Groundwater Science and Policy: An International Overview

Groundwater Science and Policy An International Overview Groundwater Science and Policy An International Overview Ed...

Author: P. Quevauviller

54 downloads 2580 Views 72MB 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

Groundwater Science and Policy An International Overview

Groundwater Science and Policy An International Overview

Edited by Philippe Quevauviller European Commission, DG Environment, Brussels, Belgium

ISBN: 978-085404-294-4 A catalogue record for this book is available from the British Library r The Royal Society of Chemistry 2008 All rights reserved Apart from fair dealing for the purposes of research for non-commercial purposes or for private study, criticism or review, as permitted under the Copyright, Designs and Patents Act 1988 and the Copyright and Related Rights Regulations 2003, this publication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission in writing of The Royal Society of Chemistry, or in the case of reproduction in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK, or in accordance with the terms of the licences issued by the appropriate Reproduction Rights Organization outside the UK. Enquiries concerning reproduction outside the terms stated here should be sent to The Royal Society of Chemistry at the address printed on this page. Published by The Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge CB4 0WF, UK Registered Charity Number 207890 For further information see our web site at www.rsc.org

Foreword One of the major questions that faces the world at the beginning of this century is the threat to our water resources, both in terms of quantity and of quality. There are today 6.2 billion inhabitants on Earth, 14% of whom are already suffering from hunger, a number that has unfortunately been increasing for the last five years, and 20% of whom lack an adequate drinking water supply. In 2050, there will most likely be 9 billion people. To provide food and an adequate domestic water supply to that many people, with their evolving diets, we will need roughly 75% more water than we use todayw, mostly for agriculture, at the same time as we will need to safeguard a biodiverse and healthy series of natural ecosystems, indispensable for the ecological equilibrium of the planet. In addition, our current urban life, our industrial activity and our agricultural practices increasingly generate sources of contaminants that are spread in the atmosphere, in surface water, on soils and in groundwater, threatening water quality. Finally, these threats must be examined in the context of climate variability, which most likely will be enhanced by impending climate changes, as indicated by the recent Fourth Assessment Report of the Intergovernmental Panel on Climate Change released on 29 January 2007z. Not only will more water be needed, but it will have to be available even during droughts. The world has always experienced great climate variability, such as the seven years of fat cows and lean cows in the Bible. For instance, some archaeological studies conducted simultaneously in Greece and China seem to show that a major drought occurred in these two countries around AD 400; in 1876–1878, and in 1898– 1900, there were severe droughts simultaneously in Brazil, China, India and Ethiopia, causing dramatic faminesy. In 1998, following a strong El Ninõ event, there were large deficits in the grain production in China and Indonesia. These two countries were able to import food from the world stocks, and no major consequences were felt, but the global food stocks fell to a very low level. It is w

See, for example, Les Eaux Continentales, coordinated by G. de Marsily, EDP Sciences, Paris, 2006; M. Griffon, Nourrir la Plane`te, Odile Jacob, Paris, 2006; International Water Management Institute, Water for Food, Water For Life: The Comprehensive Assessment of Water Management in Agriculture, Colombo, Sri Lanka, report to be published 2007. z Intergovernmental Panel on Climate Change, Fourth Assessment Report, 2007. y See, for example, A. Sen and J. Dre`ze, Omnibus, Oxford University Press, New Delhi, 1999; M. Davis, Late Victorian Holocausts, El Ninõ Famines and the Making of the Third World, Verso, London, 2001 (also available in French: Geńocides Tropicaux, La Dećouverte, Paris, 2003 and 2006).

v

vi

Foreword

most likely that such events will occur again and severely affect water resources simultaneously across different continents. As emphasised by the Stern Report released on 30 October 2006z, taking action now to combat climate change is of utmost importance, if severe economical and social crises in the course of this century are to be avoided. But this applies also to the protection of the environment and the management of water resources, as shown, for example, by the Millennium Ecosystem Assessment Report8. This book addresses one of the central issues of these concerns: the science and policy of groundwater resource management. Groundwater is and will indeed be a major world resource for both irrigation and domestic use. Aquifers supply, in fact, about one-fourth of the flow of all the rivers in the world, about 90% of it in the low-flow season, about 75% of the drinking water supply in Europe, about 50% worldwide and a majority of the irrigation water in the world; aquifers are also the major natural means of storing water during wet years and make it available during droughts. Finally, almost 20% of the world freshwater reserves are contained in the aquifers, while surface waters only contribute 0.5%; the remaining 80% are the continental ice sheets, not readily usable. Managing and protecting our groundwater resources is thus a very urgent and important task. By contrast, aquifers, because they are not visible and their functioning in general poorly understood, are currently very poorly managed worldwide, or rather not managed at all, and often exploited at the highest possible level without a thought spared for their protection. In fact, very little time is left to learn how our aquifer systems operate, what their reserves are and how to protect and to manage them in order to be able to meet tomorrow’s challenges. This book is thus a timely contribution to this endeavour. With the 2000 Water Framework Directive**, the European Community took the lead in Europe in addressing the issue of re-establishing the ecological and chemical quality of our continental waters, and with the new daughter directive on groundwaterww, adopted on 12 December 2006, the European Community is again setting out the framework for the management of groundwater in Europe. Philippe Quevauviller, who was the leader at the European Commission for the development of the daughter directive on groundwater, was in an ideal position to assemble the necessary multidisciplinary team of specialists who have contributed to this book. Groundwater is indeed a topic that requires the interaction of many disciplines for its management: geologists, hydrologists, geochemists, geobiologists, agronomists and farming experts, health specialists, z

www.sternreview.org.uk. Millennium Ecosystem Assessment, Ecosystems and Human Well-being: Synthesis, Island Press, World Resources Institute, Washington, DC, 2005 (www.maweb.org). ** Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy, Official Journal of the European Communities, L 327, 22.12.2000. ww Directive 2006/118/EC of the European Parliament and of the Council on the protection of groundwater against pollution and deterioration, L 372, 12.12.2006. 8

Foreword

vii

economists, law-makers, social scientists, etc. About 70 such experts from 13 European countries have contributed to this book, with one author from the US Environmental Protection Agency giving some perspectives on groundwater protection in the USA. With eleven different sections, the book in a coordinated fashion covers most of the topics of importance in the management of groundwater, both from the technical side and the management aspect. Bringing together often opposite views of aquifer management, that of the scientist and that of the policy maker, is a rare achievement, for which the authors and the editor must be congratulated. Science–policy integration, regulatory frameworks and stakeholder involvement are covered at the same time as groundwater characterisation, monitoring, risk assessment, remediation, modelling and management at the basin scale. The book ends with a remarkable concluding section on further policy and research objectives that need to be addressed in the coming few years, in order to put Europe at the forefront of groundwater resource management, and to meet the social, economic and ecological challenges of our water supply in the 21st century. Ghislain de Marsily French Academy of Sciences and University Paris VI, Paris, France

Preface Having been educated in geosciences, it was somehow logical that my career path would, after 25 years, bring me back to a scientific sector which opened my eyes as a researcher. Even if my actual research activities primarily focused on environmental analytical developments and applications, and later on quality assurance matters, I never lost interest in geochemistry and geology, and this is certainly what decided me to move from science to policy when the Environment Directorate-General of the European Commission searched for an officer who would develop a new groundwater directive responding to the requirements of the Water Framework Directive (WFD) 2000/60/ECw. The new Groundwater Directive (2006/118/EC)z was adopted on 12 December 2006 and it opens a new era for groundwater protection. Policy developments are also flourishing in other parts of the world as illustrated by the new groundwater rules also adopted in December 2006 by the US Environmental Protection Agencyy, and the recently published FAO legislative study on groundwater in international lawz. These regulations obviously represent progress, but we should not overlook that groundwater data are still very scarce in comparison to data gathered in the surface water sector over the last 40 years. Groundwater is a ‘‘hidden’’ resource which has essentially been monitored in the light of its uses, mainly for human consumption, over the past decades. It is only recently that the environmental value of groundwater has been put forward as a key issue, and this has been reflected in the orientations of the European Union (EU) legislation. Much work, therefore, remains ahead of us to get a better appraisal of risks affecting the qualitative and quantitative status of groundwater, perform representative monitoring programmes and establish management plans that will enable measures to be identified and operated for the sake of prevention of deterioration and enhancement of groundwater quality and quantity. This is specifically the purpose of the WFD and its ‘‘daughter’’ Groundwater Directive. This implementation work, however, will only be efficient if it is backed up by the best of scientific and technological state-of-the-art. One would think, therefore, that it would then be natural that scientific experts work hand-inhand with policy-makers to identify the most appropriate tools and methods w

Official Journal of the European Communities, L 327, 20.12.2000, p. 1. Official Journal of the European Communities, L 372, 12.12.2006, p. 19. See Chapter 3.2 of this book. z FAO Legislative Study no. 86, 2005 (ISSN 1014-6679). z y

ix

x

Preface

that would best serve policy implementation. This is happening to a certain extent, but the situation to date is far from being satisfactory. In other words, the dialogue and interactions between the scientific and policy-making communities are not as straightforward as one could expect. In the sector of groundwater, a bridge between these two communities has developed within the last five years, and greatly contributed to EU policy design and development. It is now time to put sciences and technologies at the service of policy implementation, and this is reflected by the content of this RSC book Groundwater Science and Policy. This book has been written by internationally recognised experts who have gathered experiences in policy or research developments, in particular in the framework of projects funded by the EU Framework Programme for Research and Technological Developments. It represents a unique experience of operational links among the policy and science worlds. Philippe Quevauviller

Contents xxxiii

List of Contributors 1. General Introduction Chapter 1

General Introduction: The Need to Protect Groundwater Philippe Quevauviller 1.1 1.2

Introduction The Scientific Background 1.2.1 The Hydrogeological Cycle 1.2.2 Waters in Aquifers 1.2.3 Groundwater Flows 1.2.4 Groundwater Quality 1.3 Groundwater Deterioration Risks 1.3.1 Quantitative Aspects 1.3.2 Links to Associated Ecosystems 1.3.3 Groundwater Pollution 1.4 Groundwater Risk Assessment: Implications for Policy 1.4.1 Groundwater Protection Needs 1.4.2 Assessment, Prevention and Control 1.4.3 Monitoring 1.5 Conclusions References

3 4 4 5 6 6 7 7 9 10 13 13 14 16 18 18

2. Science–Policy Integration Needs Chapter 2.1 Science–Policy Integration for Common Approaches Linked to Groundwater Management in Europe Philippe Quevauviller 2.1.1

Introductory Remarks on Science–Policy Integration Needs

xi

21

xii

Contents

2.1.2

Science Integration in the Light of Groundwater Management 2.1.3 Examples of Projects in Support of Groundwater Policy 2.1.3.1 Risk Assessment 2.1.3.2 Groundwater Remediation 2.1.3.3 Diffuse Pollution 2.1.3.4 Management Issues and Information Tools 2.1.3.5 Relevant Networks 2.1.4 A Project ‘‘Tailor-Made’’ to Support the New EU Groundwater Directive: BRIDGE 2.1.5 Conclusions: Some Research Needs References

24 25 25 26 26 26 27 27 28 29

Chapter 2.2 Transferring Scientific Knowledge to Societal Use: Clue from the AQUATERRA Integrated Project Philippe Ne´grel, Dominique Darmendrail and Adriaan Slob 2.2.1 2.2.2 2.2.3

Introduction Overview of The AQUATERRA Project Environmental Policies 2.2.3.1 Four Generations of Environmental Policies 2.2.3.2 Science in the Four Generations of Environmental Policies 2.2.4 AQUATERRA Case Studies 2.2.4.1 The Ebro Case Study 2.2.4.2 The Meuse Case Study 2.2.5 Discussions of the Contributions of AQUATERRA to Societal Use Related to Groundwater Resources Management 2.2.5.1 Stakeholder Involvement 2.2.5.2 Learning Approaches in Policy-Making 2.2.6 Conclusions References Chapter 2.3

31 33 35 35 36 38 38 43

51 54 55 56 56

Groundwater Management and Planning: How Can Economics Help? Jean-Daniel Rinaudo and Pierre Strosser 2.3.1 2.3.2

Introduction Economic Methodologies and Tools: Four Possible Ways of Supporting Groundwater Management and Planning 2.3.2.1 Economic Characterisation of Water Uses

58

60 60

xiii

Contents

2.3.2.2 2.3.2.3

Methods to Assess Environmental Costs Economic Methods for Appraisal of Groundwater Projects and Policies: Cost-effectiveness and Cost–Benefit Analysis 2.3.2.4 Economic Behavioural Models 2.3.2.5 Selected Illustrations 2.3.3 Assessing and Simulating Current and Future Socio-Economic Impact of Groundwater Deterioration 2.3.4 Cost Effectiveness Analysis of Groundwater Protection Measures: Finding the Least Costly Way to Reduce Nitrate Pollution to Groundwater 2.3.5 Cost–Benefit Analysis of Groundwater Protection: Finding the Economically Optimal Level of Groundwater Protection 2.3.6 Integrating Economic and Groundwater Models for Simulating Nitrate Pollution in the Upper Rhine Valley Aquifer 2.3.7 Designing Economic Instruments for Groundwater Management 2.3.7.1 Environment Taxes and Charges 2.3.7.2 Tradable Groundwater Licences and Rights 2.3.8 Conclusions References

61

62 63 64

64

66

69

71 73 74 76 77 79

3. Groundwater Regulatory Framework Chapter 3.1 European Union Groundwater Policy Philippe Quevauviller 3.1.1 3.1.2 3.1.3 3.1.4 3.1.5 3.1.6 3.1.7

Introduction The 1980 Groundwater Policy Framework Preliminary Assessment (1982) The Groundwater Action Programme (1996) The Groundwater Policy Framework Under the WFD The New Groundwater Directive 2006/118/EC Policy Integration 3.1.7.1 Nitrates Directive 3.1.7.2 Urban Wastewater Treatment Directive 3.1.7.3 Plant Protection Products Directive 3.1.7.4 Biocides Directive 3.1.7.5 IPPC Directive 3.1.7.6 Landfill Directive 3.1.7.7 Sewage Sludge Directive 3.1.7.8 Other Directives

85 86 88 89 91 97 98 98 99 100 101 101 102 103 103

xiv

Contents

3.1.8 Conclusions: The Need for Worldwide Cooperation References

104 105

Chapter 3.2 US Drinking Water Regulation: The Ground Water Rule Crystal Rodgers-Jenkins 3.2.1 3.2.2 3.2.3

Introduction Federal Statutory Authority Challenges in Developing the GWR 3.2.3.1 US Groundwater System Demographics 3.2.3.2 Occurrence Data 3.2.3.3 Public Health Risks 3.2.4 The Risk-Targeted Approach 3.2.5 Conclusions Acknowledgements References

107 108 108 109 110 112 113 116 116 116

4. Stakeholder Interactions Chapter 4.1 Principles of the Common Implementation Strategy of the WFD: The Groundwater Woking Group Philippe Quevauviller, Johannes Grath and Andreas Scheidleder 4.1.1

The Need for Multi-stakeholder Involvement in the Environmental Policy Development and Implementation Process 4.1.2 The WFD Common Implementation Strategy 4.1.2.1 General Principles 4.1.2.2 Supporting Activity: The Pilot River Basin Network 4.1.3 The CIS Working Group on Groundwater 4.1.3.1 Objectives 4.1.3.2 Leadership and Network 4.1.3.3 Achievements from 2003 to 2006 4.1.3.4 Perspectives for 2007–2009 4.1.4 Perspectives References

121 122 122 123 123 123 124 124 125 127 127

Chapter 4.2 The Pilot River Basin Network: Examples of Groundwater-related Activities Lorenzo Galbiati and Giovanni Bidoglio 4.2.1 4.2.2

Introduction Science–Policy Integration in the PRB Exercise Linked to Groundwater Management

