EU27:Â€agriculture, forests, combustion in industry and residential combustion, industrial emissions, road and other transport. The histogram shows the split of NOx emissions [Gg N yrâ•›−1] by country. The map shows also the emissions from international aviation (red) and navigation (blue) which are not included in the national totals. Values are less than 40 kg N km−2 yrâ•›−1. Basis:Â€Agriculture:Â€Indicator Database for European Agriculture V1, 2009; forest soils:Â€INTEGRATOR, 2009; industrial processes, combustion and fugitive emissions, and waste management:Â€EDGAR-CIRCE (Van Aardenne et al., 2009). Method:Â€(i) Agriculture:Â€emissions are calculated with the DNDC meta-model as described in Britz and Leip (2009). In the Indicator Database for European Agriculture, a correction of the NOx fluxes is applied only if a closed N-budget cannot be obtained through adjustment of N2 fluxes (considered as the weakest term in the DNDC meta-model) and N-leaching within the bounds set. Then, the loss terms NOx, N2O, N2, and N-leaching are scaled to obtain a closed N-budget. (ii) Forest soils:Â€based on results with the model PnET-N-DNDC for European forest (Kesik et al., 2005) an NOx /N2O ratio of 1.25 is used. N2O emissions are estimated from a meta-model based on PnET-N-DNDC simulations (Kesik et al., 2005). (iii) Industrial processes:Â€production data are from statistics of the US geological survey, UN industrial commodity statistics and data from SRI Consulting (SRIC, 2005). The emission factors are from EMEP/EEA (2009). emissions from industrial processes are allocated spatially based on point source maps for the most important source categories, and using population density for the remaining categories. (iv)Â€Combustion and fugitive emissions:Â€fuel consumption data by sector and fuel type (stationary) or transport mode and fuel type (mobile) is obtained from International Energy Agency (IEA) statistics (IEA/OECD, 2007). Production of crude oil is estimated based on IEA statistics (IEA/OECD, 2007), and venting/flaring is estimated based on data from the Carbon Dioxide Information Analysis Center (CDIAC, 2008), supplemented by reporting of the countries to the United Nations Framework Convention on Climate Change (UNFCCC). The emission factors are based on IPCC (2006)), EMEP/EEA (2009) and Amann et al. (2007). The emissions from stationary combustion are spatially distributed using point source maps for power plants, steel production plants, and oil refineries, and maps on urban and rural population density for the other sectors. (v) Combustion transport:Â€emissions from transportation include road and rail transportation, domestic and international navigation, domestic and international aviation and other transportation. The fuel use by each transport mode and fuel type is from IEA statistics (IEA/OECD, 2007). The fuel use in international navigation is divided between sea and port activities of 15 ships types based on Dalsøren et al. (2009). Fuel consumption in aviation is divided between landing and take-off; climbing and descent; cruise; and super-sonic based on gridded data from the AERO2K project (Eyers et al., 2004). A detailed split of the fuel used in road transportation is used in the EDGAR database considering heavy and light duty vehicles, passenger cars, buses, mopeds, and motorcycles by applying country-specific fleet distribution calculated based on registration, number of vehicles, and driven vehicle kilometres from International Road Federation (IRF, 2007). The impact of emission control measures is calculated based on European emissions standards (EURO 0Â€– EURO 4) and other regional standards, with data from CONCAWE (2001) and EMEP/EEA (2009). The emission factors are based on EMEP/EEA (2009), EIPPC BREF, IPCC (2006) and scientific literature. Emissions from aviation are spatially allocated based on AERO2K project, and presented separately for domestic and international aviation. Emissions from road transportation are spatially allocated using road density map, weighted with population in the case of passenger cars. Emissions from international navigation are spatially allocated using a ship traffic density map of Wang et al. (2007). Other transport emissions are gridded using population density. (vi) Waste management:Â€the amount of solid waste incinerated without energy recovery is estimated based on the reporting of the countries to the UNFCCC. The emission factors for solid waste are from EMEP/EEA (2009). Emissions from the waste sector are spatially distributed based on human population density.

353


Figure 16.7 Total N2O emissions in EU-27 around the year 2000. The map shows total N2O emissions from terrestrial ecosystems, industry, energy and waste for a grid of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 total area]. The pie diagram at the right side gives the split of total N2O emissions [Gg N yearâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU-27:Â€agricultural soils, manure management excluding manure spreading on soils, forest soils, energy (large scale and domestic), industry (mainly chemical industry), waste (wastewater treatment and other waste) and transport (road and non-road). The histogram shows the split of N2O emissions [Gg N yr−1] by country. Basis:Â€Agriculture:Â€Indicator Database for European Agriculture V1, 2009; forest soils:Â€INTEGRATOR, 2009; industrial processes, combustion, and solid waste management:Â€EDGARv4 (JRC/PBL, 2009); waste water systems:Â€Indicator database for European Agriculture V1, 2009. Method:Â€(i) Agriculture:Â€emissions are calculated with the DNDC meta-model as described in Britz and Leip (2009). In the Indicator Database for European Agriculture, N2O fluxes are corrected to obtain a closed N-budget only in case the correction of N2 fluxes (considered as the weakest term in the DNDC metamodel) and N-leaching alone is not possible within the bounds set. In this case, the loss terms NOx, N2O, N2, and N-leaching are scaled to obtain a closed N-budget. (ii) Forest soils:Â€emissions are estimated from a meta-model based on simulation results for European forest soils with the model PnET-N-DNDC (Kesik et al., 2005). (iii) Industrial processes:Â€production data are from statistics of the US geological survey, UN industrial commodity statistics and data from SRI Consulting (SRIC, 2005). Abatement of N2O emissions from nitric acid and adipic acid emissions is included based on the reporting of countries to the UNFCCC. The emission factors are from IPCC (2006). Emissions from industrial processes are allocated spatially based on point source maps for the most important source categories, and using population density for the remaining categories. (iv) Combustion and fugitive emissions:Â€fuel consumption data by sector and fuel type (stationary) or transport mode and fuel type (mobile) is obtained from International Energy Agency (IEA) statistics (IEA/OECD, 2007). The emission factors are based on IPCC (2006) and other sources. The emissions from stationary combustion are spatially distributed using point source maps for power plants. (v) Combustion transport:Â€emissions from transportation include road and rail transportation, domestic and international navigation, domestic and international aviation and other transportation. The fuel use by each transport mode and fuel type is from IEA statistics (IEA/OECD, 2007). The fuel use in international navigation is divided between sea and port activities of 15 ships types based on Dalsøren et al. (2009). Fuel consumption in aviation is divided between landing and take-off; climbing and descent; cruise; and super-sonic based on gridded data from the AERO2K project (Eyers et al., 2004). A detailed split of the fuel used in road transportation is used in the EDGAR database considering heavy and light duty vehicles, passenger cars, buses, mopeds, and motorcycles by applying country-specific fleet distribution calculated based on registration, number of vehicles, and driven vehicle kilometres from International Road Federation (IRF, 2007). The impact of emission control measures is calculated based on European emissions standards (EURO 0Â€– EURO 4) and other regional standards, with data from CONCAWE (2001) and EMEP/EEA (2009). The emission factors are based on IPCC (2006). Emissions from aviation are spatially allocated based on AERO2K project, and presented separately for domestic and international aviation. Emissions from road transportation are spatially allocated using road density map, weighted with population in the case of passenger cars. Emissions from international navigation are spatially allocated using a ship traffic density map of Wang et al. (2007). Other transport emissions are gridded using population density. (vi) Solid waste management:Â€the amount of solid waste composted and incinerated is estimated based on the reporting of the countries to the UNFCCC. Composting data are complemented with information from European Compost Network (ECN, 2008). The emission factors for solid waste are from IPCC (2006). Emissions from the waste sector are spatially distributed based on human population density. (vii) Waste-water systems:Â€emissions from nitrogen in effluents (0.005 kg N2O-N kg−1 N) as well as emissions from advanced sewage treatments systems (3.2 g N2O person−1 yrâ•›−1) are from IPCC (2006).

354

Adrian Leip

Figure 16.8 Total reactive N input to the hydrosphere (rivers and groundwater) in EU-27 for the year 2002. The map shows total Nr point sources from sewerage systems and diffuse sources from agriculture and forest soils and atmospheric Nr deposition to inland water surfaces for a grid at of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 yrâ•›−1 total area]. The pie diagram at the right side gives the split of N input to the hydrosphere [Gg N year−1, rounded to 10 Gg N yearâ•›−1] for EU27:Â€point sources (sewage systems) and diffuse sources (agriculture leaching, run-off, and forest soils). The histogram shows the split of Nr input to the hydrosphere [Gg N yearâ•›–1] by country. Basis:Â€Sewage systems and agriculture:Â€Indicator Database for European Agriculture V1, 2009; forest soils:Â€INTEGRATOR, 2009. Method:Â€(i) Sewage systems:Â€nitrogen in agricultural products is estimated with CAPRI, N from fish is obtained from Eurostat fish statistics and a mean N-content of 2.6%. Non-consumed proteins (waste) are assumed to be 30%. N retention in sewage systems is calculated from the percentage of people connected to sewage systems (for rural and urban population) with mechanic, biological or advance treatment and corresponding retention efficiencies is obtained from Van Drecht et al. (2009). According to IPCC (2006), industrial wastewater is assumed to be 25% of domestic nitrogen (in advanced treatment systems). Spatial downscaling is done according to the population density. (ii) Agriculture:Â€total nitrogen leaching and runoff of nitrogen from agricultural soils is estimated from the DNDC-CAPRI meta-model (2009) and integrated into the IDEAg according to Leip et al. (2009b). In addition to the data presented in Figure 16.2, the data presented here include run-off from livestock housing and manure management systems, which are estimated using the MITERRA approach (Britz and Witzke, 2008; Velthof et al., 2009). Spatial distribution is in accordance with the livestock density per grid cell by animal group. (iii) Forest soils:Â€nitrogen Nr leaching from forest soils is estimated as described in Figure 16.3. (iv) Atmospheric deposition:Â€Atmospheric Nr-deposition is calculated with the European-scale EMEP MSC-W Chemical transport model (European Monitoring and Evaluation Programme, Meteorological Synthesizing CentreÂ€– West) as described in Figure 16.4.

N2O emissions take a larger share of the total denitrification losses with generally narrow N2/N2O ratios around 3–4 in forest soils, but higher values for rough grazing. The forest type is very important in determining the rate and also the type of Nr emissions, due to its impact on litter quality and soil pH. Pilegaard etÂ€ al. (2006) carried out a detailed study of Nr fluxes in 15 forest sites throughout a year and across Europe and found that coniferous forest soils had much higher NO emissions than deciduous forest soils. On the other hand, N2O emissions were slightly higher in deciduous forests compared to coniferous ones. Atmospheric transport and atmospheric deposition is discussed in detail by Simpson etÂ€al. (2011, Chapter 14 this volume). In contrast to the figures presented there, we show here

absolute deposition fluxes per area of grid cell (Figure 16.4). For the contributions of N-deposition by ecosystem shown in the pie diagram and the histogram, the share of the various ecosystems at the grid scale is taken into account also. Even though cropland covers a smaller area in EU27 than forests, about 40% of deposition fluxes on the continent go on cropland, more than on forest land, because European forests are predominantly located in areas with smaller atmospheric Nr concentrations. Forests are the main receptor ecosystem for atmospheric Nr deposition in Scandinavian countries (Finland 70% and Sweden 66%) and alpine regions as Austria (55%). Deposition over semi-natural land is important in Mediterranean countries (Greece 35%, Spain 30%), but also in countries where conditions are too wet for other land uses

355


Figure 16.9 Total land productivity of agricultural land in EU-27 for the year 2002. The map shows total agricultural land productivity for a grid of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 total area]. The pie diagram at the right side gives the split of N productivity [Gg N yrâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU27:Â€crop products, fodder (fodder maize fodder beet and other fodder on arable land), grass and other (flowers, nurseries, etc.). The histogram shows the split of N productivity [Gg N yrâ•›−1] by country. Basis:Â€Indicator Database for European Agriculture V1, 2009. Method:Â€Land productivity is based on CAPRI regional statistics on agricultural and grassland yield with crop-specific nitrogen contents. The data include crop residues that are returned to the soil. As statistics on grassland yields are scarce, these are estimated within CAPRI on the basis of the energy requirement of livestock and energy supply from other feed available (concentrates, fodder). Spatial allocation is done on the basis of simulation results for crop-potential yields (Genovese et al., 2007) under given climatic and soil conditions.

(UK 54%, Ireland 76%). Croplands are the main receptor for atmospheric Nr deposition in Denmark (58%), Poland (52%) and Lithuania (51%). The map shows also deposition fluxes over European shelf regions and deep ocean waters, which are of equal magnitude. Together they receive about 40% of the Nr deposited over the continent. The size of the total deposition is largely controlled by agricultural emissions of NH3, which are transported over shorter distances than NOx and strongly influence local deposition rates. High deposition regions with deposition rates over 20 kg N ha−1 yr−1 are therefore associated with intensive livestock production, such as in the Netherlands and the Po Valley in Northern Italy. High deposition rates of >10 kg N ha−1 yr−1 are found almost throughout Central Europe.

16.3.2â•‡ Emissions of reactive nitrogen to the atmosphere and hydrosphere Emissions of reactive nitrogen to the atmosphere and the hydrosphere are caused by both agricultural and other land-use

356

activities as well as by fuel combustion and industrial processes. Detailed maps of the spatial distribution of total emissions of Nr to these media are helpful for understanding the occurrence of hot-spots and could be the first step in identifying appropriate and well-targeted mitigation measures. We show here the most important fluxes of Nr, i.e. total emissions of NH3 (FigureÂ€16.5), total emissions of NOx (Figure 16.6) and total emissions of N2O (Figure 16.7) to the atmosphere, as well as total Nr emissions to the hydrosphere (Figure 16.8). About 95% of NH3 emissions originate from the agriculture sector. The contribution of agriculture is rather stable across the countries in EU27, with a highest contribution in countries such as Ireland, Spain or Hungary (98%). The importance of soil/field emissions (from field application of manures, grazing and fertilizers) versus emissions from manure management systems (animal housing and manure storage) varies between 40% of emissions from soils in Denmark and 67% of emissions from soils estimated for Ireland. The reason is, of course, the importance as well as the structure of the livestock sector (grazing versus housing of the animals). As manure has higher

Adrian Leip

Figure 16.10 Total human N consumption of reactive nitrogen in EU-27 for the year 2002. The map shows human nitrogen consumption of agricultural products including food waste for a grid at of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 total area]. The pie diagram at the right side gives the split of human N consumption [Gg N yrâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU-27:Â€crop products, animal products, fish products (incl. shellfish), and the N in non-edible products (peelings, bones) and food waste. The histogram shows the split of human N consumption [Gg N yrâ•›−1] by country. Basis:Â€Indicator Database for European Agriculture V1, 2009. Method:Â€Consumption is based on CAPRI regional statistics. Spatial allocation is done on the basis of livestock density.

volatilization fractions for NH3 than mineral fertilizer, we find hotspots of NH3 fluxes where the livestock density is high, both in intensive production systems with housed animals such as in the Po Valley, Italy, Denmark, and the Netherlands, as well as in regions with a predominance of grazing animals such as in Ireland. Intensive animal production systems have mainly developed in the vicinity of large metropolitan areas, as the banlieue of the Paris area, but also west of Berlin. The metropolitan areas themselves however, are usually low emitters of NH3. Hotspots of NH3 fluxes are found in the major European plains such as the Po Valley, North Germany, the Netherlands and Bretagne, but also in hilly regions such as the Alps in Southern Germany or also, for example, the north of Andalucia. NOx emissions arise mainly from industrial and energysources accounting for 96% of emissions. Only 4% of NOx emissions are formed biogenically, with 3% from agriculture and 1% from forests, according to the estimates presented here. Butterbach-Bahl etÂ€al. (2008) give a range of 48.8–128.9 Gg N yr−1 for NOx emissions from agricultural soils depending on the methodology used, which matches well with the presented number of 76 Gg N yr−1 from agricultural soils (including

emissions from pasture), to which about 32 Gg N yr−1 emissions of NOx from manure management systems are added. For forest soils, Butterbach-Bahl etÂ€al. use the process-based model Forest-DNDC as the only approach and give a number of 75 Gg N yr−1, which is about twice as large as the 32 Gg N yr−1 in our estimate from the INTEGRATOR model. The numbers do not include NOx emissions from forest fires and burning of agricultural residues, which are available in the Global Fire Emissions Database (GFED) (van der Werft etÂ€al., 2010). According to these data, about 11 Gg of NOx-N are released to the atmosphere by burning of woodland and forests (about 75%) and agricultural waste (25%). The bulk of NOx emissions originate from combustion processesÂ€– about equal amounts of NOx-N are emitted from stationary combustion (industry and residential combustion) and mobile combustion (mainly road transport) with 42% of total emissions or about 1.5 Tg N yr−1 each. It is important to note that emissions from aviation and navigation are divided between domestic and international transport. While the emissions from international navigation/aviation are included in the spatially allocated emissions, they do not appear in national

357


Figure 16.11 Total livestock N consumption of reactive nitrogen in EU-27 for the year 2002. The map shows human nitrogen consumption of agricultural products including food waste for a grid at of 1 km × 1 km. The values are in kg N per total pixel area [kg N km−2 total area]. The pie diagram at the right side gives the split of livestock N consumption [Gg N yrâ•›−1, rounded to 10 Gg N yearâ•›−1] for EU-27:Â€grass (cutting and grazing), fodder (fodder maize, fodder beet and other fodder on arable land), crops (cereals and other non-fodder crops, mainly leguminous crops and oilseeds), concentrates (energy-rich and protein-rich concentrates, oilseed cakes, milk powder, molasse, etc.) and other (straw, animal products). The histogram shows the split of livestock N consumption [Gg N yrâ•›−1] by country. Basis:Â€Indicator Database for European Agriculture V1, 2009. Method:Â€Consumption is based on CAPRI regional statistics and Eurostat national statistics for fish products with crop-specific nitrogen contents. Spatial allocation is done on the basis of population or livestock density for the consumption of food or feed, respectively.

totals. Military emissions, which are estimated to be small, are not included. Additional details on the transport sector are given in Section 16.4. The highest share of industry-combustion to total NOx emissions is observed in Central-Eastern European countries such as Romania (43%), Poland (47%), or the Czech Republic (48%). In Poland, residential combustion is also considerable (21% of total NOx emissions). The importance of residential NOx emissions depends on the fuel mix and the energy efficiency; the share of residential emissions to total NOx emissions in Italy and France are relatively high with 15% and 20%, respectively, and low in Finland and Estonia (both 7%). Accordingly, the map of total NOx emissions shows high values in centres of energy-intensive industry, such as Sachsen-Anhalt in Germany, North Italy, the Netherlands, or along intensive traffic lines. In many cases, both have been developed along major river streams as can be observed for the Rhine where we find large industry complexes but also an important traffic axis. The main sources of N2O are biogenic sources including agricultural soils, manure management, as well as forest soils and

358

the waste sector, accounting together for 74% of all N2O fluxes in EU27. Processes leading to N2O formation in soils, as well as the upscaling of N2O fluxes from these sources for both agricultural and non-agricultural terrestrial sources, are discussed in Chapters 6 and 15 (Butterbach-Bahl etÂ€ al., 2011, this volume; De Vries etÂ€al., 2011, this volume). Agricultural soils are the main source contributing 73% of biogenic N2O sources with manure management systems, forests and the waste sector contributions at 7%, 8% and 2%, respectively. The contribution of the waste sector to biogenic N2O emissions ranges from 1%, for example in Finland, the Netherlands and Poland, to 16% in Denmark. Agricultural soils contribute to almost 90% of biogenic N2O emissions in Hungary, Greece and Finland, while the smallest contribution of soils being estimated for Estonia (45%). The most important non-biogenic sources of N2O are industrial processes, which are not caused by fuel combustion, but by the industrial processes themselves. The chemical industry is an important source of N2O emissions accounting for 20% of EU27 N2O emissions. Globally, nitric acid production is the most important N2O source within the chemical industry,

Adrian Leip

followed by adipic acid, caprolactam and glyoxal production. As industrial plants are point sources with an uneven distribution, the significance of industry ranges from 0% of national N2O emissions in Poland, Portugal and the UK to more than 60% in Denmark. Additionally, N2O is used in anaesthesia and in aerosol spray cans. The spatial structure of N2O emissions is thus a combination of the one observed for NH3 (mainly livestock) and NOx (mainly energy and transport). Thus, we find those areas which were already identified for both other gases such as the Po Valley, the Netherlands, and Sachsen-Anhalt, but also identify high fluxes from rural agricultural areas such as Hungary or Poland. The map of N2O fluxes is further complicated by its large dependency on environmental conditions. The soil type is a particularly important factor, which tends to increase fluxes in Northern Europe, where soils with a high content of organic carbon prevail, and leads to lower fluxes in Southern Europe with soils of lower organic carbon content, but this trend is overlaid with the influence of soil moisture and temperature (see Leip etÂ€al., 2010b). The input of reactive nitrogen to European aquatic systems is dominated by point sources through sewage systems, including industrial fluxes and diffuse sources from agriculture. Thus, the Nr load to rivers is highly correlated with population density as can be seen from Figure 13.10 in Billen etÂ€al. (2011, Chapter 13 this volume). This is superimposed on the pattern of agricultural nitrogen leaching, which is similar to the spatial pattern of agricultural nitrogen surplus shown in Figure 16.2. Runoff in agricultural systems from stables or manure management systems and leaching from forest soils are estimated to be of minor importance. Data on the distribution of sewage systems are available for most European countries from EUROSTAT, EEA (1998), Wieland (2003) and Jeppsson etÂ€al. (2002). The overall values for these removal fractions for a country are calculated as the weighted average of the four classes as compiled for the IMAGE model by Van Drecht etÂ€al. (2009). In the IMAGE model it is assumed that Nr emitted by people not connected to sewerage systems will be retained and does not enter the hydrosphere. In rural areas of Europe, however, the majority of human wastes will be discharged to unmonitored small sewage plants, or to septic tank/soakaway facilities. In the UK, for example, these are the dominant point sources in most rural catchments, in comparison to the major sewage treatment plants in larger towns or cities. We assumed that the Nr from such unmonitored small sewage plants or septic tanks/soakaways undergoes a ‘biological-treatment’ like transport to the river/groundwater system. A detailed study on the nitrogen removal efficiency of sewage treatment systems in the UK (Johnes, 1996) takes in to account the fact that although the process is optimized in larger sewage treatment plants in urban areas, in many of the older treatment systems, in rural areas and in most of the coastal towns and villages, volumes treated have exceeded initial design capacity through local population growth or migration. As a consequence, only mechanical treatment is available for part of the Nr, with little biological treatment of wastes, particularly

during cold, wet periods. The same is true for peak seasons, for example in tourist regions (e.g. alpine ski resorts in the winter; beach and lake vacation regions in the summer and spring), when the Nr input to sewage systems exceeds their capacity.

16.3.3â•‡ Secondary nitrogen indicators In this section we present secondary reactive nitrogen indicators that answer two important questions:Â€what is the amount of proteins (nitrogen) that European crop- and grasslands can currently produce and what is the amount of proteins (nitrogen) that European inhabitants (humans and livestock) consume? The first map shows total land productivity (Figure 16.9), defined as the sum of harvested crops, and grazed biomass according to the soil system budget. Included in the total productivity are crop residues that are or are not used for other purposes (as animal feed or bedding material, biofuels or burned), which is in contrast to the definition of nitrogen autotrophy (see Billen etÂ€ al., 2011, Chapter 13 this volume), which gives its production of food and feed only (harvested and grazed products; Billen etÂ€al., 2007, 2008). The second and third map show human consumption of nitrogen (Figure 16.10) and the consumption of nitrogen by livestock (Figure 16.11). Both are split by the main proteinsources, i.e. crops, livestock products and fishery products for human consumption and grass, crops/fodder and concentrates for livestock consumption. The sum of both maps gives the nitrogen heterotrophy in Europe. A large part of rural Europe is characterized by a high degree of regional specialization of agricultural activities. In most traditional agrarian systems in Europe, livestock farming used to be a critical component, providing a way to ensure cropland fertility by bringing to it, Nr extracted from seminatural N2 fixing areas in the form of manure. During the past half century, in parallel with increasing urbanization, many lowland rural areas of Europe have shifted either towards exclusive crop production, with very little cattle breeding, or to intensive livestock farming, supported to a large extent by feed importation. Mixed farming areas are restricted to highland or mountainous regions. As a result of this specialization, the exchange of food and feed over long distances has considerably increased and now often represents quite a significant share in the nitrogen budget of regions, or even of countries. Still, the total productivity of agricultural land is particularly high in the hinterland of large metropolitan areas (e.g. Paris, Berlin, London) and in areas of intensive animal production (the Netherlands, Belgium, Po Valley, Italy, many areas of England and Ireland). About 45% of the productivity yields crop products, while the other 50% of nitrogen are distributed over dedicated fodder production (20%, fodder maize and fodder beet) and grassland (25%). The significance of grass varies largely between 5% in Denmark and 60% in Ireland while the importance of feed production varies between 10% (Bulgaria, Greece) to over 40% (Sweden, Estonia). Human consumption obviously peaks in metropolitan areas like Paris, London and Berlin which are clearly visible in the map and consume more than 15 Mg N yr−1 per square kilometre. Outside

359


these hotspots, human Nr-consumption is generally between 200 and 2000 kg N yr−1 km−2. There is about 36% consumption of crop products and 33% consumption of animal products, with a small fraction (3%) of fish products consumed. The last third is estimated to be either non-edible or wasted. In many central-western countries intake from animal proteins dominales while in countries like Greece, Romania and Bulgaria about 70% of protein intake is from vegetable sources. Livestock Nr-consumption follows the livestock density and thus the same regions with high/low rates can be identified that were already visible particularly in the NH3 map (FigureÂ€16.5). The consumption of crop products by livestock is more than double the consumption of crop products by the human population, a value that increases to 3.8 and 9.3 in Ireland and Denmark. About the same amount of concentrates is being fed to animals as crop products, and about 40% of animal feed is stemming from grass and fodder. However, we find large differences in animal nutrition across the countries with some countries feeding almost 50% with concentrates, as in the Netherlands, Portugal and Denmark, while this share is only 12%–13% in Latvia, Romania countries and the Czech Republic. The share of grass is naturally high in mountainous countries like Austria, but also in lowland countries such as Ireland and the UK. In all cases, grass contributes to more than 40% of the protein requirements of animals. Urban areas, where food is consumed but not produced, are obviously heterotrophic. Rural regions specialized into crop production are autotrophic and export nitrogen as food and/ or feed, while those characterized by intensive animal farming sustained by imported feed are usually heterotrophic as in those regions the import of feed is usually not balanced by exported food products.

16.4â•‡ Integrated nitrogen budgets Integrated nitrogen budgets are defined here as the quantification of all major nitrogen fluxes across sectors and media within given boundaries, and fluxes across these boundaries, on an annual basis. They provide a valuable tool for optimizing the benefits of policies addressing imbalances evident in the nitrogen cascade. These policies have often been designed to achieve a specific goal, neglecting unwanted side effects such as pollution swapping. Integrated nitrogen budgets per se do not directly give a quantification of the risk of such pollution swapping effects, as these are determined by mechanistic effects and require an understanding of the dynamics of the nitrogen fluxes. However, in many instances, integrated nitrogen budgets can give a good indication of where Nr pollution is most severe and where swapping problems from one medium to another might occur. Here, a standardized integrated Nitrogen Budget (iNB) approach has been implemented to derive a suite of national integrated nitrogen budgets (NiNB) as well as the European Nitrogen Budget (ENB). Each of the NiNBs has been compiled by national experts from each country, using data available at national scale. For each iNB, five sectors are differentiated:Â€industry and energy, transport, agriculture, forestry and

360

natural terrestrial ecosystems, and waste. Each sector has a pool of Nr and is connected to the others by three transport media:Â€ the atmosphere, the hydrosphere, and the consumers who transport, for example, agricultural products to the waste management systems. Pools of Nr are hard to quantify and are often not incorporated; however stock changes are important indicators to detect possible accumulation or depletion of Nr and their implications for soil productivity, biodiversity or the development of human health problems. Conceptually, the ENB contains the same kind of information as NiNBs and they can thus be discussed together. From a methodological point of view, however, the ENB is largely model-based (see Section 16.4.2) as robust and consistent estimates for many elements of a NiNB at the European level are not currently available.

16.4.1â•‡ National integrated nitrogen budgets National integrated nitrogen budgets (NiNB) help in visualizing the main elements of the N cascade within a country into a figure that might transmit its main messages at a quick glance, but nevertheless contains sufficient detailed information for further analysis. Therefore, a NiNB is regarded as a very efficient policy instrument and an important tool to help prioritize policies. In particular, NiNBs can serve five objectives:Â€(i) they are an efficient instrument for visualizing the N cascade and its potential impact and thus help to raise awareness; (ii) NiNBs provide policy makers with information for developing efficient emission reduction measure; (iii) more importantly, they can provide a tool for monitoring the impact and environmental integrity of implemented policies; (iv) NiNBs are useful for comparisons across countries; and (v) they can help pinpoint knowledge gaps and thus contribute to improving our scientific understanding of the N cascade. Often, NiNBs have to rely on information of different origin and quality, and therefore it may not be possible to ‘close’ the budget for one or several sectors. N fluxes presented in NiNBs are ideally based on a sufficiently dense network of observational data or on detailed models calibrated and validated on national conditions; however, often data gaps have to be filled from simpler models of a broader scope such as the models used for the European Nitrogen Budget. To build a NiNB is thus a challenging task and many elements of a budget will only be quantifiable within a very high uncertainty range, for example the amount of nitrogen denitrified and released as the stable and harmless N2 gas; or sedimented and stored for potential future release in the oceans. The magnitude of the uncertainty itself is usually unquantified. Despite these difficulties, NiNBs have been developed for some countries or are in the process of being developed. • Switzerland formulated environmental targets for agriculture in 1996 based on the observation that additional efforts were required to minimize pollution of soil, air and water and to maintain biodiversity. Measures in the agriculture sector were found to be particularly cost-efficient. The recommendations built on the Swiss N-budget that had been

Adrian Leip

•

•

•

•

•

developed for the year 1994. The Swiss N-budget was updated for the year 2005 and published by the Federal Office for the Environment (BAFU, 2010). The Netherlands is a country facing significant Nr pollution problems, such as particularly high nitrate concentrations in the groundwater, as well as a decrease in biodiversity and forest vitality, high atmospheric NOx and NH3 concentrations leading to human health effects, and algal blooms in the North Sea (Erisman etÂ€al., 2005). The Dutch nitrogen budget was estimated by Erisman etÂ€al. (2005) on the basis of an analysis by van Grinsven etÂ€al. (2003) and also proposed a list of measures to address the Nr pollution problems in the Netherlands. In Germany the Federal Environment Agency (UBA) released a draft national nitrogen budget as background information to the Integrated Strategy for the Reduction of Nitrogen Emissions in April 2009 (Umweltbundesamt, 2009a,b), motivated by the fact that despite major efforts most environmental targets (halting loss of biodiversity, national emission ceilings for NOx and NH3, concentration of nitrate in drinking water, mitigation of global climate change) appeared unlikely to be met and that only an integrated approach would support the development of cost-efficient and effective solutions. Whereas the strategy contains a set of measures, the budget contains very detailed information on all quantifiable nitrogen fluxes across sectors and interfaces between environmental media above 1 Gg N yr−1. In France, the construction of the nitrogen budget has been initiated with the aim of developing an overarching vision of the nitrogen cascade between industrial sectors and environmental compartments through cooperation between various French research institutes and agencies. The development of the N-budget is ongoing and results are preliminary. The United Kingdom has built a national N-budget using theÂ€ iNB approach, based on a detailed N-budget for agriculture following the OECD approach (DEFRA, 2008), national scale N flux modelling to freshwater and coastal systems and to and from the atmospheric N pool, and published data on non-agricultural sectoral fluxes. Further work is warranted to refine and update the initial budget presented here. The Czech Republic has launched a project to estimate all N-fluxes following the German example. To that purpose, the Czech Hydrometeorological Institute (CHMI) is cooperating with the Ministry of Agriculture and with the Central Institute for Supervising and Testing in Agriculture. CHMI is providing a range of emission and deposition data. Other institutions calculate N fluxes with regard to, for example, feed, manure, agricultural products, waste and leaching. Cooperation with the Institute of Geology and other institutes is planned.

Other countries, like Turkey have recently started developing a national N-budget.

Figures 16.12–16.17 show the national integrated nitrogen budgets for countries in Europe available to date. Each NiNB is constructed from nationally available information and thus the budgets are not directly comparable. For example, river export has not been estimated in Germany, while it constitutes a significant flux in the Netherlands and Switzerland. Additional details on the data sources of the NiNBs and the outcomes of the respective projects are given in the supplementary information (see supplementary material Chapter 16, Section A). As the NiNBs are not constructed with a harmonized or even comparable methodology, a comparison of single flux estimates must be done with care. Nevertheless, they highlight the general differences across the countries, as can be seen by ranking emissions by sector and dominant Nr form of emissions. Such an assessment is shown in Figures 16.18–16.20. For example, nitrogen fluxes in the Netherlands are dominated by industrial Nr fixation. The export of Nr, mainly as fertilizer, is by far the largest N-flux and feed-imports are higher than the input of mineral fertilizer. By contrast, in Switzerland the combined estimate of atmospheric Nr deposition plus biological N fixation to agriculture is higher than the input of mineral fertilizer and feed imports. Both countries have an important exchange of Nr with other countries or the sea through river flow. In the Netherlands, the Nr transit through the country as import and export are roughly the same. However, in Switzerland, river export is the single most important sink for the country exporting more Nr than is applied to agricultural soils as mineral fertilizer. A summary of the national N-budgets presented above by main compartment or sector (Table 16.2) shows that balance is not closed for most sectors. These N-budgets give aggregated data for countries, without a spatial dimension such as was shown in the key maps above. Not all nitrogen fluxes have been (or could be) estimated yet, and, secondly, the data for different compartments and sectors have been taken from the best available, but to some extent inconsistent, data sets. Examples include atmospheric deposition (which in many countries has been obtained from the EMEP model) and atmospheric emissions, which can have data sources which are partly inconsistent with the information used in the EMEP model. The main gap between the sum of Nr emissions to and removals from the atmosphere is due to fluxes of molecular nitrogen (N2) through N fixation and denitrification, which are difficult to estimate and have not been quantified for many countries/compartments. Therefore a negative balance is observed for most countries. The agriculture sector, which is in many countries one of the best-described sectors as gross nitrogen balance calculations have been made with the OECD methodology (OECD/ Eurostat, 2003), ideally gives a closed soil N-budget, yet the link to consumers and/or the industry has often been obtained from different sources. Un-quantified stock changes in agricultural soils can potentially account for part of the positive (accumulation) or negative (depletion) balance in agriculture. The French NiNB does not yet include fluxes from agriculture to the consumer or to the hydrosphere and the data show therefore a considerable gap.

361


Figure 16.12 A national integrated nitrogen budget for Switzerland, derived from a study that was carried out between 1994 and 1996 based on a mandate of the Departments of Economic Affairs and Home Affairs of the Swiss government. The project aimed at identifying the most important N fluxes between the all compartments, at assessing the fluxes with respect to the exceedance of effects-based environmental and health quality criteria, and at elaborating a strategy for a stepwise reduction of emissions of reactive nitrogen. The results of the work of the project group are summarized in the report. “Strategy for the Reduction of Nitrogen Emissions” (BUWAL, 1996).

One of the largest fluxes is the generation of Nr in the industry and energy sector. This Nr has three important sources: (a) nitrogen fixation with the Haber–Bosch process, (b) release of nitrogen from fossil energy carriers such as coal, and (c) thermal generation of Nr at high temperatures during the burning process. For the NiNBs presented, total industrial/ energy N-fixation has been estimated as the difference between total estimated N-outputs and inputs in these sectors, to give an indication of the order of magnitude of this N-flux. A positive N-balance is found for the consumers. While food input and N-output to sewage systems is quantified in most cases, the underlying assumptions may differ. Not all biomass produced is edible; not all edible parts will be consumed; and assumptions on the fate of the produced biomass are associated with considerable uncertainty (see discussion on the European Nitrogen Budget). The waste-streams are often poorly quantified in the NiNBs presented. The consumer has a central role in national N-budgets:Â€with excess-supply in most European countries the incentives for ‘nitrogen-efficient’ behaviour are not currently well developed to affect consumers’ behaviour. The choices of consumers nonetheless steer the societal machinery with major consequences for the nitrogen budget. In addition to input of nitrogen as biomass, the consumer receives also significant amount of industrial nitrogen, as plastics, pigments or other chemical products. The

362

fate of these products is not yet quantified and appears as a (positive) balance as these substances accumulate in the anthroposphere. An overview of emissions to the atmosphere is given in Figure 16.18. Generally, agriculture is the main emitter of Nr to the atmosphere and to the hydrosphere (Figure 16.19), but taking all combustion sources together (incl. industry), atmospheric emissions from combustion are roughly at the same level as agricultural emissions, with the exception of the United Kingdom. The United Kingdom is the country where industrial and energy emissions play the biggest role:Â€emissions to the atmosphere from combustion sources are almost twice as large as the emissions from agricultural sources. Nevertheless, as extensive agricultural activities involves high mineral fertilizer input (more than 1 Tg N yr−1), many livestock and high precipitation, this country estimates the highest leaching rate to waters (almost 0.5 Tg N yr−1). Emissions to the hydrosphere have not been estimated yet in the French N-budget, however, this country is characterized by extensive agricultural activity reflected both in the input of mineral fertilizer nitrogen (2.5 Tg N yr−1) and high atmospheric emissions (> 0.7 Tg N yr−1). The high land productivity throughout almost the whole territory of France (see also Figure 16.9) makes the country independent of feed-imports (only 7% of mineral fertilizer-input or 175 Gg N yr−1).