128 129

xv

Contents

4.2.3

Managing Groundwater Bodies in the Shannon PRB for the Implementation of the WFD 4.2.3.1 Groundwater Body Delineation 4.2.3.2 Groundwater Management for the WFD 4.2.3.3 Example of Risk Assessment Methodology for Diffuse Groundwater Pollution in the Shannon PRB 4.2.4 Groundwater Natural Background Levels and Threshold Definition in the Tevere PRBs Under the BRIDGE Project 4.2.4.1 Tevere PRB: The Colli Albani Case Study 4.2.4.2 Groundwater Status Evaluation by Threshold Values 4.2.4.3 Case Study of Salone-Acque Vergini System 4.2.4.4 Case Study of the Protected Area of Castelporziano 4.2.5 Conclusions References

130 131 131

131

133 134 135 136 136 140 141

Chapter 4.3 The Harmoni-CA Initiative Geo E. Arnold, Wim J. de Lange and Michiel W. Blind 4.3.1 4.3.2 4.3.3

Introduction The Harmoni-CA Initiative Process of Bridging the Gap Between Research and Policy/Water Management 4.3.3.1 Harmoni-CA Forums and Conferences 4.3.3.2 CatchMod/Harmoni-CA Workshops 4.3.3.3 Conclusions and Lessons Learned from the Conferences and Workshops 4.3.4 Products/Tools of Harmoni-CA 4.3.4.1 WISE-RTD Web Portal 4.3.4.2 Guidance Documents, Synthesis Reports and Summaries 4.3.5 SPI-Water 4.3.6 Groundwater Directive References

142 143 143 144 145 145 146 146 147 148 149 149

Chapter 4.4 Linking Public Participation to Adaptive Management Claudia Pahl-Wostl, Jens Newig and Dagmar Ridder 4.4.1 4.4.2 4.4.3

Introduction Adaptive Management and Public Participation Rationales and Requirements for Effective Participation in Groundwater Management

150 151 155

xvi

Contents

4.4.3.1

Rationales and Goals for Public Participation 4.4.3.2 Public Participation Provisions in the WFD 4.4.3.3 German Experiences with Public Participation 4.4.3.4 Example: Regional Participation in Groundwater Protection from Agricultural Nitrate 4.4.4 Social Learning in Public Participation: Support for Adaptive Management 4.4.4.1 Appropriate Framing Conditions 4.4.4.2 Well-Designed Process Management 4.4.4.3 Well-Selected Methods and Tools 4.4.4.4 Leadership Issues 4.4.4.5 An Example of Participation-based Measures in Groundwater Management 4.4.5 Conclusions References

155 157 158

159 161 162 163 164 165 167 169 170

5. Groundwater Characterization and Risk Assessment Chapter 5.1 Groundwater Characterization and Risk Assessment in the Context of the EU Water Framework Directive Andreas Scheidleder, Johannes Grath and Philippe Quevauviller 5.1.1 5.1.2 5.1.3 5.1.4 5.1.5

5.1.6 5.1.7 5.1.8 5.1.9

5.1.10

Legal Background Groundwater Body Identification and Delineation Initial Characterisation Further Characterisation Additional Requirements of the WFD 5.1.5.1 Transboundary Groundwater Bodies 5.1.5.2 Groundwater Bodies with Lower Objectives 5.1.5.3 Interaction with Aquatic and Terrestrial Ecosystems Conceptual Model/Understanding Identification of Driving Forces and Pressures Identification of Significant Pressures Assessing the Impacts of Pressures 5.1.9.1 Tools to Assist 5.1.9.2 Scaling Issues Evaluating the Likelihood of Failing to Meet the Objectives

177 180 182 183 184 184 184 184 184 186 187 188 188 189 190

xvii

Contents

5.1.11 Reporting on the Characterisation and Risk Assessment References

191 192

Chapter 5.2 Groundwater Quality Background Levels Emilio Custodio and Marisol Manzano 5.2.1 5.2.2

Introduction Rationale to Establish the Groundwater Quality Baseline 5.2.3 Methods to Establish the Natural Baseline Quality of Groundwater 5.2.3.1 Study of Major and Trace Inorganic Component Chemistry 5.2.3.2 Organic Component Chemistry 5.2.3.3 Hydrogeochemical Modelling 5.2.3.4 Tracers and Temporal Scales 5.2.3.5 Study of Natural Baseline Trends 5.2.4 Conclusions Acknowledgements References

193 195 198 200 202 206 208 211 214 215 215

Chapter 5.3 Groundwater Age and Quality Klaus Hinsby, Roland Purtschert and W. Mike Edmunds 5.3.1 5.3.2

Introduction Groundwater Age Estimation 5.3.2.1 The Definition of Groundwater Age 5.3.2.2 Environmental Tracers for Absolute Age Estimation 5.3.2.3 Geoindicators: Estimating Environmental Change and Relative Ages 5.3.2.4 Numerical Modelling of Groundwater Age 5.3.3 Groundwater Age and Water Quality and Quantity Issues 5.3.3.1 Groundwater Quality as a Function of Age 5.3.3.2 Groundwater Age and Monitoring 5.3.3.3 Groundwater Age and Water (Over)exploitation 5.3.4 Groundwater Age and the Water Framework and Groundwater Directives 5.3.4.1 Groundwater Age and Derivation of Natural Background Levels and Threshold Values

217 219 219 219 223 225 226 226 227 228 228

228

xviii

Contents

5.3.4.2

Groundwater Interaction with Dependent Ecosystems 5.3.5 Case Studies 5.3.5.1 Examples of Danish Case Studies 5.3.5.2 Examples of Swiss and German Case Studies 5.3.5.3 Example of a British Case Study 5.3.6 Conclusions References

228 229 229 231 234 235 236

Chapter 5.4 Characterisation of Groundwater Contamination and Natural Attenuation Potential at Multiple Scales Thomas Ptak and Jerker Jarsjo¨ 5.4.1 5.4.2

Introduction The Integral Groundwater Investigation Method 5.4.2.1 Concepts and Principles 5.4.2.2 The Inversion Problem 5.4.2.3 Application of the Integral Investigation Method 5.4.3 Methodology to Consider Aquifer Parameter Uncertainty and to Delimit Contaminant Source Zones Using Integral Measurements 5.4.3.1 Principles 5.4.3.2 Decision Tree Approach 5.4.3.3 Example of Application at an Industrial Site 5.4.4 Quantification of Natural Attenuation Rates Using Integral Measurements 5.4.4.1 Principles 5.4.4.2 Example of Application at a Former Gasworks Site 5.4.5 Multilevel Integral Investigation of Contamination 5.4.5.1 The Multilevel Integral Investigation Method 5.4.5.2 Example of Application 5.4.6 Conclusions Acknowledgements References

240 241 242 244 247

249 249 255 258 259 259 260 262 262 263 265 266 267

xix

Contents

Chapter 5.5 Improved Risk Assessment of Contaminant Spreading in Fractured Underground Reservoirs Christos D. Tsakiroglou 5.5.1

Introduction 5.5.1.1 Literature Review 5.5.2 Objectives and Approach of the TRACE-Fracture Project 5.5.3 Description of Ringe Site 5.5.3.1 Conceptual Fracture Network Model 5.5.4 Hierarchical Methods for the Determination of Transport Properties 5.5.4.1 Multiphase Transport Coefficients of Single Fractures and Fracture Networks 5.5.5 Numerical Modelling of NAPL Fate in Unsaturated and Saturated Zones 5.5.6 Risk Assessment of Contaminated Sites 5.5.6.1 Site Remediation 5.5.6.2 In Situ Stimulation/Remediation of Contaminated Sites 5.5.7 Socioeconomic Relevance and Policy Implications Acknowledgements References

269 270 271 272 273 275 275 278 281 282 286 287 288 289

Chapter 5.6 Groundwater Risk Assessment at Contaminated Sites (GRACOS): Test Methods and Modelling Approaches Peter Grathwohl and Hans van der Sloot 5.6.1 5.6.2

Introduction Leaching Tests (Heavy Metals, Low-Volatility Organic Compounds) 5.6.2.1 Total Composition vs. Aqueous Concentrations 5.6.2.2 Percolation vs. Batch or Shaking Tests: Comparison to Field 5.6.2.3 Boundary Conditions Leading to Changing Release Rates 5.6.2.4 Release Kinetics 5.6.2.5 Column Tests: Standardisation and Design 5.6.2.6 Concluding Remarks for Leaching Tests 5.6.3 Groundwater Risk Assessment for Volatile Compounds 5.6.3.1 Vapour-phase Diffusion

291 293 293 294 299 299 302 304 305 305

xx

Contents

5.6.3.2

Coupled Models for Simulation of Field Sites 5.6.3.3 Concluding Remarks for Risk Assessment for Volatile Compounds 5.6.4 Modelling for Groundwater Risk Assessment of Inorganic Constituents 5.6.5 Conclusions/Recommendations Acknowledgement References

306 307 309 312 313 313

Chapter 5.7 INCORE: Integrated Concept for Groundwater Remediation Thomas Ertel and Hermann J. Kirchholtes 5.7.1 5.7.2 5.7.3

Motivation and Basic Concept Cycle I: Plume Screening Cycle II: Source Identification 5.7.3.1 GC-MS Fingerprinting for Petroleum Hydrocarbons 5.7.3.2 Isotopic Fingerprinting for Chlorinated Hydrocarbons 5.7.4 Cycle III: Remediation Strategy 5.7.4.1 Remediation Scenarios 5.7.4.2 ISIRE: In situ Remediation Technologies— Decision Support 5.7.5 Implementation 5.7.5.1 Administrative Aspects 5.7.5.2 Cost Considerations 5.7.5.3 Implementation in Projects References

316 319 320 320 329 333 333 335 337 337 337 340 340

6. Groundwater Monitoring Chapter 6.1 Groundwater Monitoring in the Policy Context Johannes Grath, Rob Ward and Andreas Scheidleder 6.1.1 6.1.2

Introduction Monitoring Requirements: Legal Background and Objectives 6.1.3 General Principles 6.1.3.1 Conceptual Model 6.1.3.2 Three-dimensional Characteristics and Variability 6.1.3.3 Aquifer Types 6.1.4 Chemical Status and Trend Monitoring

345 347 349 349 349 351 352

xxi

Contents

6.1.4.1 Overall Objectives 6.1.4.2 Surveillance Monitoring Programme 6.1.4.3 Operational Monitoring Programme 6.1.4.4 Selection of Representative Monitoring Sites 6.1.4.5 Selection of Monitoring Determinands 6.1.4.6 Monitoring Frequency 6.1.5 Quantity (Water Level) Monitoring 6.1.5.1 Overall Objective 6.1.5.2 Monitoring Parameters 6.1.5.3 Selection of Monitoring Density 6.1.5.4 Monitoring Frequency 6.1.6 Review and Update References

352 352 353 353 356 356 359 359 360 361 361 362 362

Chapter 6.2 Screening Methods for Groundwater Monitoring Catherine Gonzalez, Anne-Marie Fouillac and Richard Greenwood 6.2.1

Groundwater Monitoring Requirements and Specific Issues 6.2.2 Environmental Variability: Spatial and Temporal Groundwater Quality Variability 6.2.3 Screening Methods Towards Groundwater Monitoring Needs 6.2.3.1 Common Physicochemical Methods used for Groundwater Assessment 6.2.3.2 Emerging Screening Tools 6.2.3.3 Potential Uses of Emerging Tools 6.2.4 Screening Methods and Priority Substances 6.2.5 New Trends and Perspectives References

363 366 368 368 369 371 374 375 376

Chapter 6.3 Quality Assurance for Groundwater Monitoring Philippe Quevauviller and Ste´phane Roy 6.3.1 6.3.2 6.3.3 6.3.4 6.3.5

Need for Quality Assurance for Groundwater Analysis Within-Laboratory Quality Measures Statistical Control Comparison of Analytical Methods Interlaboratory Studies 6.3.5.1 Introduction 6.3.5.2 Scope of the Groundwater Interlaboratory Programme

378 379 379 380 380 380 381

xxii

Contents

6.3.6

Certified Reference Materials 6.3.6.1 General Principles 6.3.6.2 CRMs for Major Elements in Groundwater 6.3.6.3 CRMs for Trace Elements in Groundwater 6.3.6.4 CRMs for Bromide in Groundwater 6.3.6.5 Conclusions: Availability of Water CRMs 6.3.7 Assessment of Uncertainty Linked to Groundwater Sampling. A Case Study: The METREAU Project 6.3.7.1 Introduction 6.3.7.2 Groundwater Sampling: New Devices 6.3.7.3 Uncertainties Associated with the Sampling Stage 6.3.7.4 Conclusions Acknowledgements References

383 383 384 387 390 391 392 392 393 397 399 400 401

7. Groundwater Pollution Prevention and Remediation Chapter 7.1 Prevention and Reduction of Groundwater Pollution at Contaminated Megasites: Integrated Management Strategy and its Application on Megasite Cases Jeroen Ter Meer, Hans Van Duijne, Rob Nieuwenhuis and Huub Rijnaarts 7.1.1

Addressing Groundwater on Large (Former) Industrial Sites 7.1.2 Risk-based Approach for Contaminated Megasites 7.1.2.1 Integrated Management Strategy 7.1.2.2 Megasite Objectives 7.1.2.3 Concept for Megasites 7.1.2.4 Megasite Categories 7.1.2.5 Modelling and Monitoring 7.1.2.6 Risk Management Scenarios 7.1.3 Relevance of Risk-based Management of Megasites for the Groundwater Directive 7.1.3.1 Case Studies 7.1.4 Conclusions References

405 406 406 407 407 408 410 410 412 412 420 420

Chapter 7.2 Forecasting Natural Attenuation as a Risk-based Groundwater Remediation Strategy Ryan D. Wilson, Steven F. Thornton and David N. Lerner 7.2.1

The Nature of Groundwater Pollution from Point Sources

421

xxiii

Contents

7.2.1.1 7.2.1.2 7.2.1.3 7.2.2 Natural 7.2.2.1 7.2.2.2 7.2.2.3 7.2.2.4 7.2.2.5 7.2.3 Current 7.2.3.1

Sources and Plumes Risk Assessment: Is Clean-up Required? Options for Remediation Attenuation of Contaminants Why Natural Attenuation is Important Contributing Processes Plume Development and Life Phases Biodegradation in Detail Hydrogeology and Heterogeneity Natural Attenuation Assessment Practices Establishing Whether Natural Attenuation is Occurring 7.2.3.2 Establishing Whether Natural Attenuation is Sufficient 7.2.3.3 Steps in Natural Attenuation Assessment 7.2.4 CORONA: A New Natural Attenuation Assessment Philosophy 7.2.4.1 Core and Fringe Controlled Plumes 7.2.4.2 Preferred CORONA Site Instrumentation 7.2.4.3 CoronaScreen Models 7.2.4.4 Output Goals 7.2.4.5 Model Descriptions 7.2.5 Recommendations Acknowledgements References

421 422 423 424 424 424 427 429 432 433 435 437 439 442 442 443 443 444 446 450 451 451

Chapter 7.3 Diffuse Groundwater Quality Impacts from Agricultural Land-use: Management and Policy Implications of Scientific Realities Stephen Foster and Lucila Candela 7.3.1

7.3.2 7.3.3 7.3.4 7.3.5

Why is Agricultural Land-use the Greatest Challenge Facing the New EC Water Directives? 7.3.1.1 How Does Agricultural Land-use Impact on Groundwater Quality? 7.3.1.2 Are All Types of Groundwater Body Equally Threatened by Agricultural Practices? 7.3.1.3 What can be Done to Make Agricultural Cropping more ‘‘Groundwater Friendly’’? Guidelines on ‘‘Best Agricultural Practice’’ Reducing Overall Cultivation Intensity Constraints on Pesticide Manufacture, Sale or Use Improving Irrigation Water Use Efficiency 7.3.5.1 What are the Main Policy Implications for Groundwater Body Quality Protection?

454 457 460 462 462 463 464 465 466

xxiv

Contents

Acknowledgements References

469 469

8. Integrated River Basin Management Chapter 8.1 Integrated Management Principles for Groundwater in the WFD Context Philippe Quevauviller 8.1.1 8.1.2

Introduction Challenges Linked to Groundwater Management 8.1.3 River Basin Management Principles 8.1.3.1 Water and Its Environment 8.1.3.2 River Basin Management Objectives 8.1.4 Operational Management 8.1.4.1 Pollution Control 8.1.4.2 Voluntary Agreements 8.1.4.3 Cost Recovery 8.1.4.4 Institutional Structure 8.1.4.5 Infrastructure vs. Regulation, Financing and Empowerment 8.1.4.6 Decentralisation 8.1.4.7 Privatisation 8.1.5 Planning 8.1.5.1 Functions of Plans and Policies 8.1.5.2 The Planning Process 8.1.5.3 Planning Systems 8.1.6 Analytical Support 8.1.6.1 Analytical Support for Operational Management: Main Challenges 8.1.6.2 Analytical Support and the Strategic Level: New Directions 8.1.7 International River Basins 8.1.7.1 The Challenge 8.1.7.2 International Basins at the Global Level 8.1.7.3 International River Basin Organisations 8.1.7.4 Interbasin Co-operation (Twinning) 8.1.8 Public Participation 8.1.8.1 At European Level 8.1.8.2 At International Level 8.1.9 Conclusions References

473 474 475 475 476 476 477 478 479 479 480 481 481 482 482 482 483 483 485 486 487 487 488 488 489 489 489 491 491 492

xxv

Contents

Chapter 8.2 System Approach to Environmentally Acceptable Farming Ramon Laplana and Nadine Turpin 8.2.1 8.2.2

Introduction Integrated River Basin Management with BMPs to Mitigate NPS Pollution 8.2.2.1 What are Best Management Practices? 8.2.2.2 BMP Typology 8.2.2.3 How to Design Acceptable BMPs ? 8.2.3 How to Build a Cost-effectiveness Grid of Bundles of BMPs 8.2.3.1 Effectiveness Assessment 8.2.3.2 Implementation Costs 8.2.3.3 Building of the Comparison Grid 8.2.4 Conclusions References

494 495 495 497 497 502 502 505 507 508 509

Chapter 8.3 WATCH. Water Catchment Areas: Tools for Management and Control of Hazardous Compounds Thomas Track, Steve Setford, Sharon Huntley, Claudine Vermot-Desroches, John Wijdenes, Damia Barcelo´, Monica Rosell Linares, Peter Werner, Jens Fahl, Hans-Peter Rohns, Claudia Forner, Jesper Holm, Douglas Graham, Eckard Hitsch and Josef Lintschinger 8.3.1 8.3.2

8.3.3

8.3.4

8.3.5

8.3.6

Introduction Near-infrared Fuel Leak Sensor 8.3.2.1 Description of Results 8.3.2.2 Conclusion and Perspectives Analysis of MTBE and Related Compounds in Soil and Groundwater 8.3.3.1 Description of Results 8.3.3.2 Conclusions and Perspectives Immunoassay for Field-based Determination of MTBE 8.3.4.1 Description of Results 8.3.4.2 Conclusion and Perspectives Elucidation of Stimulating and Inhibiting Effects on Biodegradation 8.3.5.1 Description of Results 8.3.5.2 Conclusions and Perspectives Development of an Early Warning and Management System for Groundwater Resources

511 513 514 516 517 517 519 519 520 523 524 525 527 528

xxvi

Contents

8.3.6.1 Description of Results 8.3.6.2 Conclusions and Perspectives Acknowledgement and Further Information

528 531 532

9. Groundwater Status Assessment Chapter 9.1 Methodology for the Establishment of Groundwater Environmental Quality Standards Dietmar Mu¨ller and Anne-Marie Fouillac 9.1.1

Introductory Remarks on Science–Policy Integration Needs 9.1.2 Towards an Integrated Management of Groundwater Resources 9.1.3 The Framework: An Integrated Characterisation Process 9.1.4 Criteria to Assess Quality and Status of Groundwater 9.1.4.1 Natural Background Levels 9.1.4.2 Generic Reference Values 9.1.4.3 Attenuation Criteria 9.1.5 Receptor-oriented Quality Standards Derived by a Tiered Approach 9.1.6 How to Determine a Threshold Value (Example: Surface Water) 9.1.7 Compliance Regime for Groundwater Quality Standards 9.1.8 Conclusions Acknowledgement References

535 536 536 537 538 538 540 541 541 543 543 544 544

Chapter 9.2 Pesticides in European Groundwaters: Biogeochemical Processes, Contamination Status and Results from a Case Study Christophe Mouvet 9.2.1 9.2.2

Introduction Major Processes Involved in the Transport of Pesticides from the Soil to and in Groundwater 9.2.2.1 Sorption and Degradation 9.2.2.2 Biodegradation in Microcosms with Solids from the Unsaturated Zone Below the Root Zone 9.2.2.3 Sorption on Solids from the Unsaturated Zone

545 546 547

548 550

xxvii

Contents

9.2.2.4

Biodegradation in Microcosms with Solids and Water from the Saturated Zone 9.2.2.5 Sorption on Solids from the Saturated Zone 9.2.2.6 Field Studies Involving both Sorption and Biodegradation 9.2.2.7 Synthesis of Published Results on Sorption and Biodegradation below the Root Zone 9.2.3 Status of Groundwater Contamination by Pesticides at European Scale 9.2.3.1 The Waterbase Data Base of the European Environment Agency, June 2006 9.2.3.2 The Indicator Fact Sheet of the European Environment Agency 9.2.4 Status of Groundwater Contamination by Pesticides in Selected European Countries or Regions 9.2.4.1 Status in Italy (Adapted from Refs. 60 and 61) 9.2.4.2 Status in Wallony (Belgium) 9.2.4.3 Status in Sweden 9.2.4.4 Status in France 9.2.4.5 Status in the UK 9.2.4.6 Synthesis of the Data at National and Regional Scales 9.2.5 A Case Study: The Bre´villes Spring 9.2.5.1 Brief Description of the System and the Methods Used 9.2.5.2 Main Results from the Piezometer Network 9.2.5.3 Main Results from the Monitoring of the Spring 9.2.5.4 Conclusions from the Bre´villes Case Study 9.2.6 Conclusions and Perspectives Acknowledgements References

550 553 553 555 555 555 558 559 559 561 564 565 568 570 570 571 572 573 576 576 579 579

Chapter 9.3 Evaluation of the Quantitative Status of Groundwater– Surface Water Interaction at a National Scale Hans Jørgen Henriksen, Lars Troldborg, Per Nyegaard, Anker L. Højberg, Torben O. Sonnenborg and Jens Christian Refsgaard 9.3.1 9.3.2

Introduction The National Water Resource Model (DK Model) 9.3.2.1 Conceptual Model 9.3.2.2 Processes and Data 9.3.2.3 Model Code

584 587 587 587 589

xxviii

Contents

9.3.2.4 Model Parameterisation and Calibration 9.3.2.5 Model Validation 9.3.2.6 A Few Examples of Model Results 9.3.3 Criteria for Sustainable Groundwater Abstraction 9.3.4 Model Results 9.3.5 Discussion and Conclusions 9.3.5.1 Appropriateness of Approach for the WFD 9.3.5.2 Linking the Modelling Approach with Monitoring 9.3.5.3 Strengths and Weaknesses of Approach 9.3.5.4 Novelty of this Work Acknowledgements References

589 591 591 592 596 600 600 601 602 604 604 604

10. Modelling Chapter 10.1 Conceptual Models in River Basin Management Antony Chapman, Jos Brils, Erik Ansink, Cećile Herivaux and Pierre Strosser 10.1.1 Introduction 10.1.2 Integrated Water Resource Management 10.1.3 Conceptual Models in the Context of River Basin Management 10.1.3.1 What are Conceptual Models? 10.1.3.2 What Role can Conceptual Models Play? 10.1.3.3 From Conceptual Models to Quantitative/Computer-based Models 10.1.4 Building Conceptual Models in the Context of River Basin Management: Some Principles 10.1.4.1 The DPSIR Framework as a Guide 10.1.4.2 Investigating the Dynamics of River Basin Systems 10.1.4.3 Stakeholder Integration and Response 10.1.5 Experience from the Aqua terra Research Project 10.1.5.1 Context and Objectives 10.1.5.2 Identifying the Main Environmental Issues Relevant to the Case Studies 10.1.5.3 Integrating Stakeholders’ Views 10.1.5.4 Developing Simplified Representations of the Systems Investigated 10.1.5.5 Preliminary Lessons from the Experience of INTEGRATOR 10.1.6 Conclusions

611 612 613 613 614 615 617 617 618 619 620 620 621 622 622 624 625

xxix

Contents

Acknowledgements References

626 626

Chapter 10.2 Modelling Reactive Transport of Diffuse Contaminants: Identifying the Groundwater Contribution to Surface Water Quality Hans Peter Broers, Bas van der Grift, Jasper Griffioen and Ruth Heerdink 10.2.1 10.2.2 10.2.3 10.2.4

Introduction Framing a Conceptual Model Building a Regional Model for the Kempen Area Verifying the Model: Setting up a Customised Monitoring System 10.2.5 Predictions of Groundwater Contributions to Surface Water Quality 10.2.5.1 Discussion 10.2.5.2 Policy Aspects 10.2.6 Conclusions Acknowledgement References

630 631 634 637 640 641 642 643 643 643

11. Conclusions: Further Policy and Research Needs Chapter 11.1 SNOWMAN: An Alternative for Transnational Research Funding Jo¨rg Frauenstein and H. Johan Van Veen 11.1.1 Introduction 11.1.1.1 Transnational Research 11.1.1.2 Soil and Groundwater Management 11.1.1.3 Forerunning Projects and Significant Scientific Input 11.1.2 An ERA-NET Bridging the Gap Between National and European Research 11.1.3 The Way Forward 11.1.3.1 The Meaning of Cooperation 11.1.3.2 A Stepwise Approach Towards Cooperation 11.1.4 The Upcoming Coordinated Call 11.1.4.1 Objectives and Projects 11.1.4.2 The Principles of the Coordinated Call 11.1.5 Conclusions Acknowledgements References

647 647 649 651 654 654 654 657 662 663 664 668 669 670

xxx

Contents

Chapter 11.2 Incorporation of Groundwater Ecology in Environmental Policy Dan L. Danielopol, Christian Griebler, Amara Gunatilaka, Hans Ju¨rgen Hahn, Janine Gibert, F. Mermillod-Blondin, Giuseppe Messana, Jos Notenboom and Boris Sket 11.2.1 Introduction: Groundwater Science and the New Order 11.2.2 The New Groundwater Ecology: Its Interest for Water Management Projects and/or Water Policy Planners 11.2.3 The Groundwater Ecosystem Approach as a Framework for Planning Pollution Prevention and/ or Environmental Protection Strategies 11.2.4 Diversity of Groundwater Habitats and Organisms: Their Usefulness for Environmental Monitoring Programmes 11.2.5 Groundwater-Dependent Ecosystems: A Holistic Representation 11.2.6 Overview: The Expanded Order (Achievements and Future Needs) Acknowledgements References

671

672

674

680 683 684 686 686

Chapter 11.3 Towards a Science–Policy Interface (WISE-RTD) in Support of Groundwater Management Philippe Quevauviller 11.3.1 Introduction 11.3.1.1 General Needs 11.3.1.2 Different Levels of Interactions 11.3.2 Introduction to WISE 11.3.2.1 What is WISE? 11.3.2.2 Why do we Need WISE? 11.3.2.3 What is the Objective of the WISE Process? 11.3.2.4 Does WISE Already Exist? 11.3.2.5 Who will Use WISE? 11.3.2.6 What will WISE be Used for? 11.3.2.7 What are the Next Steps in WISE Development? 11.3.3 EU RTD Funding Mechanisms 11.3.3.1 FP5 Research Projects 11.3.3.2 FP6 Targeted Research and Integrated Projects

690 691 691 693 693 694 694 695 695 696 697 697 697 698

xxxi

Contents

11.3.3.3 FP6 ERA-NET Projects 11.3.3.4 Projects Issued from the Scientific Support to Policies (SSP) Priority 11.3.3.5 Orientations of the 7th Framework Programme 11.3.3.6 LIFE: Demonstration Projects 11.3.4 An Operational Web Interface: WISE-RTD 11.3.4.1 The Harmoni-CA Initiative 11.3.4.2 The WISE-RTD Web Portal 11.3.5 Conclusions: Needs for an Overall Science–Policy Integration Framework References

698

Appendix I Outline of Water Framework Directive Appendix II Outline of Groundwater Directive

704 711

699 699 700 700 700 701 701 703

Appendices

Subject Index

716

Contributors List BLIND Michiel W. RIZA P.O. Box 17 NL-8200 AA Lelystad The Netherlands Email: [email protected]

ANSINK Erik Wageningen University Environmental Economics and Natural Resources Group P.O. Box 8130 NL-6700 EW Wageningen The Netherlands Email: [email protected]

BRILS Jos Netherlands Organisation for Applied Scientific Research (TNO) Built Environment and Geosciences Business Unit Groundwater and Soil P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: [email protected]

ARNOLD Geo E. RIZA P.O. Box 17 NL-8200 AA Lelystad The Netherlands Email: [email protected]. nl BARCELO´ Damia Consejo Superior de Investigaciones Cientı´ ficas Instituto de Investigaciones Quı´ micas y Ambieltales de Barcelona Dept. de Quı´ mica Ambiental Jordi Girona, 18-26 E-08034 Barcelona Spain Email: [email protected] BIDOGLIO Giovanni European Commission, Joint Research Centre, Via E. Fermi l, TP 460 IT-21020 Ispra (VA) Italy Email: [email protected]

BROERS Hans Peter Netherlands Organisation for Applied Scientific Research (TNO) Geological Survey of The Netherlands Princetonlaan 6 P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: [email protected] CANDELA Lucila Technical University of Catalonia Dept. of Geotechnical Engineering & Geoscience Gran Capita`, s/n Ed. D-2 ES-08034 Barcelona Spain Email: [email protected]

xxxiii

xxxiv

CHAPMAN Antony r3 Environmental Technology Ltd c/o School of Horticulture and Landscape University of Reading TOB2, Earley Gate, Whiteknights Reading RG6 6AU United Kingdom Email: [email protected] CUSTODIO Emilio Technical University of Catalonia Dept. of Geotechnical Engineering Gran Capita`, s/n Ed. D-2 ES-08034 Barcelona Spain Email: [email protected] DANIELOPOL Dan L. Austrian Academy of Sciences Institute of Limnology Mondsee str. 9 AT-5310 Mondsee Austria Email: [email protected] DARMENDRAIL Dominique Bureau de Recherches Geólogiques et Minie`res (BRGM) 3, avenue Claude Guillemin FR-45060 Orl!ans c!dex 2 France Email: [email protected] DE LANGE Wim J. RIZA P.O. Box 17 NL-8200 AA Lelystad The Netherlands Email: [email protected]. nl

Contributors List

EDMUNDS W. Mike Oxford Centre for Water Research Oxford University Centre for the Environment South Parks Road Oxford 0X1 3QY United Kingdom Email: [email protected]

ERTEL Thomas Sachversta¨ndigen-Bu¨ro Boschstr. 10 DE-73734 Esslingen Germany Email: thomas@sv-ertel de

FAHL Jens University of Technology Dresden Institute of Waste Management and Contaminated Site Treatment Pratzschwitzer Str. 15 D-01796 Pirna Germany Email: [email protected]. de

FOSTER Stephen IAH President Int. Association of Hydrogeologists P.O. Box 9 Kenilworth (Warwick) CV8 1JG United Kingdom Email: [email protected]

FORNER Claudia Stadtwerke Du¨sseldorf AG Quality Control Water Wiedfeld 50 D-40589 Du¨sseldorf Germany

xxxv

Contributors List

FOUILLAC Anne-Marie Bureau de Recherches Geólogiques et Minie`res (BRGM) 3 avenue Claude Guillemin FR-45060 Orleáns ce´dex 2 France Email: [email protected] FRAUENSTEIN Jo¨rg Federal Environmental Agency Dessau Unit II 4.3 Terrestrial Ecology, Land Management, Regional Protection Concepts Wo¨rlitzer Platz 1, 06844 Dessau P.O. Box 1406 DE-06813 Dessau Germany Email: [email protected] GALBIATI Lorenzo Age`ncia Catalana de l’Aigua Provença, 204-208 ES-08036 Barcelona Spain Email: [email protected] GIBERT Janine Universite´ Claude Bernard Lyonl UMR CNRS 5023 EHF Equipe d’Hydrobiologie et Ecologie Souterraines. Baˆt FOREL 43 Bd 11/11/1918 FR-69622 Villeurbanne cedex France Email: [email protected] GONZALEZ Catherine Ecole des Mines d’Ale`s 6 avenue de Clavie`res FR-30319 Ale`s Cedex France Email: [email protected]