Adrian Leip

Figure 16.13 The Dutch national nitrogen budget is based on information from different publications, mainly originating from the National Bureau of Statistics. The data shown represent the average situation of 1995, 1997, 1998 and 1999 (van Grinsven et al., 2003).

Figure 16.14 The national nitrogen budget for Germany has been calculated by the Federal Environment Agency (Umweltbundesamt, 2009). The data is compiled of official, national emission, deposition and flux data sets for the years 2000–2004. The most important fluxes are emissions to atmosphere, deposition, input into hydrosphere and following export to coastal ecosystems.

363


Figure 16.15 National nitrogen budget for France calculated with data gathered between 2003 and 2007 (Personal Communication Groupe de travail français sur l’azote réactif, or French working group on reactive nitrogen). The construction of the French national N-budget is an ongoing process, and the figures are therefore preliminary results which are bound to be modified.

The split of atmospheric emissions over the three reactive nitrogen gases NOx, NH3, and N2O reflects again the weight of the energy versus the agriculture sectors of the considered countries (see Figure 16.20).

16.4.2â•‡ The European Nitrogen Budget (ENB) Galloway etÂ€al. (2008) formulated five vexing questions that should guide the direction of nitrogen research in the near future. The first of them relates to the ultimate fate of reactive nitrogen, and particularly the role of denitrification in soils and freshwater systems that are not well constrained. Based on the information presented above and in previous chapters, the development of an integrated N-budget for Europe is attempted. This European Nitrogen Budget. It covers the territory of the EU27 countries, with the exception of Malta and Cyprus (limited by the availability of data from the Indicator Database for European Agriculture). The European Nitrogen Budget, in contrast to most national N-budgets, is almost completely model-based, combining a model for agriculture, forestry, industrial emissions and atmospheric deposition into a common framework as given in TableÂ€16.1. The restriction to a few models reduces the number of conflicting data as each of the models ensures consistency in the data sets used and Nr fluxes estimated. Inconsistencies at the interfaces between the models can not be excluded, though

364

they are usually small as the models deal with complementary sectors. For example, the nitrogen deposition fluxes to agricultural and (semi-) natural soils obtained from the latest simulations of the EMEP model for the year 2000 differ from the figure used in the IDEAg and INTEGRATOR models, based on slightly older versions. This points towards the need for a better fine-tuning and integration of the models, even though the differences are not large. Surprisingly, the atmospheric compartment shows a good match of total Nr emissions with total Nr deposition plus net export of nitrogen, even though the emission data are obtained from the above-mentioned models and the EDGAR-CIRCE database for industrial and energy sources, which are to some extent independent of the emission data used for the EMEP model. We are not aware of a Europewide model estimating Nr fluxes for surface waters and coastal areas and here the ENB had to be complemented by individual flux estimates taken from literature. In this chapter, we focus on nitrogen fluxes in the present time (year 2000). Nevertheless, a European nitrogen budget for the year 1900 has been developed too and is presented in the supplementary information (see supplementary material Chapter 16, Section B). When looking at the ENB, one has to keep in mind that the numbers presented are associated with large uncertainties which are difficult to quantify. The extent to which errors or biases in input data, model assumptions and model parameterizations propagate to the aggregated model output is hard to assess. In

Adrian Leip

Figure 16.16 The integrated nitrogen budget of the United Kingdom builds on the UK TAPAS modelling for agriculture (DEFRA), providing data on mineral fertilizer application rates and manure management and on the national food and feed balance modelling (DEFRA, 2008), national N flux modelling to freshwaters and the coastal zone (Johnes and Butterfield, 2002), the UK national emissions inventory and atmospheric transport and deposition modelling and a range of literature sources for UK waters. The data sets are not, however, co-incident in time, and the present budget represents the period from 1995–2005. Work is currently underway to update the budget.

the absence of sufficient observational data that allow the construction of independent estimates, two routes are possible: (i) the comparison of model results driven by the same set of input data (see de Vries etÂ€al., 2011, Chapter 15 this volume) and (ii) a systematic assessment of the uncertainty of input data and model structural parameters and their impact on the model outcome at the large scale with Monte Carlo analysis. Both routes are currently followed in the NitroEurope Integrated Project and the results will shed some light on the reliability of estimated Nr fluxes. Additionally, for some Nr fluxes, it is indeed possible to use data from atmospheric concentration measurements to quantify the strength of total Nr emissions for some components (e.g. N2O) using inversion tools which are independent of the models used in the construction of the ENB; also this route is followed in the NitroEurope project (Sutton etÂ€al., 2007). Figure 16.22 gives the nitrogen budget for the main sectors and compartments considered in the ENB. The atmospheric compartment comprises only fluxes of NH3 and NOx as input and wet- and dry deposition as output. The large fluxes of molecular nitrogen, in particular N-fixation in the industry and energy sectors and denitrification in terrestrial and aquatic ecosystems are poorly quantified and would add another 20–30 Tg N yr−1 in input and output. Nitrogen fixation occurs through biological N-fixation and through the Haber–Bosch process, but there is also input of Nr from fossil energy carriers and newly formed Nr through thermal reaction. These

fluxes are included in the number presented for N-fixation in the energy and industry sector, but are not quantified for the other sectors. Consumers and the waste sector store an unquantified amount of Nr in products, but most of the consumed Nr will accumulate in the wider environment, be land-filled or incinerated. Indeed, according to the numbers presented, more than 50% of the Nr made available to consumers appears to have a purpose other than nutrition. While some information on the fate of this Nr might be available, so far we were not able integrate robust data into the European Nitrogen Budget. At the European scale, budgets of industry and energy exceed those of agriculture. Agricultural soils can act as a source or a sink for carbon and nitrogen if organic matter is being depleted or accumulated. Our data suggest that a significant part of the nitrogen lost from the agriculture sector originates from mineralization of soil organic matter. A large exchange of nitrogen takes place in coastal areas, which act as a sink of oceanic nitrogen that is denitrified to N2 and N2O in the shelf regions of Europe. As a consequence, the flux of the N2 from these regions might be the largest single nitrogen flux in absolute terms and also the estimate for the N2O flux from the shelf regions is very high and exceeds, in absolute terms, fluxes from other sectors including agriculture. The split of atmospheric emissions in EU27 countries by sectors for three reactive gases (NOx, NH3, N2O) and the total is

365


800

Switzerland

700

Netherlands

United Kingdom

500

France

400

Czech Republic

300 200

Netherlands Germany United Kingdom 400

France Czech Republic

300 200 100

100

366

Switzerland

500

Germany

600

0

600

Emissions [Gg N yr-1]

Emissions [Gg N yr-1]

Figure 16.17 The integrated nitrogen budget of the Czech Republic is being constructed by the Hydrometeorological Institute (CHMI) in cooperation with the Ministry of Agriculture and with the Central Institute for Supervising and Testing in Agriculture on the basis of data from different sources (e.g. National Statistical Office) for the years 2004–2008, but mainly for 2007.

Industry +Energy

Transport Consumer Agriculture

(Semi-) natural

Waste

Aquatic

0 Industry +Energy

Transport Consumer Agriculture (Semi-) natural

Waste

Aquatic

Figure 16.18 Absolute nitrogen emissions to the atmosphere for the main sectors/compartments.

Figure 16.19 Absolute nitrogen emissions to the hydrosphere for the main sectors/compartments.

shown in Figure 16.23. The figure shows that NOx fluxes dominate the emissions from energy-related sources, NH3 fluxes are the strongest for agricultural sources; the waste sector and aquatic systems emit mainly de-nitrification products (N2, N2O). Overall, the emissions for NOx and NH3 are roughly at the same level, with 3.5 Tg NOx-N yr−1 and 3.2 Tg NH3-N yr−1, respectively. N2O contributes 1.2 Tg N2O-N yr−1. The global warming potential of this greenhouse gas is 580 Tg CO2-eq (using a GWP of 298; IPCC, 2007). This is higher than the emissions of N2O

estimated in the European GHG inventory (EEA, 2010) for the year 2000 (412 Mt CO2-eq) which decreased to 364 Mt CO2-eq in the year 2008. However, it should be kept in mind that the ENB emissions include both anthropogenic and natural sources while the greenhouse gas inventories are restricted to anthropogenic emissions only. For example, it is likely that a significant portion of the coastal N2O fluxes of 500 Gg N2O-N yr−1 (or 230 CO2-eq yr−1) originates from nitrogen in incoming oceanic water. In addition, there are methodological differences that influence calculations. IPCC methodology includes indirect emissions

Adrian Leip NOx

70%

NH3 60%

N2O

50% 40% 30% 20% 10% 0% Switzerland

Netherlands

Germany

United Kingdom

France

Czech Republic

Figure 16.20 Split of total atmospheric emissions over the three reactive gases NOx, NH3, and N2O.

from agricultural soils only, while the estimates presented here cover virtually all available land and include implicitly all indirect N2O emissions, including those caused by deposition of NOx fluxes from the combustion processes. To account for indirect emissions from industrial and energy sources, roughly 20 Mt CO2-eq should be added. Direct N2O emissions from agricultural soils for EU27 in 2002 are estimated as 380 Tg N2O, which is about 35 Tg N2O yr−1 (or 15 Tg CO2-eq yr−1) more than reported by UNFCCC for the categories ‘direct soil emissions’ (4D1) and ‘pasture, range and paddock manure’ (4D2) (EEA, 2010). Thus, agreement for EU27 is satisfying, even though differences are larger for individual countries. Most of the data used for the ENB are based on the same models as also used for the key maps in Section 16.3. Thus the spatial variability of the most important fluxes is shown in Figures 16.1–16.11. In the following sections, the main figures presented in the European Nitrogen Budget are briefly reviewed, emphasizing those numbers and sectors which have not yet been introduced in detail elsewhere in the European Nitrogen Assessment.

Industry sector The motivation for the invention of the Haber–Bosch process to synthesize reactive nitrogen (ammonia) from atmospheric molecular nitrogen was the urgent need for nitrogen to enable sufficient agricultural food production and the provision of raw material for explosives (Erisman etÂ€al., 2008; Sutton etÂ€al., 2011, Chapter 1 this volume). As of 2008 around 48% is the nitrogen synthesized globally by the Haber–Bosch process (121 Tg N; Erisman etÂ€al., 2008). About 24 Tg N is used in various industrial processes and the production of non-fertilizer products (IFA Statistics, 2010). Several ammonia-based products are used as fertilizers, in industrial processes and in chemical products. Besides some ammonia salts, other ammonia-based industrial products that are not used as fertilizers include nitric acid, adipic acid, hydrogen cyanides, diisocyanates, acrylonitrile, melamine and others (Domene and Robert, 2001). For Europe, it is estimated that about 30% of the Nr fixed with the Haber–Bosch process is used for non-agricultural purposes, including about 4.5 Tg N in

Western Europe in 2007, and 0.7 Tg N in Central Europe, totalling about 5.2 Tg N (see Winiwarter etÂ€al., 2011, ChapterÂ€24 this volume). An accounting of the production of nitrogen-containing substances in Europe is provided in the supplementary information (see supplementary material Chapter 16, Section C). The fate of these products is unknown. The total net trade of EU27 for total N (fertilizer and non-fertilizer) is estimated to be 1631 Gg N net import. Table 16.3 shows that the trade is dominated by the import of ammonia and urea, while derived products have a slight export-surplus.

Transport sector According to the European Environment Agency (EEA) the transport sector accounts for around one third of all final energy consumption in the EEA member countries and for more than a fifth of greenhouse gas emissions (EEA, 2009b). Transport is represented by international and domestic air, sea and inland waterway, off-road and pipeline transport, rail and road transport. It is mainly characterized by the road transport sector, which, in the year 2005 contributed more than 73% to global transport fuel consumption (EDGARv4, EIA, 2007; IEA/OECD, 2007) followed by air transport (≈ 11%), sea and inland waterways (≈Â€9%), rail transport (≈ 4%) and other transport (≈ 3%). In Europe, road transport has been the dominating source for NOx emissions since 1970 (Vestreng etÂ€ al., 2009). With the implementation of strict measures and action plans in the early 1990s within the framework of the Convention on Longrange Transboundary Air Pollution, European NOx emissions were continuously reduced (Pulles etÂ€ al., 2007). These early measures in Europe complemented clean air initiatives in the US (CONCAWE, 1997) and investigations in the automobile industry enforced by legislation. The main contributors in road transport producing high NOx emissions are heavy duty vehicles (HDV) using diesel fuel and light duty vehicles using gasoline (EMEP/EEA, 2009). With the introduction of EURO standards for light (considering passenger cars) and heavy duty vehicles emission reduction has led to a substantial NOx emissions decrease for all vehicles types in Europe. In Western Europe, NOx emissions of heavy duty vehicles have been cut by 86% compared with levels in the 1990s (ACEA, 2009). In contrast, NH3 and N2O emissions generally increased in the last years due to the worldwide turnover of vehicle fleets equipped with EURO 3/III standards. Although new emissions standards introduced significant NOx emissions reductions, the age structure of a national fleet causes a significant time lag until the new standard can show an effect. Moreover, increasing diesel consumption and increasing growth rates in freight transport volume on a national base (see Lambrecht etÂ€al., 2009) prevent further NOx emissions decreases. Furthermore, one of the reasons why some air quality problems still persist, even though vehicles have become far cleaner, is that emissions in real driving conditions tend to be higher than emissions under test conditions. Consequently, of the EU27 Member States, only 15 (up from 10 in 2007) expect to be at, or below, their respective

367

Integrating nitrogen fluxes at the European scale Table 16.2 Summary of nitrogen input, output, stock changes and the nitrogen balance for the main compartments/sectors for the National integrated Nitrogen Budgets of the Netherlands, Germany, Switzerland, France, and the United Kingdom. The numbers are summarized from the above figures, thus input and output are the sums of arrows pointing to or from the respective compartments with the exception of nitrogen fixation of N2 (both in industry and biological N-fixation), which is regarded as new Nr input. Stock changes refer only to quantified stock changes in terrestrial ecosystems (soil stock changes or standing biomass in forests) and aquatic systems (sedimentation in lakes or marine waters)

Switzerland

Input

Atmosphere

Output

Stock change

Balance

254

0

255.5

0

−2

Industry +Energy +Transport

33

53

86

0

0

Consumer

12

0

0

0

12

Agriculture

92

57

171

0

−22

(Semi-) natural land

0

0

29

0

−29

Waste

0

0

48

0

−48

112

0

100

14

−2

0

0

0

0

0

Freshwater Marine water

Output

Stock change

Balance

Netherlands

Input

New Nr

Atmosphere

791

0

2975

0

−2184

0

2600

2876

0

−276

Consumer

419

0

82

0

337

Agriculture


885

13

785

0

113


62

0

4

0

58

Waste

90

0

90

0

0

515

0

378

125

12


378

0

Germany

Input

New Nr

Atmosphere

2263

0

4096

0

−1833

0

2052

2052

0

0


0 Output

0 Stock change

378 Balance

Consumer

1010

0

498

0

512

Agriculture

2202

233

2373

0

62

0

70

118

0

−48

Waste

514

0

489

0

25

Freshwater

687

0

688

0

−1

Marine water

492

0

0

0

492

France

Input

New Nr

Output

Atmosphere

1178

0

3401.26


Stock change 0

Balance −2223


0

2683

2683

0

0

Consumer

0

0

0

0

0

Agriculture

2681

569

716

0

2534

0

0

88

0

−88

(Semi-) natural land Waste

0

0

42.53

0

−43

Freshwater

0

0

797

0

−797

Marine water

797

0

0

0

797

United Kingdom

Input

New Nr

Atmosphere

1110

0

2347

0

−1238

0

1417

1417

0

0

Consumer

728

0

243

0

485

Agriculture

1178

0

1478

0

−300


368

New Nr

Output

Stock change

Balance

Adrian Leip Table 16.2 (cont.)

United Kingdom

Input


0

0

72

0

Waste

260

0

200.7

0

59

Freshwater

719

0

721

0

−2

Marine water

605

0

0

0

605

Czech Republic

Input

Atmosphere

219


0

Consumer

46

Agriculture

360


0

Waste

51


New Nr

New Nr

Output

Output

Stock change

Stock change

Balance −72

Balance

0

629

0

−409

408

408

0

0

0

51

0

−4

38

137

0

261

0

0

0

0

0

13

0

37

0

0

70

0

−70

0

0

0

0

0

Figure 16.21 Nitrogen budget for Europe (European Nitrogen Budget) for EU-27 compiled with data for the period around the year 2000. Basis:Â€(i) atmospheric transport and atmospheric deposition:Â€EMEP Unified model, rv3.1, 2009. Atmospheric transport is obtained from the source-receptor matrix available at http://www.emep.int/ ; (ii) atmospheric emissions from industry and energy, transport and solid waste systems:Â€EDGAR-CIRCE (Van Aardenne et al., 2009), (iii) industrial trade and non-fertilizer products:Â€Prud’homme, 2009; (iv) agricultural nitrogen fluxes incl. mineral fertilizer use and food and feed trade:Â€Indicator Database for European Agriculture, V1, 2009; (v) (semi-)natural systems:Â€INTEGRATOR, 2009. Export of forestry products:Â€FAOSTAT; (vi) fishery data:Â€Eurostat, 2009 (http://epp.eurostat.ec.europa.eu/portal/page/portal/statistics/search_database); (vii) sewage system fluxes including input of nitrogen and emissions:Â€Indicator Database for European Agriculture, V1, 2009; (viii) fluxes to groundwater and surface water systems including flux from surface waters to coastal zones:Â€IMAGE, 2009; (ix) N2O emissions from coastal zones:Â€Bange (2008); (x) other fluxes from coastal zones; see text.

emission ceilings by 2010 (NEC Directive status report 2008, EEA, 11/2009) and NOx emissions will be higher than expected. In addition, global emissions are likely to increase due to the strong economic growth in regions such as East Asia (Streets

and Waldhoff, 2000), Central Europe and Southeast Asia (e.g. Thailand). Emissions will also increase in the Middle East and Africa, where less policy regulations are in place (Cofala etÂ€al., 2007).

369

Integrating nitrogen fluxes at the European scale N input

30,000

N output New N

25,000

Stock changes

Emissions [Gg N yr –1]

20,000

Table 16.3 Net trade of nitrogen of Europe (EU27). Values are net export in Gg N yrâ•›−1. Negative values indicate a net import towards the European Union

Product Urea

15,000

10,000

5,000

298

AN

215

CAN

141

NH3 UAN (estimate)

–5,000 Atmosphere Industry Consumer Agriculture Forests +Energy +Transport

Waste


Figure 16.22 Nitrogen balance by main sectors/compartments for EU27 around the year 2000. The contribution of newly generated nitrogen to the N-input as well as the contribution of changes in nitrogen stocks to the N-output (for example sedimentation in lakes and in estuaries) are also shown.

100%

NOx

NH3

N2O

Total

26 −1536 70 −1631

globally by 9%. The corresponding values for navigation are 13.3% (Annex 1 countries) and 16.4% (worldwide). Also the emissions from railway transport will increase with increasing freight transport, which has been restructured in the past 20 years.

90%

Agriculture sector

80%

Agriculture is the sector with the largest source of reactive nitrogen emissions in Europe as a whole and for each of its countries. We find a high recycling of Nr between crop production and manure excretion; the livestock sector receives about the same amount of Nr in the form of domestically produced and imported feed than the grass- and crop-sector. About 60% of agricultural products (consumed or used in industry) originate from crop production. The nitrogen use efficiency (defined as Nr in useful products relative to Nr inputs) for cultivation on soils is about 60% (considering also N-input through atmospheric deposition and biological nitrogen fixation), while the nitrogen use efficiency for a farm N-budget including animal products drops to about 30% (Leip etÂ€al., 2010a). Productivity in Europe is high, as has been shown; however, to supply the protein requirements of European citizens, about 400 Gg N yr−1 of agricultural products for food and about 3 Tg N yr−1 for feed or industrial use have to be imported. For comparison, Galloway etÂ€al. (2008; UNEP and WHRC, 2007) estimate is a significant global trade in fertilizer (31 Tg N), grain (12 Tg N) and meat (0.8 Tg N), and a net import of vegetal products (2367 Gg N) and meat (110 Gg N) to Europe. The authors include in their data Eastern Europe, a region which is a large producer and exporter of mineral fertilizer. Therefore, Europe is shown to be a net exporter of fertilizer (5376 Gg N), neglecting a large internal trade in the European Union (see Table 16.5).

70% 60% 50% 40% 30% 20% 10% 0% Industry +Energy

Transport Consumer Agriculture Forests

Waste

Aquatic

Total

Figure 16.23 Split of atmospheric emissions in EU27 around the year 2000 by sector into the three reactive gases NOx, NH3, and N2O.

A particular concern is aviation, which is the fastest-Â�growing transport sector. This growth is partly driven by increasing wealth and low prices (for aviation, fuel tax is currently not considered), which underpin strong growth in tourism travel. Aviation now accounts for more than 10% of greenhouse gas emissions. Emission standards for ships and aviation are dealt with by the respective UN organizations (International Maritime Organization, IMO, and International Civil Aviation Organization, ICAO) and by international conventions including the Convention on Long-range Transboundary Air Pollution which also addresses other sectors in addition to transport. Present measures regulate emissions (NOx) on the Landing and Take Off cycle and were designed to address airport air quality problems. The Committee on Aviation Environmental Protection (CAEP) is pursuing new certification methodologies that also take account of the flight mode as well. Table 16.4 shows that Nr emissions from civil aviation (Annex 1 countries) increased by 3.5% from 2000 to 2005, and

370

−845

AS

NPK (estimate)

0

Net export

Forestry Forests are currently undergoing net growth with net immobilization of Nr in the soil of about 810 Gg N yr−1 and a Nr uptake into the above-ground biomass of 320 Gg N yr−1 (estimated with the INTEGRATOR model). National

Adrian Leip Table 16.4 Reactive N emission in the years 2000 and 2005 for civil aviation, global aviation, marine activities and navigation and the development in percent from 2000 to 2005

Sector

Unit

Civil aviation

Gg N yr

−1

2000

2005

98

101

Global aviation

Gg N yr

−1

820

894

9.1

Marine and navigation

Gg N yr−1

1966

2229

13.3

UNFCCC

Global shipping

Gg N yr−1

3398

3954

16.4

EDGAR-CIRCE

Table 16.5 Trade of nitrogen in fertilizer, vegetal products and meat with (West, Central and East) Europe (Gg N yr−1)

Export from Europe Fertilizer

Grain

Meat

North America

2352

South America

2268

Africa

359

SE-Asia

977

Australia

59

Import to Europe

639

North America

423

South America

2318

Africa

193

Russia

81

SE-Asia

100

South America

82

Russia

28

Source: From Galloway et al., 2008.

estimates of carbon sequestration in the land use/land use change and forestry (LULUCF) sector are about 25 Tg CO2-eq (EEA, 2009a). The main uncertainty in converting this value into sequestered Nr (or stock changes in forests) is the ratio of sequestered carbon in above-ground material and soils, which differ considerably on the C/N ratio. As a rough assumption we use 50% of carbon sequestered in soils, giving stock changes of about 300 Gg N for EU27. From FAO statistics and national data we obtain a figure of 380 million m3 of total roundwood production in EU27 for the year 2000. Converted to nitrogen uptake/removal from forests this gives approximately 190 Gg N, using a basic wood density of 0.4 for average temperate and boreal trees (see IPCC, 2006), a carbon/dry-biomass ratio of 0.5 and a C/N ratio around 400 (see Katri etÂ€al., 2004). Part of this wood is used for domestic burning and the resulting emissions are included in the energy/industry figures calculated in the EDGAR database. The remaining biomass will be used in paper and wood products. However, these numbers might be an under-estimation of the real removal of wood, as a comparison between satellite imagery and forest statistics in Italy has shown (Corona etÂ€al., 2007).

Change (%) 3.5

Reference UNFCCC EDGAR-CIRCE

Waste sector There is little quantitative information on sewage sludge applied to agricultural fields. The value indicated in the figure above has been obtained from the national GHG inventories of European counties to the UNFCCC. In EU15, only seven countries report that domestic or industrial sewage sludge is applied to agricultural soils. The total of 45 Gg N per year is only a small fraction of the 0.7 Tg N of sewage sludge that is assumed to enter the solid waste sector. We do not distinguish here landfilling and waste burning. The input to the waste sector is mainly determined by household wastes; the estimate for the input to the solid waste systems includes currently only agricultural wastes. The IMAGE model uses a nitrogen factor of 3–4 kg N capita−1 yr−1, a value which is also confirmed from data which gave a value of 3.94 kg N per capita for over 60 catchments in the UK (Johnes, 2007). The IMAGE data are based on measured nitrogen influent to wastewater treatment plants, divided by the number of connected people. Therefore, the estimate based on human diet may be higher. For example, Billen etÂ€al. (2008) estimate an annual food intake of 8.2 kg N capita−1 yr−1 from national French domestic consumption data; the estimate based on household consumption figures may be higher. These include products bought but not ingested, the fate of which is solid wastes instead of wastewater. The CAPRI data used here suggest higher values with about 5 Tg N offered to the consumers. This contains 6% nitrogen in non-edible products and an assumed 30% of food wastes. Thus the actual annual nitrogen consumption in Europe including consumption by pets is estimated to be 6.3 kg N capita−1 yr−1 in the input of nitrogen from consumer to the waste water treatment systems is estimated to be 3.1 Tq N yr–1.

Aquatic systems We define coastal areas as the shelf regions with a water depth of less than 200 m (Uher, 2006). This includes most of the shallow Baltic and most of the North Sea, as well as the Adriatic Sea, but excludes most of the Mediterranean Sea such as the Balearic, Ligurian and Tyrrhenian Seas. Also only narrow strips of the Atlantic coast in Spain and Portugal are included. Budgeting nitrogen fluxes in aquatic systems is one of the most difficult parts of the European Nitrogen Budget. • We assume that Nr leached from soils enters the groundwater, however, sub-surface flow does also occur but has not yet been estimated for Europe.

371


• Atmospheric Nr deposition has been estimated on an ‘area-fraction’ basis. However, Nr deposition in sealed and non-sealed urban soils will enter the aquatic system either directly or via sewage treatment systems. This has not been accounted for. • The models predict total Nr leaching fluxes; their differentiation between nitrate and organic nitrogen from diffuse sources is not possible at the moment. Voss etÂ€ al. (2011, Chapter 8, this volume) have made an attempt to quantify the global nitrogen balance in shelf regions. Most of the flux terms are associated with considerable uncertainty. However, burial in sediments and biological nitrogen fixation are both relatively small flux terms; in Europe, biological nitrogen fixation occurs mainly in the Baltic Sea and thus the value shown in the figure above is the estimate reported for the Baltic Sea (Rahm etÂ€al., 2000; Schneider etÂ€al., 2003). However, the contribution of benthic nitrogen fixation is not considered in this value, as estimates are lacking. Burial in sediments is likely to be a small loss term, which we are not able to quantify for European shelf regions.

16.5â•‡ Conclusion Environmental problems related to nitrogen concern all economic sectors and impact all media:Â€atmosphere, pedosphere, hydrosphere and anthroposphere. Therefore, the integration of fluxes presented in depth in earlier chapters for individual sectors/media is needed to get a picture of the overall problem. This chapter presents a set of high resolution maps showing key elements of the N flux budget across Europe. Additionally, comparative nitrogen budgets are presented for a range of European countries. A European Nitrogen Budget is presented on the basis of state-of-the-art Europe-wide models and databases focusing on different parts of Europe’s society. Key maps of nitrogen fluxes have been plotted from five models and databases covering together all sectors and media in Europe. These models combine a large spatial extent with a high spatial resolution of the data and a focus on nitrogen fluxes. The maps show high pressure on the environment in regions used intensively for agriculture such as the Netherlands, the Po Valley, Brittany, but also in the banlieue of large metropolitan areas such as Paris, Berlin and London. These areas have high N-input and agricultural surplus as well as NH3 and N2O fluxes. NOx emissions, on the other hand, are dominated by industrial and combustion sources and their distribution reflects the degree of industrialization and population density. The map shows hotspots in centres of energyintensive industry, such as Sachsen-Anhalt in Germany, North Italy, the Netherlands, or along intensive traffic lines. The spatial distribution of N2O fluxes shows elements of both patterns, but it is further complicated by the strong dependence of N2O emissions from soil properties and meteorological conditions. Land productivity in Europe is high and could be sufficient to sustain the protein requirement of European population. However, a large part of these resources is invested to feed the livestock, which consume three times the nitrogen that

372

humans consume but deliver only about 50% of the proteins in human’s diet in EU-27. As a consequence, large amount of feedstuff must be imported to Europe. National nitrogen budgets are difficult to compile using a wide range of data sources and are currently available only for a limited number of countries. The summary of the national N-budgets shows that the balance is not closed for several sectors. This is partly because not all nitrogen fluxes have yet been (or could be) estimated, and partly because the data for different compartments and sectors have been taken from best available, but partly inconsistent, data sets. Overall, national N-budgets have already shown themselves to be useful tools to identify the most important N-fluxes in a country, to provide an efficient visualization of the complexity of a problem and to elaborate efficient mitigation strategies. Furthermore, through the integration of data from different and independent sources, data gaps and sometimes contradicting scientific understanding of processes have been highlighted. Modelling approaches have been used to fill in the data gaps in some of these budgets, but it became obvious during this study that further research is needed in order to collect necessary data and make national nitrogen budgets inter-comparable across Europe. The European Nitrogen Budget is largely modelbased and provided a challenge in combining five Europe-wide models and databases. Results suggest that European agriculture (EU-27) receives c. 18 Tg N yr−1 reactive nitrogen, out of which only c. 7 Tg N yr−1 find their way to the consumer or are further processed in industry. Some 3.7 Tg N yr−1 of reactive nitrogen are released by the burning of fossil fuel, out of about 16 Tg N yr−1 of N2 which is fixed into Nr each year in industry and energy generation. The contribution in emissions of reactive nitrogen of the industry and energy sectors is comparable to that of the transport sector. More than 8 Tg N yr−1 of reactive nitrogen are disposed of to the hydrosphere; Europe is a net exporter of Nr through atmospheric transport of c. 2.3 Tg N yr−1. The largest single sink for Nr appears to be denitrification to N2 in European shelf regions. However, this sink is also the most uncertain one as it concerns also Nr that is imported through the exchange with the open ocean which is potentially as large as the input of Nr in mineral fertilizer. In contrast to most chapters in this assessment, the current chapter presents considerable new information that has been just lately compiled, estimated or calculated with recently improved models. The European Nitrogen Assessment aims at providing a comprehensive assessment of nitrogen, its related problems and possible solutions, but also at raising awareness on nitrogen issues and promoting the idea of integrated assessment as an important prerequisite for successful solutions. The large variety of problems associated with the excess of reactive nitrogen in the European environment requires such an integrated nitrogen management approach that would allow for creation and closure of N budgets within European environments. The first steps to reach this goal have already been taken in the process of assembling the assessment, jointly with the

Adrian Leip

contemporarily formed Task Force on Reactive Nitrogen under the UN-CLRTAP and the NitroEurope-Integrated Project. The progress of this joint effort is reflected in this chapter.

Acknowledgements The authors gratefully acknowledge support from the European Commission for the NitroEurope Integrated Project, and from the European Science Foundation for the Nitrogen in Europe (NinE) programme and COST 729.

Supplementary materials Supplementary materials (as referenced in the chapter) are Â�available online through both Cambridge University Press:Â€ www.cambridge.org/ena and the Nitrogen in Europe website:Â€www.nine-esf.org/ena .

References Aber, J. D. (1992). Nitrogen cycling and nitrogen saturation in temperate forest ecosystems. Trends Ecol., Evol., 7, 220–224. ACEA (2009). Historical series 1990Â€– 2008, average car age in Europe, publications 2009, European Automobile Manufacturers Association. Available at http://www.acea.be/index.php/collection/ statistics. Amann, M., Asman, W., Bertok, I. et al. (2007). Cost-effective emission reductions to meet the Environmental Targets of the Thematic Strategy on Air Pollution under different greenhouse gas constraints NEC, Scenario Analysis Report Nr. 5. IIASA, June 2007. Andersson-Sköld, Y. and Simpson, D. (1999). Comparison of the chemical schemes of the EMEP MSC-W and the IVL photochemical trajectory models. Atmos. Env., 33, 1111–1129. BAFU 2010: Sticksloffflüsse in der Schweiz, Stoffflussanalyse für das Jahr 2005, Heldstab J., Reusimann J., Bledermann R., Len D. Im Auftrag des Bundesamis für ummett BAFU, limwett Hissen Nr. 1018. Bern. Bange, H. W. (2008). Gaseous nitrogen compounds (NO, N2O, N2, NH3) in the ocean. In:Â€Nitrogen in the Marine Environment, 2nd edition. Capone, D. G., Bronk, D. A., Mulholland, M. R. and Carpenter, E. J. (eds.),Â€Elsevier, Amsterdam, pp. 51–94. Billen, G., Garnier, J., Mouchel, J.-M. and Silvestre, M. (2007). The Seine System:Â€introduction to a multidisciplinary approach of the functioning of a regional river system. Sci. Total Environ., 375, 1–12. Billen, G., Barles, S., Garnier, J., Rouillard, J. and Benoit, P. (2008). The food-print of Paris:Â€long-term reconstruction of the nitrogen flows imported into the city from its rural hinterland. Region. Environ.Change, http://dx.doi.org/10.1007/ s10113–008–0051-y Billen, G., Silvestre, M., Grizzetti, B. etÂ€al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W.Â€Erisman etÂ€al., Cambridge University Press. Bouwman, A. F., Kram, T. and Klein Goldewijk, K. (2006). Integrated modelling of global environmental change: an overview of IMAGE 2.4, Netherlands Environmental Assessment Agency (MNP):Â€Bilthoven, The Netherlands.

Britz, W. and Leip, A. (2009). Development of marginal emission factors for N losses from agricultural soils with the DNDC-CAPRI meta-model. Agric. Ecosyst. Environ., 133, 267–279. Britz, W. and Witzke, H.-P. (eds.) (2008). CAPRI Model Documentation 2008:Â€Version 2 p. Institute for Food and Resource Economics, University of Bonn: Bonn (available at http://www. capri-model.org/docs/capri_documentation.pdf) Butterbach-Bahl, K., Kahl, M., Mykhayliv, L. etÂ€al. (2008). A European wide inventory of soil NO emissions using the biogeochemical models DNDC/ Forest DNDC . Atmos. Environ., 43, 1392–1402. Butterbach-Bahl, K., Gundersen, P., Ambus, P. etÂ€al. (2011). Nitrogen processes in terrestrial ecosystems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman, etÂ€al., Cambridge University Press. BUWAL (1996). Strategie zur Reduktion der Stickstoffemissionen. Bericht der Projektgruppe Stickstoffhaushalt Schweiz, Schriftenreihe Umwelt Nr. 273, Bundesamt für Umwelt, Wald und Landschaft (BUWAL) and Bundesamt für Landwirtschaft (BLW):Â€Bern, Switzerland. CDIAC (2008). Fossil-fuel CO2 emissions. http://cdiac.ornl.gov/ trends/emis/meth_reg.html. Last accessed on 11.08.2009. Cofala, J., Amann, M., Klimont, Z., Kupiainen, K. and HöglundIsaksson, L. (2007). Scenarios of global anthropogenic emissions of air pollutants and methane until 2030. Atmos. Environ., 41, 8486–8499. CONCAWE (1997). Motor vehicle emission regulations and fuel specificationÂ€– part 2, detailed information and historic reviews (1970Â€– 1996), report 6/1997. CONCAWE (2001). Motor Vehicle emission regulations and fuel specificationÂ€– part 2, detailed information and historic reviews (1996Â€– 2000), report 2/2001. Corona, P., Giuliarelli, D., Lamonaca, A. etÂ€al. (2007). Confronto sperimentale tra superfici a ceduo tagliate a raso osservate mediante immagini satellitari ad alta risoluzione e tagliate riscontrate amministrativamente. Forest, 4 , 324–332. Dalsøren, S. B., Eide, M. S., Endresen, Ø. etÂ€al. (2009). Update on emissions and environmental impacts from the international fleet of ships. The contribution from major ship types and ports. Atmos. Chem. Phys., 9, 2171–2194. De Vries W., Kros, J., Reinds, G. J. etÂ€al. (2009). INTEGRATOR:Â€a modelling tool for European-wide assessments of nitrogen and greenhouse gas fluxes in response to changes in land cover, land management and climate – calculation procedures, application methodology and examples of scenario results, Alterra Report, Alterra Wageningen UR:Â€Wageningen, The Netherlands. de Vries, W., Leip, A., Reinds, G. J. etÂ€al. (2011). Geographic variation in terrestrial nitrogen budgets across Europe. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W.Â€Erisman etÂ€al., Cambridge University Press. DEFRA (2008). Summary of Food and Farming in the UK, HMSO:Â€London. Domene, L. A. F. and Robert, U. A. (2001). Nitrogen’s role in industrial systems. J. Industr. Ecol., 5, 77–103. ECN (2008). European Compost Network: Webpages EU Biowaste Treatment and Country presentations. http://www. compostnetwork.info/. Last accessed on 11.08.2009. EEA (1998). Annual topic update 1997. Topic report 02/98, European Environment Agency and European Topic Centre on Coastal and Marine Environment: Copenhagen.