GRAHAM Douglas DHI–Water and Environment DK-2970 Horsholm Denmark Email: [email protected]

GRATH Johannes Umweltbundesamt GmbH Spittelauer Laende 5 AT-1090 Wien Austria Email: Johannes.grath@ umweltbundesamt.at

GRATHWOHL Peter Centre for Applied Geoscience Universita¨t Tu¨bingen Sigwartstrasse 10 DE-72076 Tu¨bingen Germany Emai1: [email protected]

GREENWOOD Richard School of Biological Sciences University of Portsmouth King Henri Building King Henri I Street UK-Portsmouth PO1 2DY United Kingdom

GRIEBLER Christian GSF-National Research Center for Environmental and Health Institute of Groundwater Ecology Ingolsta¨dter Landstrasse 1 DE-85764 Neuherberg/Mu¨nchen Germany Email: [email protected]

xxxvi

GRIFFIOEN Jasper Netherlands Organisation for Applied Scientific Research (TNO) Geological Survey of The Netherlands Princetonlaan 6 P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: jasper.griffi[email protected]

GUNATILAKA Amara Department of Ecotoxicology Center for Public Health Medical University of Vienna Wa`hringer Strarse 10 A-1090 Vienna Austria Emai1: amarasinha.gunatilaka@ verbundplan.at

HAHN Hans Ju¨rgen Arbeitsgruppe Grundwassero¨kologie Universita¨t Koblenz-Landau, Campus Landau Abt. Biologie Im Fort 7, D-76829 Landau Germany Email: [email protected]

HEERDINK Ruth Netherlands Organisation for Applied Scientific Research (TNO) Geological Survey of The Netherlands Princetonlaan 6 P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: [email protected]

Contributors List

HENRIKSEN Hans Jørgen Geological Survey of Denmark and Greenland GEUS Øster Voldgade 10 DK-1350 Copenhagen K Denmark Email: [email protected] HERIVAUX Cećile BRGM Water Department 3 avenue Claude Guillemin BP 36009 FR-45060 Orleáns cedex France Email: [email protected] HINSBY Klaus Geological Survey of Denmark and Greenland, GEUS Øster Voldgade 10 DK-1350 Copenhagen K Denmark Email: [email protected] HITSCH Eckard Salzburg AG Centre Wasser Hagenau 1 A-5101 Bergheim Austria HØJBERG Anker L. Geological Survey of Denmark and Greenland GEUS Øster Voldgade 10 DK-1350 Copenhagen K Denmark Email: [email protected] HOLM Jesper DHI – Water and Environment DK-2970 Horsholm Denmark Email: [email protected]

xxxvii

Contributors List

HUNTLEY, Sharon Cranfield Centre for Analytical Science Cranfield University Silsoe Bedforshire MK45 4DT United Kingdom Email: s.l.huntley@cranfield.ac.uk

LINTSCHINGER Josef Salzburg AG Centre Wasser Hagenau 1 A-5101 Bergheim Austria Email: josef.lintschinger@ salzburg-ag.at

JARSJO¨ Jerker Stockholm University Dept. of Physical Geography and Quaternary Geology SE-106 91 Stockholm Sweden Emai1: [email protected]

MANZANO Marisol Technical University of Cartagena Paseo Alfonso XIII, 52 ES-30203 Cartagena Spain Email: [email protected]

KIRCHHOLTES Hermann J. Landeshauptstadt Stuttgart Amt fu¨r Umweltschutz Hermann Josef Kirchholtes 36-3.51 Gaisburgstr. 4 DE-70182 Stuttgart Germany Emai1: [email protected]

F. MERMILLOD – BLONDIN Universite´ Claude Bernard Lyou 1 UMR CNRS 5023 EHF Equipe d’Hydrobiologie et Ecologie Souterraines Ba´t FOREL 43 Bd 11/11/1998 FR-69622 Villerubanne Cedex France

LAPLANA Ramon CEMAGREF Unite´ Ader 50 avenue de Verdun FR-33612 Cestas France Emai1: ramon.laplana@cemagref. fr LERNER David N. Groundwater Protection and Restoration Group University of Sheffield Kroto Research Institute Broad Lane Sheffield S3 7HQ United Kingdom Emai1: d.n.lerner@sheffield.ac.uk

MESSANA Giuseppe Istituto per lo Studio degli Ecosistemi CNR – ISE, Sede di Firenze Via Madonna del Piano IT-50019 Sesto Fiorentino/Firenze Italy Emai1: [email protected]

MOUVET Christophe Bureau de Recherches Geólogiques et Minie`res (BRGM) 3 avenue Claude Guillemin FR-45060 Orleáns ce´dex 2 France Email: [email protected]

xxxviii

Contributors List

MU¨LLER Dietmar Umweltbundesamt GmbH Spittelauer Laende 5 AT-1090 Wien Austria Email: dietmar.mueller@ umweltbundesamt.at

NYEGAARD Per Geological Survey of Denmark and Greenland, GEUS Øster Voldgade 10 DK-1350 Copenhagen K Denmark Email: [email protected]

NEGREL Philippe Bureau de Recherches Geólogiques et Minie`res (BRGM) 3, avenue Claude Guillemin FR-45060 Orleáns ce´dex 2 France Email: [email protected]

PAHL-WOSTL Claudia Institute of Environmental Systems Research University of Osnabru¨ck Barbarastrasse 12 DE-49069 Osnabru¨ck Germany Email: [email protected]

NEWIG Jens Institute of Environmental Systems Research University of Osnabru¨ck Barbarastrasse 12 DE-49069 Osnabru¨ck Germany Emai1: jnewig@ usf.uniosnabrueck.de

PTAK Thomas University of Go¨ttingen Geosciences Center Goldschmidtstrasse 3 DE-37077 Go¨ttingen Germany Email: thomas.ptak@geo. uni-goettingen.de

NIEUWENHUIS Rob TNO Netherlands Institute of Applied Geoscience –National Geological Survey Princetonlaan 6 P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: [email protected]

PURTSCHERT Roland Climate and Environmental Physics Physics Institute, University of Bern Sidlerstrasse 5 CH-3012 Bern, Switzerland Email: [email protected]

NOTENBOOM Jos Milieu- en Natuurplanbureau Netherlands Environmental Assessment Agency. Postbus 303 NL-3720 AH Bilthoven The Netherlands Email: [email protected]

QUEVAUVILLER Philippe European Commission DG Environment (BU9 3/142) Rue de la Loi, 200 BE-1049 Brussels Belgium Email: philippe.quevauviller@ec. europa.eu

xxxix

Contributors List

REFSGAARD Jens Christian Geological Survey of Denmark and Greenland GEUS Øster Voldgade 10 DK-1350 Copenhagen K Denmark Email: [email protected] RIDDER Dagmar Institute of Environmental Systems Research University of Osnabru¨ck Barbarastrasse 12 DE-49069 Osnabru¨ck Germany Email: [email protected]

ROHNS Hans-Peter Stadtwerke Du¨sseldorf AG Quality Control Water Wiedfeld 50 D-40589 Du¨sseldorf Germany Email: [email protected]

ROSELL LINARES Monica Consejo Superior de Investigaciones Cientı´ ficas Instituto de Investigaciones Quı´ micas y Ambieltales de Barcelona Dept. de Quı´ mica Ambiental Jordi Girona, 18-26 E-08034 Barcelona Spain Email: [email protected]

RIJNAARTS Huub TNO Netherlands Institute of Applied Geoscience — National Geological Survey Princetonlaan 6 P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: [email protected]

ROY Ste´phane Bureau de Recherches Geólogiques et Minie`res (BRGM) 3, avenue Claude Guillemin FR-45060 Orleàns ce`dex 2 France Email: [email protected]

RINAUDO Jean-Daniel Bureau de Recherches Geólogiques et Minie`res (BRGM) 1034, rue de Pinville FR-34000 Montpellier France Email: [email protected]

SCHEIDLEDER Andreas Umweltbundesamt GmbH Spittelauer Laende 5 AT-1090 Wien Austria Email: andreas.scheidleder@ umweltbundesamt.at

RODGERS-JENKINS Crystal U.S. EPA 1201 Constitution Avenue, NW MC-4607M USA-Washington, DC 20460 United States of America Email: Rodgers.Crystal@epamail. epa.gov

SETFORD, Steve Cranfield Centre for Analytical Science Cranfield University Silsoe Bedforshire MK45 4DT United Kingdom Email: s.j.setford@cranfield.ac.uk

xl

SKET Boris University of Ljubljana, Biotechnical Faculty Dept. of Biology Vecna pot 111, PP2995 SI-1001 Ljubljana Slovenia Email: [email protected] SLOB Adriaan TNO Environment and Geosciences Dept. Innovation & Environment Van Mourik Broekmanweg 6 P.O. Box 49 NL-2600 AA Delft The Netherlands Email: [email protected] SONNENBORG Torben O. Geological Survey of Denmark and Greenland, GEUS Øster Voldgade 10 DK-1350 Copenhagen K Denmark Email: [email protected] STROSSER Pierre ACTeon Le Chalimont BP Ferme du Pre´ du Bois FR-68370 Orbey France Email: [email protected] TER MEER Jeroen TNO Netherlands Institute of Applied Geoscience –National Geological Survey Princetonlaan 6 P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: [email protected]

Contributors List

THORNTON Steven F. Groundwater Protection and Restoration Group University of Sheffield Kroto Research Institute Broad Lane Sheffield S3 7HQ United Kingdom Email: s.f.thornton@Sheffield.ac.uk

TRACK Thomas DECHEMA e.V. Theodor-Heuss-Allee 25 DE-60486 Frankfurt am Main Germany Email: [email protected]

TROLDBORG Lars Geological Survey of Denmark and Greenland GEUS Øster Voldgade 10 DK-1350 Copenhagen K Denmark Email: [email protected]

TSAKIROGLOU Christos D. FORTH/ICE-HT Stadiou Street, Platani P.O. Box 1414 GR-26504 Patras Greece Email: [email protected]

TURPIN Nadine UMR Me´tafort-Cemageef-Agro Paris Tech-ENITA-INRA 24 avenue des Landais BP 50085 FR-63172 Aubiere Ce´dex France Emai1: [email protected]

xli

Contributors List

VAN DER GRIFT Bas Netherlands Organisation for Applied Scientific Research (TNO) Geological Survey of the Netherlands Princetonlaan 6 P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: [email protected] VAN DUIJNE Hans TNO Netherlands Institute of Applied Geoscience – National Geological Survey Princetonlaan 6 P.O. Box 80015 NL-3508 TA Utrecht The Netherlands Email: [email protected] VAN VEEN H. Johan SKB Bu¨chnerweg 1 Postbus 420 NL-2800 AK Gouda The Netherlands Email: [email protected] VAN DER SLOOT Hans Energy Research Centre (ECN) P.O. Box 1 NL- 1755 Petten ZG The Netherlands Email: [email protected] VERMOT-DESROCHES Claudine Diaclone l Bd Fleming-BP 1985 F-25020 Besancon Cedex France

WARD Rob Environment Agency—England and Wales Olton Court Solihull West Midlands B92 7HX United Kingdom Email: rob.ward@ environment-agency.gov.uk WERNER Peter University of Technology Dresden Institute of Waste Management and Contaminated Site Treatment Pratzschwitzer Str. 15 D-01796 Pirna Germany Email: [email protected] WIJDENES John Diaclone l Bd Fleming-BP 1985 F-25020 Besancon Cedex France Email: [email protected] WILSON Ryan D. Groundwater Protection and Restoration Group University of Sheffield Kroto Research Institute Broad Lane Sheffield S3 7HQ United Kingdom Email: r.d.wilson@Sheffield.ac.uk

1. General Introduction

CHAPTER 1

General Introduction: The Need to Protect Groundwater PHILIPPE QUEVAUVILLER European Commission, DG Environment (BU9 3/142), Rue de la Loi 200, BE-1049 Brussels, Belgium

1.1 Introduction Groundwater constitutes the largest reservoir of freshwater in the world, accounting for over 97% of all freshwater available on earth (excluding glaciers and ice caps). The remaining 3% is composed mainly of surface water (lakes, rivers, wetlands) and soil moisture. Until recently, focus on groundwater mainly concerned its use as drinking water (e.g. about 75% of European Union (EU) inhabitants depend on groundwater for their water supply). Groundwater is also an important resource for industry (e.g. cooling waters) and agriculture (irrigation). It has, however, become increasingly obvious that groundwater should not only be viewed as a drinking water reservoir, but also as a critical aquatic ecosystem.1 In this respect, groundwater represents an important link of the hydrological cycle for the maintenance of wetlands and river flows, acting as a buffer through dry periods. In other words, it provides the base flow (i.e. the water which feeds rivers all year round) for surface water systems, many of which are used for water supply and recreation. In many rivers, indeed, more than 50% of the annual flow is derived from groundwater. In lowflow periods in summer, more than 90% of the flow in some rivers may come from groundwater. Hence, deterioration of groundwater quality may directly affect other related aquatic and terrestrial ecosystems. Since groundwater moves slowly through the subsurface, the impact of anthropogenic activities may last for a relatively long time, which means that pollution that occurred some decades ago—whether from agriculture, industry or other human activities—may still be threatening groundwater quality today and, in some cases, will continue to do so for several generations to come. The legacy of the past is clearly visible at large-scale contaminated sites, e.g. industrial sites or harbour areas, where it is simply not possible, with 3

4

Chapter 1

state-of-the-art technology and a proportionate use of public and/or private money, to clean up the regional contamination encountered at these locations.2 In addition, the experience of remediation of the past 20 years has shown that the measures taken have in most cases not been able to completely remove all contaminants and that pollutant sources, even if partially removed, continue to emit for long periods of time (i.e. several generations).3,4 Therefore, an important focus should be on preventing pollution in the first place. Secondly, since surface water systems receive a continuous discharge of inflowing groundwater, a deteriorated groundwater quality will ultimately be reflected in the quality of surface waters. In other words, the effect of human activity on groundwater quality will eventually also impact on the quality of associated aquatic ecosystems and directly dependent terrestrial ecosystems if so-called natural attenuation reactions such as biodegradation in the subsurface are not sufficient to contain the contaminants. Finally, groundwater is a ‘‘hidden resource’’ which is quantitatively much more significant than surface water and for which pollution prevention and quality monitoring and restoration are even more difficult than for surface waters, which is mostly due to its inaccessibility. This ‘‘hidden’’ character makes it difficult adequately to locate and quantitatively appreciate pollution impacts, resulting in a lack of awareness and/or evidence regarding the extent of risks and pressures. Recent reports, however, show that pollution from domestic, agricultural and industrial sources is, despite the progress in some fields, still a major concern, either directly through discharges (effluents) or indirectly from the spreading of nitrogen fertilisers and pesticides or through leaching from old landfills or industrial sites (e.g. chlorinated hydrocarbons, heavy metals). For example, around one-third of groundwater bodies in Europe currently exceed the nitrate guideline values.5 While point sources have caused most of the pollution identified to date, there is evidence that diffuse sources are having an increasing impact on groundwater. This chapter develops the elements discussed above as a general introduction to this book, which further elaborates issues related to groundwater policy, protection and remediation throughout the different chapters.

1.2 The Scientific Background 1.2.1

The Hydrogeological Cycle

It is estimated that roughly 22% of freshwater is stored underground, representing some 8 million km3 of 37 million km3 of freshwater found on the planet. Excluding water from polar ice, groundwater constitutes some 97% of all the freshwater that is potentially available for human use on or beneath the earth’s surface. The remainder is stored in lakes, rivers and swamps.6 Groundwater recharge is essentially ensured by rain that infiltrates through the soil into underlying layers; this recharge is occasionally augmented by streams and rivers that lose water to underground strata. Once underground,

General Introduction: The Need to Protect Groundwater

5

groundwater flows at rates which range from more than 10 metres per day to as little as 1 metre per year until it reaches an outlet, e.g. a spring or seepages at the ground surface (which actually keep rivers flowing during dry periods). The time scales at which groundwater flows hence considerably vary, depending on hydrogeological conditions. It may take years to decades for water to move through the soil to reach the water table, the level at which the ground is fully saturated, where it may remain underground for tens or even thousands of years before reappearing at the surface.6 Geological settings may also trap groundwater from both its source and its outlets. Finally, climate change may also lead to groundwater losses by depriving aquifers from recharge as it appears in a number of regions which turned into deserts. The level of available geological and hydrogeological information varies from area to area, and this has an effect on the protection schemes to be developed.7 Where the information is adequate, a comprehensive scheme, based on hydrogeological concepts, is achievable. However, as mentioned below, aquifers are rarely homogeneous and their geological variability conditions the nature of groundwater flowing through their respective lithologies and structures, which makes it difficult to establish large-scale conceptual hydrogeological models.

1.2.2

Waters in Aquifers

The nature of aquifers, consisting either of unconsolidated materials such as sand or gravel or consolidated rock such as sandstone, has a considerable influence on groundwater flows and hence on pollution pathways (see Section 1.4.3). On the one hand, unconsolidated materials, such as sands, can store up to 30% of their volume as water. On the other hand, consolidated materials may also store large volumes of water, depending upon their porosity, but groundwater flow is usually very slow owing to the small size of the pores. In some types of rocks, the capillary attraction between the groundwater and the pore surface does not allow water to be released and hence to flow. However, consolidated materials may also store water in fractures in the rock, which although they usually represent less than 1% of the total volume can be enlarged by dissolution in rocks such as limestones. Enlarged fractures enable the aquifer both to store large volumes of water and permit high groundwater flows,7 which has an impact on pollution spreading (see Chapter 5.5). Aquifers are usually bounded above by an unsaturared zone, which contains both air and water, and below by an impermeable bed constituted, for example, of clay or rock. The boundary between the unsaturared and saturated zones (water table) is found at different depths depending on the hydrogeological and climatic settings, e.g. as much as 100 m below the surface in arid areas and close to the surface in humid areas. Some aquifers are, however, bounded entirely by impermeable layers and contain groundwater under pressure (which enables water abstraction by artesian wells).

6

1.2.3

Chapter 1

Groundwater Flows

With aquifer characteristics in mind (in particular the type of materials containing the water), it is possible to better approximate groundwater flows. Groundwater moves through aquifers as a result of differences in pressure or elevation of the water table within the aquifer. The groundwater flow may be slowed down by various obstructions while moving from the point of recharge to its exit from the aquifer. In some cases, impermeable rock formations (known as aquicludes) such as shale stop completely the water flow, while other geological strata (known as aquitards), such as clay lenses embedded with sand, may slow down the groundwater flow. The groundwater flow rate depends on the permeability and porosity of the aquifer, and on the pressure gradient. As an example, highly permeable aquifers such as limestones respond rapidly to changes in recharge and abstraction rates, and groundwater levels in such areas may fluctuate by as much as 10 m a year and may change by up to 50 m a year.7 The greatest variations in groundwater flow patterns occur where changes in rock types, e.g. limestone overlying sediments and a hard crystalline rock, induce discontinuities in flow and may bring groundwater flow to the surface on the junction between the two rock types. Variations in groundwater flow may also occur within an unconsolidated alluvial aquifer, e.g. great lateral variations occur in the mix of gravel, sand and clay making up the aquifer matrix. In larger-scale alluvial aquifers, layers of sand or gravel-rich sediment interbedded with clay-rich layers induce lateral flow following the more permeable sand- and gravel-rich zones. Needless to say that groundwater flow rates are very small in comparison to those of surface water. In this respect, some groundwater from deep alluvial basins is likely to be thousands or even hundreds of thousands of years old. The slow movement of groundwater largely contributes to its purity since contaminants become highly attenuated during the usually long groundwater flowing pathway to the surface. Groundwater may also become enriched with elements that are naturally present in rocks. Furthermore, saltwater intrusion may occur near coastlines, in particular where the water table is lowered due to abstraction, and this is likely to be accentuated by rising sea levels due to climate change.

1.2.4

Groundwater Quality

Most groundwater originates from water that has permeated first the soil and then the rock below it. The soil removes many impurities and the rock through which the water then flows, perhaps for thousand of years, filters and purifies the water even further.7 It therefore usually reappears at the earth surface free of pathogenic micro-organisms. This is the reason for an increasing exploitation of groundwater resources (see Section 3.1). While groundwater is generally less easily/rapidly polluted than streams and rivers, it often contains high concentrations of dissolved elements from the rock


7

through which it has passed. Another feature is that when groundwater is polluted, many processes occur during its pathway to the surface; in particular, pollutant loads may be attenuated by adsorption by the rock itself or biochemical transformation into substances that are less harmful than the original compounds. However, severe pollutions may affect groundwater quality over long periods, i.e. once pollutants reach the water table, it may take a very long time before they are flushed out from the aquifer. Furthermore, groundwater quality affected by pollution may take a long time to recover since the water within the aquifer moves so slowly. Once polluted, aquifers are difficult—and sometimes even impossible—to clean up. The process can be likened to trying to squeeze out the last traces of soap from a sponge.7 As stressed above, the complex nature of groundwater is compounded in the context of pollution and quality problems. Let us repeat that the chemical characteristics of aquifer materials and the way pollutants react with them vary greatly. In some cases, pollutants are ‘‘filtered’’ out mechanically or through adsorption onto particles within the soil or aquifer matrix. In other cases, however, pollutants remain mobile and can rapidly spread throughout an aquifer. The aquifer matrix itself can become contaminated and pockets of pollutants can serve as continuous sources of contamination. For example, small pockets of organic solvents can remain as pollution sources virtually indefinitely because of their low solubility in water. Furthermore, changes in pH or other groundwater characteristics can cause the release of toxic materials, such as fluoride, from natural sources within aquifers. Given the hundreds of thousands of naturally occurring compounds in groundwater and aquifer materials, and the similarly large number of compounds present in waste water released to aquifers, understanding and managing pollution problems is a highly complex task. This illustrates the importance of preventing pollution of groundwater from the start rather than dealing with the consequences (Chapter 5.5). All the above considerations have an impact on the way groundwater background levels are evaluated and also on the assessment of groundwater quality either related to its use or to its environmental value. These aspects are discussed in various chapters of this book (Chapters 5.1–5.3 and 9.1).

1.3 Groundwater Deterioration Risks 1.3.1 1.3.1.1

Quantitative Aspects Over-exploitation

Groundwater is extensively used by humans throughout the world as a drinking water resource, with some countries depending almost entirely on it while others only partly using the groundwater resource for drinking water abstraction. Groundwater supplies are of obvious importance in arid areas but they are also extensively used in humid areas, largely because they provide water that requires little or no treatment and which can be cheaply produced.

8

Chapter 1

In addition, its supplies are not subject to abrupt change as a result of abnormal weather, i.e. a dry summer while affecting reservoirs will have little effect on groundwater levels. Finally, groundwater can often be tapped near to where it is needed while surface water must be either developed at the sites of natural dams or reservoirs, or piped at considerable distances to where it will eventually be used.7 However, groundwater should not be seen as a simple alternative to the use of surface water. The inadequate control of groundwater abstraction in many parts of the world usually results in some form of over-exploitation, which can lead to either reversible or irreversible damages (this consideration also applies to damages due to pollution, see Section 1.3.3). In the first case, matters can be corrected and only the question of costs is involved. In the second case, costs are also involved but, in addition, sustainability issues arise since future generations are deprived of an important resource.7 Many aquifers are being over-exploited in the sense that water is abstracted faster than the average recharge rate. This is particularly problematic in the case of fossil groundwater. The control of the balance of groundwater levels (equilibrium between abstraction and recharge) is, however, difficult to apprehend in that the recharge rate of groundwater resources is not constant and can vary considerably with the rainfall pattern. This means that what may be considered as over-exploitation in one year may be a perfectly acceptable rate of exploitation in another. To complicate matters, in some arid areas major recharge only occurs once a decade or even less frequently.7 In this circumstance, defining a sustainable abstraction rate is difficult. Adding to this, climate changes impact on the dynamic balances of groundwater resources, and these are not easily predictable.

1.3.1.2

Falling and Rising Water Tables

Over-abstraction and subsequent fall of the water table may lead to severe damage linked to ground subsidence, which is caused by water draining from the pores in underground strata, causing the rock to compact. Unconsolidated strata, especially clays which have high water content, are particularly susceptible to this phenomenon. Conversely, circumstances such as over-irrigation of land may lead to rising of the water table, leading to waterlogging of agricultural land which is often associated with salinisation. This is due to two causes: a rising water table that brings saline water into contact with plant roots; and the evaporation of irrigation water by sun, leaving the salt behind. Another occurrence of rising water table is observed in urban areas where urban recharge rates may be higher than natural (pre-natural) ones. This is not so problematic when cities consume large quantities of water (thus balancing the high recharge rates) but it may be so when groundwater is not abstracted any more, which increasingly happens owing to contaminated groundwater beneath the cities. Rising water tables under cities may lead to urban flooding, with associated costs (need to pump water out, etc.).


1.3.1.3

9

Saltwater Intrusion

Under natural conditions, coastal aquifers discharge freshwater into the sea. However, in case of (over)abstraction of groundwater in areas that are close to the coastline, this process may be reversed, leading to salt water moving inland and polluting the aquiferw. Examples of such occurrences can be found in many places of the world.7 The problem may be severe on islands where the freshwater aquifer is only a few metres thick (e.g. composed of highly permeable sediments) and surrounded by salt water; in this specific case, aquifer abstraction has to be particularly well managed.

1.3.2 1.3.2.1

Links to Associated Ecosystems Links to Associated Aquatic Ecosystems

In many areas, it is groundwater that makes the use of surface water sources possible during dry seasons. Groundwater provides the base flow to many of the world’s rivers, and this flow continues throughout the year, regardless of weather conditions. Many rivers would dry up in hot and dry summers if they would not be fed by groundwater. This is particularly important in both humid and arid regions where precipitation is highly variable. Between precipitation events, groundwater and return flows from agricultural, domestic and other users are the primary source of flows in rivers. Since return flows generally have higher pollution loads than groundwater flow, the groundwater contribution is important to both the quantity and quality of dry-season flow in surface watercourses. An evaluation of quantitative aspects of groundwater–surface interactions is described in Chapter 9.3 of this book. Productivity in coastal ecosystems is also highly dependent on the balance between freshwater inflows from surface water, groundwater discharge and saline ocean water. Disruption of this balance through diminution of groundwater contributions to base flow could have major effects on the coastal environment.

1.3.2.2

Links to Dependent Terrestrial Ecosystems

Wetlands are some of the most productive and biologically diverse inland ecosystems. In many if not most cases, water availability in wetlands depends on high groundwater levels. Consequently, the fall of the water table may have a direct impact on wetlands as the land is drying out. In this respect, a number of the world’s major wetland areas, which are sensitive ecosystems supporting a large number of plant and animal species, are now under threat due to overabstraction. In addition, groundwater pollution also represents a major threat not only to the habitat of many rare species but also in affecting the purifying role of the wetlands with respect to inland lakes. w

Note that the term ‘‘pollution’’ is appropriate when saltwater intrusion is indeed due to overabstraction, i.e. due to human activity.

10

Chapter 1

Besides wetlands, groundwater levels and quality directly influence surface vegetation communities. Phreatophytes, plants that derive a major portion of their water needs from saturated soils, can be the dominant vegetative species in ecosystems where groundwater levels are shallow. They often form critical wildlife habitat and may serve as important sources of food and timber. This vegetation also uses substantial amounts of water. Removing these species can reduce evapotranspiration and hence the demand on groundwater resources, which may cause levels to rise and thereby lead to waterlogging and other environmental problems.

1.3.3 1.3.3.1

Groundwater Pollution Introduction

Once polluted, groundwater is extremely difficult to clean up owing to its inaccessibility, its huge volume and its slow flow rates. The three major pollution threats, namely urbanisation, industrial activities and agriculture, are discussed in the following paragraphs. Let us distinguish here pollution impacts related to either human uses or the environment. In the first instance, pollutants found in groundwater are listed according to their toxic impacts on drinking water quality according to, for example, WHO drinking water quality guidelines or EU legislation (Drinking Water Directive). Pollutants may also be distinguished according to their ecotoxicological impacts, i.e. substances which are detrimental to the environment such as those pollutants listed in the EU Water Framework Directive (see Chapter 3.1). This distinction is important in that the ‘‘pollution impact’’ should be assessed differently whether it is related to a particular use of the water resource or to an impact on the aquatic or terrestrial environment. As noted in Section 2.4, groundwater may contain high concentrations of chemical substances that are present naturally (due to interactions of the groundwater with the soil or surrounding rocks) and which as such does not correspond to a pollution (i.e. due to human activities) but which may hamper the use of the groundwater for drinking water abstraction; this does not mean, however, that the groundwater is of ‘‘bad environmental quality’’. These two aspects of groundwater quality and related needs for protection against pollution are further discussed in the policy context in Chapters 3.1 and 3.2. Furthermore, issues of natural attenuation, risk assessment at contaminated sites (including megasites), remediation and prevention are extensively described in Chapters 5.4, 5.6, 5.7, 7.1–7.3).

1.3.3.2

Urbanisation and Related Discharges

Urbanisation introduces many changes to the aquifers that lie under cities. Natural recharge mechanisms are modified or replaced and new ones are introduced. Leakages and seepages from mains water and sanitation systems become an important part of the hydrological cycle in the urban environment. In this respect, many sub-city aquifers are polluted with human wastes,


11

particularly where there is insufficient connection to mains sewerage. In Europe, the situation has improved with the implementation of the Urban Wastewater Treatment Directive, but in many developing countries, septic tanks, cesspits and latrines are still common in major cities. Septic tanks, when properly operated, produce an effluent of acceptable quality in areas of low population density. In practice, they are, however, often overloaded and operate inefficiently. Effluent is often discharged directly into inland waterways, whence pollutants find their way into the underlying aquifer.7 This pollution leads to increasing occurrence of pathogens in groundwater (in particular helminths, protozoa, bacteria and viruses), which may have a direct impact on the bacteriological quality of water abstracted for human consumption (in particular when drinking water is provided by shallow private boreholes with insufficient sanitary controls). Domestic effluents are also responsible for increasing nitrate concentrations in groundwater. It should be noted, however, that while sewage and urban wastewater is generally regarded as a major source of pollution, it is also considered as a large and important resource, i.e. in many arid areas, it is used with minimal, if any, treatment to irrigate crops, including some intended for direct human consumption.7 The water used also supplies crops with essential elements such as nitrogen and phosphate which would otherwise have to be added as artificial fertiliser. This re-use is often debated as regard to its safety, and many experts consider that a better use for urban wastewater is probably to recharge the aquifer from which it came. During the recharge process, the water is considerably purified. If it is required for irrigation, it can then be abstracted either from irrigation wells or from streams whose flows has been increased by the recharge. Letting sewage water stand in shallow surface ponds and filter down through the soil and the aquifer below can be an effective means of treatment. The more slowly this is done, and the more that the surface ponds are rested between treatments, the more complete will be the treatment. Allowing ponds to dry out regularly leads to a breakdown of nitrates in the sewage, with the release of harmless nitrogen gas. With careful control, nitrogen concentration in the recharge water can be reduced to below 5 mg L 1. At the same time, most bacteria and protozoa are eliminated, and levels of organic compounds and phosphates are greatly reduced. This infiltration treatment, although it uses land areas, presents the advantage of providing a cheap underground storage system from which water can be pumped for non-potable uses.7 Solid waste disposals represent another source of major urban groundwater pollution. The worst risks occur where uncontrolled tipping, as opposed to controlled sanitary landfill, is practised, and where hazardous industrial wastes, including drums of liquid effluents, are disposed of at inappropriate sites which are selected on the basis of their proximity to where the waste is generated rather than their suitability as landfill sites. Often no record is kept of the nature and quantity of wastes disposed of at a given site and abandoned sites represent a potential hazard to groundwater for decades. To make matters worse, disposal is often on low ground where the water table is high and direct contamination of shallow groundwater likely.7

12

1.3.3.3

Chapter 1

Industry

Nearly all industries produce liquid effluents, which according to legislation have to be properly treated before they are allowed to be discharged to a water course. There are, however, still many illegal discharges (in particular from small industries producing paper and textiles, processing leather, metals and other materials and repairing vehicles, as well as small service industries such as metal workshops, dry cleaners, photo processors, etc.), e.g. of acids, oils, fuels and solvents which have a direct impact on water courses, particularly canals, or disposed into the ground and finding their ways to groundwater. Chlorinated solvents are particularly insidious pollutants because of their persistency, toxicity and the way they travel in aquifers. In this respect, many groundwater supplies are contaminated by such substances, a common cause of which is leaking storage tanks. Unfortunately, cleaning up a polluted aquifer— usually by removing contaminated soil and continuous pumping of the aquifer—is extremely difficult, very costly and takes a great deal of time.7 Industrial effluents also often contain high levels of metals such as iron, zinc, chromium and cadmium, many of which are highly toxic, even carcinogenic. A specific industrial pollution is related to mining and petroleum extraction. Quarrying and open-case mining, for example, remove the protective layer above an aquifer, leaving it more vulnerable to pollution. Deep mines or oil fields may produce fluids that are disposed of at the surface and may therefore pollute shallow aquifers, and pollutants from spoil heaps may leach into groundwater. Finally, rising water levels in abandoned mines produce acid mine drainage with subsequent mobilisation of oxidised metal ores, leading to increasing concentrations of sulfate, iron, manganese and other metals, which can cause serious groundwater pollution. Details on risk assessment of industrial pollution can be found in Chapters 5.6 and 7.1.