373

Integrating nitrogen fluxes at the European scale EEA (2009a). NEC Directive status report 2008, Reporting by the Member States under Directive 2001/81/EC of the European Parliament and of the Council of 23 October 2001 on national emission ceilings for certain atmospheric pollutants EEA Technical report series:Â€European Environment Agency: Copenhagen. EEA (2009b). Annual European Community greenhouse gas inventory 1990–2007 and inventory report 2009, submission to the UNFCCC Secretariat, European Environment Agency: Copenhagen. EEA (2010). Annual European Community greenhouse gas inventory 1990–2008 and inventory report 2010, submission to the UNFCCC secretariat, European Environment Agency: Copenhagen. EIA (2007). International energy statistics. U.S. Energy Information Administration: Washington DC, USA. http://www.eia.doe.gov/ emeu/international/contents.html. Last accessed on 2007. EMEP (2008). EMEP measurement database. The Co-operative Programme for the Monitoring and Evaluation of the Long-Range Transmission of Air Pollutants in Europe. Available at http://www. emep.int/. EMEP/EEA (2009). EMEP/EEA air pollutant emission inventory guidebook: 2009, EEA Technical Report. Available at http:// www.eea.europa.eu/publications/emep-eea-emission-inventoryguidebook-2009. Erisman, J. W., Domburg, P., de Haan, B. J. etÂ€al. (2005). The Dutch nitrogen cascade in the European perspective, ECN-CÂ€– 05–007, Energy Research Centre of the Netherlands (ECN):Â€Petten, The Netherlands. Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z. and Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geosci., 1, 636–639. Eyers, C. J., Addleton, D., Atkinson, K. etÂ€al. (2004). AERO2k Global Aviation Emissions Inventories for 2002 and 2025, QINETIQ/04/01113, December 2004. Available at http://www. cate.mmu.ac.uk/aero2kreports/AERO_2k_Global_Aviation_ Emissions_Inventories_for_2002_and_2025_V1_1.pdf. Galloway, J. N. and Cowling E. B. (2002). Reactive nitrogen and the world:Â€200 years of change. Ambio, 31, 64–71. Galloway J. N., Cowling E. B. and Kessler E. (2002). Reactive nitrogen. Ambio 31,Â€59. Galloway, J. N., Aber, J. D., Erisman, J. W. etÂ€al. (2003). The Nitrogen Cascade. BioScience, 53, 341–356. Galloway, J. N., Townsend, A. R., Erisman, J. W. etÂ€al. (2008). Transformation of the nitrogen cycle:Â€recent trends, questions, and potential solutions. Science, 320, 889–892. Genovese, G., Baruth, B., Royer, A. and Burger, A. (2007). Crop and yield monitoring activities: MARS STAT action of the European Commission. Geoinformatics, 10, 20–22. Heckelei, T., Mittelhammer, R. C. and Britz, W. (2005). A Bayesian alternative to generalized cross entropy solutions to underdetermined models. In: 89th EAAE Symposium Modelling Agricultural Policies:Â€State of the Art and New Challanges, 3–5 February, Parma, Italy. Hertel, O., Reis, S., Skjøth, C. A. etÂ€al. (2011). Nitrogen turnover processes in the atmosphere. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erismon etÂ€al., Cambridge University Press. IEA/OECD (2007). Energy balances of OECD and non-OECD countries on-line data service, 20037 http://data.iea.org. IFA (2010). Global Fertilizers and Raw Materials Supply and Supply/ Demand Balances, 2010–2014. Prud’homme, M., June 2010.

374

IPCC (2006). 2006 IPCC guidelines for national greenhouse gas inventories, prepared by the National Greenhouse Gas Inventories Programme, IGES: Tokyo, Japan. IPCC (2007). Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Working Group IÂ€– The Physical Science Basis. Summary for policymakers. Available at http://www.ipcc.ch/ WG1_SPM_17Apr07.pdf. IRF (2007). World Road Statistics 2007. Jeppsson, U., Alex, J., Pons, M. N., Spanjers, H. and Vanrolleghem,Â€P.Â€A. (2002). Status and future trends of ICA in waste water treatment- a European perspective. Water Sci. Technol., 45, 485–494. Johnes, P. D. (1996). Evaluation and management of the impact of land use change on the nitrogen and phosphorus load delivered to surface waters:Â€the export coefficient modelling approach. J. Hydrol., 183, 323–349. Johnes, P. J. (2007). Meeting ecological restoration targets in European waters:Â€a challenge for animal agriculture. In:Â€Redesigning Animal Agriculture, ed. D. Swain, E. Charmley, J. Steel and S. Coffey, CAB International: Wallingford, UK, pp. 185–203. Johnes, P. J. and Butterfield, D. (2002). Landscape, regional and global estimates of N flux from land to ocean:Â€errors and uncertainties. Biogeochemistry, 57/58, 429–476. JRC/PBL (2009). The EDGAR database EDGAR v4.0, available at http:\\edgar.jrc.ec.europa.eu\ Katri, K., Seija, K., Pekka, S. etÂ€al. (2004). Effect of elevated [CO2] on stem wood properties of mature Norway spruce grown at different soil nutrient availability. Glob. Change Biol., 10, 1526–1538. Kesik, M., Ambus, P., Baritz, R. etÂ€al. (2005). Inventory of N2O and NO emissions from European forest soils. Biogeosciences, 2, 353–357. Klimont, Z. and Brink, C. (2004). Modelling of emissions of air pollutants and greenhouse gases from agricultural sources in Europe, IIASA Interim Report IR-04–048, International Institute for Applied Systems Analysis:Â€Laxenburg, Austria. Lambrecht, M., Erdmenger, C., Bölke, M. etÂ€al. (2009). Strategie für einen nachhaltigen Güterverkehr (Summary:Â€Strategy for Sustainable Freight Transport in English), UBA Texte 18/2009. Leip, A., Marchi, G., Koeble, R. etÂ€al. (2008). Linking an economic model for European agriculture with a mechanistic model to estimate nitrogen and carbon losses from arable soils in Europe. Biogeosciences, 5, 73–94. Leip, A., Weiss, F. and Britz, W. (2009a). Agri-environmental nitrogen indicators for EU27. In:Â€Proceedings of the Conference on Integrated Assessment of Agriculture and Sustainable Development:Â€Setting the Agenda for Science and Policy (AgSAP 2009). Egmond aan Zee, The Netherlands, 10–12 March 2009, ed. M. K. Van Ittersum, J. Wolf and H. H. Van Laar, Wageningen University and Research Centre:Â€Wageningen, the Netherlands, pp.Â€184–185. Leip, A., Weiss, F., Orlandini, L. and Marchi, G. (2009b). Nitrous oxide emissions from agriculture in Europe as calculated within the Indicator Database for European Agriculture. In: Fifth International Symposium on Non-CO2 Greenhouse Gases (NCGG-5) Science, Reduction Policy and Implementation, Wageningen, the Netherlands, 30 JuneÂ€– 3 July, 2009. Leip A., Britz W., de Vries W. and Weiss F. (2011). Farm, land, and soil nitrogen budgets for agriculture in Europe. Environ. Pollu. (in press). Leip, A., Busto, M. and Winiwarter, W. (2010b). Developing stratified N2O emission factors for Europe. Environ. Pollut. (in press).

Adrian Leip Li, C. (2000). Modeling trace gas emissions from agricultural ecosystems. Nutr. Cycl. Agroecosyst., 58, 259–276. Metzger, S. M., Dentener, F. J., Lelieveld, J. and Pandis, S. N. (2002). Gas/aerosol partitioning 1. A computionally efficient model. J. Geophys. Res., 107 (ACH16). NitroEurope (2010) http://www.nitroeurope.eu/ (accessed 1 June 2010). OECD/Eurostat (2003). OECD/Eurostat gross nitrogen balances handbook. Organisation for Economic Co-operation and Development, Paris. Olivier, J. G. J., van Aardenne, J., Dentener, F. etÂ€al. (2005). Recent trends in global greenhouse gas emissions:Â€regional trends 1970–2000 and spatial distribution of key sources in 2000. Environ. Sci., 2, 81–99. Pilegaard, K., Skiba, U., Ambus, P. etÂ€al. (2006). Factors controlling regional differences in forest soil emission of nitrogen oxides (NO and N2O). Biogeosciences, 3, 651–661. Prud ’ homme, M. (2009). IFA Production and International Trade Committee, personal communication. Pulles, T., van het Bolscher, M., Brandt, R. and Visschedijk,Â€A. (2007). Assessment of global emissions from fuel combustions in the final decades of the 20th century DG Research, Retro Project, No. 034.64145, TNO report, 2007-A-R0132/B. p. 41. Rahm, L. etÂ€al. (2000). Nitrogen fixation in the Baltic proper:Â€an empirical study. J. Mar. Sys., 25, 239–248. Schneider, B. etÂ€al. (2003). The surface water CO2 budget for the Baltic Proper:Â€a new way to determine nitrogen fixation. J. Mar. Syst., 42, 53–64. Seitzinger, S., Harrison, J. A., Böhlke, J. K. etÂ€al. (2006). Denitrification across landscapes and waterscapes:Â€a synthesis. Ecol. Appl., 16, 2064–2090. Simpson, D., Winiwarter, W., Börjesson, G. etÂ€al. (1999). Inventorying emissions from nature in Europe. J. Geophys. Res., D104, 8113–8152. Simpson, D., Fagerli, H., Jonson, J. etÂ€al. (2003). The EMEP unified Eulerian model. model description, EMEP Report 1/2003, The Norwegian Meteorological Institute,Â€Oslo, Norway. Simpson, D., Butterbach-Bahl, K., Fagerli H. etÂ€al. (2006). Deposition and emissions of reactive nitrogen over European forests:Â€a modelling study. Atmos. Environ., 40, 5712–5726. Simpson, D., Aas, W., Bartnicki, J. etÂ€al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. SRIC (2005). Adipic acid, nitric acid and caprolactam production data 1974–2004, Tables from the Directory of Chemical Producers, dated 2 December 2005. SRI Consulting,Â€Gaithersburg, MD. Stoddard, J. L. (1994). Long-term changes in watershed retention of nitrogen:Â€its causes and aquatic consequences. In:Â€Environmental Chemistry of Lakes and Reservoirs, ed. L. A. Baker. Washington (DC):Â€American Chemical Society, pp.Â€223–284. Streets, D. G. and Waldhoff, S. T. (2000). Present and future emissions of air pollutants in China:Â€SO2, NOx, and CO. Atmos. Environ., 34, 363–374. Sutton, M. A., Nemitz, E., Erisman, J. W. etÂ€al. (2007). Challenges in quantifying biosphere–atmosphere exchange of nitrogen species. Environ. Pollu., 150, 125–139.

Sutton, M. A., Howard, C. M., Erisman J. W. etÂ€al. (2011). Assessing our nitrogen inheritance. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Svirejeva-Hopkins, A., Reis, S., Nardoto G. etÂ€al. (2011). Nitrogen flows and fate in urban landscapes. In:Â€The European Nitrogen Assessment, ed. M. A., Sutton, C. M. Howard, J. W. Erisman etÂ€al. Cambridge University Press. Uher, G. (2006). Distribution and air-sea exchange of reduced sulphur gases in European coastal waters. Estuar. Coast. Shelf Sci., 70, 338–360. Umweltbundesamt (2009a). Integrierte Strategie zur Minderung von Stickstoffemissionen, Umweltbundesamt. Available at http://www.umweltbundesamt.de/luft/downloads/emissionen/ stickstoffemissionsminderungsstrategie.pdf, last accessed on 26.11.2008. Umweltbundesamt (2009b). Hintergrundpapier zu einer multimedialen Stickstoffemissionsminderungsstrategie, Umweltbundesamt. Available at http://www. umweltbundesamt.de/luft/downloads/emissionen/ hg-stickstoffemissionsminderungsstrategie.pdf, last accessed on 26.11.2008. UNEP and WHRC (2007). Reactive nitrogen in the environment: too much or too little of a good thing: United Nations Environment Programme: Paris. Van Aardenne, J. A., Dentener, F. J., Olivier, J. G. J., Klein Goldewijk,Â€C.Â€G. M. and Lelieveld, J. (2001). A high resolution dataset of historical anthropogenic trace gas emissions for the period 1890–1990. Glob. Biogeochem. Cycles, 15, 909–928. Van Aardenne, J., Doering, U., Monni, S. etÂ€al. (2009). Emission Inventory for period 1990–2005 on 0.1x0.1 grid. Report to the Sixth Framework Programme Project No. 036961-CIRCE, 23 January 2009. Van der Werft, G. etÂ€al. (2010). http://www.falw.vu/~gwerf/GFED/ index.html (site accessed 1 June 2010). Van Drecht, G., Bouwman, A. F., Harrison, J. and Knoop, J. M. (2009). Global nitrogen and phosphate in urban wastewater, between 1970 and 2050. Glob. Biogeochem. Cycles, 23 (GB0A03). van Egmond, K., Bresser, T. and Bouwman, L. (2002). The European Nitrogen Case. Ambio, 31, 72–78. van Grinsven, J. J. M., van Schijndel, M. W., Schotten, C. G. J. and van Zeijts, H. (2003). Integrale analyse van stikstofstromen en stikstof beleid in Nederland, RIVM report 500003001, RIVM,Â€Bilthoven, The Netherlands. Velthof, G., Oudendaag, D., Witzke, H.-P. etÂ€al. (2009). Integrated assessment of nitrogen emissions from agriculture in EU-27 using MITERRA-EUROPE. J. Environ. Qual., 38, 402–417. Vestreng, V., Ntziachristos, L., Semb, A. etÂ€al. (2009). Evolution of NOx emissions in Europe with focus on road transport control measures. Atmos. Chem. Phys., 9, 1503–1520. Voss, M., Baker, A. Bange, H. W. (2011). Nitrogen processes in coastal and marine systems. In:Â€The European Nitrogen Assessment, ed. M.Â€A.Â€Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Wang, C., Corbett, J. J. and Firestone, J. (2007). Improving spatial representation of global ship emissions inventories. Environ. Sci. Technol., 42, 193–199. Weigelt, A., Weisser, W. W., Buchmann, N. and Scherer-Lorenzen,Â€M. (2009). Biodiversity for multifunctional grasslands:Â€equal

375

Integrating nitrogen fluxes at the European scale productivity in high-diversity low-input and low-diversity highinput systems. Biogeosciences, 6, 1695–1706. Weiss, F., Leip, A., 2010. Methodology for quantification of greenhouse gas and ammonia emissions from the livestock sector in EU27. In: Leip, A., Weiss, F., Wassenaar, T., Pexz, I., Fullmann,Â€T., Londjani, P., Tubiesso, F., Grandgirand, D., Monni,Â€S., Bicüa,Â€K. (eds.) Evaluation of the livestock sector’s contribution to the EU greenhouse gas emissions (GGELS) – final report. European Commission, Joint Research Centre.

376

Wieland, U. (2003). Water use and waste water treatment in the EU and in candidate countries. Statistics in Focus, Theme 8–13/2003, Eurostat. Winiwarter, W., Hettelingh, J. P., Bouwman, L. etÂ€al. (2011). Future scenarios of nitrogen in Europe. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press.

Part

IV

Managing nitrogen in relation to key societal threats

Chapter

17

Nitrogen as a threat to European water quality Lead author: Bruna Grizzetti Contributing authors: Fayçal Bouraoui, Gilles Billen, Hans van Grinsven, Ana Cristina Cardoso, Vincent Thieu, Josette Garnier, Chris Curtis, Robert Howarth and Penny Johnes

Executive summary Nature of the problem • Anthropogenic increase of nitrogen in water poses direct threats to human and aquatic ecosystems. High nitrate concentrations in drinking water are dangerous for human health. In aquatic ecosystems the nitrogen enrichment produces eutrophication, which is responsible for toxic algal blooms, water anoxia, fish kills and habitat and biodiversity loss. • The continuous nitrogen export to waters reduces the capacity of aquatic ecosystems to absorb, reorganise and adapt to external stress, increasing their vulnerability to future unexpected natural or climate events.

Key findings/state of knowledge • Nitrogen concentrations in European rivers, lakes, aquifers and coastal waters are high in many regions. In addition nitrate concentrations are increasing in groundwaters, threatening the long term quality of the resource. • In Europe, nitrogen pressures occur over large areas, implying elevated costs for meeting the long-term good chemical and ecological water quality requirements. A significant part of the European population could be potentially exposed to high nitrate values in drinking water if adequate treatments were not in place. Furthermore many of European aquatic ecosystems are eutrophic or at risk of eutrophication. • Nitrogen pressures have reduced biodiversity and damaged the resilience of aquatic ecosystems and continue to pose a threat to the aquatic environment and to the provision of goods and services from the aquatic ecosystems. • Even under favourable land use scenarios the nitrogen export to European waters and seas is likely to remain significant in the near future. The effects of climate change on nitrogen export to water are still uncertain.

Major uncertainties/challenges • Policy tools are available within the European Union and under international conventions to mitigate the nitrogen pollution in water, but their full implementation has not been achieved yet throughout Europe. • In many cases a delay in the water quality response to the implementation of measures have been observed, due to previous nitrogen accumulation in soils, sediments or aquifers or to inadequate design of the mitigation plans. • The issue of pollution swapping between environmental compartments has appeared as an important element to be considered by both the scientific and policy prospective.

Recommendations • To protect and enhance the European water resources the full implementation of the existing regulations related to nitrogen is necessary, in addition to an efficient environmental monitoring. • Moreover, positive synergies could be obtained by encouraging the integration in the sectoral policies and enhancing interdisciplinarity in the scientific research, especially in support of regional assessments and pollution swapping evaluations.


379

Nitrogen as a threat to European water quality

17.1â•‡ Introduction Human activities are responsible for consistent Nr export to the environment (Vitousek etÂ€ al., 1997; Schlesinger, 2009; Erisman etÂ€al., 2011, Chapter 2 this volume). The enrichment of nitrogen in the aquatic system impairs the water quality of rivers, lakes, aquifers and coastal and marine waters, and contributes to the phenomenon of eutrophication (European Environment Agency, 2001; Durand etÂ€al., 2011, Chapter 7 this volume; Voß etÂ€al., 2011, Chapter 8 this volume; Billen etÂ€al., 2011, ChapterÂ€ 13 this volume). Nr is fundamental for global food production and is still insufficient in many world regions (Sanchez and Swaminathan, 2005). However, the significant anthropogenic nitrogen mobilisation through agricultural activities, waste water discharges and fossil fuels combustion produces detrimental impacts on the aquatic environment and affects both human and ecosystem health (Lavelle etÂ€al., 2005). Estimates for year 2000 indicate that Europe is exporting 4.7 Tg of nitrogen per year to its seas (Bouraoui etÂ€al., 2009; Billen etÂ€ al., 2011, Chapter 13 this volume) and trends show that the production of Nr and its emission to the environment is accelerating because of the rise of agricultural demands and commercial energy production (Galloway etÂ€al., 2008). The aquatic ecosystems are able to remove a significant part of incoming nitrogen load but this capacity is not unlimited and strongly depends on the local ecosystem characteristics (Howarth etÂ€al., 1996; Kronvang etÂ€al., 1999a; Alexander etÂ€ al., 2000; Mulholland etÂ€ al., 2008; Durand etÂ€al., 2011, Chapter 7 this volume). Consequently there is a lot of uncertainty on the amount of self-purification of aquatic ecosystems (Seitzinger etÂ€al., 2006; Hejzlar etÂ€al., 2009) and on the capacity to absorb nitrogen pollution without undergoing radical changes (Millennium Ecosystem Assessment, 2005). In addition, nitrogen can build-up slowly in soil and water systems such as aquifers and reservoirs, and actual remediation practices might produce their effects only in the long-term (Jackson etÂ€al., 2008). Europe is thus pouring nitrogen in its water resources affecting human and ecological systems at a rate that is unlikely to reverse in the near future and with consequences that are only partially understood. The socio-ecological systems have some capacity to absorb the pollution, reorganise and adapt to the external change, but this capacity is not unlimited and a slow change of the nitrogen pool may result soon or later in some chronic or drastic unexpected effects (Carpenter and Folke, 2006). In this context of uncertainty and variability, the challenge is to better understand the extent of nitrogen enrichment in water systems and the threats it poses for human and ecosystem health in the prospective of current changing drivers, such as climate change, land use change, pollution and economic growth, and to consider which mechanisms of adaptation and mitigation the science-policy interactions need to produce. In this chapter we will try to address this challenge. The paper starts by illustrating the trend of nitrogen in European rivers, aquifer and coastal waters, in order to understand the intensity and the location of the problem. Next, the threats posed by nitrogen enriched waters to human health and aquatic ecosystem functioning are considered and then analysed further in the light of major future drivers, notably land-cover and

380

climate changes. Finally, the paper proposes a reflection on the current adaptation and mitigation strategies and on the possibilities and positive synergies of science-policy interactions.

17.2â•‡ Nitrogen enrichment in European waters 17.2.1â•‡ Nitrogen trends in European surface waters In Europe some efforts have been made during the last two decades to reduce the nutrient input from waste water discharge, but diffuse pollution from agriculture still remains a major threat for waters (European Environment Agency, 2005). Figure 17.1 shows the observed annual nitrates concentrations reported by OECD (2008) at the mouth of some major European rivers. Trend analysis indicates that in Europe between 1992 and 2005 nitrogen and phosphorus concentrations remained relatively constant in lakes while they decreased in rivers (European Environment Agency, 2009). In fact, there was a slight decrease of nitrogen concentrations in European rivers compared to the values of the 1990s, except in southern Europe, and phosphorus concentrations have significantly declined, reflecting the general improvement in wastewater treatment and the reduction of phosphates in detergents (European Environment Agency, 2005). Yet, in Europe, trends in nutrient concentrations vary according to the different regions depending on local conditions. According to the information provided by the Member States on surface water quality (European Commission, 2007 COM(2007)120), between the two periods 1996–1999 and 2000–2003, the nitrate concentrations decreased in 55% of the monitoring stations and were stable in 31%. However, in 14% of the monitoring locations nitrate concentrations were increasing. Stations reporting increasing trends were located in Luxembourg, France, United Kingdom, Portugal and Belgium, while decreasing and stable trends were found in Denmark, Austria, Ireland, Sweden, Germany and the Netherlands (European Commission, 2007 COM(2007)120). These trends need to be evaluated regionally and considering the contemporary changes in nitrogen sources. In fact, between the two above mentioned periods, nitrogen input of mineral fertiliser and manure declined by 6% and 5%, respectively (European Commission, 2007 COM(2007)120), atmospheric deposition slightly decreased (Simpson etÂ€al., 2011, Chapter 14 this volume) and nutrients point discharges were reduced by improving waste water treatments (European Environment Agency, 2005). Therefore, in certain areas nitrate concentration in surface waters may have remained constant in spite of some reduction in nitrogen inputs. According to data provided by Member States (EU27) in the last Nitrates Directive reporting exercise covering 2004–2007, nitrate concentration is increasing in 30% of the monitoring stations, while it is stable or decreasing in 70% of the stations (European Commission 2010 COM(2010)47). The current concentrations in rivers generate significant nitrogen loads to the seas (Billen etÂ€al., 2011, Chapter 13

Bruna Grizzetti

Figure 17.1 Annual nitrate concentrations (in mgN/l) in surface water at the mouth of some major European rivers (from OECD, 2008).

this volume). In European coastal waters, nitrate concentrations have remained generally stable in the Baltic, North and Celtic Seas and have increased in some Italian coastal areas (FigureÂ€ 17.2; European Environment Agency, 2005). Artioli etÂ€al. (2008) compared nitrogen budgets for European seas over three periods:Â€ before eutrophication, during severe eutrophication and in current situation. According to their study in the Baltic Proper nitrogen and phosphorus riverine loads remained stable since the eutrophication period (1955–1985), in the Coastal North Sea nutrient inputs have declined after the severe eutrophication period (ending around 1990), and in the Northern Adriatic Sea riverine loads have increased for nitrogen, while they have been halved for phosphorus as a result of phosphate banning policies in detergents. For more details

on nitrogen trends in the European Seas see Voss etÂ€al., 2011 (Chapter 8 this volume).

17.2.2â•‡ Nitrogen accumulation in aquifers Groundwater is an important resource in Europe, providing water for domestic use for about two third of the population but groundwater is a finite and slowly renewed resource and overexploitation associated with a degradation of water quality is putting in danger an important source of drinking water. In Europe, groundwater nitrate concentrations have remained stable and high in some regions (European Environment Agency, 2005). In the Third Assessment Report on the Implementation of the Nitrates Directive, the European Commission (European

381

Nitrogen as a threat to European water quality Figure 17.2 Nitrate, phosphate and N:P ratio trends observed in the European coastal waters. Source:Â€EEA web site http://www.eea.europa.eu Copyright EEA, Copenhagen, 2006.

Commission, 2007 COM(2007)120) reports that for the period 2000–2003 about 17% of the wells in EU15 exhibit a concentration of nitrate above the limit of 50 mg/l. An additional 7% were in the range between 40 and 50 mg/l, while about 60% were below 25 mg/l. Analysing the trend between the third and second assessment report, the European Commission found that even though 30% of the reported wells show an improvement in their concentration of nitrate, an alarming 36% show an increasing trend (European Commission, 2007 COM(2007)120). According to the last assessment report, covering the period 2004–2007, nitrate pollution in groundwater is still observed in 34% of the monitoring stations, with 15% of the stations with nitrate concentrations above 50 mg/l (European Commission, 2010 COM(2010)47). Background concentrations of nitrate in groundwater are very low. Most of the nitrates found in groundwater are thus of anthropogenic origin and mostly related to agricultural activities. Van Drecht etÂ€ al. (2003) estimated the total leaching of nitrogen to groundwater at 55 Tg/yr at the global scale with a contribution of 8 Tg/yr for Europe, of which 40% will reach the rivers outlets. Contribution of deep aquifers mostly affected by historical use of fertiliser was estimated at 10% of the total load of nitrogen. These calculations were made at the global scale and might hide some spatial and temporal variations, however they agree with some more detailed estimates. Behrendt etÂ€al. (2003) estimated the groundwater contribution to total load of nitrogen to be 48% for the Danube for the period 1998–2000. Schreiber etÂ€al. (2003) analysing all German catchments found a groundwater contribution ranging from 38% to 69%. Palmeri etÂ€al. (2005) estimated the contribution of groundÂ�water to total nitrogen load in the Po Valley to be around 36%. Even though highly variable and dependent on the degree of agriculture intensification and hydrogeological properties of the aquifers, groundwater is a significant source of nitrogen at the catchments outlets.

382

The quantity of nitrate present in the groundwater is strongly linked to the amount of nitrogen applied in agricultural land, and to the nitrogen surplus in particular. Indeed nitrogen surplus in agricultural land can be removed by surface runoff, leaching to the aquifer, and loss to the atmosphere or can be stored in the soil–water system. Bouraoui etÂ€al. (2009) estimated the surplus of nitrogen for Europe at 11.5 Tg for year 2000 and 10 Tg for 2005 (Figure 17.3). This surplus was calculated without considering volatilisation from manure as this pathway is an additional pressure on the environment (details on surplus computation are given in the legend of Figure 17.3). At European scale, there is a decrease of nitrogen surplus for many countries (see Figure 17.3). Dramatic decreases are observed in the Netherlands, Denmark, and Germany where the nitrogen surplus is back to the level of that of 1970. However, there is still no strong evidence that the groundwater level is responding to the decrease of nitrogen surplus. The most striking cases are those of the Eastern countries that have seen a decrease by half of the nitrogen surplus, due to the economic and political changes at the beginning of the 1990s. Improvement in the water quality observed in streams is yet to respond to these changes, as large quantities of nitrate are stored in the aquifers and are released slowly depending on groundwater residence time, which may vary from weeks to several thousands of years (Alley etÂ€al., 2002; Schlesinger, 2009). In addition, nitrogen stored in the soil system might be released slowly due to the mineralisation process (Stalnacke etÂ€al., 2004; Grimvall etÂ€al., 2000), and nitrate residence time in the unsaturated zone. For example, Sileika etÂ€al. (2006), analysing long term data on nitrate concentrations in Nemunas River (Lithuania), noted a strong increase of nitrates in surface water from the Soviet period despite the large drop in fertilisation, due among other to a large storage and accumulation of soil nitrogen during the Soviet period. Even though no clear conclusion can be drawn on the response time of aquifers to changes in fertiliser application,

Bruna Grizzetti

Figure 17.3 Estimated nitrogen surplus for European and some Mediterranean countries (kg N per ha of agricultural land). The red bar indicates the year of implementation of the Nitrates Directive (1991/676/EEC). Nitrogen surplus is computed using a simple national mass balance approach. The inputs considered include:Â€mineral application of nitrogen (source FAO), nitrogen from manure application calculated using animal number (source FAO) and excretion coefficients non-corrected for volatilisation, atmospheric nitrogen deposition (source EMEP), symbiotic biological fixation (calculated as the nitrogen in crop harvest for soybean and pulses), non-symbiotic fixation (estimated at 25 kg/ha for rice and 5 kg/ha for other upland crops). The output considered was crop harvest taken as the crop yield (source:Â€FAO) multiplied by nitrogen crop content coefficients.

383


it can be expected that past unbalanced fertiliser strategies will impact for a long time the quality of European groundwater and in turn the surface water quality. Wriedt and Bouraoui (2009) estimated the average residence time for continental Europe for elementary river basin of about 100 km2 based on river density, slope, and parent material properties. According to these estimates, northern countries are characterised by shorter residence time (less than five years) and one can expect a faster reaction of the groundwater to any improved fertilisation strategy. Longer residence time are calculated in Southern Europe, Southern England (Chalk area), Poland, and along the North Sea Coast of Europe and thus any management decision to reduce nitrate load to groundwater might take decades to see any positive effect. All this evidence indicate that in Europe the past and current anthropogenic activities are impacting and might impact water resources for the years and decades to come. Indeed, nitrogen surplus from agriculture is still high in many countries and huge quantities of nitrogen are stored in the soil or aquifers. There are some major concerns as the Eastern countries will probably intensify their agriculture and thus their fertilisation rate in the near future, and some countries of Western Europe have not seen their nitrogen surplus decrease but rather stabilise at high levels. Efforts have been taken, through conventions or the application of binding Directives, and still Europe’s waters are suffering from excess nitrogen. It is a complex task to estimate how and how long it will take to restore Europe’s waters to good quality. So it is still a priority to assess how this excess nitrogen is affecting both human and ecosystems health, and evaluate how this impact will vary in a changing environment.

17.3â•‡ Threats for human and ecological aquatic systems 17.3.1â•‡ The human–ecological system The human and the environmental systems are strictly interconnected. Humans are altering the natural nitrogen cycle to increase their benefits from nature, but they are affected by the changes they are causing in the environment. This section analyses the threats induced by the nitrogen enrichment in European waters considering humans and aquatic ecosystems as the principal receptors of impacts in a context of inherent mutual relation for which human actions affect ecosystems and the impaired ecosystems affects human health and well-being. The Millennium Ecosystem Assessment has highlighted the benefits that people obtain from nature, the ecosystem services, and the tight link between human and ecological systems (Millennium Ecosystem Assessment, 2005). Ecosystem services include provisional services, such as food, fresh water, wood and fibre; regulating services, which affect water purification and regulation of climate, flood and disease; cultural services, which provide recreational, aesthetic and spiritual benefits; and supporting services, such as nutrient cycling, soil formation and primary production (Millennium Ecosystem

384

Assessment, 2005). These services support livelihood and development of human society and their sustainable use is fundamental for the human wealth and security (Folke etÂ€al., 2002). Water and nitrogen are directly or indirectly involved in all the ecosystem services. Nitrogen in aquifers and reservoirs impairs water quality for drinking purpose, affecting directly the human health. Nitrogen enrichment in lakes, rivers and coastal and marine waters may produce the phenomenon of eutrophication which has detrimental effects for aquatic ecosystems. Consequently, the ecosystem per se is damaged and some services such as fish provision or aesthetic and recreational uses are directly affected. Moreover, through the intensive mobilisation, the nitrogen cycle and primary production may become distorted and some regulating services may be reduced or compromised. For example evidence shows that busting denitrification through additional nitrogen input in the water system may increase the emission of N2O to the atmosphere, which acts as a strong greenhouse gas affecting the climate. Studies have highlighted that the efficiency of water purification may be reduced increasing the total amount of nitrogen input in the river system (Mulholland etÂ€ al., 2008), and that aquatic ecosystems significantly impacted by eutrophication are more vulnerable to flood events and diseases spreads (Folke etÂ€ al., 2004; McKenzie and Townsend, 2007). Thus, altering the nitrogen availability in water is likely to significantly affect the social and environmental system, reducing many ecosystem benefits. The socio-ecological system has some capacity, resilience, to ‘absorb disturbance and reorganise while undergoing change so as to still retain essentially the same function, structure, identity and feedbacks’ (Walker etÂ€ al., 2004). Resilience is often associated with diversity, such as biological species and economic options, to support the ecosystem capacity to renew and reorganise into a desired state although under pressure (Carpenter and Folke, 2006). Moreover, the actors of the socio-ecological system have the possibility to adapt to ongoing changes in order to moderate the undesired effects and as well to try to reverse them. However, this ecosystem ability to absorb stress and recuperate is not linear and is possible only until a certain threshold (tipping point), beyond which the recovery is difficult or impossible, leading to a regime shift. Human degradation of the environment has reduced the ecosystems resilience, shrinking the ability to mitigate natural hazards (Carpenter and Folke, 2006) and increasing the likelihood of drastic changes to less desired capacity to generate ecosystem services (Scheffer etÂ€al., 2001; Folke etÂ€al., 2004). There are various ways in which nitrogen enrichment of water can affect human and aquatic ecosystem health. Direct effects from use of drinking water have been described extensively in the scientific literature and are the subject of current policies. Indirect effects such as from eutrophication are less well known. In the following part we describe the threats on human health due to nitrogen in drinking water and then the threats on aquatic ecosystems and their consequent indirect effects on human health and well being, providing, where possible, estimates at European scale.

Bruna Grizzetti

17.3.2â•‡ Effects of nitrogen rich drinking water on human health Currently in many countries there are strict limits on the permissible concentration of nitrate in drinking water and in many surface waters. The limit is 50 mgNO3/l in the European Drinking Water Directive (Directive 98/83/EC) and 44 mgNO3/l in the United States (equivalent to 11.3 mgN/l and 10 mgN/l, respectively). These limits are in agreement with WHO recommendations established in 1970 and recently reviewed and reconfirmed (WHO, 2007; the exact formulation of the standard is that the sum of NO3/50 + NO2/3 should not exceedÂ€1). The European Nitrates Directive also sets a limit concentration of 50 mgNO3/l for groundwater and surface water, as a threshold value for Member States to protect water bodies. There are two main health issues related to nitrate in drinking water:Â€the linkage with infant methaemoglobinaemia, also known as blue baby syndrome, and with cancers, for example of the digestive tract (Ward etÂ€al., 2005). The evidence for nitrate as a cause of these serious diseases is controversial (Powlson etÂ€al., 2008; Salomez and Hofman, 2003). In addition there is evidence for increased cardiovascular health with increased nitrate intake (Webb etÂ€al., 2008). Presently, it is widely accepted that methaemoglobinaemia in Europe is rare, and that in general incidence is related to presence of pathogens in drinking water rather than to nitrate or nitrite (Addiscott, 2005). The emerging and returning question is whether nitrate in drinking water is harmful to humans, and if the drinking water standard in some cases could be increased (van Grinsven etÂ€al., 2006; L’hirondel and L’hirondel, 2002). Such an increase would have great implications for policies, measures and costs related to water treatment and to fertiliser and manure application. To answer this question, the following points are relevant. • Only a small proportion ( 1.5

Values are based on literature (Guidance Document on Eutrophication from Intercalibration Group; Vollenweider etÂ€al., 1976; OECD, 1982; Cardoso etÂ€al., 2001).

at European scale. Three classes of potential risk were established based on the literature (Vollenweider etÂ€al., 1976; OECD, 1982; Cardoso etÂ€ al., 2001) and on the information available from the Water Framework Directive Intercalibration Exercise (Guidance Document on Eutrophication from Intercalibration Group). The classification of total nitrogen concentration in classes of potential risk is shown in Table 17.2. These values, which are in agreement with the ecological thresholds reported in other studies (Durand etÂ€al., 2011, Chapter 7 this volume), were combined with estimates of total nitrogen concentrations in European surface waters (Bouraoui etÂ€ al., 2009) to derive a European map of potential risk of eutrophication related to nitrogen (Figure 17.9). The map (Figure 17.9) shows that large parts of European water courses may be threatened by potential risk of eutrophication due to nitrogen concentrations. The risk varies with countries but is generally lower in the Scandinavian region (Figure 17.10). Figure 17.11 reports the potential risk of eutrophication per country for European inland wetlands using the same nitrogen concentration classification. The Water Framework Directive requires that Member States evaluate the ecological status of their surface waters based on the deviation in the status of a number of biological elements (such as phytoplankton and macrophytes) from the water body type specific reference condition. The division of water bodies into types and the establishment of type specific reference conditions should allow for better resolving the effects on the biological elements (biological Figure 17.9 Map of potential risk of eutrophication for surface freshwater based on estimated total nitrogen concentrations. The map shows three classes of nitrogen concentration estimated in surface waters associated with a potential risk of eutrophication. The related total nitrogen concentration per class of risk is reported in Table 17.2.

390

Bruna Grizzetti Figure 17.10 Percentage of surface of inner waters per class of potential risk of eutrophication related to total nitrogen concentration (see TableÂ€17.2) per country. Inner waters refer to Corine Land Cover 2000 classes 40 and 41 (water courses and water bodies, respectively).