1.3.3.4

Agriculture

Agriculture is responsible for one of the main pollution groundwater threats. The main source arises from the intensive use of nitrogen-rich fertilisers and of pesticides, a problem that has spread from the industrialised countries to developing ones. The high levels of nitrate and, in some areas, pesticides in groundwater are clearly linked to agricultural activities. This pollution is generally worse where the soil is very permeable, allowing agricultural chemicals to be quickly washed down to underlying aquifers. However, not all nitrates in groundwater are due to agriculture, as we have seen that much of it also originates from untreated sewage. Monitoring techniques can hardly distinguish between the different agricultural and sewage nitrates (besides isotopic measurements). The leaching of nitrate from fields not only leads to pollution but is also a serious source of waste: nitrate that percolates down into aquifers has done nothing to stimulate plant growth. Other components of fertilisers, including potassium and chloride, also find their way from fields to aquifers.


13

Regarding pesticides and herbicides, substances currently in use are designed to be toxic and, sometimes, persistent. There is no doubt that pesticides are leached through the soil and carried down to underlying aquifers (see Chapter 9.2). In some circumstances, soils can adsorb or immobilise a large fraction of such agricultural chemicals. Many pesticides and herbicides, however, break down slowly under aquifer conditions and, as a result can persist over long time periods. In any case, groundwater pollution data are generally scarce, and the extent of pollution in Europe is hence not accurately known. Further considerations on diffuse groundwater impacts from agricultural land use are discussed in Chapters 7.3 and 8.2.

1.4 Groundwater Risk Assessment: Implications for Policy 1.4.1

Groundwater Protection Needs

Public and policy-maker perceptions of groundwater represent an important root cause of emerging problems.8 In many cases, regardless of the degree of formal education individuals have had, perceptions of groundwater resource dynamics are partial at best. Groundwater is often viewed, for example, as an inexhaustible resource, cleaned by the filtering action of aquifers and held, as in a ‘‘bowl’’ or ‘‘lake’’, or ‘‘underground river’’. These perceptions do not reflect reality, and often result in use patterns that cause unanticipated problems. Most misunderstandings relate to the scale of aquifer systems, the distribution of groundwater within them and the timescales on which groundwater systems function (see Chapter 5.5). Groundwater protection relies on two closely interlinked components:7 (i) land surface protection, based on hydrogeological concepts and information particularly regarding aquifers and vulnerability; and (ii) groundwater protection responses for potentially polluting activities, giving guidelines on the acceptability of the activities, investigation requirements and, where appropriate, the likely planning or licensing controls. Groundwater protection schemes enable authorities to take account of (i) the potential risks to groundwater resources and sources; and (ii) geological and hydrogeological factors, when considering the control and location of potentially polluting activities.7 Although groundwater is one of the world’s key natural resources, it is still not sufficiently well protected and poorly controlled. In many instances, the extent of groundwater pollution still remains to be evaluated or even detected, given the slow rates of groundwater movement and the volume of storage involved. In 1996, UNEP indicated that what we know of pollution levels in aquifers may only be the tip of an underground iceberg,6 which consideration is still valid ten years later. There is, in any case, no doubt that this important resource needs to be better protected from both over-exploitation and pollution. This is the aim of the developing groundwater legislation which is described in Chapter 3.1.

14

1.4.2

Chapter 1

Assessment, Prevention and Control

Assessment and appropriate controls of major threats are highly necessary particularly in areas subject to irreversible side effects such as saltwater intrusion and land subsidence (in the case of over-abstraction), but also in the case of high aquifer vulnerability, taking into account pollutant loads (in the case of pollution). Vulnerability assessments and controls require a number of legal and administrative steps, examples of which are described in the light of the EU Water Framework Directive and associated Groundwater Directive (see Chapter 3.1). In the case of pollution, there is a need to distinguish between point sources of pollution, such as landfills and specific industrial discharges, and diffuse sources of pollution linked to agricultural activities and to a lesser extent atmospheric deposition. Efforts should be made to reduce pollution from pollution sources, paying particular attention to practices in areas where aquifers are highly vulnerable. This means that land-use planning regulations have to be enforced in the most sensitive areas. This also concerns of course agricultural practices (see Chapter 8.2). A key step in assessing pollution risks is based on the analysis of the groundwater vulnerability (ease with which groundwater may be contaminated by human activities). It depends on the time of travel of infiltrating water (and contaminants), the relative amount of contaminants and the contaminant attenuation capacity of the geological materials through which the water and contaminants infiltrate (see further discussion in Chapters 5.4 and 9.2). As all groundwater is hydrologically connected to the land surface, it is the effectiveness of this connection that determines the relative vulnerability to contamination. Groundwater that readily and quickly receives water (and contaminants) from the land surface is considered to be more vulnerable than groundwater that receives water (and contaminants) more slowly and in lesser quantities. The travel time, attenuation capacity and quantity of contaminants are a function of the following natural geological and hydrogeological attributes of any area: (i) the subsoils that overlie the groundwater; (ii) the type of recharge, whether point or diffuse; and (iii) the thickness of the unsaturated zone through which the contaminant moves. In general, little attenuation of contamination occurs in bedrocks via fissures, and high attenuation will be possible in subsoils (sands, gravels, glacial tills (or boulder clays), peat, alluvial silts and clays). Groundwater is most at risk where the subsoils are absent or thin and, in areas of karstic limestone, where surface streams sink underground at swallow holes. Vulnerability may be mapped according to the elements in Table 1.1. Establishing such maps is an important part for deciding upon a groundwater protection scheme and an essential element in the decision-making on the location of potentially polluting activities.7 Firstly, the vulnerability rating for an area indicates, and is a measure of, the likelihood of contamination. Secondly, the vulnerability map helps to ensure that a groundwater protection scheme is not unnecessarily restrictive on human economic activity. Thirdly, the vulnerability maps help in the choice of preventive measures and enable developments, which have a significant potential to contaminate, to be located

15


Vulnerability mapping guidelines (adapted from Ref. 7).

Table 1.1

Hydrogeological conditions Vulnerability rating

Extreme High Moderate Low

Subsoil permeability (type) and thickness

Unsaturated zone

Kast features

High permeability (sand/gravel)

Moderate permeability (e.g. sandy subsoil)

Low permeability (e.g. clayey subsoil, clay, peat)

(Sand/gravel aquifers only)

(o30 m radius)

0–3 m 43 m N/A N/A

0–3 m 3–10 m 410 m N/A

0–3 m 3–5 m 5–10 m 410 m

0–3 m 43 m N/A N/A

– N/A N/A N/A

Risk to Groundwater

Hydrogeological Factors

Vulnerability to contaminants

Figure 1.1

Groundwater Value

Other Factors

Contaminant Loading

Preventive Measures

Factors under consideration for groundwater risk assessment.

in areas of lower vulnerability. This assessment strongly relies on proper characterisation of groundwater settings and related risks, an issue that is discussed in Chapter 5.1. In this respect, the risk depends on (i) the hazard afforded by a potentially polluting activity, (ii) the vulnerability of groundwater to contamination, and (iii) the potential consequences of a contamination event. The hazard depends on the potential contaminant loading. The natural vulnerability of the groundwater dictates the likelihood of contamination if a contamination event occurs. The consequences to the target depends on the value of the groundwater, which is normally indicated by the aquifer category (regionally important, locally important or poor) and the proximity to an important groundwater abstraction source (e.g. a public supply well). The risk assessment encompasses geological and hydrogeological factors and factors that relate to the potentially polluting activity. This is illustrated in Figure 1.1. In the light of the above, prevention of groundwater contamination is of critical importance and must be a key aim for the following reasons:7 Once groundwater contamination occurs, the consequences last far longer than surface water contamination (months, years and sometimes

16

Chapter 1

decades) because groundwater moves slowly through the soil and unsaturated zone of the aquifer. Remediation is frequently not practical or is very expensive. Also, it is both impractical and a poor environmental strategy to provide comprehensive treatment to remove certain pollutants, such as pesticides and other trace organics. It is therefore preferable to prevent or reduce the risk of groundwater contamination than to deal with its consequences. Groundwater is an important resource used for drinking water, industry and agriculture, and should be protected for present and future usage. Groundwater provides the base flow (i.e. the water which feeds rivers year-round, and upon which flood flows are superimposed) to surface water systems, many of which are used for water supply and recreational purposes.

1.4.3

Monitoring

The assessment and controls of groundwater quantity and quality have to be supported by representative and reliable monitoring data. Groundwater monitoring is largely a national responsibility but, because groundwater does not respect national boundaries, monitoring has to be conceived at national, regional and international levels. This is exactly the principle of the Water Framework Directive (see Chapter 3.1). Improved monitoring systems are needed to provide information not only on groundwater quantity and quality in the light of its use as drinking water resources, but also for the evaluation of its environmental quality in relation to associated aquatic and directly dependent terrestrial ecosystems. In particular, existing monitoring networks hardly provide early warning of pollution, and networks are needed to include monitoring of pollution loads, particularly in vulnerable recharge areas. Recommendations have been given by UNEP in this respect6 (see Table 1.2). Groundwater quality monitoring has at least four objectives that need to be carefully distinguished in the design of monitoring systems: definition of the extent of groundwater pollution (analysis of pressures and impacts); quality control of groundwater used as drinking water (drinking water supply surveillance); early discovery of groundwater pollution from a given activity (offensive detection monitoring); and provision of advance warning of the arrival of polluted water at important sources of supply (defensive detection monitoring). Detailed monitoring guidelines in support of policy implementation, more specifically to the EU Water Framework Directive and new Groundwater Directive (see Chapter 3.1), have been developed and are summarised in Chapter 6.1.

a

FC: faecal coliforms.

Chlorinated hydrocarbons, other organic compounds, metals

Cl, NO3, NH4, SO4, FC

Domestic wastewatera Industrial effluent

Pollution indicators

High

Bacteriological Monitor at water table

High

Chemical

Pollution risk

Early warning monitoring required

Hours to weeks

Fractured

Unconfined

Aquifer type

Cl, NO3, NH4, SO4, FC Metals, range of organic compounds

Monitor at water table

Moderate

High for mobile compounds

Days to months

Thin unsaturated zone

Granular

Persistent organics (metals)

Cl, NO3, NH4, SO4, FC

(1) Monitor unsaturated zone and (2) at water table

Low

High for mobile and persistent compounds

Years to decades

Deep unsaturated zone

Key elements in an early warning monitoring strategy (adapted from Ref. 6).

Travel time from surface to saturated aquifer zone

Table 1.2

Persistent organics (metals)

Cl, NO3, NH4, SO4

Monitor semiconfining layer and aquifer

Moderate for persistent compounds only Very low

Decades +

Semi-confined

General Introduction: The Need to Protect Groundwater 17

18

Chapter 1

1.5 Conclusions Groundwater protection is essential in relation to the intrinsic water resource quality for various uses, and its environmental value. Basic environmental principles include the following: Sustainable development principles, seeking to ensure that economy and society can develop to their full potential within a well-protected environment, and with responsibility towards present and future generations and the wider international community. Precautionary principle, requiring that emphasis should be placed on dealing with the causes, rather than the results, of environmental damage and that, where significant evidence of environmental risk exists, appropriate action should be taken even in the absence of conclusive scientific proof of cause. Polluter pays principle, of which the objective is to allocate correctly the costs of pollution, consumption of energy and environmental resources, and production and disposal of waste to the responsible polluters and consumer, rather than to society at large or future generations, which in turn provides an incentive to reduce pollution and consumption.

References 1. J. Gibert, in Groundwater Ecology: A Tool for Management of Water Resources, ed. C. Griebler, D. L. Danielopol, J. Gibert, H. P. Nachtnebel and J. Notenboom, European Commission, EUR 1987, 2001, p. 413. 2. S. Heidrich, M. Schirmer, H. Weiss, P. Wycisk, J. Grossmann and A. Kaschl, Toxicology, in press. 3. P. Grathwohl, Diffusion in Natural Porous Media, Contaminant Transport, Sorption/Desorption and Dissolution Kinetics, Kluwer Academic, 1998. 4. G. Teusch, H. Ruegner, D. Zamfirescu, M. Finkel and M. Bittens, Land Contamin. Reclam., 2001, 9, 1. 5. EEA, Groundwater Quality and Quantity in Europe, Environmental Assessment Report 3, Copenhagen, 2000. 6. UNEP, Groundwater: a threatened resource, UNEP Environment Library 15, Nairobi, Kenya, 1996. 7. Irish EPA, Groundwater Protection Schemes, 1999 (ISBN 1-899702-22-9). 8. J. J. Burke and M. H. Moench, Groundwater and Society: Resources, Tensions and Opportunities, United Nations, 2000 (ISBN 92-1-104485-5).

2. Science–Policy Integration Needs

CHAPTER 2.1

Science–Policy Integration for Common Approaches Linked to Groundwater Management in Europew PHILIPPE QUEVAUVILLER European Commission, DG Environment (BU9 3/142), Rue de la Loi 200, BE-1049 Brussels, Belgium

2.1.1

Introductory Remarks on Science–Policy Integration Needs

It is now well recognised that a better understanding of environmental problems requires an improved awareness of multidisciplinary scientific developments.1 Awareness by itself is, however, not sufficient, and a better research integration is also required at the various stages of policy developments (design, development, implementation and review). Ideally, the relevant research for any environmental policy should be feeding the policy-making process directly in a ‘‘tailor-made fashion’’ so that results may be used in the right way at the right time (in relation to the policy agenda). In many instances, however, this is still far from being the case. In the sector of European Union (EU) water policies, on-going discussions are taking place among the scientific community (representatives of research consortia and of the European Commission’s Research Directorate-General) and the policy-making community (representatives of EU member states’ environment agencies or ministries and the European Commission’s Environment Directorate-General) to examine research–policy coordination needs. These discussions have highlighted the key importance of improving/increasing the information and communication flow within and between these w

The views expressed in this chapter are purely those of the author and may not in any circumstances be regarded as stating an official position of the European Commission.

21

22

Chapter 2.1

communities.2 On the one hand, the implementation of existing policies and the development of new ones are constantly fed by scientific inputs, which must fulfil specific thematic and timing requirements in order to be of direct value to the policy-making process. On the other hand, research and technological development (RTD) activities aiming to support relevant environmental policies have to take into account the policy-making agenda and its specific needs to adapt the objectives and the scope of their related work programmes. Experience has shown, however, that this interrelationship is not as efficient as it could/should be, owing to a lack of a clear coordination mechanism. In fact scientists view the end-user in the research project as the ‘‘legitimate’’ client for their research results, but on the ground there is a significant lack of transfer mechanisms that would allow passing on of the relevant information to other stakeholders or policy makers.2 The latter often do not have the time, the capacity to translate research results into policy, or even simple access to specific technical journals and thus relevant information mostly remains within the specialised scientific community. Communication difficulties are also linked to the different ‘‘jargon’’ used in the different communities. One problem hampering proper science–policy integration is related to different timing concepts of research and policy. Policy tends to operate on the short or medium term with clearly defined milestones, while science is generally developed on a long-term basis. In addition, policy tries in most instances to achieve an acceptable (political) compromise (which is, sometimes somewhat critically, labelled as the ‘‘least common denominator’’), whereas the scientific community strives to obtain objective, scientific facts and wants to understand a phenomenon to the greatest possible detail. In addition, scientific career reviewing schemes rarely give credit to works for the integration of knowledge to fulfil policy objectives. As a result, scientists are not sufficiently motivated to perform this type of work and in most cases they may not be aware about the specific research issues needed at a certain time to effectively support policy development.2 In many instances, knowledge is available but is either not widely known or is insufficiently transferred from the scientific community to the other relevant parties. Furthermore, policy-makers, planners and implementers are often not sufficiently ‘‘entrepreneurial’’ with respect to the selection and use of state-of-the-art research projects that would directly feed the policy objectives (which is essentially due to time pressures and lack of streamlined information). As is discussed below, this is often due to a lack of dialogue or an inappropriate communication of relevant information. In the best case, policy-makers are able to define a list of open questions and then, on this basis, open a discussion with the scientific community in order to understand if the questions are addressing the actual research needs or whether the subject matter has already been treated sufficiently in the past so that responses may be obtained from readily available results. Such a mechanism, however, is certainly not straightforward and often not sufficiently operational. The development of environmental policies mixes legal requirements with issues of technical feasibility, scientific knowledge and socioeconomic aspects and requires intensive multi-stakeholder consultations. In this context, the

23

Science–Policy Integration for Common Approaches POLICY IMPLEMENTATION POLICY DEVELOPMENT

RESEARCH, SCIENTIFIC PROGRESS, POLICY INTEGRATION

POLICY REVIEW

DESIGN OF POLICY

Figure 2.1.1

Integration of scientific progress into the policy-making process.

consideration of scientific progress represents one of the key aspects for the design of new policies and the review of existing ones. Within the EU, this consideration is fully embedded into the Sixth Environmental Action Programme which stipulates that ‘‘sound scientific knowledge and economic assessments, reliable and up-to-date environmental data and information, and the use of indicators will underpin the drawing-up, implementation and evaluation of environmental policy’’.3 This requires, therefore, that scientific inputs constantly feed the environmental policy process. This integration also involves various players, namely the scientific and policy-making communities but also representatives from industry, agriculture, non-governmental organisations etc. (Figure 2.1.1). The problem of non-coincidence of research and policy agendas depends upon the stage of development of the policy. It is more acute at the starting phase of policy design such as thematic strategies defined by the EU 6th Environment Action Programme (e.g. Soil Thematic Strategy).3 In comparison, the Water Framework Directive (WFD),4 along with its related Common Implementation Strategy (see Chapter 4.1), provides a stable platform which allows the building of strong partnerships among policy and scientific communities. In the context of the WFD, we may distinguish R&D needed in support of policy development (e.g. background information required for designing ‘‘daughter directives’’ such as the new Groundwater Directive; see Chapter 3.1) and R&D directly feeding implementation (short- or medium-term research and demonstration projects supporting specific milestones such as e.g. monitoring, integrated management). Long-term R&D is of course also necessary for reviews and possible improvement of the legislation (such as the technical adaptations covered by the WFD and the review periods under the River Basin Management plans). At the present stage, efforts are lacking for presenting results of research and demonstration projects in a form that policy-makers can easily use, e.g. ‘‘science-digested’’ policy briefs. Adversely, one may stress that the consideration of research results by the policy-making community is not straightforward, mainly for political reasons and difficulties in integrating the latest research developments in legislation. The difficulty is enhanced by the fact that the policy-making community is probably not defining its role as ‘‘science

24

Chapter 2.1

customer’’ sufficiently well. In other words, the dialogue and communication are far from being streamlined to allow for an efficient flow of information. In this respect, improvements could be achieved through the development of a ‘‘science–policy interface’’ based on a coordination of relevant programmes/ projects with direct relevance to the WFD implementation. This issue is discussed in Chapter 11.3. Integration in a broad sense, as perceived for the environmental policy sector, goes of course beyond the science–policy issues. It concerns the need to consider the environment as it appears in all relevant policies, interactions of various environmental compartments, socioeconomic aspects, etc. This chapter actually focuses on integration of scientific and technological progress into the policy-making and implementation process, in particular in the groundwater policy sector, illustrating the necessity and complexity of the knowledge-based approach. The above considerations are largely inspired by recent papers discussing this issue.2,5

2.1.2

Science Integration in the Light of Groundwater Management

When dealing with groundwater science and policy, the interdependent nature of hydrological, hydrogeological and water-use systems makes the continuum between data, information and knowledge of particular importance.6 Understanding systemic interactions is essential as a basis of identifying groundwater management options and generating sufficient social consensus to implement them. Developing this understanding requires a steady flow of hydrological data as well as data on water use. It also requires continuous refinement of the scientific foundation, both physical and social, upon which policy and management solutions rely. Examples of this type of progressive approach to groundwater information needs are quoted by Burke and Moench:6 e.g. the Groundwater Forum in the UK7 and the Groundwater Foundation in the USA represent initiatives to address, respectively, groundwater research needs and public education. Another example is the Working Group on Groundwater of the Common Implementation Strategy of the WFD, which is described in Chapter 4.1. Lack of data and scientific understanding of groundwater resources often represents a critical gap undermining the development of groundwater management approaches and institutions.6 The absence of data often limits the degree to which hydrogeologists are able to quantify and describe complex aquifer dynamics. Equally important are the ways in which raw data and information are treated, presented and used. Information is only useful if it used. For this to occur, the information must be accessible to potential users and presented in a manner they can understand. This is discussed in Chapter 11.3 with regard to the development of the EU Water Information System for Europe (WISE). In many instances, however, the absence of information (or the non-use of essential information) creates situations in which emerging

Science–Policy Integration for Common Approaches

25

problems and management options are poorly understood. As a result, the essential nature of basic data and research on hydrogeology should be clear. Generating the technical information required to meet emerging groundwater management needs depends on at least three types of scientific data collection and analytical activities:6 long-term baseline monitoring for understanding the dynamics of hydrogeological systems and providing warning of emerging problems; targeted research on basic processes; and site-specific analysis of problems and management options at local levels. Technicians often make a strong plea for more data, and when they obtain them, they frequently present tentative, difficult-to-interpret scenarios and request financing to collect more data in order to strengthen their interpretations.6 Policy-makers and the general public, on the other hand, often have unrealistic expectations regarding the ability of technical and scientific analyses to provide straightforward answers, particularly given the general paucity of groundwater data and the limited means with which to gather and analyse them. Hydrogeological systems are rarely simple. The importance of bridging this gap between the data available and their policy implications relates directly to some of the fundamental challenges facing the development of management institutions.

2.1.3

Examples of Projects in Support of Groundwater Policy

As highlighted in Section 1, the relevant research for any environmental policy should ideally be feeding the policy-making process directly in a ‘‘tailor-made fashion’’ so that results may be used in the right way and at the right time (in relation to the policy agenda).2 In many instances, however, this is far from being the case. This paragraph lists some RTD projects directly or indirectly related to EU groundwater policy and examines how they were related to policy implementation (Directive 80/68/EEC8 and/or Water Framework Directive) or development (proposal for the new Groundwater Directive described in Chapter 3.1).9 Details on the RTD funding mechanisms and on the priorities under which the projects discussed below have been funded are given in Chapter 11.3. It should be noted that the examples given below do not prejudge about the way project outputs were effectively transferred to the user community (an impact assessment analysis has not been done so far to enable this issue to be concretely discussed), nor do they pretend to provide an exhaustive list of ongoing and terminated projects.

2.1.3.1

Risk Assessment

Research supporting improvements of groundwater (pollution) risk assessment has been flourishing over the past few years (from 1998 onward). Examples are

26

Chapter 2.1

illustrated in several chapters of this book. Most of these projects directly or indirectly support groundwater characterisation needs as well as the future programmes of measures required under the WFD.4 They concern, for example, studies of contaminant spreading in fractured underground reservoirs leading to conceptual geological modelling and database development for organic pollutants in groundwater (TRACE Fracture project; contact: [email protected]) (see Chapter 5.6); and groundwater risk assessment at contaminated sites leading to guidelines (GRACOS project; contact: [email protected]).

2.1.3.2

Groundwater Remediation

Examples of projects of direct applicability to groundwater remediation concern, for example, integrated concepts for groundwater remediation, including guidelines and studies on natural attenuation of organic pollutants in groundwater (INCORE project; contact: [email protected]) (see Chapter 5.8); on-site remediation of groundwater contaminated by polar organic compounds using a new adsorption technology (OROGONATE project; contact: [email protected]); and protection of groundwater at industrially contaminated sites (PURE project; contact: [email protected]).

2.1.3.3

Diffuse Pollution

Linked to groundwater analyses of pressures and impacts, some projects examine issues such as, for example, actual status and scenarios of pesticides in European groundwaters (PEGASE project; contact: [email protected]), integrated soil and water protection from diffuse pollution (SOWA project; contact: [email protected]), etc. The two projects are described, respectively, in Chapters 9.2 and 5.7.

2.1.3.4

Management Issues and Information Tools

R&D projects also support groundwater management and information tools that are directly or indirectly in support of the WFD river basin management planning. Examples concern, for example, the development of an online portal in the form of a web-based information platform for soil, groundwater and contaminated land (EUGRIS project; contact: [email protected]); integrated management system (IMS) for the prevention and reduction of contamination at large-scale contaminated sites (WELCOME project; contact: [email protected]) (see Chapter 7.1); tools for management and control of hazardous compounds in water catchment areas (WATCH project; contact: [email protected]) (see Chapter 8.3); etc. The scientific basis for improved river basin management through a better understanding of the river– sediment–soil–groundwater system as a whole, at different temporal and spatial scales, is studied in an integrated project, the AQUATERRA project (contact: [email protected]) (see Chapter 2.2).


2.1.3.5

27

Relevant Networks

Besides research projects, networks involving the scientific community, industrial stakeholders, representatives of international associations, NGOs, etc. are directly or indirectly contributing to networking activities that are relevant to groundwater policy development and implementation. Examples of such networks provide exchange of innovative know-how in the field of applied research for contaminated land and groundwater issues (ANCORE network: www. ancore.org); develop technical recommendations for sound decision-making on the rehabilitation of contaminated sites in Europe (CLARINET network: www.clarinet.at); and discuss issues of industrially contaminated land in Europe (NICOLE network: www.nicole.org). Other networks are focused on training, e.g. on innovative management of groundwater resources in Europe (IMAGETRAIN project: www.image-train.net), or on the coordination of national research programmes, e.g. on sustainable management of soil and groundwater under the pressure of soil pollution and soil contamination (SNOWMAN project; contact: [email protected]) (see Chapter 11.1).

2.1.4

A Project ‘‘Tailor-Made’’ to Support the New EU Groundwater Directive: BRIDGE

The BRIDGE (background criteria for the identification of groundwater thresholds) project was designed to develop a common methodology, intended for possible use by member states, on ‘‘how to derive groundwater threshold values’’. The project was developed in 2004–2006 and has just been concluded. It has been carried out at European level, involving a range of stakeholders and efficiently linking the scientific and policy-making communities. The different objectives were: To evaluate and assemble scientific outputs to set out criteria for the assessment of the chemical status of groundwater. These criteria are data for characterisation of natural and anthropogenic pollutants, parameters indicative for pollution, data for characterisation of groundwater bodies as hydrologic and hydrogeological parameters. To derive a plausible general approach, how to structure relevant criteria appropriately with the aim to set representative groundwater threshold values scientifically sound and defined at national river basin district or groundwater body level. To check the applicability and validity of this approach by means of case studies at the European scale, and to carry out an environmental impact assessment taking into account the economic and social impacts. The final methodology for the derivation of environmental thresholds for pollutants at a national or regional level of groundwater bodies or river basins has been developed in close consultation with representatives from member

28

Chapter 2.1

states’ environment ministries and agencies, and stakeholders from the CIS Working Group on Groundwater. It had to take due considerations of the negotiation of the new Groundwater Directive which was running at the same time, which represented an additional challenge above the sole scientific one. The final meeting was held in Paris on the 15 December 2006. The proposed method for deriving groundwater threshold values will now be directly communicated to the member state experts for policy discussions and expected adoption before summer 2007. The research will therefore fulfil one of the requirements of the new Groundwater Directive, requiring member states to establish groundwater threshold values by the end of 2008, following a common methodological approach. More details on the project are given in Chapter 9.1. The project was terminated at the end of 2006 and a CD-ROM provides a copy of the different findings and workpackage reports to interested stakeholders. Contact: AnneMarie Fouillac (email: [email protected]).

2.1.5

Conclusions: Some Research Needs

Some basic research needs in support of groundwater management and policy have been expressed by Burke and Moench6 and are outlined below. Basic hydrological and hydrogeological data collection and research, including long-term monitoring of groundwater (levels and quality) and surface water (flows and quality); the synthesis of background information (hydrogeological and socioeconomic data); and detailed investigations in areas where indicators signal the emergence of specific problems. Operational and basic research activities are also essential in order to document and evaluate the results of management attempts and in order to understand key elements of the hydrogeological processes that underpin various management options. Monitoring the resource use and potential sources of pollution. Basic information on how groundwater resources are being used and who is using them is as essential for management as scientific data on the aquifer system itself. Without understanding how groundwater resources are being used, particularly for irrigated agriculture, it is impossible to identify points of management leverage. Devising management systems requires detailed knowledge of such factors as the actual locations where groundwater extraction is occurring, the efficiency with which it is used and its role in agriculture and other water-use systems. Programmes for registering wells and estimating, either directly or indirectly, the amount of groundwater extracted are essential steps. Similarly, it is important to identify, as far as possible, other activities that could have substantial impacts on groundwater conditions. This includes the basic information on potential point and non-point sources of pollutants arising from industry, domestic sewage and agricultural chemical use.


29

Besides these research needs, which are still valid at the time of publication of this book, other issues have been highlighted during the development of the new Groundwater Directive, in particular: research on groundwater ecosystems (see Chapter 11.2), and on interactions among groundwater and associated aquatic and dependent terrestrial ecosystems to strengthen the knowledge about environmental groundwater quality; research on groundwater conceptual modelling, georeferencing system, visualisation, vulnerability assessment, enabling to better assess quantitative and qualitative pressures on groundwater systems, as well as the effectiveness of programmes of measures; improved risk assessment and characterisation methods (e.g. support to standardisation of assessment methods for pollutant mobilisation from different types of pollution sources), integrated management approaches, innovative measures for a better protection of groundwater against pollution, etc.; R&D linked to improved programmes of measures, including integration of requirements of different regulations (see Chapter 3.1), and considering identified pressures and the way to tackle them with technically feasible, cost-effective and socially acceptable measures complying with current legislation; and development of new monitoring methods (e.g. in situ techniques, field monitoring) and improvement of data quality and comparability, and pre-normative research related to operationally defined operations (e.g. sampling) and parameters (e.g. extractable contents of chemical substances), etc. The EU 7th Framework Programme of Research and Technological Development10 will take some of these considerations into account in the Theme 6 ‘‘Environment’’. Research outputs will then need to be efficiently transferred to policy-makers, along the principles described in Chapter 11.3.