Figure 17.11 Percentage of surface of inner wetlands per class of potential risk of eutrophication related to total nitrogen concentration (see Table 17.2) per country. Inner wetlands refer to Corine Land Cover 2000 classes 35 and 36 (inland marshes and peat bogs, respectively).

communities) from the variation associated to natural pressures (such as natural water fluctuations) and thus a more precise evaluation of the ecological condition. Figure 17.12 shows the ecological status of the European lakes, based on data that have been collated as part of the Water Framework Directive Intercalibration exercise (2004–2007). The ecological status of lakes was evaluated by chlorophyll-a values by the Mediterranean, Atlantic, Central/Baltic and Northern Geographic Intercalibration Groups (GIG), while for the Northern GIG the ecological status evaluation included chlorophyll-a values and indicators of the status of macrophytes communities’ composition. Concerning water acidification, within Europe, nitrogen and sulphurs emission ceilings are set to prevent critical load

exceedance under both the Gothenburg Protocol of the UN-ECE Convention on Long-range Transboundary Air Pollution (CLRTAP) and the EU National Emissions Ceiling Directive (NECD) (see Hettelingh etÂ€al., 2007, 2008). Five countries currently submit freshwater critical loads data to the international mapping and modelling programme under the CLRTAP; the UK, Norway, Sweden, Finland and Switzerland (Canton Ticino). However, impacted acid sensitive lakes and streams are present in many other countries including France, Spain, Italy, Austria, Poland, Slovakia, Czech Republic, Romania, Bulgaria and Germany (Curtis etÂ€al., 2005b; Evans etÂ€al., 2001). The critical load of total nitrogen deposition, when excluding the effects of sulphur deposition, is < 400 eq/ha per year across large areas of Scandinavia and parts of the UK and the

391

Nitrogen as a threat to European water quality Figure 17.12 Ecological status of European lakes according to the harmonised methodology of the Water Framework Directive Intercalibration Group (in most of the region the ecological status is evaluated by chlorophyll-a values, explanation in the text). The lakes are subdivided in three main classes of High, Good and Less Good ecological status (Lake Intercalibration data provided by the Lake Geographical Intercalibration Groups and available at CIRCA folders:Â€http://circa.europa.eu/ Members/irc/jrc/jrc_eewai/library. Data organised by Sandra Poikane, Joint Research Centre).

Swiss Alps and these deposition levels are exceeded across large regions of Europe (EMEP, 2009).

17.4â•‡ Prospects for European water quality 17.4.1â•‡ Impact of climate change on freshwater quality Most striking expected climate change is a continental global warming from 1 °C to 4 °C more pronounced during winter in Eastern Europe and in summer for Southern and Western Europe. Precipitation exhibits a strong north to south gradient with an increase in northern Europe and a decrease going south (Alcamo etÂ€al., 2007) and a more pronounced seasonality. In addition there is a high probability of increase of extreme events. These changes will affect directly and indirectly the quality of freshwater and coastal and marine waters in Europe (Battarbee etÂ€al., 2008; European Environment Agency, 2008). Indeed the predicted climate change will affect the nutrient cycling in the water bodies, but will also affect the generation and transport processes of nitrogen originating from land based sources. The impact of climate change on water quantity has been studied extensively while fewer researches have focused on the impact of water quality. Water quality and nutrient concentrations in particular, will be affected by water quantity due to dilution or concentration effects, mobility, residence time and water temperature impact (Ducharne etÂ€ al., 2007; Whitehead etÂ€al., 2009). Direct impacts of increased temperatures will be

392

accelerated nutrient cycling (Whitehead etÂ€al., 2009; Murdoch etÂ€al., 2000). Furthermore, climate change or global change in general often results in synergistic effects of the different stressors (temperature increase, precipitation and land use change). It is thus very difficult to isolate the single effect of each stressor on the fate of nitrogen. In southern European countries it is expected that climate change will result in lower runoff, exacerbated by increased water abstraction for irrigation purposes and growing human consumption. An increase in the number of intermittent streams and drought periods is also expected. The higher population release of nutrients due to eating habits with an associated increased industrial production will release increased amount of nutrients in the lower water volume resulting in both higher nutrient loads but also higher nutrient concentrations in the water courses due to point source discharges, unless additional treatment is applied (Eisenreich, 2005). Higher nutrient concentrations downstream of point discharges (industries, concentrated animal management operations) are also predicted by Whitehead etÂ€al. (2009) and Murdoch etÂ€ al. (2000). Concerning agriculture, even though in Southern countries the decrease in water precipitation is expected to result in lower nutrient loads, the concentration of nutrients might increase due to lower water volume circulating in the streams (Mimikou etÂ€ al., 2000). Therefore, for southern regions, an increased contribution of point sources relative to diffuse sources is anticipated with associated management problems.

Bruna Grizzetti

In northern regions, higher precipitation may also increase nutrient loads, and higher temperatures are likely to enhance nitrogen mineralisation. Ducharne etÂ€al. (2007), studying the Seine river basin, predicted that an increase of temperature by 2.3 °C will result in an increased soil nitrogen mineralisation from 8% to 26%. Taking into account all processes enhancing or decreasing nitrogen production from mineralisation, they expected a net increase of 20% for nitrate leaching and nitrate in streams if appropriate farming practices are not implemented. Bouraoui etÂ€ al. (2002), evaluating the impact of six climate change scenarios for the River Ouse (UK), found that climate change will increase nutrient losses to surface water due to the combined effects of increased mineralisation and the increased amount of water circulating through the soil profile. Wilby etÂ€al. (2006), who studied the impact of climate change on water resources in the River Kennet (UK), also predicted increasing concentrations of nitrates and ammonium until 2050. They estimated that due to the increased occurrence of summer drought, nitrates will build up in the soils and will be flushed into the streams when droughts break. Bouraoui etÂ€al. (2004) evaluated the impact of observed climate change occurring in Finland and predicted that the increase in temperature associated with the increased winter runoff due to accelerated snowmelt results in higher nitrogen loading by about 3% with pronounced seasonality. They found increased losses of nitrogen during the winter and reduced loads during the traditional snow melt period. Similar conclusions were reached by Arheimer etÂ€ al. (2005) when studying the Ronnea catchment (Sweden). The authors predicted an increased nitrogen loading due to increased winter precipitation and increased mineralisation from soil organic matter despite an increase in surface and groundwater nitrogen retention. The increased retention under the climate change scenarios is attributed to increased temperature, increased nitrogen concentration and drier summer conditions yielding a longer transit time in the summer, the most active season for nitrogen removal process. They also noticed an extension of the areas contributing to the total nitrogen loads, with obvious management implications. Assessing the impacts of climate change on nutrient loads or concentration in freshwater and coastal ecosystems is not straightforward as many other factors impact their status and health. Eutrophication is not only controlled by nutrient availability but is also affected by light conditions, temperature, residence time, flow regime. Invasion of non-native species will obviously depend on temperature, flow conditions but also on nutrient availability and cycling (Murdoch etÂ€al., 2000; Rahel and Olden, 2008).The impact of climate change on freshwater ecosystems has been studied in controlled experiments in mesocosms tanks (Feuchtmayr etÂ€ al., 2007; Christoffersen etÂ€al., 2006; Moss etÂ€al., 2003). Feuchtmayr etÂ€al. (2007) showed in their experiment that higher nutrient concentrations would result in increased population of four out of eleven of the macroinvertebrates used in their study. McKee etÂ€al. (2003) measured an increase in zooplankton abundance under increased nutrient addition. Moss etÂ€al. (2003) showed that nutrient addition (unlike warming) increased phytoplankton chlorophyll a concentration and total algal biovolume, and did not affect the

number of species. In northern boreal regions where lentic and lotic ecosystems are oligotrophic, an increase of nutrient loads might lead in earlier stages to an increase of biodiversity. Conversely, in southern boreal countries where freshwater ecosystems are eutrophic, an increase in nutrient loads could result in a decrease of biodiversity. Under the combination of increased temperature and nutrient inputs, arctic lakes will exhibit an increase algal production and biomass leading to a potential colonisation of predatory fish (Flanagan etÂ€al., 2003). The impact of climate change on ecosystems has been studied through the detailed analysis of dry years and even droughts. Many authors report that under increased temperature, blooms of the harmful of cyanobacteria will likely increase (Johnk etÂ€al., 2008; Paerl and Huisman, 2008). Many reasons explain this increase in bloom:Â€higher temperatures favour the cyanobacteria growth, the warming of the surface water will reducing vertical mixing, and there will be an increase of the growth period to an earlier shift of stratification in spring and late destratification in autumn. Mooij etÂ€ al. (2005) reviewing the impact of climate change on lakes in the Netherlands also report the increased presence of cyanobacteria and their predominance in the phytoplankton community. The increased loading of nutrients for instance under severe storms or wet winters followed by dry or drought condition and associated increased residence time will increase algae blooms (Paerl and Huisman, 2008). Arheimer etÂ€ al. (2005) concludes also that under climate change, the increased amount of inorganic nitrogen entering the Ronnae lake (Sweden) will stimulate algae growth resulting in increased concentrations of cyanobacteria, zooplankton and detritus. Similar effects are reported by Whitehead etÂ€al. (2009). For additional information on the effects of climate change on coastal and marine waters and the implication for the link between nitrogen and carbon cycle see Voß etÂ€al., 2011 (Chapter 8 this volume). Studies at the European scale have shown that climate changes are expected to alter the soil nitrogen cycle with great regional variability (Bouraoui and Aloe, 2007), and to exacerbate the problem of eutrophication, enhancing algal blooms and new harmful invasive algae (project EURO-LIMPACS, Battarbee etÂ€al., 2008). All these studies are affected by uncertainty, but some of the climate change effects are already being observed in European water ecosystems (Battarbee etÂ€al., 2008; European Environment Agency, 2008).

17.4.2â•‡ Nitrogen fate under future land use changes Due to the interdependence of the world economic and ecological systems, prospective scenarios of human activities in watersheds should be conceived at the global level. The Millenium Ecosystem Assessment has provided the story lines of four scenarios of the world future named Global Orchestration (GO), Order from Strength (OS), Technogarden (TG), and Adapting Mosaic (AM) (Alcamo etÂ€al., 2006). The scenarios differ in terms of environmental management (pro-active or reactive) and in the degree of connectedness among and within institutions across country borders (globalisation or regionalisation). Technogarden and

393

Nitrogen as a threat to European water quality Table 17.3 Total nitrogen and total phosphorus river load to European coastal waters calculated by the GlobalNEWS models for 2000 and for 2030–2050 according to the two extreme Millenium Ecosystem Assessment scenarios

Total nitrogen load (TgN/yr)

Total phosphorus load (TgP/yr)

2000 reference

4.04

0.59

2030 Global Orchestration

4.01

0.56

2050 Global Orchestration

3.99

0.54

2030 Adapting Mosaic

3.64

0.55

2050 Adapting Mosaic

3.50

0.53

Scenario

Adapting Mosaic were developed assuming pro-active environmental management, while Order from Strength and Global Orchestration assume reactive environmental management. Global Orchestration and Technogarden reflect trends towards globalisation, while regionalisation is assumed in Order from Strength and Adapting Mosaic. The IMAGE and GlobalNEWS models of watershed nutrient fluxes have been used to calculate nitrogen and phosphorus delivery at the outlet of 5700 world watersheds (Seitzinger etÂ€ al., 2009). The results for European watersheds are shown in Table 17.3 for the two more contrasted scenarios GO and AM. The trends shown are the results of several opposite drivers. The number of inhabitants with a sewage connection will further increase in all scenarios, but the removal of nitrogen in wastewater treatment also increases. On the other hand, the overall nitrogen use efficiency in agricultural production in Europe increases in all scenarios, particularly Adapting Mosaic (AM), and this leads to an overall reduction of nitrogen river export. In the GO scenario there is also an increasing production. In AM, with a faster population growth than in GO between 2000 and 2030, the nitrogen export will decrease. This is caused by a combination of lower meat consumption and a major effort in better incorporating animal manure in the agricultural system. Although global scenarios are best able to take into account the interconnected nature of world economies, there is also a need to increase the spatial resolution of their simulated results, in order to examine them at the sub-regional scale, closer to that at which management decisions are to be taken. With this in mind, Thieu etÂ€al. (2010a) proposed an interacting approach in which a sub-regional watershed model makes use of a background of ‘large-scale-scenario’ constraints provided by global models and enhances them by integrating sub-regional dynamics. Applying this approach to downscale the Global-NEWS predictions of the Millennium Ecosystem Assessment scenarios to the Seine, Somme and Scheldt watersheds, they predicted an overall increase of nitrogen delivery to the Southern Bight of the North Sea at the 2050 horizon in the GO scenario while the predictions indicate a decrease by about 20% in the AM scenario. Several studies have looked at the impact of future land use and land management on nutrient losses. Bouwman etÂ€al.

394

(2005a) predicted that from 1995 to 2030 in Western Europe grassland area will decrease from 60 Mha to 53 Mha and arable land will decline from 86 Mha to 76 Mha. On the other hand for transition countries (including new member states and the old states of the USSR) for the same period grassland will increase from 90 Mha to 96 Mha and arable land areas from 266 Mha to 273 Mha. At the same time, the nitrogen recovery has steadily increased for Western Europe from 44% in 1970 to a predicted 58% in 2030. A similar trend is foreseen for transition countries with a predicted change of nitrogen recovery from 38% in 1970 to 58% in 2030. This indicates an increase in productivity for both Western Europe and transition countries. Indeed for Western Europe, the level of fertiliser use in 2030 is predicted to be similar to that of 1995. For transition countries, there should be an increase of nitrogen fertiliser use by about 25%. Similar conclusions are reported by the EFMA in its 2008 outlook for 2018 (EFMA, 2008). EFMA (2008) expects that crop yield will increase for all major crops including wheat, barley, maize, potatoes and oilseed rape. It also predicts a shift in the agricultural production more oriented to grain and oil seed rape productions. This should lead to an increase of nitrogen consumption of 8% and 23%, for wheat and oil seed rape respectively. EFMA (2008) expects a decrease of nitrogen consumption in fodder and grassland production. Overall it is predicted that for 2018, the nitrogen consumption will increase by 3.8% in Europe, mostly in the new Member States. Indeed, there is a continuing decreasing trend of nitrogen consumption in EU15 with the exception of Austria, Denmark and Sweden due to an increase in the area devoted to the production of energy crops. On the other hand for EU12 there should be an increase of nitrogen consumption by about 17% compared to the actual level. The impact of the changing environment on nitrogen losses are reported in Bouwman etÂ€al. (2005b, 2005c). Because of the predicted increase in nitrogen recovery, there should be no increase of total nitrogen emissions (sum of denitrification, leaching and volatilisation) in 2030 when compared to the actual levels. Land use and climate changes are inherently interconnected by nature, and are both affected by the economic development and the mitigation and adaptation strategies adopted by the human society. Therefore future changes will be the result of complex interactions and will depend as well on societal choices. An actual example in this sense is the controversial issue of biofuels, which are motivated by economic and energy policies and present implications for both land use and climate changes, with expected negative consequences for water availability and nitrogen losses (Howarth etÂ€al., 2009).

17.4.3â•‡ Possible management scenarios Global scenarios address the effects on the environment of possible future economic development at global scale. At local scale, scenarios of implementation of practical measures to reduce nitrogen pollution may be evaluated to plan effective remediation strategies at river basin level. A number of modelling studies have been published to assess the effect of measures to reduce nitrogen loads to waters, such as improved wastewater

Bruna Grizzetti

treatment, or from adoption of agro-environmental measures affecting agricultural practices or landscape management. The severely eutrophied continental coast of the Channel and the Southern North Sea is particularly well documented from that respect. Cugier etÂ€al. (2005) explored the effect of phosphorus and/ or nitrogen tertiary treatment of urban wastewater in the Seine watershed, assuming constant agricultural diffuse sources, on the conditions of algal growth in the Seine Bight. Phosphorus treatment of wastewater (reducing the loading by 90%) appeared as quite an effective measure to reduce the potentially harmful algal blooms in the Seine Bight, with a 10-fold reduction for maximum dinoflagellates biomass with respect to reference levels. To reach a result similar to that for phosphorus, the reduction of point sources nitrogen should be carried up to 70%, which is technically feasible but extremely expensive. A reduction by 90% of point source nitrogen inputs would be required to lead the trophic state of the Seine Bight back to levels of flagellate development comparable to those of preindustrial periods. More recently, Thieu etÂ€ al. (2010b) generalised the same approach to the three main basins responsible for nutrient enrichment of the French and Belgian coastal zone of the North Sea (Seine, Somme and Scheldt, assuming that the Rhine plume is most of the time flowing northwards along the Netherlands coast). They showed that, although phosphorus abatement from the major point sources of wastewater is a useful measure to balance phosphorus fluxes with respect to silica inputs at the coastal zone, the generalisation of denitrification in wastewater treatment will not bring a substantial benefit. Implementation of ‘good agricultural practices’ (catch crop, reduction of fertilisation, extensification of cattle farming) would lead to a significant decrease of nitrogen fluxes exported to the sea, ranging from 14% to 23% in wet and dry years respectively. The results of these scenarios were coupled with a model of Phaeocystis development in the Southern North Sea (Lancelot etÂ€ al., 2005, 2007). The combination of wastewater treatment improvement and good agricultural practices leads to a decrease of the Phaeocystis bloom duration from 25 to 20 days, and of the biomass peak from 35·106 to 30·106 cells/l. Previous works (Rousseau etÂ€al., 2000) showed, however, that the good status of the coastal marine ecosystem requires that the biomass of Phaeocystis cells never increases above 5·106 cells/l, a threshold for the formation of ungrazable colonies (Lancelot etÂ€al., 2009). Nitrogen contamination of surface and groundwater as well as of coastal sea water will thus remain a major problem even if ‘good agricultural practices’ were generalised. To further reduce nitrogen export, and more specifically to satisfy to the OSPAR (2005) recommendation of a 50% reduction of nitrogen input to the sea, more drastic changes in agricultural practices should be envisaged. Essentially the same conclusion was reached by other authors working in different context and at various scales (Western France Kervidy catchment:Â€ Durand, 2004; England and Wales:Â€ Johnes etÂ€ al., 2007 and Johnes 2007; The Elbe river:Â€ Kersebaum etÂ€ al., 2003; Danish Gjern river:Â€ Kronvang etÂ€al., 1999b; Netherlands:Â€Wolf etÂ€al., 2005).

17.5â•‡ Policies for managing threats to European water quality 17.5.1â•‡ Policy and regulatory context At the beginning of the 1990s, the European Commission enforced various regulations to control and reduce nutrient loads into surface water, groundwater and coastal and marine water. In 1991, with the Nitrates Directive (Directive 91/676/EEC) and the Urban Waste Water Treatment Directive (Directive 91/271/EEC) the Commission started an ambitious plan to reduce the nutrient diffuse pollution originated from agriculture and the nutrient point pollution generated by urban waste water discharges. Then, the Commission established regulations for industrial emissions in 1996 (Directive 96/61/EC) and updated and reinforced the protection of drinking water in 1998 (Directive 98/83/EC). A comprehensive regulation for European water was enforced in year 2000, through the Water Framework Directive (WFD, Directive 2000/60/ EC) and its daughter directives. The WFD aims at protecting all the waters, including inland and coastal surface waters and groundwater and to achieve a good ecological status by 2015. It combines emission limit values with environmental quality standards. The WFD complements and integrates the European water legislation on nutrient reduction, in particular the Nitrate Directive and the Urban Waste Water Directive, into a coherent prospective of river basin management (Bloch, 2001). The Groundwater Directive (Directive 2006/118/EC) completes and specifies the WFD concerning the protection of aquifers, and the Freshwater Fish Directive (Directive 2006/44/ EC) sets physical and chemical parameters for fresh waters that support fish life, including among others limits for nitrite, total ammonium and non-ionised ammonia, which can be toxic for fishes. Finally, in 2008 the Marine Strategy Directive (Directive 2008/56/EC) was enforced to protect the marine environment, aiming to achieve or maintain a good environmental status for the European seas by 2021. Other European policies directly or indirectly influence the nitrogen enrichment in water, especially those related to the driving forces for nitrogen emission in the environment, such as the Common Agricultural Policy and the National Emission Ceiling Directive (Directive 2001/81/EC). The Commission supports the implementation of the WFD and related policies through the Common Implementation Strategy, a joint work programme, which involves Member States, other countries, stakeholders and NGOs to promote dialogue, common understanding and best practice exchange. An example is the Pilot River Basin activity, where, among other issues, a network of European river basins is sharing knowledge and experience in developing and implementing measures to reduce nutrient pollution (Charlet, 2007; http:// prb-water-agri.jrc.ec.europa.eu/). Similarly, Member States are working together to establish common ways to evaluate the ecological status of water bodies through the Intercalibration process (European Commission, 2008). The latter compares national ecological assessments in order to ensure that ‘good ecological status’ means the same in all the European countries. Indeed, to ensure a similar level of ambition in setting

395


the environment objectives in the European Union, the WFD mandated the intercalibration of the results of the national assessment methods. The boundary values for good ecological status for the different water types are established by the intercalibration exercise and will be the basis for setting environmental objectives for the management and programme of measures for the European river basins. The first step of the intercalibration exercise included a common agreement of reference condition criteria, and of the acceptable departures from the reference conditions. It was followed by the application of these criteria to a benchmark dataset to establish ecological rules, or if not possible statistical ones, for setting good status boundaries. An example is the data displayed in Figure 17.12 for lakes. Such data were obtained from national sources and from on-going EU projects that compiled EU wide biological datasets. (Further information can be obtained from the intercalibration technical report for lakes, the Commission decision on intercalibration and the following web address:Â€ http://circa.europa.eu/Members/irc/ jrc/jrc_eewai/library.) When the common implementation strategy and the dialogue fail, the Commission can use its powers under the Treaty and take the non-compliant Member States to court. In addition to European legislation, policies and interventions have been planned to protect European seas through international conventions:Â€HELCOM in the Baltic Sea, OSPAR in the North Sea, the Barcelona convention MEDPOL in the Mediterranean Sea and the Bucharest convention in the Black Sea. For the river basins discharging in the North Sea, the OSPAR Convention established the reduction of inputs of nitrogen and phosphorus to areas affected or likely to be affected by eutrophication in the order of 50% compared to input levels in 1985, to be achieved by 1995 (PARCOM Recommendation 88/2 and 89/4). Similarly, in the Baltic Sea within the HELCOM convention Contracting Parties committed to reduce nitrogen and phosphorus loads to the sea according to country specific reduction targets set in the recent Baltic Sea Protection Plan. A comprehensive discussion of European Directives and international actions addressing the nitrogen pollution through all the environmental compartments is presented in Oenema etÂ€ al., 2011 (Chapter 4 this volume).

17.5.2â•‡ Scientific and technical knowledge The research community has developed a sound understanding of sources and pathways causing water nitrogen enrichment and of the consequent processes of transformation, transport, storage and removal (Durand etÂ€al., 2011; Voss etÂ€al., 2011; Chapters 7 and 8, this volume). Effective tools that the scientific community can offer for understanding the impact of human disturbance and the potential success of restoration interventions include the construction of nutrient budgets at river basins scale (Billen etÂ€al., 2011, Chapter 13, this volume) and the spatial evaluation of pressures and sources contribution to nitrogen export (Johnes, 1996; Johnes and Heathwaite, 1997; Behrendt etÂ€al., 2003; Grizzetti etÂ€al., 2008) together with development and the evaluation of future scenarios (Bouwman

396

etÂ€ al., 2005c; Bouraoui and Aloe, 2007; Velthof etÂ€ al., 2009). Moreover, the development of integrated indicators of nutrient pressure, eutrophication status and ecosystem functioning offers useful tools to monitor in space and in time nutrient trends and measures effectiveness (European Environment Agency, 2005). Extensive knowledge is available on interventions to alleviate nitrogen (and phosphorus) pollution for the water system. Measures can be targeted on the sources, on the landscape or directly on the water body management (Novotny and Olem, 1994). They include among others:Â€ optimal fertilisation, where fertiliser application are reduced to match the crop requirements; spatial nutrient management, which implies lower application in areas with high erosion and runoff, use of catch crops to reduce erosion and nutrient leaching; improvement in livestock and manure surplus management; implementation of advanced treatments for waste water discharges; optimisation of sewer systems; creation of riparian strips, sedimentation ponds, appropriate drainage systems; and finally restoration of wetlands and floodplains to increase denitrification and protect wildlife habitat and biodiversity. A European initiative on this field is the COST Action 869 (COST 2010a) on mitigation options for nutrient reduction in surface water and groundwater. It provides an extended review of all the measures implemented in European countries describing for each measure main benefits, reported efficacy, region of application, likely disadvantages and potential costs (COST 2010b). Economic tools, such as cost-benefit analysis (CBA), have been developed to support policy decisions on controversial environmental issues (OECD, 2006). CBA for environmental resolutions consists of calculating, commonly in currency units, the net benefits generated by the policy or project at each point in time, based on the analysis of all benefits and costs. Although widely employed, environmental CBA is controversial (Ludwig etÂ€al., 2005), for the way the discount rates are set and for the methods to account and monetise environmental externalities in the economic models. In fact, the evaluation of environmental effects and ecosystem services is not always possible in monetary terms, and implies a vision of the system and the actors concerned, involving definitely a discussion of values. Similarly, establishing discount rates requires a judgment in a time perspective. In economic models, discount rates indicate the way costs and benefits are weighted over time. When present benefits are weighted higher than future ones, ecosystem services are consumed faster, while the contrary produces the conservation of the natural capital with eventual wealth loss for the present generation (Ludwig etÂ€ al., 2005). Therefore, the choice of the appropriate discount rate is not straightforward, involves a system of values and is undermined by the high uncertainty related to long-term pollution effects on ecosystems. However, evidence provides support for policies that maintain ecosystem services over the long term (Ludwig etÂ€al., 2005). Brink etÂ€al., 2011 (Chapter 22 this volume) provide a CBA of nitrogen in the environment with a European perspective.

Bruna Grizzetti

17.5.3â•‡ Effects of implementation of measures To reduce eutrophication in estuaries and coastal waters programmes of measures implemented in their river basins will be essential to restrict anthropogenic nutrient inputs (Smith and Schindler, 2009). The WFD requires Member States to prepare river basins management plans including the analysis of nutrient pressures and the plan for implementing mitigation measures, in order to achieve by 2015 a ‘good ecological status’ of all the water bodies. The latter should be achieved considering the reference status, which refer to water body conditions prior to significant anthropogenic pollution. However, the concept of good ecological status implies some room for interpretation. In fact, even consistent nutrient abatement may not lead the water bodies to the desired status as many ecosystems present a hysterisis behaviour. Duarte etÂ€al. (2009) illustrated the trajectories of restoration in four North European coastal ecosystems (Marsdiep, Netherlands; Helgoland, Germany; Odense Fiord, Denmark and Gulf of Riga, Latvia/Estonia). They argued that in addition to nutrient enrichment other human induced changes, such as climate changes, population growth, freshwater withdrawal, may affect many fundamental factors of ecosystem functioning, producing baseline conditions different from those of the ‘reference’ status, even returning to pristine nutrient inputs. They observed that the restoration pathways of the monitored ecosystems followed dynamic trajectories of ‘regime shift and shifting baseline’. As a consequence, when setting targets of restoration, emphasis should be put on the values ensuring reliable provision of ecosystem services and good ecosystem functioning rather than focusing on particular past conditions (Duarte etÂ€ al., 2009). This does not mean that anthropogenic nutrient input should not be reduced; on the contrary these observations show a clear awareness on how human impacts can be difficult to reverse. In Europe only 30% of surface water bodies have been identified as not being a risk of failing to achieve the WFD environmental objectives by 2015, while 40% are at risk and for the rest 30% data are not sufficient for evaluation. The lack of information regards especially coastal and transitional water (European Commission 2007 SEC(2007)362). Similar figures are reported for groundwater bodies, with 25% not at risk, 30% at risk and for the remaining 45% the evaluation is not possible for the lack of data (European Commission 2007 SEC(2007)362). Nutrient diffuse pollution has been identified as one of the most significant and widespread pressures on water ecosystems in Europe (European Commission 2007 COM(2007)128). In general during the past two decades the nutrient pollution from urban areas and point sources has been significantly reduced through the implementation of waste water treatment plants, while diffuse sources originate from agriculture remain a problem (European Environment Agency, 2005). According to the monitoring information transmitted by Member States, after almost 15 years from the enforcement of Nitrates Directive, in EU27 monitoring stations with average annual nitrate concentrations above 50 mgNO3/l were 15% for groundwater and 3% for surface water. Member States with the highest proportion of sampling points with nitrate

concentrations above 50Â€ mg/l were Estonia, the Netherland, Belgium, England, France, Northern Italy, North-East of Spain, Slovakia, Romania, Malta and Cyprus (European Commission 2010 COM(2010)47). In 2003, the level of compliance with the Urban Waste Water Treatment Directive in the EU15 was 79% in normal areas and 84% in sensitive areas (European Commission 2007 SEC(2007)363). There were 17 ‘big cities’ still without wastewater treatment and some countries, mainly in Southern Europe, presented areas with inadequate or lacking wastewater treatment (European Commission 2007 SEC(2007)363). Measures to reduce nitrogen and phosphorus pollution in water bodies have already been introduced in many European countries under the European legislation, international conventions and national plans. The assessment report on the implementation of the Baltic Sea Action Plan, under the HELCOM Convention, indicates that since 1990 nitrogen and phosphorus diffuse and point source loads have been slightly decreasing in the Baltic Sea catchment, however the target input levels foreseen in the Action Plan have not been met and additional reductions are needed (HELCOM, 2009). In the areas under the OSPAR Convention, the source reduction of 50% (compared to the level of 1985) has been met for phosphorus, but not completely for nitrogen. In fact, the target for nitrogen source reduction was achieved only by Denmark (in 2003), Germany and the Netherlands (both in 2005), although progress in this direction has been made also by the other Contracting Parties (OSPAR, 2008). See also Voss etÂ€al., 2011 (Chapter 8 this volume). The implementation of programmes of measures for nutrient reduction varies in the European countries. Moreover, the water quality response to the mitigation programmes has been variable, depending on the design of the plans, the specific pollution conditions and the environmental characteristics. In general a delay has been observed between remediation actions and water response. In Denmark, the implementation of targeted regulations and nitrogen efficiency measures has reduced nitrogen loads to waters by 32%, while maintaining the crop yield and increasing the livestock production (Kronvang etÂ€al., 2008). In Norway, results from long-term monitoring show that in spite of changes in management practices driven by subsidies and production conditions few correspondent trends were registered in nutrient losses (Bechmann etÂ€al., 2008). In part of England, the effect of actual measures introduced in the Nitrogen Vulnerable Zones is not evident and a time lag is expected because of the specific soils and aquifers characteristics (Jackson etÂ€al., 2008; Worrall etÂ€al., 2009). In Finland, no clear reduction of nutrient loads or water quality improvements were observed although a large-scale programme to reduce nutrient emissions from agriculture has been introduced since 1995 (Ekholm etÂ€al., 2007). In general, scientific evidence shows that the adopted policies to reduce anthropogenic nutrient inputs to European seas were more effective in abating point sources than diffuse sources and more successful for phosphorus rather than for nitrogen, leading to the increase of the N:P ratio in anthropogenic inputs (Artioli etÂ€ al., 2008). Indeed, assessing policy effectiveness

397


in reducing loads of nitrogen is controversial and presents regional differences. This is related to the diffuse nature of the sources, the tight connections with lifestyles, notably human diet, and the economic implications due to the links with agriculture and livestock production. Moreover, long-retention times of groundwater may retard the system recovery (Artioli etÂ€al., 2008). In addition, the problem of ‘pollution swapping’ needs to be taken into wider consideration from both the scientific and the policy prospective (Stevens and Quinton, 2009). The term ‘pollution swapping’ is used to indicate when a mitigation measure reduces a targeted pollutant while increasing the level of another pollutant. Unfortunately, this side effect is present in many commonly implemented measures, but weakly considered in scientific studies and management plans. For example, constructed wetlands and riparian buffer zones are widely used to remove nitrogen but at the same time they promote denitrification, increasing the emissions of N2O, a strong greenhouse gas, and thus swapping the pollution from water to the air compartment. If not correctly understood the mechanisms of pollution swapping may lead to contradicting interventions and unsuccessful regulations (Stevens and Quinton, 2009). An enlightening example in this sense is the model MITERRAEUROPE (Velthof etÂ€ al., 2009), which provides an integrated assessment of the effect of nutrient mitigation measures in EU27 at country and regional level, considering the effects of mitigation measures contemporary on nitrogen emissions to atmosphere and to groundwater.

17.5.4â•‡ Opportunities for policy and science integration In developing mitigation options for nitrogen pollution in water, the way forward is to adopt a more holistic approach in the policy frame, undertaking an integration effort in legislation tools and an evaluation of the potential ‘ecosystem service swapping’, and in the scientific research, promoting interdisciplinary studies including more forms of pollutants and all the environmental compartments (Stevens and Quinton, 2009; Collins and McGonigle, 2008). Effective strategies to reduce eutrophication in aquatic ecosystems need to consider the whole land-ocean continuum and to control both nitrogen and phosphorus (Conley etÂ€al., 2009). A stronger integration between the Marine Strategy and relÂ� evant sectoral policies, such as the Common Agricultural Policy and the Fisheries Policy, would be beneficial for the sustainable management and protection of the marine ecosystem and its resources (Salomon, 2009). Similar positive synergies could derive from an improved coordination between policies to prevent water nutrient pollution and the Common Agricultural Policy and among the different nitrogen related regulations, in order to avoid problems of pollution swapping (Oenema etÂ€al., 2011, Chapter 4 this volume). The Commission already supports the integration between the implementation of different policies, encouraging Member States to use EU financing instruments available under the Common Agricultural Policy and the Cohesion Policy for improvements in the water field

398

(European Commission 2007 COM(2007) 128), or the further integration of climate change mitigation and adaptation strategies into the implementation of EU water policy (see European Climate Change Programme). Finally, a strategic choice for the future will be to create positive synergies between River Basin Plans foreseen by the European WFD by December 2009 and international instruments, such as the Baltic Sea Action Plan (HELCOM Convention) or the OSPAR Resolutions (OSPAR Convention). An effort to improve integration and interdisciplinarity is required also in the scientific research. Indeed, there is a need for integrated assessments to enable comparison between different regions for promoting regional assessment, and to evaluate the potential synergies between different type of policies or remediation measures, taking into account future scenarios. There is a clear need to promote science–policy–society interactions to build reciprocal trust and understanding, and to produce a new type of transdisciplinarity knowledge to support a sustainable management of the ecosystem resources. In fact, the complexity of interactions of the human–ecological system cannot be addressed by traditional disciplines, but requires a new kind of interdisciplinary science, concerned by problem-solving aspects (Carpenter etÂ€al., 2009). This new type of knowledge production, also referred to as sustainability science, seeks to understand the interactions between nature and society, addressing the dynamics of the interactions, the longterm trends, the vulnerability and resilience of the nature–Â� society system and the opportunities for adaptive management and societal learning (Kates etÂ€al., 2001). To address problems related to sustainable development, the integration between natural science and social science is a key point and cooperation is necessary between the academic disciplines and the different parts of society (Tappeiner etÂ€al., 2007).

17.6â•‡ Conclusions and way forward From the present overview it appears that the anthropogenic increase of nitrogen in water, together with other nutrients, causes many direct and indirect biogeochemical and ecologÂ� ical responses in the aquatic ecosystems, most of which are unÂ�desired and detrimental for the human-ecological system. In spite of some encouraging trends, nitrogen concentrations in rivers, lakes, aquifers and coastal waters are generally high and stable in many regions, and even increasing in some areas. In addition, evidence shows that there is gradual and increasing nitrogen enrichment of groundwater resources across Europe. This poses direct threats to human health and ecosystem functioning, reducing the actual provision and the future reliability of ecosystem services. A large part of European freshwaters and coastal waters are affected by eutrophication and current global drivers such as climate and land-use changes could exacerbate the situation in the near future. An additional challenge will be represented by the economic development of Eastern Europe, which could potentially lead to additional nitrogen loadings to the Baltic and Black Seas.

Bruna Grizzetti

Policy tools are available within the European Union and under international conventions to mitigate the nitrogen pollution in water. Their full implementation has not been achieved yet throughout Europe, but plans of measures to reduce nitrogen losses to water have already been implemented in many European countries, producing some encouraging results. However, in many cases a delay in the water quality response to the implementation of measures have been observed, due to previous accumulation of anthropogenic nitrogen in soils, sediments or aquifers or to inadequate design or targeting of the mitigation plans. At European level some regional differences have emerged in the sensitivity of coastal ecosystems to nutrient loads and on the effect of policy measures in changing the N:P ratio. Finally, the issue of pollution swapping has appeared as an important element to be considered by both the scientific and policy perspective. To support the sustainable management of the human-ecological system and promote the protection of water resources in relation to the threats posed by nitrogen in European water the full implementation of the regulations is necessary, combined with an efficient environmental monitoring. Moreover, positive synergies could be obtained by encouraging integration in the sectoral policies and enhancing interdisciplinarity in the scientific research, especially in support of regional assessments and pollution swapping evaluations. The continuous nitrogen export to waters directly or indirectly threatens the biodiversity in the aquatic ecosystems, and slowly and gradually erodes the resilience of the aquatic ecosystems, increasing their vulnerability to other unexpected stresses. Water eutrophication and aquatic biodiversity loss have economic and political implications. According to Folke etÂ€al. (2002) two errors underpinned the past policies for managing natural resources. The first was the assumption that the ecosystem response to anthropogenic pressures is linear and the second was the lack of recognition of the mutual interdependence of the human and ecological system. Drastic changes in the ecosystem status may imply elevated costs forÂ€the direct loss of ecological and economic recourses and for the actions required for restoration. Building and maintaining the ecosystem resilience result in a wise investment for future human wealth, especially in the context of uncertainty and global environmental changes. A policy aiming at good ecological status can certainly contribute in this direction by investing in substantially improving nitrogen use efficiency and cleaning waste waters, in spite of the recovery time and costs.

Acknowledgements This chapter was prepared with the support of the NinE Programme of the European Science Foundation, the NitroEurope Integrated Project (funded by the European Commission) and the COST Action 729.

References Addiscott, T. M. (2005). Nitrate, Agriculture and Environment, CAB International, Wallingford, UK. Alcamo, J., Van Vuuren, D. and Cramer W. (2006). Changes in ecosystem services and their drivers across the scenarios.

In:Â€Ecosystems and Human Well-Being:Â€Scenarios, ed. S. R. Carpenter etÂ€al., pp. 279–354, Island Press, Washington, DC. Alcamo, J., Moreno, J. M., Nováky, B. etÂ€al. (2007). Europe, Climate Change 2007:Â€Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK, pp. 541–580. Alexander, R. B., Smith, R. A. and Schwarz, G. E. (2000). Effect of stream channel size on the delivery of nitrogen to the Gulf of Mexico. Nature, 403, 758–761. Alley, W. M., Healy, R. W., LaBaugh, J. W. and Reilly, T. E. (2002). Flow and storage in groundwater systems. Science, 296, 1985–1990. Anderson, D. M., Glibert, P. M. and Burkholder, J. M. (2002). Harmful algal blooms and eutrophication:Â€Nutrient sources, composition, and consequences. Estuaries, 25, 704–726. Arheimer, B., Andreasson, J., Fogelberg, S. etÂ€al. (2005). Climate change impact on water quality:Â€model results from southern Sweden. Ambio, 34(7), 559–566. Artioli, Y., Friedrich, J., Gilbert, A. J. etÂ€al. (2008). Nutrient budgets for European seas:Â€a measure of the effectiveness of nutrient reduction policies. Marine Pollution Bulletin, 56, 1609–1617. Battarbee, R., Kernan, M., Livingstone, D. etÂ€al. (2008). Euro-limpacs Position Paper: Impact of climate change on European freshwater ecosystems:Â€consequences, adaptation and policy. Bechmann, M., Deelstra, J., Stålnacke, P. etÂ€al. (2008). Monitoring catchment scale agricultural pollution in Norway:Â€policy instruments, implementation of mitigation methods and trends in nutrient and sediment losses. Environmental Science and Policy, 11, 102–114. Behrendt, H., Bach, M., Kunkel, R. etÂ€al. (2003). Nutrient emissions into river basins of Germany on the basis of a harmonised procedure. Federal Environmental Agency (Umweltbundesamt). Billen, G. and Garnier, J. (2007). River basin nutrient delivery to the coastal sea:Â€assessing its potential to sustain new production of nonsiliceous algae. Marine Chemistry, 106, 148–160. Billen, G., Silvestre, M., Grizzetti, B. etÂ€al. (2011). Nitrogen flows from European watersheds to coastal marine waters. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Bloch, H. (2001). EU policy on nutrients emissions:Â€legislation and implementation, Water Science and Technology, 44, 1–6. Bouraoui, F., Galbiati, L. and Bidoglio, G. (2002). Climate change impacts on nutrient loads in the Yorkshire Ouse catchment (UK). Hydrology and Earth System Sciences, 6, 197–209. Bouraoui, F., Grizzetti, B., Granlund, K., Rekolainen, S. and Bidoglio,Â€G. (2004). Impact of climate change on the water cycle and nutrient losses in a Finnish catchment. Climatic Change, 66, 109–126. Bouraoui, F. and Aloe, A. (2007). European agrochemicals geospatial loss estimator:Â€model development and application. EUR Report 22690 EN, European Commission Joint Research Centre, Luxembourg. Bouraoui, F., Grizzetti, B. and Aloe, A. (2009). Nutrient discharge from river to seas for year 2000, EUR Report 24002 EN, European Commission Joint Research Centre, Luxembourg. Bouwman, A. F., van Drecht, G. and van Der Hoek, K. W. (2005a). Global and regional surface nitrogen balances in intensive agricultural production systems for the period 1970–2030. Pedosphere, 15, 137–155. Bouwman, A. F., van Drecht, G. and van der Hoek, K. W. (2005b). Surface N balances and reactive N loss to the environment from

399

Nitrogen as a threat to European water quality global intensive agricultural production systems for the period 1970–2030. Science in China, Series C, 48, 767–779. Bouwman, A. F., van Drecht, G., Knoop, J. M., Beusen, A. H. W. and Meinardi, C. R. (2005c). Exploring changes in river nitrogen export to the world’s oceans. Global Biogeochemical Cycles, 19, 1–14. Brink, C., van Grinsven, H., Jacobsen, B. H. etÂ€al. (2011). Costs and benefits of nitrogen in the environment. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Burkholder, J. M., Tomasko, D. A. and Touchette, B. W. (2007). Seagrasses and eutrophication. Journal of Experimental Marine Biology and Ecology, 350, 46–72. Caddy, J. F. (1993). Toward a comparative evaluation of human impacts on fishery ecosystems of enclosed and semi-enclosed seas. Reviews in Fisheries Science, 1, 57–95. Camargo, J. A. and Alonso, A. (2006). Ecological and toxicological effects of inorganic nitrogen pollution in aquatic ecosystems:Â€a global assessment. Environment International, 32, 831–849. Cardoso, A. C., Duchemin, J., Magoarou, P. and Premazzi, G. (2001). Criteria for the identification of freshwaters subject to eutrophication: their use for the implementation of the NitratesÂ€and Urban Waste Water treatment Directives. EUR Report 19810 EN, European Commission Joint Research Centre, Luxembourg. Carpenter, S. R. and Folke, C. (2006). Ecology for transformation. Trends in Ecology and Evolution, 21, 309–315. Carpenter, S. R., Caraco, N. F., Correll, D. L. etÂ€al. (1998). Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications, 8, 559–568. Carpenter, S. R., Mooney, H. A., Agard, J. etÂ€al. (2009). Science for managing ecosystem services:Â€beyond the Millennium Ecosystem Assessment. Proceedings of the National Academy of Sciences of the United States of America, 106, 1305–1312. Charlet, M. (2007). Experiences in the analysis of pressures and impacts from agriculture on water resources and developing a related programme of measures, European Commission Joint Research Centre, Luxembourg. Christoffersen, K., Andersen, N., Søndergaard, M., Liboriussen, L. and Jeppesen, E. (2006). Implications of climate-enforced temperature increases on freshwater pico- and nanoplankton populations studied in artificial ponds during 16 months. Hydrobiologia, 560, 259–266. Cloern, J. E. (2001). Our evolving conceptual model of the coastal eutrophication problem. Marine Ecology Progress Series, 210, 223–253. Collins, A. L. and McGonigle, D. F. (2008). Monitoring and modelling diffuse pollution from agriculture for policy support:Â€UK and European experience. Environmental Science and Policy, 11, 97–101. Conley, D. J., Paerl, H. W., Howarth, R. W. etÂ€al. (2009). Ecology: controlling eutrophication –Â€nitrogen and phosphorus. Science, 323, 1014–1015. COST (2010a). http://www.cost869.alterra.nl COST (2010b). http://www.cost869.alterra.nl/Fs/List_of_options. htm#NM Cugier, P., Billen, G., Guillaud, J. F., Garnier, J. and Ménesguen,Â€A. (2005). Modelling eutrophication of the Seine Bight under present, historical and future Seine river nutrient loads. Journal of Hydrology, 304, 381–396. Curtis, C. J., Evans, C., Helliwell, R. C. and Monteith, D. (2005a). Nitrate leaching as a confounding factor in chemical recovery from acidification in UK upland waters. Environmental Pollution, 137, 73–82.

400

Curtis, C. J., Botev, I., Camarero, L. etÂ€al. (2005b). Acidification in European mountain lake districts:Â€a regional assessment of critical load exceedance. Aquatic Sciences, 67, 237–251. De Roos, A. J., Ward, M. H., Lynch, C. F. and Cantor, K. P. (2003). Nitrate in public water supplies and the risk of colon and rectum cancers. Epidemiology, 14, 640–649. Diaz, R. J. and Rosenberg, R. (2008). Spreading dead zones and consequences for marine ecosystems. Science, 321, 926–929. Directive 91/271/EEC (1991) concerning urban waste water treatment. (OJ (1991) L271/40). (Urban Waste Water Directive). Directive 91/676/EEC (1991) concerning the protection of waters against pollution caused by nitrates from agricultural sources (OJ (1991) L375/1). (Nitrates Directive). Directive 96/61/EC (1996) concerning integrated pollution prevention and control (IPPC Directive). Directive 98/83/EC (1998) on the quality of water intended for human consumption. (Drinking Water Directive). Directive 2000/60/EC (2000) establishing a Framework for Community Action in the Field of Water Policy. (OJ (2000) L327/1). (Water Framework Directive). Directive 2001/81/EC (2001) on national emission ceilings for certain atmospheric pollutants. Directive 2006/44/EC (2006) on the quality of fresh waters needing protection or improvement in order to support fish life (OJ (2006) L264/20). Directive 2006/118/EC (2006) on the protection of groundwater against pollution and deterioration (OJ L 372, 27.12.2006, p.19). Directive 2008/56/EC (2008) establishing a framework for community action in the field of marine environmental policy (Marine Strategy Framework Directive) (OJ L 164, 25.6.2008, pp. 19–40). Dodds, W. K., Bouska, W. W., Eitzmann, J. L. etÂ€al. (2009). Eutrophication of U. S. freshwaters:Â€Analysis of potential economic damages. Environmental Science and Technology, 43, 12–19. Duarte, C. M., Conley, D. J., Carstensen, J. and SanchezCamacho,Â€M. (2009). Return to Neverland:Â€shifting baselines affect eutrophication restoration targets. Estuaries and Coasts, 32, 29–36. Ducharne, A., Baubion, C., Beaudoin, N. etÂ€al. (2007). Long term prospective of the Seine River system:Â€confronting climatic and direct anthropogenic changes. Science of the Total Environment, 375(1–3), 292–311. Durand, P. (2004). Simulating nitrogen budgets in complex farming systems using INCA:Â€calibration and scenario analyses for the Kervidy catchment (W. France). Hydrology and Earth System Sciences, 8, 793–802. Durand, P., Breuer, L., Johnes, P. etÂ€al. (2011). Nitrogen processes in aquatic ecosystems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. EFMA (2008). Forecast of food, farming and fertiliser use in the European Union 2007–2017. Eisenreich, S. J. (ed.) (2005). Climate Change and the European Water Dimension. European Commission, Joint Research Centre: Ispra, Italy. Ekholm, P., Granlund, K., Kauppila, P. etÂ€al. (2007). Influence of EU policy on agricultural nutrient losses and the state of receiving surface waters in Finland. Agricultural and Food Science, 16, 282–300. EMEP (2009). Transboundary acidification, eutrophication and ground level ozone in Europe in 2007, EMEP Status Report 2009, Norwegian Meteorological Institute, Oslo, Norway.

Bruna Grizzetti Erisman, J.W., van Grinsven, H., Grizzetti, B., etÂ€al. (2011). The European nitrogen problem in a global perspective. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J.Â€W. Erisman etÂ€al., Cambridge University Press. European Commission (2007). The quality of drinking water in the European Union, Synthesis Report on the Quality of Drinking Water in the European Union period 2002–2004. (Directives 80/778/EEC and 98/83/EC). European Commission (2007) COM(2007) 120. Report from the Commission to the Council and the European Parliament on the implementation of the Council Directive 91/676/EEC concerning the protection of the waters against pollution caused by nitrates from agricultural sources for the period 2000–2003. European Commission (2007) COM(2007) 128. Communication from the Commission to the European Parliament and the Council ‘Towards Sustainable Water Management in the European Union’ First stage in the implementation of the Water Framework Directive 2000/60/EC. European Commission (2007) SEC(2007) 362. Accompanying document to the communication from the Commission to the European Parliament and the Council ‘Towards Sustainable Water Management in the European Union’ First stage in the implementation of the Water Framework Directive 2000/60/EC. European Commission (2007) SEC(2007) 363. Annex to the Communication from the Commission to the European Parliament and the Council ‘Towards Sustainable Water Management in the European Union’ First stage in the implementation of the Water Framework Directive 2000/60/ EC. 4th Commission Report (Executive Summary) on the Implementation of the Urban Waste Water Treatment Directive. European Commission (2008). Commission Decision (2008/915/EC) of 30 October 2008 establishing, pursuant to Directive 2000/60/ EC of the European Parliament and of the Council, the values of the Member State monitoring system classifications as a result of the intercalibration exercise (notified under document number C(2008) 6016). OJ. L 332/20, 10/12/2008. European Commission (2010) COM(2010) 47. Report from the Commission to the Council and the European Parliament on the implementation of the Council Directive 91/676/EEC concerning the protection of the waters against pollution caused by nitrates from agricultural sources based on Member States reports for the period 2004–2007. European Environment Agency (2001). Eutrophication in Europe’s coastal waters, EEA, Copenhagen. European Environment Agency (2005). The European environment: state and outlook 2005, EEA, Copenhagen. European Environment Agency (2008). Impacts of Europe’s changing climate, 2008 indicator-based assessment: Joint EEAJRC-WHO report. EEA Report 4/2008, EEA, Copenhagen. European Environment Agency (2009). Progress towards the European 2010 biodiversity target. EEA Report 4/2009, EEA, Copenhagen. Evans, C. D., Cullen, J. M., Alewell, C. etÂ€al. (2001). Recovery from acidification in European surface waters. Hydrology and Earth System Sciences, 5, 283–297. Feuchtmayr, H., McKee, D., Harvey, I. F., Atkinson, D. and Moss,Â€B. (2007). Response of macroinvertebrates to warming, nutrient addition and predation in large-scale mesocosm tanks. Hydrobiologia, 584, 425–432. Flanagan, K. M., McCauley, E., Wrona, F. and Prowse, T. (2003). Climate change:Â€the potential for latitudinal effects on algal

biomass in aquatic ecosystems. Canadian Journal of Fisheries and Aquatic Sciences, 60, 635–639. Folke, C., Carpenter, S., Elmqvist, T. etÂ€al. (2002). Resilience and sustainable development:Â€building adaptive capacity in a world of transformations. Ambio, 31, 437–440. Folke, C., Carpenter, S., Walker, B. etÂ€al. (2004). Regime shifts, resilience, and biodiversity in ecosystem management. Annual Review of Ecology, Evolution, and Systematics, 35, 557–581. Galloway, J. N., Townsend, A. R., Erisman, J. W. etÂ€al. (2008). Transformation of the nitrogen cycle:Â€recent trends, questions, and potential solutions. Science, 320, 889–892. Glibert, P. M., Harrison, J., Heil, C. and Seitzinger, S. (2006). Escalating worldwide use of urea: a global change contributing to coastal eutrophication. Biogeochemistry, 77, 441–463. Griesenbeck, J. S., Steck, M. D., Huber, J. C. etÂ€al. (2009). Development of estimates of dietary nitrates, nitrites, and nitrosamines for use with the short Willet food frequency questionnaire. Nutrition Journal, 8, 16. Grimvall, A., Stalnacke, P. and Tonderski, A. (2000). Time scales of nutrient losses from land to sea: a European perspective. Ecological Engineering, 14, 363–371. Grizzetti, B., Bouraoui, F. and De Marsily, G. (2008). Assessing nitrogen pressures on European surface water. Global Biogeochemical Cycles, 22, GB4023. Gulis, G., Czompolyova, M. and Cerhan, J. R. (2002). An ecologic study of nitrate in municipal drinking water and cancer incidence in Trnava District, Slovakia. Environmental Research, A, 88, 182–187. Hartikainen, H., Johnes, P. J., Monteith, C. and Okamura, B. (2009). Bryozoan populations reflect nutrient enrichment and productivity gradients in rivers. Freshwater Biology, pp. 1–15. Hejzlar, J., Anthony, S., Arheimer, B. etÂ€al. (2009). Nitrogen and phosphorus retention in surface waters:Â€an inter-comparison of predictions by catchment models of different complexity. Journal of Environmental Monitoring, 11, 584–593. HELCOM (2009). Euthrophication in the Baltic Sea: an integrated thematic assessment of the effects of nutrient enrichment and euthrophication in the Baltic Sea region: Executive Summary, Helshinki Commission, Helsinki. Hettelingh, J.-P., Posch, M., Slootweg, J. etÂ€al. (2007). Critical loads and dynamic modelling to assess European areas at risk of acidification and eutrophication. Water, Air and Soil Pollution, 7, 379–384. Hettelingh, J.-P., Posch, M. and Slootweg, J. (2008). Critical load, dynamic modelling and impact assessment in Europe:Â€CCE Status Report 2008. PBL Report 500090003, www.mnp.nl/cce, Coordination Centre for Effects, Bilthoven, The Netherlands. Howarth, R. W. (1988). Nutrient limitation of net primary production in marine ecosystems. Annual Review of Ecology and Systematics, 19, 89–110. Howarth, R. W., Billen, G., Swaney, D. etÂ€al. (1996). Regional nitrogen budgets and riverine N and P fluxes for the drainages to the North Atlantic Ocean:Â€natural and human influences. Biogeochemistry, 35, 75–139. Howarth, R. W., Bringezu, S., Bekunda, M. etÂ€al. (2009). Rapid assessment on biofuels and the environment:Â€overview and key findings. In:Â€R. W. Howarth and S. Bringezu (eds.), Biofuels:Â€Environmental Consequences and Interactions with Changing Land Use, Proceedings of the Scientific Committee on Problems of the Environment (SCOPE) International Biofuels Project Rapid Assessment, 22–25 September 2008, Gummersbach

401

Nitrogen as a threat to European water quality Germany, pp. 1–13, Cornell University, Ithaca, NY, (http://cip. cornell.edu/biofuels/) IARC (2006). Ingested nitrates and nitrites. http://monographs.iarc.fr/ ENG/Meetings/94-nitratenitrite.pdf Jackson, B. M., Browne, C. A., Butler, A. P. etÂ€al. (2008). Nitrate transport in chalk catchments:Â€monitoring, modelling and policy implications. Environmental Science and Policy, 11, 125–135. Johnes, P. J. (1996). Evaluation and management of the impact of land use change on the nitrogen and phosphorus load delivered to surface waters:Â€the export coefficient modelling approach. Journal of Hydrology, 183, 323–349. Johnes, P. J. and Heathwaite, A. L. (1997). Modelling the impact of land use change on water quality in agricultural catchments. Hydrological Processes, 11, 269–286. Johnes, P. J. (2007). Meeting ecological restoration targets in European Waters:Â€a challenge for animal agriculture. In:Â€Redesigning Animal Agriculture:Â€The Challenge for the 21st Century, pp. 185–203, CAB International, Wallingford, UK. Johnes, P. J., Foy, R., Butterfield, D. and Haygarth, P. M. (2007). Land use for Good Ecological Status:Â€an evaluation of scenarios for water bodies in England and Wales. Soil Use and Management, 23, 176–194. Johnk, K. D., Huisman, J., Sharples, J. etÂ€al. (2008). Summer heatwaves promote blooms of harmful cyanobacteria. Global Change Biology, 14, 495–512. Johnson, P. T. J., Chase, J. M., Dosch, K. L. etÂ€al. (2007). Aquatic eutrophication promotes pathogenic infection in amphibians. Proceedings of the National Academy of Sciences of the United States of America, 104, 15781–15786. Kates, R. W., Clark, W. C., Corell, R. etÂ€al. (2001). Environment and development:Â€Sustainability science. Science, 292, 641–642. Kersebaum, K. C., Steidl, J., Bauer, O. and Piorr, H. P. (2003). Modelling scenarios to assess the effects of different agricultural management and land use options to reduce diffuse nitrogen pollution into the river Elbe. Physics and Chemistry of the Earth, 28,Â€537–545. Kronvang, B., Hoffmann, C. C., Svendsen, L. M. etÂ€al. (1999a). Retention of nutrients in river basins. Aquatic Ecology, 33, 29–40. Kronvang, B., Svendsen, L. M., Jensen, J. P. and Dørge, J. (1999b). Scenario analysis of nutrient management at the river basin scale. Hydrobiologia, 410, 207–212. Kronvang, B., Andersen, H. E., Børgesen, C. etÂ€al. (2008). Effects of policy measures implemented in Denmark on nitrogen pollution of the aquatic environment. Environmental Science and Policy, 11, 144–152. L’ hirondel, J. and L ’ hirondel, J. L. (2002). Nitrate and Man:Â€Toxic, Harmless or Beneficial?, CAB International, Wallingford, UK. Lancelot, C., Spitz, Y., Gypens, N. etÂ€al. (2005). Modelling diatom and Phaeocystis blooms and nutrient cycles in the Southern Bight of the North Sea:Â€the MIRO model. Marine Ecology Progress Series, 289, 63–78. Lancelot, C., Gypens, N., Billen, G., Garnier, J. and Roubeix, V. (2007). Testing an integrated river-ocean mathematical tool for linking marine eutrophication to land use:Â€the Phaeocystis-dominated Belgian coastal zone (Southern North Sea) over the past 50 years. Journal of Marine Systems, 64, 216–228. Lancelot, C., Rousseau, V. and Gypens, N. (2009). Ecologically based indicators for Phaeocystis disturbance in eutrophied Belgian

402

coastal waters (Southern North Sea) based on field observations and ecological modelling. Journal of Sea Research, 61, 44–49. Lavelle, P., Dugdale, R., Scholes, R. etÂ€al. (2005). Nutrient Cycling, Ecosystems and Human Well-Being:Â€Current State and Trends, Washington, DC. Millennium Ecosystem Assessment. Island Press, 23. Lotze, H. K., Lenihan, H. S., Bourque, B. J. etÂ€al. (2006). Depletion degradation, and recovery potential of estuaries and coastal seas. Science, 312, 1806–1809. Ludwig, D., Brock, W. A. and Carpenter, S. R. (2005). Uncertainty in discount models and environmental accounting. Ecology and Society, 10(2). McKee, D., Atkinson, D., Collings, S. E. etÂ€al. (2003). Response of freshwater microcosm communities to nutrients, fish, and elevated temperature during winter and summer. Limnology and Oceanography, 48, 707–722. McKenzie, V. J. and Townsend, A. R. (2007). Parasitic and infectious disease responses to changing global nutrient cycles. EcoHealth, 4, 384–396. Millennium Ecosystem Assessment (2005). Ecosystems and Human Well-Being:Â€Synthesis, Island Press, Washington, DC. Mimikou, M. A., Baltas, E., Varanou, E. and Pantazis, K. (2000). Regional impacts of climate change on water resources quantityÂ€and quality indicators. Journal of Hydrology, 234, 95–109. Mooij, W. M., Hülsmann, S., De Senerpont Domis, L. N. etÂ€al. (2005). The impact of climate change on lakes in the Netherlands:Â€a review. Aquatic Ecology, 39, 381–400. Moss, B., McKee, D., Atkinson, D. etÂ€al. (2003). How important is climate? Effects of warming, nutrient addition and fish on phytoplankton in shallow lake microcosms. Journal of Applied Ecology, 40, 782–792. Mulholland, P. J., Helton, A. M., Poole, G. C. etÂ€al. (2008). Stream denitrification across biomes and its response to anthropogenic nitrate loading. Nature, 452, 202–205. Murdoch, P. S., Baron, J. S. and Miller, T. L. (2000). Potential effects of climate change on surface-water quality in North America. Journal of the American Water Resources Association, 36, 347–366. National Research Council (2000). Clean Coastal Waters:Â€Understanding and Reducing the Consequences of Nutrient Pollution, National Academy Press, Washington, DC. Nixon, S. W. (1988). Physical energy inputs and the comparative ecology of lake and marine ecosystems. Limnology and Oceanography, 33, 1005–1025. Novotny, V. and Olem, H. (1994). Water Quality: Prevention, Identification, and Management of Diffuse Pollution, Van Nostrand Reinhold, New York. Oczkowski, A. and Nixon, S. (2008). Increasing nutrient concentrations and the rise and fall coastal fishery; a review of data from the Nile Delta, Egypt. Estuarine, Coastal, and Shelf Science, 77, 309–319. OECD (1982). Eutrophication of Water, Monitoring, Assessment and Control, Organisation for Economic Co-operation and Development, Paris. OECD (2006). Cost-Benefit Analysis and the Environment, Organisation for Economic Co-operation and Development, Paris. OECD (2008). OECD Environmental Data COMPENDIUM 2006– 2008, Organisation for Economic Co-operation and Development, Paris.

Bruna Grizzetti Oenema, O., Bleeker, A., Braathen, N. A. etÂ€al. (2011). Nitrogen in current European policies. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. OSPAR (2005). Common Procedure for the Identification of the Eutrophication Status of the OSPAR Maritime Area, OSPAR, London. OSPAR (2008). Nutrients in the Convention Area: Assessment of implementation of PARCOM Recommendations 88/2 and 89/4, OSPAR, London. Paerl, H. W. and Huisman, J. (2008). Climate:Â€blooms like it hot. Science, 320, 57–58. Palmeri, L., Bendoricchio, G. and Artioli, Y. (2005). Modelling nutrient emissions from river systems and loads to the coastal zone:Â€Po River case study, Italy. Ecological Modelling, 184, 37–53. Phillips, G., Pietilainen, O. P., Carvalho, L. etÂ€al. (2008). Chlorophyllnutrient relationships of different lake types using a large European dataset. Aquatic Ecology, 42, 213–226. Powlson, D. S., Addiscott, T. M., Benjamin, N. etÂ€al. (2008). When does nitrate become a risk for humans? Journal of Environmental Quality, 37, 291–295. Rahel, F. J. and Olden, J. D. (2008). Assessing the effects of climate change on aquatic invasive species. Conservation Biology, 22, 521–533. Rousseau, V., Becquevort, S., Parent, J. Y. etÂ€al. (2000). Trophic efficiency of the planktonic food web in a coastal ecosystem dominated by Phaeocystis colonies. Journal of Sea Research, 43, 357–372. Salomez, J. and Hofman, G. (2003). Nitrates:Â€a mixed blessing? Food Science and Technology, 17, 44–45. Salomon, M. (2009). Recent European initiatives in marine protection policy:Â€towards lasting protection for Europe’s seas? Environmental Science and Policy, 12, 359–366. Sanchez, P. A. and Swaminathan, M. S. (2005). Cutting world hunger in half. Science, 307, 357–359. Scheffer, M., Carpenter, S., Foley, J. A., Folke, C. and Walker, B. (2001). Catastrophic shifts in ecosystems. Nature, 413, 591–596. Schlesinger, W. H. (2009). On the fate of anthropogenic nitrogen. Proceedings of the National Academy of Sciences of the United States of America, 106, 203–208. Schmidt, C. K. and Brauch, H. J. (2008). N,N-Dimethylsulfamide as precursor for N-Nitrosodimethylamine (NDMA) formationÂ€upon ozonation and its fate during drinking water treatment. Environmental Science and Technology, 42, 6340–6346. Schreiber, H., Constantinescu, L. T. etÂ€al. (2003). Harmonised inventory of point and diffuse emissions of nitrogen and phosphorus for a transboundary river basin, http://danubis. icpot.org Seitzinger, S., Harrison, J. A., Böhlke, J. K. etÂ€al. (2006). Denitrification across landscapes and waterscapes:Â€a synthesis. Ecological Applications, 16, 2064–2090. Seitzinger S. P., Mayorga E., Kroeze C. etÂ€al. (2009). Global river nutrient export trajectories 1970–2050:Â€a Millennium Ecosystem Assessment scenario analysis. Global Biogeochemical Cycles, doi:10.1029/2009GB003587 Sileika, A. S., Stalnacke, P., Kutra, S., Gaigalis, K. and Berankiene,Â€L. (2006). Temporal and spatial variation of nutrient levels in the Nemunas River (Lithuania and Belarus). Environmental Monitoring and Assessment, 122, 335–354.

Simpson, D., Aas, W., Bartnicki, J. etÂ€al. (2011). Atmospheric transport and deposition of nitrogen in Europe. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Smith, V. H. (2003). Eutrophication of freshwater and coastal marine ecosystems:Â€a global problem. Environmental Science and Pollution Research, 10, 126–139. Smith, V. H. and Schindler, D. W. (2009). Eutrophication science:Â€where do we go from here? Trends in Ecology and Evolution, 24, 201–207. Stalnacke, P., Vandsemb, S. M., Vassiljev, A., Grimvall, A. and Jolankai, G. (2004). Changes in nutrient levels in some Eastern European rivers in response to large-scale changes in agriculture. Water Science and Technology, 49, 29–36. Stevens, C. J. and Quinton, J. N. (2009). Policy implications of pollution swapping. Physics and Chemistry of the Earth, 34, 589–594. Tappeiner, G., Tappeiner, U. and Walde, J. (2007). Integrating disciplinary research into an interdisciplinary framework: AÂ€caseÂ€study in sustainability research:Â€introduction to the special issue. Environmental Modeling and Assessment, 12, 253–256. Thieu, V., Mayorga, E., Billen, G. and Garnier, J. (2010a). Subregional and downscaled global scenarios of nutrient transfer in river basins:Â€Seine–Somme–Scheldt case study. Global Biogeochemical Cycles, 24, GB0A10. Thieu, V., Garnier, J. and Billen, G. (2010b). Assessing impact of nutrients mitigation measures along rivers continuum to southern bight of the North Sea. Science of the Total Environment, 408, 1245–1255. UNESDA (2009). Union of European Beverages Associations. van Drecht, G., Bouwman, A. F., Knoop, J. M., Beusen, A. H. W. and Meinardi, C. R. (2003). Global modeling of the fate of nitrogen from point and nonpoint sources in soils, groundwater, and surface water. Global Biogeochemical Cycles, 17, 26–31. van Grinsven, H., Ward, M. H., Benjamin, N. and de Kok, T. M. C. M. (2006). Does the evidence about health risks associated with nitrate ingestion warrant an increase of the nitrate standard for drinking water? Environmental Health, 5, 5–26. van Grinsven, H., Rabl, A. and de Kok, T. M. C. M. (2010). Estimation of incidence and social cost of colon cancer due to nitrate in drinking water in the EU: Quantative cost–benefit assessment. Environmental Health, 9(58). Velthof, G. L., Oudendag, D., Witzke, H. P. etÂ€al. (2009). Integrated assessment of nitrogen losses from agriculture in EU-27 using MITERRA-EUROPE. Journal of Environmental Quality, 38, 402–417. Vitousek, P. M., Aber, J. D., Howarth, R. W. etÂ€al. (1997). Human alteration of the global nitrogen cycle:Â€sources and consequences. Ecological Applications, 7, 737–750. Vollenweider, R. A. (1976). Advances in defining critical loading levels for phosphorus in lake eutrophication. Memorie dell’ Istituto Italiana di Idrotiologia, 33, 53–83. Voss, M., Baker, A., Bange, H.W. etÂ€al. (2011). Nitrogen processes in coastal and marine systems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Walker, B., Holling, C. S., Carpenter, S. R. and Kinzig, A. (2004). Resilience, adaptability and transformability in social-ecological systems. Ecology and Society, 9(2).

403

Nitrogen as a threat to European water quality Ward, M., de Kok, T., Levallois, P. etÂ€al. (2005). Drinking water nitrate and health: recent findings and research needs. Environmental Health Perspectives 113, 1607–1614. Webb, A. J., Patel, N., Loukogeorgakis, S. etÂ€al. (2008). Acute blood pressure lowering, vasoprotective, and antiplatelet properties of dietary nitrate via bioconversion to nitrite. Hypertension, 51, 784–790. Whitehead, P. G., Wilby, R. L., Battarbee, R. W., Kernan, M. and Wade, A. J. (2009). A review of the potential impacts of climate change on surface water quality. Hydrological Sciences Journal, 54,Â€101–123. WHO (2007). Public Water Supply and Access to Improved Water Sources,World Heath Organization, Geneva. Wilby, R. L., Whitehead, P. G., Wade, A. J. etÂ€al. (2006). Integrated modelling of climate change impacts on water resources and quality in a lowland catchment:Â€River Kennet, UK. Journal of Hydrology, 330, 204–220.

404

Wolf, J., Rotter, R. and Oenema, O. (2005). Nutrient emission models in environmental policy evaluation at different scales: experience from the Netherlands. Agriculture, Ecosystems and Environment, 105, 291–306. Worm, B., Barbier, E. B., Beaumont, N. etÂ€al. (2006). Impacts of biodiversity loss on ocean ecosystem services. Science, 314, 787–790. Worrall, F., Spencer, E. and Burt, T. P. (2009). The effectivenessÂ€ofÂ€nitrate vulnerable zones for limiting surface water nitrate concentrations. Journal of Hydrology, 370, 21–28. Wriedt, G. and Bouraoui, F. (2009). Towards a large scale assessment of water availability in Europe, European Commission Joint Research Centre, Luxembourg. Zhao, Y.-Y., Boyd, J. M., Woodbeck, M. etÂ€al. (2008). Formation of N-nitrosamines from eleven disinfection treatments of seven eifferent surface waters. Environmental Science and Technology, 42, 4457–4862.

Chapter

18

Nitrogen as a threat to European air quality Lead authors: Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Contributing authors: David Simpson, Till Spranger, Wenche Aas, John Munthe and Ari Rabl

Executive summary Nature of the problem • Atmospheric emissions of nitrogen oxides and ammonia are contributing to a number of negative effects to human health and ecosystems. These effects include both effects of the primary emissions but more importantly through actions of secondary pollutants such as ground level ozone (O3) and secondary particulate matter (PM). • The main air pollution effects include effects of nitrogen dioxide to human health, effects from ground level ozone to human health and vegetation and effects from particulate ammonium and nitrate to human health. There is a difficulty of ascribing health effects to NO2 per se at ambient levels rather than considering NO2 as a surrogate for a traffic-derived air pollution mixture.

Approaches • The chapter gives a brief review of our current understanding of the mechanisms and processes regarding N-containing air pollutants and their effects on human health, vegetation (effects of reactive nitrogen on ecosystems through eutrophication and acidification is treated in Dise etÂ€al., 2011; Velthof etÂ€al., 2011, Chapters 20 and 21, this volume) and materials. It presents historical development, current situation and outlines future perspectives of reactive nitrogen related air pollution and its effects in Europe in relation to national and EU legislation on emission limitation and air quality control.

Key findings/state of knowledge • In the EU-27 countries, 60% of the population lives in areas where the annual EU limit value of NO2 is exceeded. Air quality standards for nitrogen dioxide are exceeded mainly in urban areas. Concentrations have decreased since 1990, although the downward trends have been smaller or even disappeared after 2000. • Episodic ozone concentrations have decreased over Europe since 1990 due to VOC and NOx control. In the same time tropospheric background and continental background concentrations have increased. Present concentrations are still a threat to both human health and vegetation. • Ammonium and nitrate comprise substantial fractions of PM10 and PM2.5 (sometimes more than 1/3 and control of these compounds is important for meeting air quality standards). • It is very likely that sensitive species are and will be negatively affected by emissions of ammonia almost everywhere in western, central and parts of southern Europe, at least in areas with intensive animal husbandry.

Major uncertainties/challenges • There are large uncertainties with respect to further developments of continental-background ozone concentrations in the atmosphere due to uncertainties in future emissions of methane and nitrogen oxides. • The role of particulate ammonium and particulate nitrate regarding human health effects is still under discussion. The long-term effects of NO2 on human health found in epidemiological studies reflect rather effects of combustion or traffic related air pollution than effects of NO2 per se, making it difficult to evaluate the health effects of NO2 as such.

Recommendations • The role of ammonia and nitrogen oxides regarding PM exposures needs to be further investigated, in particular with respect to their importance for health effects. • The low success in controlling ammonia emissions needs to be further assessed, in particular in connection with the development of new agricultural policies.


405

Nitrogen as a threat to European air quality Table 18.1 The role of N containing compounds and ozone in air pollution effects. The threats to ecosystems from N deposition are discussed in Grizzetti etÂ€al., 2011 and Dise etÂ€al., 2011 (Chapter 17 (threats to water)) and Chapter 20 (threats to biodiversity))

Effects Compounds

Human health

Nitrogen dioxide

X

Ammonia

Ecosystems X

Particles NH4 /NO3

X

Ozone

X

+

−

N deposition

Visibility

X

X

X X

X

X

X

X (acidification, eutrophication)

18.1â•‡ Introduction Air pollution is a major threat to human health and ecosystems in Europe. The EU Thematic Strategy on Air Pollution (TSAP) (CEC, 2005) estimated that air pollution in the year 2000 caused between 300 000 and 400 000 premature deaths, mainly due to particles but also with a significant contribution from ozone. Reactive nitrogen contributes significantly to formation of both these air pollutants. European ecosystems are also threatened from air pollution through deposition of N and S containing compounds and through direct effects on vegetation. In addition to the effects on human health and ecosystems there are also significant air pollution effects on materials an visibility. The European Union has through several directives regulated emissions of air pollution. These regulations include directives on emission from combustion plants, motor vehicles, off-road machinery, industrial processes, etc., but also the emissions ceilings directive. Air quality is further regulated through air quality standards (EC, 2008a, b, c). In addition there is a protocol on national emissions ceilings under the Convention on Long-Range Transboundary Air Pollution (LRTAP Convention) regulating emissions of sulphur dioxide, nitrogen oxides, ammonia and volatile organic compounds. All these efforts have caused a decrease in emissions since 1990. Even if the emissions are going down and are expected to be further reduced over the next decade, air pollution will still be a significant threat to European population and ecosystems over the next decade. Nitrogen oxides and ammonia emissions play an important role in these effects both directly through the action of the primary emissions but also indirectly through actions of secondary air pollutants and through their deposition to the ground (Table 18.1). In this chapter we will mainly assess the role of nitrogen in threats to human health, direct effects on vegetation (effects on acidification and eutrophication of ecosystems are treated in Dise etÂ€ al., 2011, Chapter 20 this volume) and effects on materials. Combustion processes, e.g. road traffic and industry are large contributors to emissions of nitrogen oxides (NOx), mainly in the form of nitric oxide, NO. As mentioned in Hertel etÂ€al., 2011 (Chapter 9, this volume), NO is rapidly oxidised to nitrogen dioxide, NO2, in the atmosphere. Nitrogen dioxide is a

406

Materials

strong oxidant that absorbs visible light and hence may form a brownish red colour layer during high concentration episodes. Additionally, nitrogen dioxide is a toxic gas that can cause both long term and short term effects on health. As already mentioned in Hertel etÂ€al., 2011 (Chapter 9, this volume), the threats of nitrogen oxides to air quality do not only concern NO and NO2 themselves. Emission of NOx also contributes to the formation of secondary pollutants, i.e. pollutants that are formed in the atmosphere, such as ozone (O3) and secondary particulate matter (PM). The exact formation pathways are covered in detail in Hertel etÂ€ al., 2011 (Chapter 9, this volume), and hence only briefly presented here. Ozone is formed photochemically in the presence of NO2 and volatile organic compounds (VOC). However, ozone can also be destroyed by reaction with NO and therefore low ozone concentrations are often observed close to the NOx sources while the high concentrations are observed further from the sources in urban background air. Ozone is one of the most important of the global air pollutants in terms of impacts to human health, croplands and natural plant communities, and may become more important in the future. Tropospheric ozone has also impacted on climate; according to IPCC (2007) the year 2005 radiative forcing caused by ozone formed from anthropogenic emissions was the third largest (0.35 W/m2) after that of anthropogenic CO2 and methane (1.66 and 0.45 W/m2 respectively). Particulate nitrate can be formed from oxidation of NO2 to nitric acid (HNO3) that can further react with ammonia to form ammonium nitrate or can be absorbed on existing particulate matter. Nitrate and ammonium are two of the major inorganic components in urban aerosol particles. In the atmosphere NH3 reacts not only with HNO3 but also with other acid gases such as H2SO4 and HCl, and aerosols, forming ammonium (NH4+) containing particles. Oxides of nitrogen can also contribute to formation of Secondary Organic Aerosol particles (SOA) in photochemical smog. Atmospheric particles have an adverse impact on both climate and health. The climate effect is both direct via absorbing terrestrial radiation and scattering solar radiation and indirect, e.g. by influencing clouds. Both effects lead to cooling of the climate (IPCC, 2007) and will be covered in more detail in Butterbach-Bahl etÂ€al., 2011 (Chapter 19, this volume). The fact that particles scatter/absorb light also affects visibility in cities and scenic areas.

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson

Particulate matter (PM) is the most important contributor to adverse health effects of air pollution (WHO, 2005a). Before assessing effects of reactive nitrogen through contribution to the PM-mass we will give a brief overview of PM properties, sources and sinks. Particles can be both natural and anthropogenic in origin. Soil erosion, sea spray, volcanic eruptions and oxidation of biogenic VOC are examples of natural sources, while e.g. biomass burning and fossil fuel combustion are anthropogenic sources. Particles are also addressed as being primary or secondary depending on how they arise in the atmosphere. If the particles are emitted directly from its source they are referred to as primary, whereas the term secondary is used for particles that are formed in the atmosphere via gas-to-particle conversion, often induced by chemical reactions. The formation of particulate nitrate is an example of a secondary particle formation process initiated by the oxidation of NO2 to HNO3. Particles are not only classified regarding their origin but also by size. The particle size range is divided into coarse particles, i.e. particles with a diameter of >2.5 μm, and fine particles < 2.5 μm. The fine fraction is further divided into accumulation mode (100 nm–2.5 μm), ultrafine mode (10–100 nm) and nucleation mode particles ( 2.5 μm) may preferentially affect the airways and lungs, while fine particles (diameter > 0.1 μm) may preferentially affect the cardiovascular system. Ultrafine particles (UFP, diameter > 0.001 μm) may also migrate via the lung to other organs, including the liver, spleen, placenta and foetus, or via the nerve system to the brain. The health implications of these observations remain unknown since there are not yet enough epidemiological studies to be able to determine the exposure-response relationship for fine and ultrafine particles. This is why there are currently no guidelines for UFP exposure. Smaller particles have larger relative surface areas and therefore commonly induce more inflammation (Diociaiuti etÂ€ al., 2001; Pozzi etÂ€al., 2003). Health risks of PM in terms of increase of the diurnal average PM10 by 10 μg/m3 are, according to an analysis done by the WHO (2000), an increase in relative risk of mortality by 0.6%– 1.6%, an increase in occurrence of asthma related problems and medication usage by 3%–5% and increase of the number of daily hospital admission due to respiratory causes by 0.8%. As the long-term exposure to PM results in a substantial reduction in life expectancy, the long term effects clearly have greater significance to public health than the short-term effects. PM2.5 shows the strongest association with mortality indicating a 6% increase in the risk of deaths from all causes per 10â•›μg/m3 increase in long-term PM2.5 concentration. The estimated relative risk amounts to 12% for deaths from cardiovascular diseases and 14% for deaths from lung cancer per 10â•›μg/m3 increase in PM2.5 (Poppe etÂ€ al., 2002, 2004) (Table 18.7). Other effects related to long-term exposure include increases in lower respiratory symptoms and chronic obstructive pulmonary disease and reductions in lung function in children and adults. Studies on large populations show a strong effect of PM2.5 on mortality, and have been unable to identify a threshold concentration below which ambient PM has no effect on health:Â€a no-effect level. After a thorough review of recent scientific evidence, a WHO working group therefore concluded that, if there is a threshold for PM, it lies in the lower band of currently observed PM concentrations in the European Region. The chemical composition of particles may also influence their health effects. The primary, carbon-centred, combustionderived particles have been found to have considerable inflammatory potency (Armstrong etÂ€al., 2004; Mudway etÂ€al., 2004). One of the hypotheses considered for PM’s mechanisms of action is the oxidative potential of the particles or specific components. PM from traffic sites seems to have high oxidative

Table 18.7 Estimates of increase of long-term mortalities attributable to an increase of the PM exposure by 10 μg/m3 and their confidence intervals (CI) (Poppe etÂ€al.., 2002)

Long-term mortality

Age group (years)

Relative risk per 10 μg/m3 (95% CI)

Mortality, cardiopulmonary

≥30

1.08 (1.02Â€– 1.14)

Mortality, lung cancer

≥30

1.13 (1.04Â€– 1.22)

Mortality, total â•… (excluding violent death)

≥30

1.06 (1.02Â€– 1.10)

activity, and emissions from road traffic have been linked with a wide range of health effects, including effects on the cardiovascular and respiratory systems, and on atopic sensitisation to allergens in outdoor air. There is substantial epidemiological evidence of associations between health and sulphates that suggest that if sulphates are reduced (as part of the reduction of a mixture) then there will be real benefits to health. There is not much evidence for toxicity of airborne nitrates, which may be at least partly due to difficulties with measuring nitrates. Problem arises also with difficulty of epidemiological studies to distinguish effects of different pollutants in ambient air and of toxicological studies to describe effects across all sensitive groups in the population. The situation is very well summed up by this passage from a recent review paper by Reiss etÂ€al. (2007):Â€‘For nitrate-containing PM, virtually no epidemiological data exist. Limited toxicological evidence does not support a causal association between particulate nitrate compounds and excess health risks. There are some possible indirect processes through which sulfate and nitrate in PM may affect health-related endpoints, including interactions with certain metal species and a linkage with production of secondary organic matter. There is insufficient evidence to include or exclude these processes as being potentially important to PM-associated health risk.’

18.2.3â•‡ Effects on vegetation Emissions of reactive nitrogen lead to increased atmospheric deposition into ecosystems. Various effects such as acidification and eutrophication of soils and waters, reduced biodiversity and formation of marine algal blooms are summarised in Butterbach-Bahl etÂ€al., Durand etÂ€al., Voß etÂ€al., Grizzetti etÂ€al., Dise etÂ€al. and Velthof etÂ€al., 2011 (Chapters 6, 7, 8, 17, 20 and 21, this volume). These long-term effects are mostly associated with critical loads of nitrogen, 10–100 years deposition levels of total Nr that are set as upper thresholds below which negative effects do not occur in specific ecosystems. In this chapter only the direct effects of the primary and secondary gas-phase pollutants related to emissions of Nr to the air, i.e. NO, NO2, NH3 and ozone, will be assessed. These effects are associated with critical levels of individual air pollutants, i.e. short-term air concentration levels (1 h to 1 year means) ‘fixed on the basis of scientific knowledge, above which direct adverse effects may occur’ (EC, 2008c). Legislation for the protection of natural vegetation against the direct effects from air pollution has developed together

413

Nitrogen as a threat to European air quality

with legislation for air pollution effects on human health. The Â�vegetation-related concentration limits for primary and secondary air pollutants related to reactive nitrogen as set in Directive 2008/50/EC were presented in Table 18.4.

Ammonia, NH3 Exposure to ammonia leads to a mixture of direct, indirect, primary and secondary effects on vegetation and ecosystems (Cape etÂ€ al., 2009; Sutton etÂ€ al., 2009). There are clear indications that the pathway of uptake of ammonia via leaf uptake from atmosphere, i.e. the direct effect of ammonia, is dominant (as opposed to the indirect effect which is via root system from the soil) (Sutton etÂ€al., 2009). Ammonia acts as a macro-nutrient and at low exposure levels plants respond by increasing their biomass production. Growth stimulation is also considered as potentially adverse for (semi-) natural vegetation because plant growth is often limited by the supply of nutrient nitrogen, and so any increases in growth may lead to negative effects on community composition. The fertilisation effect can at higher exposure levels lead to secondary long-term adverse effects including increased susceptibility to abiotic (drought, frost) and biotic stresses. In addition, various primary toxic effects are known (Cape etÂ€al., 2009; Sutton etÂ€ al., 2009). The critical levels have been revised in recent years on the basis of experimental data (Table 18.8). Annually averaged concentrations below 1 µg/m³ will protect (1) sensitive lichen communities and bryophytes and (2) ecosystems where sensitive lichens and bryophytes are an important part of the ecosystem integrity. Based on data from heathlands and forest ground flora, 3 µg/m³ (uncertainty estimate 2–4 µg/m³) are assumed to protect higher plants. The critical levels given in Table 18.8 apply for native and forest species. A monthly average critical level 23 µg/m³ was retained to deal with the possibility of high peak emissions during periods of manure application.

Oxides of nitrogen, NOx Oxides of nitrogen can have a fertiliser effect, but can also be toxic to plants, depending on concentrations. The critical levels for NOx are based on the sum of the NO and NO2 concentrations because there is insufficient knowledge to establish separate critical levels for the two pollutants, although some evidence indicates that at low concentrations typical of ambient conditions, NO is more phytotoxic than NO2 (Mills, 2004). Since the type of response varies from a fertiliser effect to toxicity depending on concentration, all effects have been considered to be adverse. As for ammonia, the growth stimulation was also considered as potentially adverse for (semi-) natural vegetation owing to potential negative effects on community composition. In the past, the critical level for nitrogen oxides referred only to NO2. However, because of new evidence of the toxicity of nitric oxide (NO), the critical level now refers to NOx, defined as the combined concentrations of NO and NO2. The critical level value remains 30 μg/m3 (as NO2 equivalent) as an annual mean and 75â•›μg/m3 as a 24 hour mean. As for ammonia, UN/ECE Working Group on Effects strongly recommended the

414

Table 18.8 Critical levels (CL) of ammonia (Cape etÂ€al., 2009)

Averaging period

Critical value µg/m3

1 month

23

Provisional value for all â•‡ plants

1 year

1

Lichen communities and â•‡bryophytes; ecosystems where these are a key part of ecosystem integrity

1 year

3 (2–4)

Higher plants (heath â•‡land, grassland and forest ground flora)

Receptor

use of the annual mean value, as this parameter is much more reliable than shorter-term averages, and the long-term effects of NOx are thought to be more significant than the short-term effects (Mills, 2004).

Ozone Ozone damage to vegetation has been recognised and studied for many decades (Benton etÂ€ al., 2000; Matyssek and Innes, 1999; Skärby etÂ€ al., 1998). Today ozone is considered to be the most important gaseous pollutant causing effects on vegetation in Europe. It enters plants through leaf stomata and oxidises plant tissue, causing changes in biochemical and physiological processes and eventually death of the injured plant cells. Besides visible injuries on leafs and needles, ozone also causes premature leaf loss, reduced photosynthesis and reduced leaf, root, and total dry weights in sensitive plant species. This leads to significant decrease in productivity of some agricultural crops and to reduced forest production. In addition, many native plants in natural ecosystems are sensitive to ozone. The developments in ecotoxicology led to development of critical levels for ozone effects on plants. Traditionally, these were related to ozone concentrations, using simple mean values for different time windows (day, month, etc.), as still is the case for other gaseous air pollutants (NOx, SO2, NH3). In 1990s a new kind of critical level for ozone was elaborated, based on accumulated exposure over a concentration threshold, following the rationale that higher ozone concentrations are believed to be more damaging to plants. Example of such metrics is AOT40 used in Europe (Fuhrer etÂ€al., 1997). The LRTAP Convention’s mapping manual (Mills, 2004) suggested two main metrics for use in performing regional scale risk assessments of ozone damage:Â€the AOT40 index and the flux based AFstY approach (Accumulated stomatal Flux over thresholds of Y nmol/m2/s). The flux-based approach relates risk to the absorbed ozone dose rather than ambient ozone concentration, through the use of stomatal conductance algorithms. These flux-based metrics have been defined in detail in the revised UN-ECE Mapping Manual (Mills, 2004) and are strongly recommended rather than use of AOT40. Table 18.4 presents a summary of the limit and target values of the EU Directive 2008/50/EC related to effects on vegetation.


18.2.4â•‡ Effects on materials Atmospheric pollution is an important factor in material deterioration including degradation of systems used for material protection and cultural heritage materials. Corrosion of materials was originally mostly associated with air pollution by sulphur dioxide (SO2); however the more recent studies have shown that nitric acid (HNO3), ozone and particulate matter contribute significantly to the negative effect of air pollution on materials. The lifetime of technological products is shortened because of air pollution. Buildings and other structures, as well as objects of cultural heritage, exposed to the atmosphere deteriorate more rapidly. The resulting physico-chemical and economic damage can be significant, not to mention the loss of unique parts of our cultural heritage and hazards due to decreased reliability of complicated technological devices. Also, as the result of weathering, especially that caused by acidifying pollutants, a significant part of the metals used in construction and manufactured products are released to the biosphere with a potential hazard to the environment. Deterioration rates can be calculated using dose–response functions. The recommended functions have been derived from field research programmes undertaken as part of the UN-ECE ICP Materials Exposure Programmes. Two sets of dose–response functions have been derived. One was developed for SO2-dominated situations taking into account the synergic effect of exposure to ozone and the effect of acid rainfall in combination with climatic parameters. The second was developed for multi-pollutant situations combining effects of gaseous SO2, NOx, O3, HNO3 and particulate matter together with acid rainfall in combination with climatic parameters. The impact of wet deposition of acidic species on sensitive materials is considered as an effect of the total load of H+ deposition, and impact of deposition of the sea salt as an effect of the total load of Cl– deposition. The dose–response functions use annual average concentrations of air pollutants and are available for limestone, sandstone, copper, bronze, zinc, steel and aluminium (ICP, 2010). The recommended unit of corrosion attack is the surface recession R (in μm) with the exception of aluminium, where the mass loss ML (in gm–2) should be used instead. Table 18.9 summarises the available dose–response functions for materials and lists the dependencies of these functions. For zinc, bronze and limestone the multi-pollutant function should be used when levels of HNO3 and/or particulate matter are expected to be high, as is the case with most European cities today with pollution dominated by traffic. Dose–response functions for paint coatings are also available, expressed as lifetime equations for the coatings. Because atmospheric deterioration of materials is a cumulative, irreversible process, which proceeds even in the absence of pollutants, ‘critical’ values are not as easily defined as for some natural ecosystems. Some rate of deterioration must be defined which may be considered ‘acceptable’ based on technical and economic considerations. This approach provides the basis for mapping ‘acceptable areas’ for corrosion, and deriving areas where the acceptable pollution level/load is exceeded, in an analogous way to the maps produced for natural ecosystems.

The term ‘acceptable’ is reserved for materials used in technical constructions, while ‘tolerable’ is used in connection with the degradation of cultural heritage. Based on maintenance intervals and tolerable corrosion attack before maintenance for cultural heritage objects, tolerable corrosion rates have been determined. These corrosion rates are 2.5 times higher than background corrosion values and values of tolerable corrosion for the first year exposure of some materials are listed in Table 18.9. Particles also contain soiling materials, and the tolerable PM10 level for soiling of three selected materials is 12–22 μg/m3 based on reasonable cleaning intervals.

18.3â•‡ Historical trends in air pollution and their current effects on health, vegetation and materials After an improvement of the air pollution situation in Europe at the beginning of nineties the trends in concentrations have been more or less stagnating for many air pollutants during the decade beginning at late 1990s. Overall exposure of Europe’s population to pollutants with a health impact has not improved since the late 1990s; however, there have been some pollutant-specific exceptions. Figure 18.1 shows that increasing part of urban population is exposed to ozone and PM concentrations over the target values. Whilst exposure to high levels of NO2 has steadily decreased, up to 30% of Europe’s urban population may still be exposed to concentrations in excess of limit values. Thus, determined effort is still required if ambient air concentration and exposure targets are to be met. For ozone there was considerable variation between years. Usually, a maximum of 25% of the urban population was exposed to concentrations above limit values. In 2003Â€– a year with extremely high ozone concentrationsÂ€ – this fraction increased to approximately 60%. For PM10 the urban population potentially exposed to ambient air concentrations in excess of the EU limit value varied between 23% and 45% between 1997 and 2004. There was no discernible trend over the period (EEA, 2010). Acidifying emissions in Europe have declined substantially since 1990. As sulphur emissions have fallen, nitrogen has become the predominant acidifying agent. Ozone concentrations have remained largely unchanged in recent years, even though emissions of precursor gases have been falling. Exposure of vegetation to ozone exceeds criteria for protection over very large areas of central and southern Europe. Declining concentrations of acidifying air pollutants has resulted in decreased observed corrosion rates of materials at the ICP Materials sites, by about 50% on average in the period 1987–1997. The corrosion rate of carbon steel decreased further in 1997–2003 (Figure 18.2), though the rates for zinc and limestone increased slightly. Nitric acid and particulate matter currently contribute to corrosion, in addition to sulphur Â�dioxide. Exceedances of tolerable levels of corrosion for cultural heritage materials were frequent. For 1990, it was estimated that air pollution caused € 1.8 billion of materials damage. Emission reductions envisaged under the Gothenburg Protocol are

415

416 

Multi

a

If RH < 60%, counts as RH = 0.

SO2 



multi

Sandstone



SO2

Limestone





multi SO2

Aluminium



SO2

Bronze





Multi SO2



SO2

Copper

Zinc



SO2

Steel

μg/m3

SO2

Situation

Material



O3







PM10





HNO3



















mg/m2/ year

H+





Cl−



















%

a

RH

C





















o

T

7

8

0.22

0.6

0.8

1.1

20

μm/year

Tolerable corrosion

Table 18.9 Functional dependencies of the dose–response functions for materials (UN/ECE ICP-materials, 2007). Situation (of air pollution) is SO2-dominated (SO2) or multi-pollutant (multi); dependencies of dose–response functions on concentrations of gaseous pollutants, wet deposition of acidity (H+), chlorides (Cl−), relative humidity (RH) and temperature (T) are based for annual mean values of these environmental variables


and 3 μg/m3 for herbaceous plants; Cape etÂ€ al., 2009) reveals the actual risk of elevated NH3 emissions on terrestrial ecosystems. Despite uncertainties, it is very likely that bryophytes and lichens are negatively affected almost everywhere in Western, Central and parts of Southern Europe, and herbaceous plants in areas with intensive animal husbandry (NW France, Flanders, Netherlands, Denmark, parts of Germany, Switzerland and Northern Italy). Not surprisingly, these are also the areas with highest nitrogen deposition rates and exceedances of critical loads, with additional adverse effects on sensitive ecosystems (see Dise etÂ€al., 2011, Chapter 20 this volume). The contribution of ammonium to the PM concentrations can be estimated from measurements of PM chemical composition. Putaud etÂ€al. (2004) reviewed composition of PM measured over the last decade at 24 sites situated in natural, rural, near-city, urban, and kerbside areas in Europe. Figure 18.4 shows the contribution of ammonium to PM2.5, PM10 and coarse PM at the different types of locations. The figure indicates that contribution of ammonium to PM2.5 in Europe is around 8%. The ammonium contribution to PM is by Amann etÂ€al. (2008) assumed proportional to the PM-related health effects calculated for the National Emission Ceilings analysis. Loss in statistical life expectancy attributable to the exposure of PM2.5 is between 6 and 36 months in central Europe (Figure 18.51).

expected to improve materials damage across Europe by more than € 1 billion (UN/ECE ICP-materials, 2007).

18.3.1â•‡ Ammonia Regional ammonia (NH3) concentrations in Europe have been calculated with the EMEP MSC-West model; Figure 18.3 shows the yearly mean concentrations for the years 1990, 1995, 2000 and 2005. Note that with a grid size of 50 km × 50 km, the model cannot capture the large gradients in ammonia that exist at small spatial scales. The model overestimates low ammonia concentrations (for instance at the forest sites from the EU project NOFRETETE) and underestimates high concentrations (e.g. EMEP sites, which are situated in rural areas, often surrounded by agricultural activities). However, the model calculations did not show any systematic deviation for ammonia with respect to seasons. Comparing the calculated NH3 concentration with the updated critical levels (1 μg/m3 for lichens and bryophytes % of urban population 100 80 60

18.3.2â•‡ Oxides of nitrogen

40

The concentrations of NO2 are spatially very variable in the urban environment, depending on time of the day, season, reactivity and meteorological factors. The natural background annual mean of NO2 is between 0.4–9.4â•›μg/m3, the urban annual mean is usually 20–90â•›μg/m3 (WHO, 2006) and for the rural background it is 15–30â•›μg/m3. Hence the WHO guideline, as well as the EU limit value concerning NO2, is exceeded in many larger cities and, as emissions of NOx are strongly traffic-related, there is a rising concern regarding NO2 concentrations in growing cities with high traffic density. For instance in a megacity like Beijing the annual average NO2 concentration in 2002 was 76 μg/m3 (Molina and Molina, 2004). In the EU-27

20 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 NO2

PM10

O3

Figure 18.1 Percentage of the urban population in Europe (EEA member countries) potentially exposed to pollutant concentrations over selected limit/ target values (only pollutants (partly) related to NR) (NO2:Â€annual mean of 40 μg/m3 NO2 not to be exceeded; PM10:Â€24 h average of 50 μg/m3 not to be exceeded more than 35 times a calendar year; O3:Â€daily maximum of 8 h mean of 120 μg/m3 O3 not to be exceeded more than 25 days per calendar year, averaged over three years) (EEA, 2010a).

80

60 50 40 30 20 10

Lisbon (P)

Chaumont (F)

Lahemaa (Est)

Moscow (Rus)

Madrid (E)

Toledo (E)

Lincoln Cath. (GB)

Aspvreten (S)

Birkenes (N)

Stockholm (S)

Oslo (N)

Venice (I)

Milan (I)

Rome (I)

Cassacia (I)

Bottrop (D)

Langenfeld (D)

Ahtäri (Fin)

Waldhof (D)

Kopisty (Cz)

0 Prague (Cz)

m/year

Figure 18.2 Carbon steel corrosion rates for sites in 21 European cities for years between 1987 and 2005 (see legend). The grey dashed line is the tolerable corrosion rate 20 μg/year (figure prepared from data in UN/ECE ICP-Materials, 2006, and 2007).

1987/88 1992/93 1994/95 1996/97 1997/98 2000/01 2002/03 2005/06

70

417


1990

1995

8

8

4

4

3

3

2

2

1

1

0.5

0.5

2005

2000

8

8

4

4

3

3

2

2

1

1

0.5

0.5

Figure 18.3 NH3 concentrations (in μg/m3over Europe calculated with the EMEP model for the years 1990, 1995, 2000 and 2005 using a legend with limits of 1â•›μg/m3 for lichens and bryophytes (≥ darkgreen area) and of 3 μg/m3 for herbaceous plants (≥ dark orange area).

countries, 60% of the population lives in areas where the annual EU limit value of NO2 is exceeded. The exceedance is primarily associated with the urban environment and its local traffic (Mol etÂ€al., 2010). Urban pollution is assessed in detail in SvirejevaHopkins etÂ€al., 2011 (Chapter 12 this volume). Also in suburban areas some Member States are still having problems with attaining the annual limit value as can be seen in Figure 18.5. In Europe (the EEA32 countries) the NOx emission have decreased by 18% in the period 1997–2006; traffic related

418

emissions showed an even larger decrease of 28%. In line with this, the ambient NOx concentrations showed decreasing trend throughout the region with decrease of about 22% at rural stations and 27% at urban/traffic stations between 1997 and 2007 (Figure 18.6). While the decrease in NO2 concentrations at rural and urban stations was quite similar to the decrease in NOx during this period, reduction in NO2 concentrations at traffic stations was small (6%) (Figure 18.6; Mol etÂ€al., 2009). This can be explained by two processes that are important only


very close to the sources. Firstly, at a constant oxidant level the NO2/NOx ratio increases with lowering NOx concentration due to the NO titration. Secondly, there are clear indications that the fraction of directly emitted (primary) NO2 in the total NOx emission from road transport is increasing. The main reason is high proportion of the primary NO2 in emissions from the growing fleet of diesel cars of Euro 3 and III standard and

Figure 18.4 Contribution of ammonium to PM2.5, PM10 and coarse PM (=PM10−PM2.5) at natural and rural background, near city and urban background, and kerbside measurement stations in Europe (24 in total). Data for years 1991–2001 were analysed after the data from Putaud etÂ€al. (2004).

higher, equipped with oxidation catalysts or particle traps incorporating oxidation catalysts (Sjödin etÂ€al., 2009).

18.3.3â•‡ Ozone Rural O3 concentrations had doubled from about 10–15 ppb in rural Europe at the end of the nineteenth century to 20–30 ppb in the 1980s. Volz and Kley (1988) were able to show that by analysing the more than 100 years old ozone data from around the turn of the nineteenth and twentieth centuries made at Montsouris by old-day measurement methodologies and comparing it with more modern techniques. Since the 1980s, rural O3 concentrations have increased in many areas (Staehelin and Schnadt Poberaj, 2008) with quite different rates of change at different locations (The Royal Society, 2008). The majority of ozone precursor emissions originate from anthropogenic sources. A recent review of the Gothenburg Protocol (TFIAM, CIAM, 2007) showed that the emissions of the O3 precursors NOx and NMVOC have declined substantially as a result of emissions controls. In 2005, NOx and NMVOC emissions were 30% and 38% lower than 1990 levels for the European countries within the Protocol. Figure 18.5 Annual mean concentration map of NO2 (μg/m3) in 2008; the two highest concentration classes correspond to the limit value (40 μg/m3) and limit value plus margin of tolerance (44 μg/m3), respectively (Mol etÂ€al., 2010).

419


The decrease in NOx and NMVOC emissions in Europe has resulted in a reduction in the magnitude of short-term peak O3 concentrations during episodes, with declines in daily peak concentrations of around 30 ppb. Reductions in peak O3 concentrations have been observed widely in Europe, both in urban and rural areas. The temporal pattern of O3 concentrations, however, reveals several additional changes during the period in which emissions of O3 precursors over Europe have declined. In particular, the lower percentiles of the Â�frequency

Figure 18.6 Relative changes in annual mean concentrations of NO2 and NOX to year 2007 for the three stations types of the AirBase network:Â€rural (a), urban (b) and traffic (c) (data from Mol etÂ€al., 2009).

(a)

distribution and even the mean concentration at many sites have been growing. These effects are illustrated in Figure 18.7a and Figure 18.7b, from Jenkin (2008). This figure shows the changes in O3 concentrations at an urban site (Leeds) and a rural site (Lullington Heath) in the UK, and show similar trends to those observed at other sites across the UK, and more widely in Northern Europe (The Royal Society, 2008).

Effects on health The highest levels of exposure to ozone are estimated for southern Europe, with the highest levels found in northern Italy. The estimated population exposure indicates that large regions fail to meet environmental objectives and a notable fraction of the urban population, typically around 25%, is exposed to elevated ozone. Extreme conditions in 2003 pushed this to approximately 60%. Regional differences in exposure levels across Europe are shown in Table 18.10. These are expected to diminish in the next decade. Exposures in continental Europe are projected to fall by 20%–30% in southern France, Germany, northern Italy and Switzerland and to rise in the United Kingdom and Scandinavia (WHO, 2008). It can be expected that population exposure will increase regardless of currently planned precursor emission reductions, owing to increasing background levels and reduced ozone depletion in urban areas. Current exposures to ozone in Europe are associated with premature mortality and morbidity. Effects include 21 000 premature deaths, 14 000 hospital admissions for respiratory disease and more than 100 million person-days of restricted activity per year in the EU 25 (WHO, 2008). These figures are underestimates, as they do not account for possible effects at levels below 70 μg/m3. The heat wave of summer 2003 gave an indication of the potential impact of a future warmer climate on air-pollution related health effects. Bell etÂ€al. (2007) investigated how climate

(b) 350

200 Leeds Centre

300

160

250

120

95 %-ile 90 %-ile

80

75 %-ile 40

50 %-ile 25 %-ile 10 %-ile 5 %-ile 1995

2000

2005

[ozone] (µg m–3)

[ozone] (µg m–3)

max

0 1990

Lullington Heath

max

200 150 120

95 %-ile 90 %-ile 75 %-ile 50 %-ile 25 %-ile 10 %-ile 5 %-ile

100 80 60 40 20 0 1990

1995

2000

2005

Figure 18.7 Changes in ground-level O3 concentrations at (a) urban (Leeds centre) and (b) rural (Lullington Heath) sites in the UK, showing a decline in peak values and increases in the mean and lower percentiles of the distribution. Trend in hourly mean O3 distributions based on data over the periods 1993–2006 and 1990–2006, respectively. The solid lines are linear regressions of data indicating the average trend over the period (Jenkin, 2008). With permission from Elsevier.

420

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Table 18.10 Fractions of populations exposed to ozone levels exceeding the EU directive target value of 120 μg/m3 for more than 25 days a year, by region (WHO, 2008)

Regiona

2002

2003

Northern Europe

0%

0%

Northwestern Europe

0–10%

40–50%

Central and eastern Europe

20–30%

80–90%

Southern Europe

60–70%

60–70%

a

orthern Europe:Â€Denmark, Estonia, Finland, Iceland, Latvia, Lithuania, N Norway, Sweden; Northwestern Europe:Â€Belgium, France (north of 45ºN), Ireland, Luxembourg, Netherlands, United Kingdom; Central and eastern Europe:Â€Austria, Czech Republic, Germany, Hungary, Poland, Slovakia, Switzerland; Southern Europe:Â€Cyprus, France (south of 45ºN), Greece, Italy, Malta, Portugal, Slovenia, Spain.

change could affect ambient ozone concentrations, using an hourly concentration model for 50 United States cities for 1990 and 2050. Future concentrations were based on the IPCC A2 scenario and the impact of altered climate on ozone was estimated. The maximum 1 h ozone levels were estimated to increase on average by almost 10 μg/m3 (maximum 19.2 μg/m3), the highest increases occurring in cities with current high pollution levels.

Effects on vegetation Simpson etÂ€al. (2007) used the EMEP chemical transport model to map the different indicators of ozone damage across Europe for two illustrative vegetation types, wheat and beech forests. Figure 18.8a illustrates the AOT40 index for forests as calculated with the EMEP model for the year 2000, with Figure 18.8b showing the ratio of this AOT40 value to the recommended critical level (CL), 5000 ppb.h. Firstly, we can note that the spatial gradients of AOT40 are very large, with typically a factor 10 difference between AOT40 values in southern Europe and those in the Nordic countries. Exceedance of the 5000 ppb.h CL for AOT40 is widespread, with only a few areas (mainly in Northern Europe) experiencing lower values. Relative exceedances of the CL of more than a factor of 10 occur in southern Europe (Figure 18.8b). Figure 18.8c,d present the corresponding results for AFst1.6, the relevant flux-based statistic for deciduous forests. The spatial pattern of AFst1.6 is rather different from that of AOT40. Whereas AOT40 clearly shows maxima in southern Europe, with much lower values in the Nordic countries, the spatial gradients in AFst1.6 are much smaller. Although the highest AFst1.6 values are still seen in parts of southern Europe, the difference between the Mediterranean and southern Sweden or Finland is typically less than a factor of two. Indeed, for the great majority of Europe, AFst1.6 values lie between 8 and 16â•›mmol/m2. The suggested CL of 4 mmol/m2 for AFst1.6 seems to be exceeded over essentially all of Europe. As noted in the Royal Society’s synthesis on ozone (The Royal Society, 2008), there is a substantial body of evidence from North America and Europe, supported by some work in Asia, Africa and Latin America, that elevated O3 levels cause reductions in the yield of sensitive crop species (Mauzerall and

Wang, 2001; Emberson etÂ€al., 2003), and some estimates have been made of the economic impacts of crop loss due to ambient O3 levels. The annual cost of arable crop production lost due to O3 was estimated to be $2–4 billion in the USA in the 1980s, with an equivalent estimate for the EU of €6.7 billion (90% confidence interval €4.4–9.3 billion per year) in 2000 (Holland etÂ€al., 2006). This is equivalent to 2% of arable agricultural production, but does not account for a range of other effects, including those on crop quality, visible injury, and susceptibility to pests and diseases. The greatest economic losses in Europe were predicted to be in Mediterranean countries, together with France and Germany, because the assessment used a concentrationbased exposure index. Wheat, tomatoes, vegetables and potatoes were the crops with the greatest yield losses. Van Dingenen etÂ€al. (2009) recently provided the first global estimate of crop yield loss for four major commodities (wheat, rice, maize, soybean), of $14–26 billion in the year 2000. This is significantly higher than present day losses to crops projected to occur as a result of climate change. The Royal Society (2008) report further noted that all of these estimates are primarily based on data from field chamber experiments which may under-estimate the real effects of O3 in the field. For example, the decrease in soybean yield under open-air conditions in the Soy Free Air Concentration Enrichment (FACE) experiment, conducted in the USA, was greater than predicted by a synthesis of previous chamber studies (Morgan etÂ€al., 2006). In one season, this was partly because O3 exposure increased the impact of a major defoliating hail event. More field release experiments, in which O3 is released over a crop which is not enclosed in chambers, are therefore needed to reduce the uncertainty in future estimates of loss in crop productivity. These need to be in a range of locations and to cover different cropping systems. According to EEA (2007) large parts of the EEA 32 countries currently exceed exposure criteria for forests. More than half of the agricultural area exceeds criteria for crop protection, total crop yield losses reaching an estimated €3 billion per annum in 2000. Since 2000 the exposure of crops has not been reduced. Furthermore, in a number of areas ozone concentrations have actually increased in recent years as can be seen in Figure 18.9. Adverse meteorology and the changing balance of airborne pollutants lie behind this.

18.3.4â•‡ PM Particles have been addressed as being one of the most important air pollutants regarding adverse human health impacts. The largest health impact estimates are for long term effects of PM2.5. De Leeuw and Horalek (2009) estimated the number of premature deaths in EU-27 countries from PM2.5 to be almost 500 000. The number of deaths that can be attributed to long-term PM10 exposure in the EU-25 countries is about 350 000– 370 000 premature (CEC, 2005; EEA, 2009a,b) (impacts from PM10 and PM2.5 are not additive). Figure 18.10 shows the fraction of the European population that is exposed to a certain concentration range of PM10 (both annual mean and the 36th highest daily mean) and PM2.5. WHO (2006) set an air quality guideline

421

Nitrogen as a threat to European air quality (a)

(b)

100

100 60000

10

40000

80

80

8

20000

6

10000

60

60

4

5000 3 3000

40

40

2

1000

1.5

500

20

20 1.2

200

40

60

80

100

120

40

(c)

60

80

100

120

(d)

100

100 24 16

80

10 80

8

12 8

60

6 60

4

4 2

40

3 40

2

1 0.5

20

0.2

40

60

80

100

120

1.5 20 1.2

40

60

80

100

120

Figure 18.8 Calculated metrics for ozone damage to forests as calculated with the EMEP model for the year 2000:Â€(a) AOT40 values (ppb.h), (b) ratio of calculated AOT40 to critical level of 5000â•›ppb.h, ( c) AftY1.6 values (mmol/m2), (d) ratio of AfstY1.6 to critical level of 4 mmol/m2 (from Simpson etÂ€al., 2007).

of 10 μg/m3, the lowest level at which total, cardiopulmonary and lung cancer mortality have been shown to increase with confidence in response to PM2.5. Only about 9% of the population is exposed to concentration below this guideline. Besides the guideline, the WHO has defined three interim targets, 15, 25 and 35 μg/m3. The air quality guideline for PM10 is 20 μg/ m3 (WHO, 2000). About ¾ of the total European population is exposed to concentrations that are above this guideline for annual mean of PM10. Figure 18.11a shows the annual mean concentrations of PM10. A statistical analysis of the monitoring data indicated that the daily PM10 limit value corresponds with an annual mean of 31 μg/ m3, although regional differences may occur (Mol etÂ€al., 2010, and references there in), so both the exceedances of the annual limit

422

value and of the short-term (daily) limit value can be derived from the figure. The map indicates that both limit values have been exceeded in many countries across Europe. Figure 18.11b shows the annual mean concentrations of PM2.5 and enables a comparison with the PM2.5 target value of 25 μg/m3. Data from the AirBase network show exceedance of PM10 limit values of both daily and annual means at all types of stations with increasing numbers from rural background to urban background to traffic stations. The extent of exceedance of the daily limit value is larger than of the limit value for annual mean. The daily limit value is frequently exceeded at urban background stations (about 28% of stations) and at traffic stations (more than 32% of stations) (Mol et al., 2010). Regarding PM2.5, the AirBase data show that at 6%, 14%, 5%

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson O3 AOT40 (May-July)

30000

(a)

PM10 annual mean

AOT40 ( g/m3.h)

24000 18000

EU Target level

45 g/m3

0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Figure 18.9 Annual variation in the ozone AOT40 value (May–July in μg/ m3.h). Average values over all AirBase rural stations which reported data over at least six years in the period 1996–2006. The orange line corresponds to the 5 year averaged value (EEA, 2010b).

and 10% of the rural, (sub)urban background and traffic stations and industrial sites the PM2.5 target value of 25 μg/m3 has been exceeded (Mol et al., 2010). There are regional and seasonal variabilities in PM10 and PM2.5. In the Mediterranean region high concentrations of PM are often associated with inter-continental transport during the spring/summer (Saharan dust and Asian continental outflow) and it is therefore more common that PM concentrations are exceeded during spring and winter. For the rest of Europe, where PM is dominated by regional sources, winter time is the most common season for high PM concentrations which are a combination effect of higher emissions and less effective dispersion in wintertime. The annual average PM concentrations in continental Europe are shown in Table 18.11 for different environments. Highest concentrations are often measured at the road side. The Europe-wide tendencies in annual mean PM10 concentrations for the time period 1997–2007 are shown in Figure 18.12. The following observations can be made:Â€ PM10 concentrations in 2004 were approximately 25% lower than in 1997. However, an actual net tendency in PM10 concentrations from 1997 onwards cannot be discerned due to the large inter-year variations over the entire period. Variations in meteorological conditions between years can explain part of these variations. Urban and rural background concentrations trends follow each other closely. The rural background concentration provides the dominating contribution to total urban PM10. PM10 concentrations at street level are on average approximately 8 μg/m3 higher than the average concentrations measured at 301 urban stations in 19 countries. Figure 18.S1 in supplement shows map with loss in statistical life expectancy attributable to the exposure to PM2.3 which is highest in central Europe with statistical life expectancy loss between about 6 and 36 months.

Chemical composition of particulate matter Nitrate and ammonium contribute significantly to the particulate matter in Europe. From the EMEP measurements (rural stations, EMEP, 2008), NO3− and NH4+ contributed by between 6%–19% and 5%–9% respectively to the PM10. This is in line with earlier assessments by Putaud etÂ€al. (2004). The

(b)

PM10 36th highest daily mean

65 g/m3

(c)

PM2.5 annual mean

35 g/m3

Figure 18.10 Exposure of European inhabitants in year 2005 to (a) PM10, annual mean value; (b) PM10, 36th highest daily mean value; (c) PM2.5, annual mean value concentrations (sources:Â€(a) and (b) EEA, 2009a; (c) de Leeuw and Morálek, 2009).

relative quantity of nitrate tends to increase with increasing PM10 resulting in a higher relative contribution on exceedance days (daily average above 50 μg/m3) compared to annual averages. Putaud etÂ€al. (2004) found that on these days nitrate was a major component of PM10 and PM2.5 together with organic matter. The more than proportional rise in nitrate levels as a function of PM10 appears to be valid for the whole of north-

423

424 (b)

Figure 18.11 (a) Annual mean concentration map of PM10 (μg/m3); the two highest concentration classes corresponds to the annual limit value (40 μg/m3) and to a statistically derived level (31 μg/m3) corresponding to the short-term limit value). (b)Â€Annual mean concentrations of PM2.5 (from Mol et al., 2010).

(a)

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Table 18.11 Average PM10 and PM2.5 levels in Europe for 2002 (WHO, 2006)

Type of PM value PM10 annual average

PM10 daily average

PM2.5 annual average

μg/m3

Site

21.7

rural

26.3

Urban background

32

In street

38.1

rural

43.2

Urban background

51.8

In street

11–13

Rural background

15–20

Urban background

20–30

In street PM10, Traffic PM10, Urban PM10, Rural

125 100 75 50 1996

1998

2000

2002

2004

2006

2008

Figure 18.12 Inter-annual variation of PM10, 1997–2007 annual means (from EEA, 2009a, b).

western Europe. The reason for the extra high nitrate levels on exceedance days is that these days are often associated with very stagnant conditions. Ammonium nitrate, which has a more local character than sulphate, can build up fast, whereas the levels of pollutants from long range transport are not specifically enhanced. There are no clear long-term trends in NO3− concentrations. NO3− concentrations do not follow general PM10 and PM2.5 trends (Putaud etÂ€al., 2004). The highest concentrations found in Europe are in the Po Valley, where there are large concentrations of NOx as well as NH3. Some of the smallest concentrations of NO3− are found in rural and natural backgrounds that may be due to lack of local sources of NOx. However, the concentrations near cities are often higher than at kerbside sites, which may be due to the time needed to form NO3− from NOx or due to low concentrations of NH3 (Figure 18.13).

18.4â•‡ Future perspectives, national ceilings 18.4.1â•‡ Ammonia In its Thematic Strategy on Air Pollution (TSAP), the European Commission outlined the strategic approach towards cleaner air in Europe (CEC, 2005) and established interim environmental objectives for the year 2020. Current emission ceilings for ammonia that should be met in 2010 were developed by 2006 (Klimont etÂ€al., 2007). In order to meet new objectives for eutrophication, acidification and for particulate matter the following policies were suggested. (1) A possible extension of the Integrated Prevention and Pollution Control (IPPC) Directive, to include installations

Figure 18.13 Mean contributions of NH4+ and NO3− ions to PM10 (a) and PM2.5 (b) at 16 measurement sites evaluated by Putaud for time period 1991–2001 and in case of PM10 also at EMEP sites in 2006 (12 sites with NH4+ data and 25 sites with NO3− data, EMEP 2008).

for intensive cattle rearing and a possible revision of the current thresholds for installations for the intensive rearing of pigs and poultry. (2) In the context of the current rural development regulation and the Commission proposals for rural development for 2007–2013, the Commission encourages the Member States to make full use of the measures related to farm modernisation, meeting standards and agro-environment to tackle ammonia emissions from agricultural sources. The cost-effective emission ceilings for ammonium that would lead to achievement of the agreed interim objectives in 2020 and that are based on energy projections corresponding to the recent Climate and Energy Package of the European Commission and the national projections of agricultural activities were examined by Amann etÂ€al. (2008). The ‘Current policy’ (CP) scenario was adopted as the starting point for the optimisation of additional emission-control measures to achieve the TSAP objectives. The ‘Current policy’ scenario considers implementation rates of already decided emission control legislation as currently laid down in national laws and implementation of the recent Commission proposals on the introduction of EURO-VI standards for heavy duty vehicles (EC, 2009) and on the revision of the Integrated Pollution Prevention and Control (IPPC) Directive for large stationary sources (EC, 2008a). To achieve the environmental objectives of the TSAP in 2020, further emission reduction measures on top of those considered in the ‘Current policy’ scenario would be needed to increase the reduction efforts for NH3 emissions from 8% to 22%. Figure 18.14a shows the national emissions of NH3 (Gg/y) in 2007, the national emissions ceilings for year

425

Nitrogen as a threat to European air quality (a)

800 2007 NEC 2010

600

NEC 2020

Gg NH3

TSAP 2020 400

200

Germany

Finland

Spain

EU-27/10

Germany

Finland

Spain

EU-27

Austria

Romania

Latvia

Greece

France

Italy

Slovakia

UK

Denmark

Sweden

Lithuania

Netherlands

Poland

Estonia

Ireland

Slovenia

Belgium

Luxembourg

Hungary

Malta

Bulgaria

Cyprus

Portugal

Czech Rep.

0

NEC 2010

(b)

Figure 18.14 (a) National emissions of NH3 (Gg/y) in 2007 (EMEP, 2010), the national emissions ceilings for year 2010 (NEC 2010) and the ‘Current policy’ scenario for 2020 for the 27 EU MemberStates. The scenario with measures optimised to achieve the TSAP objectives in 2020 (TSAP 2020, the cost-effective scenario) is also shown. The EU-27 emission totals are divided by 10. (b ) Distances of the year-2007 national NH3 emissions from values for year 2007 on the straight-line emission trajectories from year 2000 to year 2010 (NEC 2010, brown) and to year 2020 (‘Current policy’ scenario, blue), respectively, expressed as difference in percent of the year2007 value on the respective emission trajectory. The areas with green smiles indicate EU countries with NH3 emissions that in 2007 were below the emission trajectory to NEC 2010 (afore the brown line) and to ‘Current policy’ 2020 (blue line), respectively. The total EU-27 NH3 emissions in year 2007 were below the emission trajectory to NEC 2010 but above the emission trajectory to ‘Current Policy’ 2020. All emission scenarios come from Amann etÂ€al. (2008).

"Current policy" 2020 40% 20% 0% –20%

2010 and the ‘Current policy scenario’ for 2020, representing suggested NEC 2020, for the 27 EU Member States’. Distances of the year-2007 national NH3 emissions from the emission trajectories going from the individual Member States emissions in 2000 to their respective NEC values are shown both for the NEC for 2010 and for the suggested NEC for 2020 in Figure 18.14b. It can be seen that only five Member States are above their emission trajectories to the 2010 NEC while more than a third are above their trajectories to the 2020 NEC. According to the national projections for 2010 published in the NEC directive status report for 2008 (EEA, 2009b) will only two Member States, Germany and Spain, miss their national ceilings.

18.4.2â•‡ Oxides of nitrogen The decrease in NOx emissions during the 1990s is partly due to the introduction of emission standards for road transport in Europe that included both light-duty vehicles (petrol and Â�diesel) and heavy duty vehicles, but also regulations targeting other sectors have contributed. In May 2005 the 1999 Gothenburg Protocol came into force regarding emission ceilings for (inter

426

Austria

Romania

Greece

Latvia

France

Italy

Slovakia

Denmark

UK

Sweden

Lithuania

Netherlands

Poland

Estonia

Ireland

Slovenia

Luxembourg

Belgium

Hungary

Bulgaria

Malta

Cyprus

Czech Rep.

–60%

Portugal

–40%

alia) NOx. Compared to the level of 1990, Europe’s emissions of NOx should by 2010 be reduced by 41% (VOC 40%, NH3 17%). Additionally, the EU introduced a national emission ceilings (NEC) Directive, 2001/81/EC, concerning NOx and other pollutants. Compared to the Gothenburg Protocol, this directive puts more pressure on some of the member states than others. The emission ceilings must be attained by 2010. While land-based NOx emissions are expected to decrease in the coming decades, emissions from shipping are projected to increase. In 2000 the emissions from international maritime shipping in the seas surrounding the European Union (i.e. Baltic, North Sea, Northeast Atlantic, and Mediterranean Sea) amounted to approximately 30% of the land-based emissions in the EU-25. Legislation is in force to control emission of SO2 and to some extent also NOx from international shipping; however, Annex 6 of the MARPOL protocol regulates NOx emissions only on newly installed engines after 2016. Considering the long lifetime of ship engines, this legislation will impact the NOx emissions only in distant future. Further, the expected increase in the volume of ship movements will compensate for the environmental benefits of these measures and will lead to

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson 12 Land-based sources, CP scenario

Mt NOx/year

10

Land-based sources, TSAP target International shipping baseline

8 6 4

International shipping MTFR

2

National shipping baseline

0 2000

2005

2010

2015

2020

Figure 18.15 Emissions of nitrogen oxides from shipping (baseline scenario considering MARPOL Annex 6) compared with the emissions from landbased sources in the EU25 (‘Current policy scenario’). Emissions at Maximal Technology-Feasible Reduction are shown for 2020 (after Cofala etÂ€al., 2007).

a continued growth in ship emissions. By 2020 emissions from maritime activities would come close to the projected baseline emission levels from land-based sources and surpass the target levels established by the European Commission for land-based sources (Figure 18.15). Airport studies confirm that aircraft continue to be a relatively small contributor to regional pollution although aircraft-related NOx contributions could increase as air traffic increases and other non-aircraft emission sources become progressively cleaner. The European Union have decided on a 10% substitution of traditional fuels in the road transport sector (petrol and conventional diesel) by alternative fuels before the year 2020 (Directive 2003/30/EC) in order to meet the challenges with increased transportation, decreased oil resources and enhanced greenhouse gas emissions. Impact of this directive on the NOx emissions have not been fully investigated yet as the emission factors for many alternative fuels and their blends are not known. Up-coming Euro 5 and 6 standards (EC 715/2007, 1999/96/EC), embodying after-treatment technologies will be implemented 2010–2015 with the aim of further reductions of the NOx emissions. The usage of diesel is increasing in Europe and this may have implications on the NO/NO2 ratio as diesel vehicles, especially when equipped with modern technology such as diesel particulate filters, emit more NO2 compared to petrol vehicles. This effect should be pronounced only close to the source region. In some regions it is evident today that NO2 is not decreasing at the same pace as NOx. Combined measurement and modelling efforts have not been able, however, to prove that the potentially increasing share of NO2 in NOx emissions due to the emissions from diesel cars could alone explain this trend. Still, in future assessment strategies concerning vehicle emissions and NOx this has to be taken into account. Additionally, it takes time to change the on-road fleet, hence, even though the latest vehicle technology means reduction in some pollutants, it may take time before that becomes a reality. Projections of future NOx emissions and their impact on future air pollution situation in Europe have been assessed by Amann etÂ€ al. (2008) with GAINS model. The ‘Current policy’ scenario, i.e. scenario implementing all agreed legislation, projects a decrease of emissions of NOx by about 48% in 2020 compared to the year 2000. This scenario approaches emission

levels of the National Emission Ceilings. These emission ceilings must be attained by the EU member states in 2010. In Figure 18.16 is shown comparison of the NEC values for the Member States to the national emission values reported to EMEP for the year 2007. Comparison with target values of the TSAP is shown as well. Figure 18.16 indicates that almost half of the Member States were in 2007 above their emission trajectories leading from their year-2000 NOx emissions to their NEC and large majority were above their respective trajectory to the planned NEC for 2020, in figure represented by the ‘current policy’ scenario for 2020 (scenarios from Amann etÂ€al., 2008). The national projections for 2010 published in the NEC directive status report for 2008 (EEA, 2009b) give a similar picture of attainment of the 2010 NEC in the EU-27 Member States.

18.4.3â•‡ Ozone Owing to the presence of stringent emission control legislation, ozone precursor emissions are expected to decline in the EU over the coming decade. VOC emissions are expected to decrease in EU15 by 33% in 2010 and 41% in 2020 compared to 2000. However, a lack of equivalent legislation will not prevent further increases in precursor emissions in other countries that are Parties to the Convention on LRTAP. This growth in emissions is expected to increase hemispheric ozone background concentrations. Furthermore, climate change could lead to higher biogenic emissions in the future (The Royal Society, 2008). At the urban scale, projected O3 precursor emissions controls (especially reductions in NOx emissions) and changes in background O3 concentrations will lead to changes in urban O3 concentrations, with potentially large increases by the end of the century depending on the future scenario. As described earlier, the changes in urban O3 concentrations result partly from reductions in the titration of urban O3 by reaction with emitted NO, but are also influenced by changes in the background O3 concentrations. This is because the reduction in the titration effect increases O3 concentrations up to the background level (Jonson etÂ€al., 2006). Figure 18.17 shows projected exposure trends as expected changes in SOMO35 for 2020 compared to similar calculations with 2000 emissions. In some areas, such as parts of the Russian Federation and Scandinavia, SOMO35 levels are seen to increase in 2020 compared to 2000, although absolute levels are relatively low (around 2000 μg/m3·days) in both cases. In those areas that had high levels of SOMO35 in 2000 (e.g. Italy and much of southern Europe), 2020 levels are seen to be significantly lower than 2000 levels, although this still leaves levels of around 4000–5000 μg/m3 days in these areas (WHO, 2008). Reduction in ozone exposure resulting from current policies, and thus in the health impact by 2020, is estimated to be small. Population ageing will increase susceptible groups and background risks in Europe in the foreseeable future.

18.4.4â•‡ PM Projections for the future have been based on current legislation, i.e. ‘business–as-usual’, and emissions of PM2.5 are

427


(a)

1600 2007

1400

NEC 2010

1200

NEC 2020 TSAP 2020

Gg NOx

1000 800 600 400 200

Malta

Austria

EU-27

Austria

EU-27

France

Spain

Ireland

France

Ireland

Malta

Belgium Belgium

Spain

UK

Greece

UK

Greece

Italy

Denmark Denmark

Poland

Italy

Lithuania

Germany

Latvia

Bulgaria

Hungary

Romania

Czech

Cyprus

Finland

Estonia

Portugal

Sweden

Slovakia

Slovenia

Netherlands

Luxembourg

0

Figure 18.16 (a) National emissions of NOx (Gg/year) in 2007 (EMEP, 2010), emissions ceilings for year 2010 (NEC 2010) and ‘Current policy’ scenario for 2020 (representing suggested NEC 2020) for the 27 EU Member States. The scenario with measures optimized to achieve the TSAP objectives in 2020 (TSAP 2020, the cost-effective scenario) is also shown. The EU-27 emission totals are divided by 10. (b) Distances of the year 2007 national NOx emissions from values for year 2007 on the straight-line emission trajectories from year 2000 to year 2010 (NEC 2010, brown) and to year 2020 (‘Current policy’ 2020, blue), respectively, expressed as difference in percent of the year-2007 value on the respective emission trajectory. The areas with green smiles indicate EU countries with NOx emissions that in 2007 were below the emission trajectory to NEC 2010 (before the brown line) and to ‘Current policy’ 2020 (blue line), respectively. The year 2007 NOx emissions of Germany, Denmark and Belgium that are above the NEC 2010 trajectory fall below the emission trajectory for ‘Current policy’ 2020. All emission scenarios come from Amann etÂ€al. (2008).

NEC 2010

(b) "Current policy" 2020 75% 50% 25% 0%

Poland

Lithuania

Germany

Bulgaria

Latvia

Hungary

Romania

Cyprus

CzechRep.

Estonia

Finland

Portugal

Slovakia

Sweden

Slovenia

Netherlands

–50%

Luxembourg

–25%

Figure 18.17 Calculated changes in SOMO35 values (μg/m3 × days) by 2020 compared with 2000 emissions (meteorology from 1997) (WHO, 2008).

428

projected to decrease by about 45% in 2020 compared to the year 2000. The main reason for this projected decline is expected to be improved vehicle technology, implemented in Euro standards 5 and 6. The PM10 is projected to decrease by 39% for the same time period which is mainly as a result of reductions in the transportation sector and power generation (CEC 2005). Subsequently, the projected life years lost due to PM2.5 is decreased by around 32% and premature deaths by 21% (CEC 2005). However, PM2.5 is still projected to account for 271â•›000 premature deaths in 2020. Table 18.12 summarises the estimated effects of air pollution on health. For 2020 it is projected that PM will cause less harm regarding both mortality and morbidity. Loss in statistical life expectancy attributable to the exposure to PM2.5 in Europe projected for year 2020 in the current policy scenario is shown together with the year 2000 situation in Figure 18.S1 in the supplementry material. Taking the projected small reductions in ammonia comparing to SO2 and NOx, it can be expected that contribution of ammonia to the secondary (and total) PM will increase.

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Table 18.12 Estimations of the effect of air pollution on health for the years 2000 (baseline) and 2020 (Current legislation in 2020 including Climate Policy) (PyeÂ€and Watkiss 2005).

Health end-point

Units (1000)

Pollutant

2000

2020

Acute mortality

Premature deaths

O3

21,4

20,7

Respiratory hospital admissions Minor Restricted Activity Days (MRADs)

Cases

O3

14

20

Days

O3

53,924

42,227

Respiratory medication use (Children)

Days

O3

21,413

12,897

Respiratory medication use (Adults)

Days

O3

8,837

8,136

Cough and â•‡ lower respiratory symptoms (children)

Days

O3

108,056

64,955

Chronic mortalitya

Life years losta

3,001

1,900

Chronic mortality

Premature deaths

PM

288

208

Infant mortality

Premature deaths

PM

0.6

0.3

Chronic bronchitis

Cases

PM

135,7

98,4

Respiratory hospital admissions

Cases

PM

51,4

32,6

Cardiac hospital admissions

Cases

PM

31,7

20,1

Restricted activity days (RADs)

Days

PM

288,292

170,956

Respiratory medication use (children)

Days

PM

3,510

1,549

Respiratory medication use (adults)

Days

PM

22,990

16,055

Cough and lower respiratory symptoms (children)

Days

PM

160,349

68,819

Lower respiratory symptoms â•‡ (adults with chronic symptoms)

Days

PM

236,498

159,724

a

a

PM a

N ote two alternative metrics are used for the presentation of chronic mortality from PM. Firstly in terms of years of life lost and secondly in terms of numbers of premature deaths. These are not additive.

18.5â•‡ Conclusions As shown in this chapter, nitrogen air pollution is a serious threat to human health and ecosystems, both through direct effects from the emitted compounds but more seriously through the secondary compounds formed in the atmosphere. Legislation to improve air quality is in place which results in improving air quality, however, there are still large exceedances of air quality standards and critical levels in Europe. Even if present and proposed legislation is fully implemented, emissions of nitrogen oxides and ammonia will still pose a problem in Europe. • Air quality standards for nitrogen dioxide are exceeded in many urban areas in Europe. The exceedance is primarily associated with the local traffic. The year-2007 NO2 emissions of 12 out of 27 EU Member States were above the emission trajectory from year 2000 to the NEC for 2010. • NEC for ammonia will be met in 2010 by most of the Member States; the ‘Current policy’ scenario, i.e. the scenario where all already agreed legislation has been implemented, is, however, above the emission level that would meet the TSAP objectives for 2020. It is very likely that sensitive species are and will be negatively affected by emissions of ammonia almost everywhere in Western, Central and parts of Southern Europe, at least in areas with intensive animal husbandry.

• There is still a need for better data on emissions and atmospheric concentrations, in particular for ammonia. • Reduction in PM2.5 mortality in 2020 for the ‘Current policy’ scenario gives lower than the TSAP goal for 2020, which is 47% reduction of the long-term exposure mortality taking year 2000 as a base. Considering the high contribution of Nr to PM2.5, further reductions of Nr emissions could have a positive effect on reducing PM-related health effects and would contribute to meet the TSAP goal. Further, deposition of Nr in Europe also leads to exceedances of critical loads for nutrition nitrogen in large parts of Europe, both under the current situation and under the ‘Current policy’ scenario in 2020. • The new findings on long-term exposure effects of ozone on mortality with the relative risk increase per additional 10 μg/m3 ozone adding on the relative risk increase per 10 μg/m3 PM2.5 would make it difficult to meet the TSAP target of 47% reduction in mortality. As only small decreases are projected for ozone concentrations, the mortality from ozone would remain high in the ‘Current policy’ scenario in 2020. • The uncertainties of the health impacts of nitrogen species (NO2, NH4+, NO3−, HNO3 and others) are very large and Â�further epidemiological and toxicological research is needed to obtain more reliable exposure–response functions. In particular, the damage cost of NH3 emissions is extremely uncertain.

429


Acknowledgements The authors gratefully acknowledge support from the Nitrogen in Europe (NinE) programme of the European Science Foundation, from the NitroEurope IP funded by the European Commission and from the COST Action 729.


References Amann, M., Bertok, I., Cofala, J. etÂ€al. (2008). National Emission Ceilings for 2020 Based on the 2008 Climate and Energy Package, NEC Scenario Analysis Report No. 6. International Institute for Applied Systems Analysis (IIASA), Laxenburg, Austria, http:// www.iiasa. Armstrong, B., Hutchinson, E., Unwin, J. and Fletcher, T. (2004). Lung cancer risk after exposure to polycyclic aromatic hydrocarbons. Environmental Health Perspectives, 112, 970–978. Bell, M., Peng, R. and Dominici, F. (2006). The exposure–response curve for ozone and risk of mortality and the adequacy of current ozone regulations. Environmental Health Perspectives, 114, 532–536. Bell, M. L., Goldberg, R., Hogrefe, C. etÂ€al. (2007). Climate change, ambient ozone, and health in 50 US cities. Climate Change, 82, 61–76. Benton, J., Fuhrer, J., Gimeno, B. etÂ€al. (2000). An international cooperative programme indicates the widespread occurrence of ozone injury on crops. Agriculture, Ecosystems and Environment, 78, 19–30. Butterbach-Bahl, K., Gundersen, P., Ambus, P. etÂ€al. (2011). Nitrogen processes in terrestrial ecosystems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Bylin, G. and Forsberg, B. (2009). Utmhusmiljön: påverkan på allergi och astma. In: Allergi och astma, ed. G. Hedlin and K. Larsson, Studentlitteratur, Lund, Sweden, pp. 159–174. CEC (2005). The Communication on Thematic Strategy on Air Pollution and the Directive on Ambient Air Quality and Cleaner Air for Europe. Impact Assessment. Sec. (2005) 1133. http://ec.europa.eu/governance/impact/ia_carried_out/docs/ ia_2005/sec_2005_1133_en.pdf. Cape, J. N., Van der Eerden, L. J., Sheppard, L. J., Leith, I. D. and Sutton, M. A. (2009). Evidence for changing the critical level for ammonia. Environmental Pollution, 157, 1033–1037. Cofala, J., Amann, M., Heyes, C. etÂ€al. (2007). Final Report:Â€Analysis of Policy Measures to Reduce Ship Emissions in the Context of the Revision of the National Emissions Ceilings Directive, International Institute for Applied Systems Analysis (IIASA),Â€Laxenburg, Austria, www.iiasa.ac.at. de Leeuw, F. and Horálek, J. (2009). Assessment of the Health Impacts of Exposure to PM2.5 at a European Level, ETC/ACC Technical Paper No. 2009/1. Diociaiuti, M., Balduzzi, M., De Berardis, B. etÂ€al. (2001). The two PM(2.5) (fine) and PM(2.5–10) (coarse) fractions:Â€evidence of different biological activity. Environmental Research, 86, 254–262. Dise, N. B., Ashmore, M., Belyazid, S. etÂ€al. (2011). Nitrogen as a threat to European terrestrial biodiversity. In:Â€The European Nitrogen

430

Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Durand, P., Breuer, L., Johnes, P. etÂ€al. (2011). Nitrogen processes in aquatic ecosystems. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. EC (1994). Directive 94/12/EC of the European Parliament and the Council of 23 March 1994 relating to measures to be taken against air pollution by emissions from motor vehicles and amending Directive 70/220/EEC. http://ec.europa.eu/enterprise/sectors/ automotive/documents/directives/motor-vehicles/index_en.htm EC (1996a). Directive 1996/61/EC on Integrated Pollution Prevention and Control (ICCP). http://eur-lex.europa.eu/LexUriServ/ LexUriServ.do?uri= CELEX:31996L0061:EN:NOT EC (1996b). Directive 96/69/EC of the European Parliament and of the Council of 8 October 1996 amending Directive 70/220/EEC on the approximation of the laws of the Member States relating to measures to be taken against air pollution by emissions from motor vehicles. http://ec.europa.eu/enterprise/sectors/automotive/ documents/directives/motor-vehicles/index_en.htm EC (1996c). Directive 96/62/EC on ambient air quality assessment and management (Air Quality Framework Directive). http://ec.europa. eu/environment/air/legis.htm EC (1997a). Directive 97/68/EC of the European Parliament and of the Council of 16 December 1997 on the approximation of the laws of the Member States relating to measures against the emission of gaseous and particulate pollutants from internal combustion engines to be installed in non-road mobile machinery. http:// ec.europa.eu/environment/air/transport/non_road.htm EC (1997b). Council Decision 97/101/EC establishing a reciprocal exchange of information and data from networks and individual stations measuring ambient air pollution within the Member States (EoI Decision). http://ec.europa.eu/environment/ air/legis.htm EC (1998a). Directive 98/69/EC of the European Parliament and of the Council of 13 October 1998 relating to measures to be taken against air pollution by emissions from motor vehicles and amending Council Directive 70/220/EEC. http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EC (1998b). Directive 98/70/EC of the European Parliament and of the Council of 13 October 1998 relating to the quality of petrol and diesel fuels and amending Council Directive 93/12/EEC. http:// ec.europa.eu/environment/air/legis.htm EC (1999a). Directive 1999/96/EC of the European Parliament and of the Council of 13 December 1999 on the approximation of the laws of the Member States relating to measures to be taken against the emission of gaseous and particulate pollutants from compression ignition engines for use in vehicles, and the emission of gaseous pollutants from positive ignition engines fuelled with natural gas or liquefied petroleum gas for use in vehicles and amending Council Directive 88/77/EEC. http://ec.europa.eu/enterprise/ sectors/automotive/documents/directives/motor-vehicles/ index_en.htm EC (1999b) Directive 1999/30/EC relating to limit values for sulphur dioxide, nitrogen dioxide and oxides of nitrogen, particulate matter and lead in ambient air (First Daughter Directive). http://ec.europa.eu/environment/ air/legis.htm EC (2000). Directive 2000/76/EC of the European Parliament and of the Council of 4 December 2000 on the incineration of waste. http://ec.europa.eu/environment/air/legis.htm

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson EC (2001a). Directive 2001/81/EC on National Emission Ceilings (NEC) http://ec.europa.eu/environment/air/legis.htm EC (2001b). Directive 2001/80/EC of the European Parliament and of the Council of 23 October 2001 on the limitation of emissions of certain pollutants into the air from large combustion plants. http:// ec.europa.eu/environment/air/legis.htm EC (2002). Directive 2002/3/EC of the European Parliament and of the Council relating to ozone in ambient air (Third Daughter Directive). http://ec.europa.eu/environment/air/legis.htm EC (2004). Commission Decision 2004/461/EC laying down a questionnaire for annual reporting on ambient air quality assessment under Council Directives 96/62/EC and 1999/30/EC and under Directives 2000/69/EC and 2002/3/EC of the European Parliament and of the Council. http://ec.europa.eu/environment/ air/legis.htm EC (2005a). Directive 2005/55/EC of the European Parliament and of the Council of 28 September 2005 on the approximation of the laws of the Member States relating to the measures to be taken against the emission of gaseous and particulate pollutants from compression-ignition engines for use in vehicles, and the emission of gaseous pollutants from positive-ignition engines fuelled with natural gas or liquefied petroleum gas for use in vehicles. http://ec.europa.eu/enterprise/sectors/automotive/ documents/directives/motor-vehicles/index_en.htm EC (2005b). Directive 2005/78/EC of 14 November 2005 implementing Directive 2005/55/EC of the European Parliament and of the Council on the approximation of the laws of the Member States relating to the measures to be taken against the emission of gaseous and particulate pollutants from compressionignition engines for use in vehicles, and the emission of gaseous pollutants from positive ignition engines fuelled with natural gas or liquefied petroleum gas for use in vehicles and amending Annexes I, II, III, IV and VI thereto. http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EC (2007). Regulation (EC) No 715/2007 of the European Parliament and of the Council of 20 June 2007 on type approval of motor vehicles with respect to emissions from light passenger and commercial vehicles (Euro 5 and Euro 6) and on access to vehicle repair and maintenance information. http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EC (2008a). Directive 2008/1/EC of 15 January 2008 concerning integrated pollution prevention and control. http://ec.europa.eu/environment/air/legis.htm EC (2008b). Commission Regulation (EC) No 692/2008 of 18 July 2008 implementing and amending Regulation (EC) No 715/2007 of the European Parliament and of the Council on type-approval of motor vehicles with respect to emissions from light passenger and commercial vehicles (Euro 5 and Euro 6) and on access to vehicle repair and maintenance information. http://ec.europa. eu/enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EC (2008c). Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe. http://ec.europa.eu/environment/air/legis.htm EC (2009). Regulation (EC) No 595/2009 of the European Parliament and of the Council of 18 June 2009 on type-approval of motor vehicles and engines with respect to emissions from heavy duty vehicles (Euro VI) and on access to vehicle repair and maintenance information and amending Regulation (EC) No 715/2007 and Directive 2007/46/EC and repealing Directives 80/1269/EEC, 2005/55/EC and 2005/78/

EC. http://ec.europa.eu/enterprise/sectors/automotive/documents/ directives/motor-vehicles/index_en.htm ECE (2009). Health risks of air pollution from biomass combustion. Report by the Task Force on Health, ECE/EB.AIR/WG.1/2009/12, 9 July 2009. EEA (2007). Air Pollution in Europe 1990–2004. EEA Technical Report No. 2/2007, EEA, Copenhagen. EEA (2009a). Spatial Assessment of PM10 and Ozone Concentrations in Europe 2005. EEA Technical Report No. 1/2009, EEA, Copenhagen. EEA (2009b). NEC Directive status report 2008. EEA Technical Report No. 11/2009, EEA, Copenhager. EEA (2010a). http://www.eea.europa.eu/data-and-maps/figures/ percentage-of-urban-population-resident-in-areas-wherepollutant-concentrations-are-higher-than-selected-limit-targetvalues-eea-member-countries-1997 EEA (2010b) http://eea.europa. eu/data-and-maps/figures/annualvariation-in-the-ozone-aot40-value-may-july-1996-2006 EEC (1970). Council Directive 70/220/EEC of 20 March 1970 on the approximation of the laws of the Member States relating to measures to be taken against air pollution by gases from positive-ignition engines of motor vehicles. http://ec.europa.eu/enterprise /sectors/automotive/documents/directives/motor-vehicles/ index_en.htm EEC (1988). Council Directive 88/77/EEC of 3 December 1987 on the approximation of the laws of the Member States relating to the measures to be taken against the emission of gaseous pollutants from diesel engines for use in vehicles. http://ec.europa.eu/ enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EEC (1991). Council Directive 91/441/EEC of 26 June 1991 amending Directive 70/220/EEC on the approximation of the laws of the Member States relating to measures to be taken against air pollution by emissions from motor vehicles. http://ec.europa. eu/enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm EEC (1993). Council Directive 93/59/EEC of 28 June 1993 amending Directive 70/220/EEC on the approximation of the laws of the Member States relating to measures to be taken against air pollution by emissions from motor vehicles. http://ec.europa. eu/enterprise/sectors/automotive/documents/directives/motorvehicles/index_en.htm Emberson, L., Ashmore, M. and Murray, F. (2003). Air Pollution Impacts on Crops and Forests: A Global Assessment, Imperial College Press, London. EMEP (2008). Transboundry Particulate Matter in Europe: Status Report 2008. EMEP Report 4/2008. http://tarantula.nilu.no/ projects/ccc/reports/emep4–2008.pdf EMEP (2010). www.emep.int Filleul, L., Rondeau, V., Vandentorren, S. etÂ€al. (2005). Twenty five year mortality and air pollution:Â€results from the French PAARC survey. Occupational and Environmental Medicine, 62, 453–60. Fuhrer, J., Skärby, L. and Ashmore, M. (1997). Critical levels for ozone effects on vegetation in Europe. Environmental Pollution, 97, 91–106. Gauderman, W. J., McConnell, R., Gilliland, F. etÂ€al. (2000). Association between air pollution and lung function growth in southern California children. American Journal of Respiratory and Critical Care Medicine, 162, 1383–1390. Grizzetti, B., Bouraoui, F., Billen, G. etÂ€al. (2011). Nitrogen as a threat to European water quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press.

431

Nitrogen as a threat to European air quality Hinds, W. C. (1999). Aerosol Technology, Properties, Behavior, and Measurement of Airborne Particles, 2nd ed. New York WileyInterscience, 442 pp. Hertel, O., Reis, S. and Ambelas Skjøth, C. (2011). Nitrogen processes in the atmosphere. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Hoek, G., Brunekreef, B., Goldbohm, S., Fischer, P. and van den Brandt, P. (2002). Association between mortality and indicators of traffic-related air pollution in the Netherlands:Â€a cohort study. Lancet, 360, 1203–1209. Holland, M., Kinghorn, S., Emberson, L. etÂ€al. (2006). Development of a framework for probabilistic assessment of the economic losses caused by ozone damage to crops in Europe, CEH Project No. C02309NEW, Department of Environment, Food and Rural Affairs, London. ICP (2010). www.corr-institute.se/ICP-Materials IMO (2006). MARPOL, (consolidated edition). IMO, ISBN 978–92– 801–42167, www.imo.org IMO (2009). Revised MARPOL Annex VI: Regulations for the Prevention of Air Pollution from Ships, 2008. IMO, ISBN 978–92– 801–42433, www.imo.org IPCC (2007). Climate Change 2007:Â€The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, ed. S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor and H. L. Miller, Cambridge University Press. Ito, K., DeLeon, S. F. and Lippmann, M. (2005). Associations between ozone and daily mortality:Â€analysis and meta-analysis. Epidemiology, 16, 446–457. Jenkin, M. E. (2008). Trends in ozone concentration distributions in the UK since 1990:Â€local, regional and global influences. Atmospheric Environment, 42, 5434–5445. Jensen, L. S., Schjoerring, J. K., van der Hoek, K. etÂ€al. (2011). Benefits of nitrogen for food fibre and industrial production. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Jerrett, M., Burnett, R. T., Pope, C. A. III, etÂ€al. (2009). Long-term ozone exposure and mortality. New England Journal of Medicine, 360, 1085–1095. Jonson, J. E., Simpson, D., Fagerli, H. and Solberg, S. (2006). Can we explain the trends in European ozone levels? Atmospheric Chemistry and Physics, 6, 51–66. Katsouyanni, K., Touloumi, G., Samoli, E. etÂ€al. (2001). Confounding and effect modification in the short term effects of ambient particles on total mortality:Â€results from 29 European cities within the APHEA2 project. Epidemiology, 12, 521–531. Klimont, Z., Asman, W. A. H., Bertok, I. etÂ€al. (2007). Measures in Agriculture to Reduce Ammonia Emission. International Institute for Applied Systems Analysis (IIASA), Laxenburg, www.iiasa.ac.at. Levy, J. I., Chemerynski, S. M. and Sarnat, J. A. (2005). Ozone exposure and mortality:Â€an empiric bayes metaregression analysis. Epidemiology, 16, 458–468. Matyssek, R. and Innes, J. (1999). Ozone: a risk factor for trees and forests in Europe? Water, Air and Soil Pollution, 116, 199–226. Mauzerall, D. and Wang, X. (2001). Protecting agricultural crops from the effects of tropospheric ozone exposure:Â€reconciling science and standard setting in the United States, Europe and Asia. Annual Review of Energy and Environment, 26, 237–268.

432

Mills, G. (ed.) (2004). Mapping critical levels for vegetation. In: Mapping Manual 2004, UN/ECE Convention on Long-range Transboundary Air Pollution, Manual on Methodologies and Criteria for Mapping Critical Loads and Levels and Air Pollution Effects, Risks and Trends (UBA ed.), http://icpmapping.org/cms/ zeigeBereich/5/manual-und-downloads.html Mol, W. J. A., van Hooydonk, P. R., and de Leeuw, F. A. A.Â€M. (2009). European Exchange of Monitoring Information and State of the Air Quality in 2007. ETC/ACC Technical paper No.2009/3, http://air-climate.eionet.europa.eu/reports/ ETCACC_TP_2009_3_EoI_AQ_meta_info2007 Mol, W. J. A., van Hooydonk, P. R. and de Leeuw, F. A. A. M. (2010). European Exchange of Monitoring Information and State of the Air Quality in 2008, ETC/ACC Technical Paper No.2010/1, http://air-climate.eionet.europa.eu/reports/ ETCACC_TP_2009_3_EoI_AQ_meta_info2007 Molina, M. J. and Molina, L. T. (2004). Megacities and atmospheric pollution. Journal of the Air and Waste Management Association, 54, 644–680. Morgan, P. B., Mies, T. A., Bollero, G. A., Nelson, R. L. and Long, S. P. (2006). Season-long elevation of ozone concentration to projected 2050 levels under fully open-air conditions substantially decreases the growth and production of soybean. New Phytologist, 170, 333–343. Mudway, I. S., Stenfors, N., Duggan, S. T. etÂ€al. (2004). An in vitro and in vivo investigation of the effects of diesel exhaust on human airway lining fluid antioxidants. Archives of Biochemistry and Biophysics, 423, 12. Nafstad, P., Haheim, L. L., Wisloff, T. etÂ€al. (2004). Urban air pollution and mortality in a cohort of Norwegian men. Environmental Health Perspectives, 112, 610–615. Poppe, C. A., Burnett, R. T., Thun, M. J. etÂ€al. (2002). Lung cancer cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. Journal of the American Medical Association, 287, 1132–1141. Poppe, C. A., Burnett, R. T., Thurston, G. D. etÂ€al. (2004). Cardiovascular mortality and long-term exposure to particulateÂ€matter air pollution:Â€epidemiological evidence of general pathophysiological pathways of disease. Circulation, 109, 71–77. Pozzi, R., De Berardis, B., Paoletti, L. and Guastadisegni, C. (2003). Inflammatory mediators induced by coarse (PM2.5–10) and fine (PM2.5) urban air particles in RAW 264.7 cells. Toxicology, 183, 243–254. Putaud, J. P., Raes, F., van Dingenen, R. etÂ€al. (2004). A European aerosol phenomenology-2:Â€chemical characteristics of particulate matter at kerbside, urban, rural and background sites in Europe. Atmospheric Environment, 38, 2579–2595. Pye, S. and Watkiss, P. (2005). CAFE CBA:Â€Baseline Analysis 2000 to 2020, AEAT Report AEAT/ED51014/Baseline Scenarios, http://www.iiasa.ac.at/ docs/HOTP/Mar05/cafe-cba-baselineresults.pdf. Reiss, R. Anderson, E. L., Cross, C. E. et al. (2007) Evidence of health impacts of sulfate- and nitrate- containing particles in ambient air inhalation. Toxicology, 19, 419–449. Samoli, E., Touloumi, G., Zanobetti, G. etÂ€al. (2003). Investigating the dose–response relation between air pollution and total mortality in the APHEA-2 multicity project. Occupational and Environmental Medicine, 60, 977–982. Seaton, A. and Dennekamp, M. (2003). Hypothesis:Â€ill health associated with low concentrations of nitrogen dioxideÂ€– an effect of ultrafine particles? Thorax, 58, 1012–1015.

Jana Moldanová, Peringe Grennfelt and Åsa Jonsson Simpson, D., Emberson, L., Ashmore, M. and Tuovinen, J. (2007). A comparison of two different approaches for mapping potential ozone damage to vegetation: a model study. Environmental Pollution, 146, 715–725. Sjödin, Å ., Jerksjö, M., Sandström, C. etÂ€al. (2009) Implementering av ARTEMIS Road Model i Sverige, IVL Report No.B1886, www.ivl.se. Skärby, L., Ro-Poulsen, H., Wellburn, F. and Sheppard, L. (1998), Impacts of ozone on forests:Â€a European perspective. New Phytologist, 139, 109–122. Staehelin, J., and Schnadt Poberaj, C. (2008). Long-term tropospheric ozone trends:Â€a critical review. Advances in Global Change Research, 33, 271–282. Stieb, D. M., Judek, S. and Burnett, R. T. (2002). Meta-analysis of time-series studies of air pollution and mortality:Â€effects of gases and particles and the influence of cause of death, age, and season. Journal of the Air and Waste Management Association, 52, 470–484. Sutton, M. A., Reis, S. and Baker, S. M. H. (eds.) (2009). Atmospheric Ammonia: Detecting Emission Changes and Environmental Impacts, Results from a CLRTAP Expert Workshop, Springer, New York. Svirejeva-Hopkins, A., Reis, S., Magid, J. etÂ€al. (2011). Nitrogen flows and fate in urban landscapes. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. TFIAM, CIAM (2007). Review of the Gothenburg Protocol. Taskforce on Integrated Assessment Modelling of the UNECE Convention on Long-range Transboundary Air Pollution (TFIAM) and the Centre for Integrated Assessment Modelling (CIAM). CIAM Report No. 1/2007. The Royal Society (2008). Ground-Level Ozone in the 21st Century: Future Trends, Impacts and Policy Implications, Science Policy Report No. 15/08, The Royal Society, London. www.royalsociety.org. Townsend, A. R., Howarth, R. W., Bazzaz, F. A. etÂ€al. (2003). Human health effects of a changing global nitrogen cycle. Frontiers in Ecology and The Environment, 1, 240–246. UN/ECE (1979). 1979 Convention on Long-Range Transboundary Air Pollution, www.unece.org/env/lrtap/lrtap_h1.htm UN/ECE (1988). Protocol to the 1979 Convention on Long-range Transboundary Air Pollution on the Reduction of Emissions of Nitrogen Oxide or their Transboundary Fluxes by at least 30 per cent, www.unece.org/env/lrtap/ nitr_h1.htm UN/ECE (1999). Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution to Abate Acidification, Eutrophication and Ground-level Ozone, www.unece.org/env/lrtap/multi_h1.htm UN/ECE ICP-Materials (2006). Trends in Corrosion and Air Pollution in 1987–2003, Programme Centre of the International Cooperative Programme on Effects of Air Pollution on Materials, including Historic and Cultural Monuments (ICP Materials), ECE/EB.AIR/WG.1/2006/7. UN/ECE ICP-Materials (2007). Dose–response functions for the multipollutant situation. Water, Soil and Air Pollution Focus, 7, 249–258.

Van Dingenen, R., Dentener, F. J., Raes, F. etÂ€al. (2009). The global impact of ozone on agricultural crop yields under current and future air quality legislation. Atmospheric Environment, 43, 604–618. Velthof, G., Barot, S., Bloem, J. etÂ€al. (2011). Nitrogen as a threat to European soil quality. In:Â€The European Nitrogen Assessment, ed. M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. Volz, A. and Kley, D. (1988). Evaluation of the Montsouris series of ozone measurements made in the 19th century. Nature, 332, 240–242. Voss, M., Baker, A., Bange, H. W. (2011). Nitrogen processes in coastal and marine systems. In:Â€The European Nitrogen Assessment, ed.Â€M. A. Sutton, C. M. Howard, J. W. Erisman etÂ€al., Cambridge University Press. WHO (2000). Air Quality Guidelines for Europe, 2nd ed., WHO Regional Publications, European Series, No. 91., WHO Regional Office for Europe, Copenhagen, www.euro.who.int/__data/assets/ pdf_file/0005/74732/E71922.pdf WHO (2003). Health Aspects of Air Pollution with particulate Matter, Ozone and Nitrogen Dioxide. Report on a WHO Working Group, Bonn, Germany, 13–15 January 2003, WHO Regional Office for Europe, Copenhagen, document EUR/03/5042688, www.euro. who.int/document/e79097.pdf. WHO (2004). Health Aspects of Air Pollution: Answers to Follow-up Questions from CAFE Report on a WHO Working group, Bonn, Germany, 15–16 January 2004, WHO Regional Office for Europe, Copenhagen, document EUR/04/5046026, www.euro.who.int/ document/E82790.pdf. WHO (2005a). Particulate Matter Air Pollution: How It Harms Health. Fact sheet of WHO Regional Office for Europe, Copenhagen, document EURO/04/05 WHO (2005b). Effects of Air Pollution an Children’s Health and Development: A Review of the Evidence. Copenhagen, WHO Regional Office for Europe. http://www.euro.who.int/document/ E86575pdf. WHO (2006). Air Quality Guidelines Global Update 2005, WHO Regional Office for Europe, Copenhagen, www.euro.who.int/__ data/assets/pdf_file/0005/78638/E90038.pdf. WHO (2007). Health Relevance of Particulate Matter from Various Sources, Report on a WHO Workshop, Bonn, Germany, 26–27 March 2007, WHO Regional Office for Europe, Copenhagen, document EUR /07/5067587, www.euro.who.int/__data/assets/ pdf_file/0007/78658/E90672.pdf WHO (2008). Health Risks of Ozone from Long-Range Transboundary Air Pollution, WHO Regional Office for Europe, Copenhagen, www.euro.who.int/__data/assets/pdf_ file/0005/78647/E91843.pdf

433

Chapter

19

Nitrogen as a threat to the European greenhouse balance Lead authors: Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle Contrubuting authors: Gilles Billen, Pascal Boeckx, Jan Willem Erisman, Josette Garnier, Rob Upstill-Goddard, Michael Kreuzer, Oene Oenema, Stefan Reis, Martijn Schaap, David Simpson, Wim de Vries, Wilfried Winiwarter and Mark A. Sutton

Executive summary Nature of the problem • Reactive nitrogen (Nr) is of fundamental importance in biological and chemical processes in the atmosphere–biosphere system, altering the Earth’s climate balance in many ways. These include the direct and indirect emissions of nitrous oxide (N2O), atmospheric Nr deposition and tropospheric ozone formation (O3), both of which alter the biospheric CO2 sink, Nr supply effects on CH4 emissions, and the formation of secondary atmospheric aerosols resulting from the emissions of nitrogen oxides (NOx) and ammonia (NH3). • Human production and release of Nr into the environment is thus expected to have been an important driver of European greenhouse balance. Until now, no assessment has been made of how much of an effect European Nr emissions are having on net warming or cooling.

Approaches • This chapter summarizes current knowledge of the role of Nr for global warming. Particular attention is given to the consequences of atmospheric Nr emissions. The chapter draws on inventory data and review of the literature to assess the contribution of anthropogenic atmospheric Nr emissons to the overall change in radiative forcing (between 1750 and 2005) that can be attributed to activities in Europe. • The use of Nr fertilizers has major additional effects on climate balance by allowing increased crop and feed production and larger populations of livestock and humans, but these indirect effects are not assessed here.

Key findings/state of knowledge • Due to its multiple, complex effects on biospheric and atmospheric processes, the importance of Nr for the European greenhouse gas balance has so far received insufficient attention. • The main warming effects of European anthropogenic Nr emissions are estimated to be from N2O (17 (15–19) mW/m2) and from the reduction in the biospheric CO2 sink by tropospheric O3 (4.4 (2.3–6.6) mW/m2). The main cooling effects are estimated to be from increasing the biospheric CO2 sink by atmospheric Nr deposition at −19 (−30 to −8) mW/m2 and by light scattering effects of Nr containing aerosol (−16.5 (−27.5 to −5.5) mW/m2), in both cases resulting from emissions of NOx and NH3. • The production of O3 from European emissions of NOx is estimated to have a modest warming effect (2.9 (0.3–5.5) mW/m2), which is largely offset by the cooling effect of O3 in reducing the atmospheric lifetime of CH4 (−4.6 (−6.7 to −2.4) mW/m2), giving an uncertain net warming of +1.7 (−6.4 to +3.1) mW/m2). • Overall, including all of these terms, European Nr emissions are estimated to have a net cooling effect, with the uncertainty bounds Â�ranging from a substantial cooling effect to a small warming effect (−15.7 (−46.7 to +15.4) mW/m2).

Major uncertainties/challenges • The largest uncertainties concern the aerosol and Nr fertilization effects, and the estimation of the European contributions within the global context. • Published estimates suggest that the default N2O emission factor of 1% used by IPCC for indirect emissions from soils following Nr deposition is too low by at least a factor of two. • The wider effects of fertilizer Nr, in allowing increased biospheric C cycling, food and feed production and populations of livestock and humans are a major uncertainty. Industrial production of Nr can be considered as having permitted increased overall consumption (of food, feed and fuel) with major net warming effects. These interactions remain to be investigated. The European Nitrogen Assessment, ed. Mark A. Sutton, Clare M. Howard, Jan Willem Erisman, Gilles Billen, Albert Bleeker, Peringe Grennfelt, Hans van Grinsven and Bruna Grizzetti. Published by Cambridge University Press. © Cambridge University Press 2011, with sections © authors/European Union.

434

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

Recommendations • The contribution of anthropogenic alteration of the nitrogen cycle to the radiative balance needs to be specifically accounted for in any greenhouse gas reporting (e.g., UNFCCC). • Although individual components of Nr emissions have cooling effects, there are many opportunities for ‘smart management‘ in linking the N and C cycles. These can help mitigate greenhouse gas emissions, while reducing the other Nr-related environmental threats such as eutrophication, acidification, air quality and human health.

19.1â•‡ Introduction and objectives This chapter aims to characterize how inputs of reactive nitrogen compounds (Nr) to the biosphere have affected the concentration of atmospheric trace substances and particles that are important for the radiative balance of the earth system. Based on our current understanding, and with a specific focus on Europe, this chapter furthermore evaluates how historic, present day and future changes in biospheric Nr inputs have and will feedback on the European greenhouse gas (GHG) balance. By including the additional cooling effects of aerosol, we extend the GHG estimates to assess the overall effect on radiative balance. The pathways of Nr input to the biosphere and how they are influencing atmospheric composition, and thus the radiative balance, are complex. They involve microbiological, plantphysiological, animal-physiological and physico-chemical processes, as well as manure management, industrial processes or atmospheric chemistry (Figure 19.1). Sources of anthropogenic Nr additions to the global biosphere are primarily related to fertilizer production, combustion processes, including the transport sector, or cultivation of leguminous plants (Galloway etÂ€al., 2004). Once Nr has entered the biosphere, it can directly or indirectly affect the radiative balance of the earth by various processes. Direct effects are generally related to the formation of N2O, a greenhouse gas which is approximately 296 times as powerful as CO2 on a 100-yr timescale and per unit of weight (IPPC, 2007). The dominant source of both, natural or anthropogenic emissions is microbial production by nitrification and denitrification (see Butterbach-Bahl etÂ€ al., 2011, Chapter 6, this volume). Indirect effects of Nr additions on the radiative balance involve a multitude of mainly biological processes on the ecosystem scale, but also physicochemical processes in the atmosphere (Figure 19.1), with the most prominent ones being as follows. (a) Changes in ecosystem C fluxes and C sequestration, affecting CO2 exchange. (b) Changes in ruminant and ecosystem CH4 production and consumption. (c) Changes in N2O production and emission. (d) Changes in atmospheric chemistry and specifically nitrogen oxides (NOx), ammonia (NH3) increasing aerosol formation and associated changes in the oxidative capacity of the troposphere with relevant feedbacks on biospheric processes, e.g., tropospheric ozone (O3) and plant growth. The primary effects of Nr inputs (which are easy to understand, but not to quantify) are increased emissions of N trace gases

(N2O, NH3, NOx) to the atmosphere. The processes driving the biosphere-atmosphere exchange of these compounds, such as nitrification and denitrification (N2O and NO) or volatilization (NH3) depend significantly on the availability of Nr in the plant–soil system. Thus, increased Nr inputs to agricultural systems (with livestock farming systems having the highest Nr use intensity in Europe) lead to increased losses of N trace gases (NH3, NOx, N2O) at the site of Nr input. However, following the cascade of nitrogen downwind or downstream into other ecosystems, N2O emissions affect a broader regional scale (Davidson, 2009; Oenema etÂ€al., 2009). Thus, Nr trace gas emissions from natural and semi-natural terrestrial ecosystems, as well as emissions from water bodies, such as lakes, rivers or coastal waters, need to be considered (Figure 19.1). As a macro-nutrient, Nr positively affects photosynthesis and thus, the assimilation of atmospheric CO2 in plant biomass (Figure 19.2) (Liu and Greaver, 2009). Furthermore, Nr can stimulate the growth of the soil microbial community and in particular stimulate low affinity CH4 oxidation in rice paddies or inhibit high affinity CH4 oxidation in upland soils (Bodelier and Laanbroek, 2004; Figure 19.2), i.e., the availability of Nr affects the tendency or strength of binding of the enzyme CH4-monooxygenase for catalyzing the oxidation of CH4 to methanol. Even the process of methane production in anaerobic sediments (but also in the enteric fermentation in ruminants) is affected by the addition of Nr, as the stimulation of plant growth is a positive feedback upon rhizodeposition of C compounds and plant litter production, both of which serve as substrates for methanogenesis. Since Nr is highly mobile within the biosphere, it also affects aquatic ecosystems, e.g., with regard to eutrophication and the biosphere–atmosphere exchange of CO2 and CH4 or by changing the source strength of coastal waters for N2O. Besides these biological processes driving biosphere–Â� atmosphere exchange of CO2, CH4 and N2O (NOx), the importance of NH3 volatilization and industrial processes, as well as soil NOx emissions for particle formation and tropospheric O3 concentrations needs to be considered. Similarly, feedback loops need to be addressed for the quantification of the net effect of Nr on the GHG balance. In the case of O3, not only is it important as a greenhouse gas, but it also has a detrimental effect on plant productivity and, thus on atmospheric CO2 removal by terrestrial ecosystems, which needs to be accounted for. Therefore, to characterize and quantify effects of Nr on the radiative balance at continental to global scales, it is necessary to evaluate the effects of Nr on N2O-, CO2-, CH4-exchange as well as on exchange of NOx- and NH3 and their consequences for aerosol and O3 formation.

435

Nitrogen as a threat to the European greenhouse balance

Atmosphere

–

OH -

CH4

N2O

–

strat. O3

+

+

trop. O3

CO2

Aerosols

+/–

+

CO2

CH4

N2O

CO2 – +

+/– Growth

Nr

– + +/– Growth +/– + Soil R h

CO2

+

–

Fossil Fuel Burning CH4

+/–

+/–

Nr

Fertiliser Production

Growth + +/– Soil R h

Industry & Transport

Nr

+/–

CH4 CO2

Agricultural Ecosystems

CH4

(Semi-)Natural Ecosystems

Wetlands & Reparian Areas

Nr

Sewage Plants

+/– R h

Nr Transport in Rivers & Groundwater Figure 19.1 Effects of reactive nitrogen (Nr) on various biospheric processes in terrestrial and aquatic ecosystems and on atmospheric chemistry. Feedbacks on the production and consumption of atmospheric compounds directly or indirectly affecting the global radiative balance are indicated by arrows, where the thickness of the arrow gives an indication about the relative importance of a particular process. (Black arrows:Â€Nr fluxes; green arrows:Â€effects (in terms of positive or negative feedbacks); Radiatively active compounds red or fluxes (arrows) of them are marked red (blue) if they tend to increase (decrease) the radiative forcing. Dashed arrows:Â€Direct effects from anthropogenic additions of Nr; Black arrows:Â€Nr fluxes; Green arrows:Â€Effects (in terms of positive or negative feedback); Red arrows:Â€compounds that increase the radiative forcing (warming); Blue arrows:Â€compounds that decrease the radiative forcing (cooling).)

This chapter estimates, for the first time, the effect of European nitrogen usage on the climate system, taking into account:Â€ (i) direct emissions of the long-lived GHG nitrous oxide (N2O), (ii) the effect of Nr on the biospheric control of other GHGs, and (iii) the effect of Nr emissions on long-lived (e.g., methane, CH4) and short-lived radiative forcing agents (e.g., O3, particles). Several different metrics are used to quantify the climate effect of a change, e.g., in atmospheric composition or land cover. The most commonly used are the radiative forcing (RF) and the global warming potential (GWP). The RF is the global, annual mean radiative imbalance to the Earth’s climate system caused by human activities. The GWP of a trace gas is defined as the instantaneous mass emission of carbon dioxide that gives the same time-integrated radiative forcing as the instantaneous emission of unit mass of another trace gas (such as CH4 or N2O), when considered over a given time horizon. Thus, this metric is particularly useful in quantifying and comparing the future climate impact that is due to current emissions of longlived GHGs. However, GWP is less suited to quantifying the impact of short-lived agents. These radiative forcing metrics do not account for the climate sensitivity to the forcing. For example, climate sensitivity of radiative forcing due to changes

436

in O3 may not be the same as for RF due to CO2 and the sensitivity might differ geographically (Hansen etÂ€al., 1997; Joshi etÂ€al., 2003). Both GWP and RF are used in this chapter:Â€ the GWP is applied to assess the impact of Nr on current GHG emissions in Europe, and the RF concept to assess the indirect effects on air chemistry and to quantify the integrated effect that European Nr emissions have had on the global climate system.

19.2â•‡ Effects of reactive nitrogen on net N2O exchange

The major driver for changes in atmospheric N2O concentrations is the increased use of Nr fertilizer, which on the one hand allowed humans to dramatically increase global agricultural production and, thus, to feed the current global world population (Erisman etÂ€al., 2008; Jensen etÂ€al., 2011, Chapter 3, this volume), but on the other hand increased Nr availability and thus microbial N2O production. Owing to cascading of applied Nr onto landscape, regional and even global scales following the volatilization of NH3 and NOx, leaching of nitrate to water bodies or erosion processes, fertilizer Nr has also affected the source strength of non-agricultural terrestrial and aquatic

Klaus Butterbach-Bahl, Eiko Nemitz and Sönke Zaehle

CO2 CO2

N 2O

CH4 Photosynthesis (+)

ANPP (+) C allocation

Autotrophic respiration (+)

N input

BNPP (+/–) Root uptake

C supply (+/–) C : N (–)

N toxicity (+) Denitrifying bacteria (+)

Heterotrophic respiration (+/–)

c

i ob

r Ae

– +/

e ( ob al) r c r i M ene (g

CO2

)

Organic matter

Me

An

C substrate supply (+/–)

ae

rob tha ic arc nog ha eni ea c (+)

N2 NH4+

CH4

Aerobic Methanotropic bacteria (–)

N 2O

NO

Anaerobic –

NH2ONO

NO2

NO3–

N2O Nitrifying bacteria (+)

Aerobic

C cycle N cycle SOC

DIC/DON

DIN/DON

+ Positive feedback – Negative feedback

Figure 19.2 Potential mechanisms regulating the terrestrial ecosystem responses of CO2, CH4 and N2O production and consumption to increased availability of Nr (figure adapted from Liu and Greaver, 2009). (ANPP, aboveground net primary productivity; BNPP, belowground net primary productivity; SOC, soil organic carbon; DOC, dissolved organic carbon; DIN, dissolved inorganic nitrogen; DON, dissolved organic nitrogen.)

systems for N2O (Galloway etÂ€al., 2004; Butterbach-Bahl etÂ€al., 2011, Chapter 6, this volume). Also the emission of Nr from industry and combustion (including transport) has directly and indirectly contributed to changes in the global atmospheric N2O source strength. In addition, emissions of NOx from all combustion processes has resulted in huge increases in the atmospheric loading of Nr, with consequences for Nr deposition to terrestrial and aquatic ecosystems and thus also for Nr availability for microbial processes and finally N2O production (Sutton etÂ€al., 2007; Simpson etÂ€al., 2011, Chapter 14, this volume). This section evaluates separately each source category for N2O and how the source strength may have changed with time.

19.2.1â•‡ Direct N2O emissions from agricultural activities N2O emissions from the agricultural sector are mainly related to direct N2O emissions from soils following the application of NrÂ€– either in the form of synthetic fertilizer or in the form of manureÂ€– or from N2O emissions related to livestock production,

specifically during manure storage, livestock grazing or from paddocks. The IPCC (2006) guidelines specifically list all of these sources and provide emission factors for estimating N2O emissions from them. Other top-down or bottom-up approaches have investigated only some of these sources or have amalgamated several sources together. Therefore, we provide a general overview of results for different approaches in Table 19.1. N2O emissions from soils are mostly estimated by using emission factor approaches (EFs), expressing proportionality between N2O efflux and fertilizer N input rate. However, it needs to be noted that these factors have a wide range of uncertainty (Eggleston etÂ€al., 2006), and their use can underestimate the cumulative effect fertilizer N production may have on worldwide N2O formation. While using the IPCC default factor (1% of Nr applied being directly emitted as N2O plus indirect N2O emissions following Nr cascading downwind/ downstream of ecosystems due to volatilization/deposition, leaching/run-off or sewage emissions, see below) may still reflect the average situation at the plot scale, its overall effect on the global emission situation may be underestimated as the increase in atmospheric concentrations is observed to be more

437

438

52

Oenema etÂ€al. (2007)

72.0

59.2

71.1

24.6

Sewage plants and waste disposal

30.7

approx. 10 kg N will lead to an enrichment of the ecosystem with Nr, i.e., increased Nr availability in the soil-plant system. Indications for Nr enrichment are a reduction of the C:N ratio of the litter, forest floor or mineral soil, increased concentrations of nitrate and ammonium in the soil solution (Kristensen etÂ€al., 2004; Mannig etÂ€al., 2008), as well as increased emissions of N2 and Nr trace gases from the soil. Several studies show that Nr deposition and N2O as well NO emissions from forest soils are positively correlated (Brumme and Beese, 1992; Brumme etÂ€ al., 1999; Papen and Butterbach-Bahl, 1999; Van Dijk and Duyzer, 1999; Butterbach-Bahl etÂ€ al., 1997; Pilegaard etÂ€ al., 2006; Skiba etÂ€al., 2006). The observed stimulation of fluxes is mainly attributed to the increased availability of Nr (as NH4+ and NO3−) for the microbial processes of nitrification and denitrification (Rennenberg etÂ€al., 1998; Corré etÂ€al., 1999), i.e., the key microbial processes responsible for N trace gas production in soils. A possible further explanation for increased N2O emissions due to ecosystem Nr enrichment was recently provided by Conen and Neftel (2007). They speculated that increased Nr availability may have reduced N2O reduction in soils via denitrification, i.e., that the ratio of N2O:N2 increases with increasing Nr availability. Since increased Nr deposition also affects nitrate leaching and runoff (Dise etÂ€al., 1998; Borken and Matzner, 2004), indirect N2O emissions from water bodies due to Nr deposition to natural systems also need to be considered. However, a thorough evaluation and quantification of Nr deposition effects on soil N trace gas emissions remains difficult, since environmental conditions, such as meteorology or

soil and plant properties, significantly affect the magnitude, temporal course and composition of the emitted N gases. Having in mind these difficulties, there have been several attempts to estimate the stimulating effect of Nr deposition on N2O emissions from forest soils. Skiba etÂ€al. (2006) used a gradient approach, with measuring sites located in a mixed forest at increasing distances from a poultry farm, i.e., a strong NH3 source. They estimated that >3% of the N deposited to the woodland sites was released as N2O. Butterbach-Bahl etÂ€al. (1998) used a regression type approach, time series of nitrogen deposition throughfall data and continuous N2O and NO emission measurements at the long-term monitoring site at Höglwald Forest for estimating Nr-deposition driven N2O losses. Their estimate is comparable to that in the study by Skiba etÂ€ al. (2006), i.e., 1.4% for coniferous forests and 5.4% for deciduous forest. Also, a literature review by Denier van der Gon and Bleeker (2005) showed that Nr deposition to forests stimulates N2O emissions within the same range; they concluded that the stimulating effect was higher for deciduous forests (5.7% of deposited N is lost as N2O) than for coniferous forests (3.7%). In a scenario study at the EU scale, Kesik etÂ€al. (2005) estimated Nr deposition effects on forest soil N2O emissions by running the biogeochemical model Forest-DNDC either with best estimates of atmospheric Nr deposition or by assuming that Nr deposition was zero. The results indicated that, across Europe, 1.8% of atmospheric Nr deposition was returned to the atmosphere as N2O. All published estimates, therefore, show that the default N2O emission factor of 1% used by IPCC for indirect emissions from soils following N deposition (Mosier etÂ€al., 1998; IPCC, 2006) is most likely too low by at least a factor of two.

19.2.6â•‡ Indirect N2O emissions from riparian areas, rivers and coastal zones Although direct emission from agricultural soils is the dominant process responsible for N2O emission by the agricultural sector, indirect emissions linked to the cascading of agricultural Nr ‘downstream’ from the fields might also play a significant role. Reactive nitrogen inputs to rivers and coastal waters include both natural and anthropogenic components; the latter is dominated by applied fertilizer Nr lost through leaching and runoff, followed by sewage and atmospheric sources (Seitzinger and Kroeze, 1998). Most of the Nr may already be denitrified in riparian areas in the direct vicinity of the sites of Nr application (e.g., arable fields). From a review of available data on N2O emissions from riparian wetlands, Groffman etÂ€al. (2000) concluded that although current data are inadequate to propose a quantitative emission factor for Nr entering riparian areas, these emissions are likely to be significant in many regions. A nitrogen budget of the Seine hydrographical network (Billen etÂ€al., 2001, 2007, 2009), reveals that up to 25%–30% of the Nr input to surface water from agricultural soils is denitrified to N2O and N2 in riparian zones, compared to only 5%–10% in-stream. If the percentage of N2O loss with respect to nitrate denitrified in riparian zones is the same as in the drainage network, N2O

441

Nitrogen as a threat to the European greenhouse balance

emissions from riparian zones would represent about 10% of the estimated total direct N2O emissions from the agricultural soils of the watershed (Garnier etÂ€al., 2009). Also with regard to N2O losses from coastal areas, estimates are highly uncertain. The total flux of Nr and the fraction of fertilizer N reaching coastal waters is both variable and difficult to estimate. Early studies of individual rivers and/or Â�estuaries, targeted both measured and estimated inputs from various land use activities such as Nr in sewage and atmospheric Â�deposition (Billen etÂ€ al., 1985; Larsson etÂ€ al., 1985; Jaworski etÂ€al., 1992; Boynton etÂ€al., 1995; Nixon etÂ€al., 1995; Howarth etÂ€ al., 1996). Other work examined dissolved Nr export in relation to specific watershed characteristics, such as human population and energy use (Cole etÂ€al., 1993), and point and non-point sources (Cole and Caraco, 1998). Another study modelled river and estuarine N2O production globally, using functions of nitrification and denitrification that were related to external N loading rates derived by adapting local/ regional models of watershed environmental parameters to global databases (Seitzinger and Kroeze, 1998). The results indicated (i) that ~ 8% of the Nr input to terrestrial ecosystems is exported as dissolved inorganic nitrogen (DIN) in rivers; (ii) the DIN export to estuaries globally (year 1990) is ~20.8 Tg Nr per yr; (iii) about 1% of the Nr input from fertilizers, atmospheric deposition, and sewage to watersheds is lost as N2O in rivers and estuaries; hence rivers and estuaries might account for 20% of current global anthropogenic N2O emissions and are thus similar in magnitude to previously identified sources such as direct anthropogenic N2O emissions from soils (Seitzinger and Kroeze, 1998). Such model-derived estimates should, however, be approached with caution. First, for denitrification the heterogeneity of microbial ecosystem structure (Rich and Myrold, 2004), oxygen status (Helder and De Vries, 1983) and physicochemical aspects such as sediment porosity and grain size (Garcia-Ruiz etÂ€al., 1998), ambient temperatures, pH and water content (Berounsky and Nixon, 1990), and levels of suspended particulate matter (Owens, 1986) all co-vary to constrain N2O production. Second, the simple linear functions relating DIN loading to N2O in the global-scale models (Seitzinger and Kroeze, 1998; Seitzinger etÂ€ al., 2000) are not well supported by individual studies, which reveal a wide range. For example, previous work in the Humber estuary (UK) implied that ~25% of the terrestrial DIN input converts to N2O via sediment denitrification (Barnes and Owens, 1999), a far larger conversion than the mean of about 0.15% employed in the global-scale models (Seitzinger and Kroeze, 1998; Seitzinger etÂ€ al., 2000). By contrast, a dynamic model of estuarine DIN cycling in the Tyne estuary (UK) showed only 3.9% of the DIN load to be nitrified, in comparison to a value of 60% assumed in the global scale models, with only 0.009% of the DIN load converted to N2O (Rodrigues etÂ€al., 2007). A corresponding value for the Scheldt estuary was 0.17% of nitrified Nr being converted to N2O, much closer to the global scale average (Rodrigues etÂ€al., 2007). Nevertheless, this study concluded that the amount of atmospheric N2O derived from agricultural sources in general, including estuarine transformations of N, might need to be

442

revised downward, consistent with constraints set by atmospheric N2O growth. Although the global models (Seitzinger and Kroeze, 1998; Seitzinger etÂ€ al., 2000) included some major European river/ estuaries in their development, direct comparisons of DIN load to N2O production were not readily available. Given the variability among the few direct comparisons that have been made and ranges of more than an order of magnitude in both DIN inputs and nitrification rates in European estuaries (Rodrigues etÂ€ al., 2007), a meaningful representative mean value for the ratio for DIN input to N2O production for European estuaries appears to remain not well constrained. Seitzinger and Kroeze (1998) presented the first comprehensive estimate of N2O emissions from terrestrial and aquatic sources at the scale of Europe (more specifically the watersheds of North Eastern Atlantic, the Baltic, the Black Sea and the Mediterranean). They calculated N2O emissions from rivers, estuaries and shelf areas as a percentage (0.3 or 3%) of nitrification and denitrification rates in rivers and estuaries which were, in turn, calculated as a function of external inputs of dissolved inorganic nitrogen (DIN). Although the method is rather rough and questionable, it provides a first order of magnitude of the fluxes at European scale:Â€0.23 Tg N2O-N per yr from rivers and 0.11 Tg N2O-N per yr from estuaries and shelves. Based on a compilation of available data of N2O concentration in estuarine and coastal marine waters, Bange (2006) arrived at overall higher figures, based on component estimates of 0.13–0.16 and 0.20–0.41 Tg N2O-N per year for European shelf and estuarine waters, respectively. However the major merit of the approach used by Seitzinger and Kroeze (1998) is to explicitly link N2O emissions from aquatic systems to the nitrogen load reaching surface waters from agricultural sources, as is suggested by the site specific data presented above. Even if one assumes that 50% of emissions are natural, the remaining magnitude of N2O emissions from coastal zones in Europe would still be at least in the same magnitude as indirect emissions from leaching and Nr deposition (Table 19.1).

19.3â•‡ Effects of Nr on net CH4 exchange

Reactive nitrogen availability has been reported to affect both the capacity of upland soils to serve as a sink for atmospheric CH4 and as a source of CH4 emissions from wetlands (Figure 19.2). Furthermore, indirect effects of Nr on animal feed quality, in this case, changes in indigestible carbohydrates and crude protein contents, can also affect ruminant CH4 emissions. These three aspects are briefly discussed here with their potential feedback on the EU radiative balance. A summary of all effects is presented in Table 19.3.

19.3.1â•‡ Net CH4 oxidation by upland soils

Globally, biological methane oxidation is estimated at 17 ± 9Â€Tg CH4-C per yr (Dutaur and Verchot, 2007). By comparison, Boeckx and Van Cleemput (2001) estimated the total EU15 CH4 oxidation (grassland and agricultural land) to be 0.2 Tg C per yr. In general, higher uptake rates than for grasslands and agricultural land are reported for other upland ecosystem

443

The European Nitrogen Assessment: Sources, Effects and Policy Perspectives

Nanotechnology assessment and perspectives

Flavonoids: Biosynthesis, Biological Effects and Dietary Sources

The European Impact Assessment and the Environment

Emissions Trading for Climate Policy: US and European Perspectives

European Union: Power and Policy-Making (European Public Policy)

The Roman Inquisition, the Index and the Jews: Contexts, Sources and Perspectives (Studies in European Judaism)

Urban and Regional Policy and Its Effects

Asbestos: Risk Assessment, Epidemiology, And Health Effects

Nanotechnology: Assessment and Perspectives (Ethics of Science and Technology Assessment)

Drug Policy Harmonization and the European Union

Russian Foreign Policy: Sources and Implications

The Obama Moment: European and American Perspectives

Social Policy. Essential Primary Sources

The European Union and National Industrial Policy

Neighbourhood Effects Research: New Perspectives

Understanding European Foreign Policy

Contemporary European Foreign Policy

European Perspectives on BE

Working Abroad European Perspectives

European Readmission Policy

The World of Nitrogen

The Nitrogen Fix

The Nitrogen Fix

Nitrogen Fix

Internet Policy and Economics: Challenges and Perspectives

Investment, Growth and Employment: Perspectives and Policy

Regional Policy (Spicers European Policy Reports)

Youth Policy (Spicers European Policy Reports)

Assessment Reform in Education: Policy and Practice

The European Nitrogen Assessment: Sources, Effects and Policy Perspectives

Nanotechnology assessment and perspectives

Flavonoids: Biosynthesis, Biological Effects and Dietary Sources

The European Impact Assessment and the Environment

Emissions Trading for Climate Policy: US and European Perspectives

European Union: Power and Policy-Making (European Public Policy)

The Roman Inquisition, the Index and the Jews: Contexts, Sources and Perspectives (Studies in European Judaism)

Urban and Regional Policy and Its Effects

Asbestos: Risk Assessment, Epidemiology, And Health Effects

Nanotechnology: Assessment and Perspectives (Ethics of Science and Technology Assessment)

Drug Policy Harmonization and the European Union

Russian Foreign Policy: Sources and Implications

The Obama Moment: European and American Perspectives

Social Policy. Essential Primary Sources

The European Union and National Industrial Policy

Neighbourhood Effects Research: New Perspectives

Understanding European Foreign Policy

Contemporary European Foreign Policy

European Perspectives on BE

Working Abroad European Perspectives

European Readmission Policy

The World of Nitrogen

The Nitrogen Fix

The Nitrogen Fix

Nitrogen Fix

Internet Policy and Economics: Challenges and Perspectives

Investment, Growth and Employment: Perspectives and Policy

Regional Policy (Spicers European Policy Reports)

Youth Policy (Spicers European Policy Reports)

Assessment Reform in Education: Policy and Practice

Recommend Documents