References 1. UK Department for Environment, Food and Rural Affairs, Science Meets Policy in Europe, DEFRA Report, London, 2005. 2. Ph. Quevauviller, P. Balabanis, C. Fragakis, M. Weydert, M. Oliver, A. Kaschl, G. Arnold, A. Kroll, L. Galbiati, J. M. Zaldivar and G. Bidoglio, Environ. Sci. Pol., 2005, 8, 203. 3. European Commission, 6th Environment Action Plan 2001–2010, 2001. 4. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy, Official Journal of the European Communities, L 327, 22.12.2000, p. 1. 5. Ph. Quevauviller, J. Environ. Monitor., 2005, 7(2), 89.

30

Chapter 2.1

6. J. J. Burke and M. H. Moench, Groundwater and Society: Resources, Tensions and Opportunities, United Nations, 2000 (ISBN 92-1-104485-5). 7. D. R. C. Grey et al., Groundwater in the United Kingdom. A Strategic Study. Issues and Research Needs. Groundwater Forum Report FG/GF 1, Foundation for Water Research, Marlow, UK, 1995. 8. Council Directive 80/68/EEC of 17 December 1979 on the protection of groundwater against pollution, Official Journal of the European Communities, L 20, 26.1.1980, p. 43. 9. Directive of the European Parliament and of the Council on the protection of groundwater against pollution and deterioration, 2006. 10. 7th Framework Programme for Research and Technological Development (2006–2013), European Commission, rue de la Loi 200, B-1049 Brussels, 2006.

CHAPTER 2.2

Transferring Scientific Knowledge to Societal Use: Clue from the AQUATERRA Integrated Project PHILIPPE NE´GREL,a DOMINIQUE DARMENDRAILa AND ADRIAAN SLOBb a

Bureau de Recherches Geólogiques et Minie`res (BRGM), 3 avenue Claude Guillemin, FR-45060 Orleáns ce´dex 2, France; b TNO Environment and Geosciences, Dept Innovation & Environment, Van Mourik Broekmanweg 6, PO Box 49, NL-2600 AA Delft, The Netherlands

2.2.1

Introduction

Scientific knowledge and data are of significant importance in a global information society. They should both be used for promoting innovation and economic development, for efficient and transparent decision-making, particularly at the governmental level, and for education and training. However, scientific data and information should be as widely available and affordable as possible. The more people are able to share them, the greater the positive effects and returns to society will be. Scientific knowledge is a ‘‘public good.’’ The inherent function of scientific investigations is to carry out a comprehensive questioning of nature leading to further knowledge. It is this new knowledge that allows cultural and intellectual enrichment and leads to the technological advances and benefits arising from science. Promoting fundamental research is a priority towards achieving development and progress. There can be no applied science if there is no science to apply and thus science is for knowledge; knowledge is for progress as stated by the World Science Forum (2003; http://www.sciforum.hu/index.php). In the field of water resources management, policy plays a key role. The policy approach is becoming more holistic and the trend is to address the whole 31

32

Chapter 2.2

groundwater–river–sediment–soil system. This widens the quite narrow defined issue of groundwater to a variety of related issues, like nature conservation, the use of space, and economic and social issues. It also confronts groundwater policy with new stakeholders that were not involved in groundwater issues before. So new stakeholders with interests from nature conservation to entrepreneurial interests are entering the policy arena. The until now usual technical approach of the groundwater issue does not fit with these developments, so a new policy approach as well as new ways of involving science in policy are needed. River basins are facing different types of pressures, such as impacts of agriculture, industries or urban waster water treatments. The drivers–pressures–state–impact–response (DPSIR) approach as illustrated by Figure 2.2.1 gives insight into environmental processes and the links between human activities and their impact on the environment. Economic activities (driving forces), such as industry, agriculture and tourism, lead to increasing pressures on the natural environment as these activities result in use of natural resources and/or emissions (accidental or controlled) of waste to water (surface and/or groundwater), soil and sediment. The use of resources and/or emissions will change the state of these environments in quantity and/or quality: sediment, water and soil resources are depleted and/or they are loaded (contaminated) with hazardous substances originating from economic activities. Above a certain level of depletion and/or contamination the environment may be impacted, i.e. loss of biodiversity, vulnerability to floods and landslides, decreased chemical and/or ecological water, soil or sediment quality and/or a shortage of these resources. Thus the DPSIR framework provides a helpful, conceptual framework for further improvement of system understanding.

Figure 2.2.1

Pressures on river basins.

Transferring Scientific Knowledge to Societal Use

33

In the past, water quality studies and monitoring in the major river basins1 have developed according to: (i) the water demand, which is increasing exponentially; (ii) the development of key issues, such as eutrophication, acidification, salinisation; (iii) new pressures (radionuclides since the 1950s, pesticides since the 1980s, endocrine disruptors and pharmaceuticals more recently); and (iv) the development of new technical tools such as analytical chemistry, in situ monitoring or models. Management of river basins will certainly benefit from an increased scientific understanding of the functioning of the biophysical system (i.e. water–sediment–soil interactions) and of its relation to the societal system like the application of scientific knowledge in river basin management and in policymaking and implementation. This chapter deals with the preliminary results issued from the 6th Framework Programme (FP) Integrated Project AQUATERRA and the transfers from the scientific research area to a policy and societal use.

2.2.2

Overview of the AQUATERRA Project

The European Commission (EC) anticipated the need to improve our common understanding of the functioning of river systems, especially in relation to changes in land use and climate. Hence, under their RTD Framework Programmes several research projects have been funded, aimed at improving this understanding, and at supporting the Water Framework Directive (WFD) implementation. AQUATERRA, the full title of which is ‘‘Integrated modelling of river– sediment–soil–groundwater systems: advanced tools for the management of catchment areas and river basins in the context of global change,’’ is one of the first Integrated Projects within the 6th European Union (EU) Framework Programme (see website at http://www.eu-aquaterra.de/ for detailed information). AQUATERRA, including a multidisciplinary team of 45 partner organisations in 12 EU countries (researchers, but also practitioners and end-users such as policy-makers, river basin managers and regional and urban land planners), has been active since 1 June 2004 and will run until May 2009. AQUATERRA aims to provide the scientific basis for an improved river basin management through a better understanding of the river–sediment–soil– groundwater system as a whole, by integrating both natural and socioeconomic aspects at different temporal and spatial scales.2 This should be applicable to European contexts facing modifications or changes due to climate change, land use and pollution of soil and water. The principal task of AQUATERRA is to provide the foundations for an improved understanding of the behaviour of environmental pollutants in order

34

Chapter 2.2

to better evaluate the evolution of water quality at different scales, from local to global scales. New field and laboratory as well as historical data will be assembled and addressed in four European river basins (Ebro, Meuse, Elbe and Danube) and a small French catchment (Bre´villes). Based on these biogeochemical, climatological and material flux data, new simulation models will help to outline trends and pollutant transport behaviour with respect to soil functioning and the water cycle. These models will integrate key biogeochemical and hydrological processes from the laboratory to the river basin scale. AQUATERRA is divided in ten sub-projects with different functions to provide information and logistical support to each other as illustrated in Figure 2.2.2. Note that the whole project is presented in a flyer (available at http:// www.attempto-projects.de/aquaterra/uploads/media/Flyerfinal_13_10_05.pdf). Some sub-projects have a scientific focus (i.e. TREND, FLUX, COMPUTE, BIOGEOCHEM, BASIN, HYDRO, MONITOR) while others are more focused on the social sciences and European policies (i.e. EUPOL and INTEGRATOR). Finally, KNOWMAN has the role of knowledge transfer and dissemination. Some new developments have been done in INTEGRATOR and EUPOL sub-projects which combine hard (hydrogeology, chemistry, geochemistry, etc.) and soft (policy, socioeconomics) sciences. Using two of the five river basins studied in AQUATERRA as case studies, we will explore the new role of science in the societal information use and the development of the Environment Policy, and in particular the Water Resources legal framework.

Impact of Global Change on Soil and Water

TREND Future trends and impacts

HYDRO Global climate Water cycle

FLUX Intercompartement mass fluxes BIOGEOCHEM Key processes Transport functions

Figure 2.2.2

BASIN - Applications • Brevilles • Ebro • Meuse • Elbe • Danube EUPOL - Policies • EU policy framework • R & D requirements INTEGRATOR • Economic and social aspects • Stakeholder needs

Bench scale

MONITOR Screening tools Pollutants

Catchmentscale

COMPUTE Integrated soil-water numerical models

Basin scale

Scientific Methodology

KNOWMAN • Dissemination activities • Knowledge transfer

The ten sub-projects of AQUATERRA working on different levels and scales.


2.2.3

Environmental Policies

2.2.3.1

Four Generations of Environmental Policies

35

Environmental policies have been developed continuously from the 1960s, when environmental problems were recognised. Various authors have described the development of environmental policy.3–5 In this chapter we distinguish four generations of environmental policies that developed subsequently.6 These generations are all needed to tackle environmental problems and thus are not replacing or competing with each other but are complementary. The first generation started in the 1960s/1970s. Because of poor environmental conditions and health problems in some regions, an environmental consciousness awoke in the public, and governments adopted new sets of environmental regulations and rules. For the most part, environmental measures involved end-of-pipe techniques to reduce emissions. The focus of environmental policies in the first generation was on regulatory measures: laws and rules. At the beginning of the 1980s, awareness increased that regulatory measures on their own were not enough to tackle the environmental problems adequately. This cleared the way for voluntary measures. The second generation of environmental policies focused strongly on prevention of environmental damage by a mix of voluntary measures and regulations. The focus of these voluntary and regulatory measures was, for example, on energy saving and waste reduction. Environment had become a topic in quality and safety systems of businesses and environmental management systems had become quite popular. So in the second generation of environmental policies, the attention shifted from end-of-pipe techniques to process-integrated measures designed to prevent pollution being emitted to the environment. In the 1990s, attention shifted from prevention within the factory to prevention of emissions within the whole production–consumption chain. Product oriented policies and chain management are typical examples of the third generation of environmental policies. The environmental aspects of products were measured during their entire life cycle and incorporated in chain management and the product design process. Major efforts were put into the communication with external stakeholders concerning the environmental performance. The policy instruments that were being used were of a voluntary nature, sometimes supported by legislative and/or financial measures. In the beginning of the 2000s even new strategies were needed because persistent environmental problems still were present. Some problems like acidification had been tackled quite well, but more persistent problems, like the emission of CO2 or the spreading of chemicals in the environment, were harder to tackle. So in the beginning of the present century the need for a fourth generation of environmental policy was derived. Although this generation of environmental policies is still developing, we can see already contours of it. In the last Dutch Environmental Policy Plan 4 (2001) ‘‘transition management’’ is presented as a key concept to tackle persistent environmental problems. The policy approach is rather process-oriented instead of contentoriented and has an interactive character.

36

Chapter 2.2

Table 2.2.1 presents an overview of the four generations of environmental policies. When we examine the characteristics of the generations more closely, a clear trend becomes apparent. We can see an increasing complexity of the environmental problems which is reflected in an increasing complexity in environmental policies. Where in the first generation goals are set quite straightforwardly, legislative instruments are being used and the number of actors is limited, in the fourth generation goals are discussed and formed with societal actors over a long period, participation is the most important instrument and the number of actors is numerous. Whereas the policy in the first generation is directly aimed at cleaning the environment in a straightforward way, in the fourth generation this aim has shifted towards upgrading environmental quality by means of system innovations and societal change.

2.2.3.2

Science in the Four Generations of Environmental Policies

What does the development of environmental policies described above mean for knowledge production in the four generations of environmental policy? In the first generation the demand for knowledge comes only from government, and is quite technical in nature. In the fourth generation, the strategy is directed towards societal change to diminish environmental damage and the number of actors involved is quite numerous. The demand for knowledge comes now from different actor groups. Therefore, the new policy approach in the third, but especially in the fourth, generation will call for a new way of dealing with knowledge in the policy process. This fits well with the ideas of Gibbons et al.7 who reflected on the role of science in society. According to them, science is undergoing a major shift from mode 1 science—the traditional way of production of scientific knowledge—to mode 2 science. In this mode 2, the societal context is very important for knowledge production. Mode 2 science has five characteristics that distinguish it from mode 1 science (see Table 2.2.2). Mode 2 science is generated in the context of application, whereas mode 1 science is produced in the ‘‘classic’’ academic environment. ‘‘The context of application describes the total environment in which scientific problems arise, methodologies are developed, outcomes are disseminated, and uses are defined.’’8 Furthermore, mode 2 science is produced in a ‘‘trans-disciplinary’’ way, ‘‘by which is meant the mobilisation of a range of theoretical perspectives and practical methodologies to solve problems. But, unlike inter- or multidisciplinarity, it is not necessarily derived from pre-existing disciplines nor does it always contribute to the formation of new disciplines.’’8 Mode 1 science is produced in the academic setting (one place, a certain time), is organised in a hierarchical way and involves academic peers to control the quality of the output. Mode 2 science is developed close to the place of application, is organised in a heterarchical manner and quality control is performed by societal actors. Mode 2 science matches quite well to the fourth generation of environmental policy because of the social accountability, the

Staff member

Legislation and external pressure New technology, registration, monitoring

Low

Actors involved

Drivers for actors

Complexity

Actions

Substances, emissions

Process changes, communication (internal and external) Moderate

Efficiency

Whole company

Processes

Regulation, voluntary measures

Legislation and regulation

Scope

Prevention

Second generation

Clean-up operations

First generation

Characteristics of four generations of environmental policies.

Means to reduce environmental damage Policy instruments

Table 2.2.1

High

Product design, balanced scorecard, covenants, etc.

Companies and stakeholders, consumers Strategic performance

Voluntary measures, financial instruments, regulations Products, production chain processes

Chain management

Third generation

Very high

Sustainability, licence to operate Societal dialogue, institutional change, etc.

Sustainability, societal processes, system innovations Societal groups

Participatory instruments

Network management

Fourth generation

Transferring Scientific Knowledge to Societal Use 37

38

Chapter 2.2

Table 2.2.2

Characteristics of mode 1 and mode 2 science.7

Knowledge developed Knowledge production Place and way of knowledge production Organisation Quality control

Figure 2.2.3

Mode 1

Mode 2

Academic context (Mono)disciplinary Homogeneous (one place, a certain time)

Context of application Transdisciplinary Heterogeneous (knowledge developed close to the place of application) Heterarchical, transient

Hierarchical, preserves form Academic, peer review

Socially accountable, reflexive

General map of the Ebro basin and its distribution within the Communinades Automonas (FR, France; AND, Andorra; CV, Valencia; PV, Pais Vasco; CM, Castilla la Mancha; CN, Galicia; AR, Aragoń; CA, Catalunã; CL, Castilla y Leon; Na, Navarra; LR, La Rioja).

fact that it recognises the importance of stakeholder views and, more generally, because of the emphasis on the context of application.

2.2.4

AQUATERRA Case Studies

2.2.4.1

The Ebro Case study

2.2.4.1.1

Context

The Ebro river basin is located in northeast Spain (Figure 2.2.3). The river itself is approximately 930 km long and drains a basin of approximately 85 500 km2 in area, or 17.3% of the surface area of Spain. The total basin area is drained by 347 main tributaries with a combined length of approximately 12 000 km. The

39


river ends at the Ebro delta, one of the most important wetlands in Europe, covering an area of about 320 km2 of sediments with wetlands and coastal lagoons, which are valuables in terms of natural resources and related economic activities. The total mean annual runoff is approximately 6.837 106 m3 and the total water storage capacity of the 187 reservoirs that regulate the water flow is around 57%. The northern tributaries provide a greater contribution to the discharge than those in the south because the former drainage basins are located in higher rainfall areas. Monthly water discharge is quite irregular, with a significant decrease during the 20th century attributed mainly to increased water use for human activities (agricultural irrigation, reservoirs, electricity production and domestic consumption). Large hydroelectric power plants in the Ebro system supply 50% of Spain’s electricity. Irrigation is responsible for an important hydraulic deficit. An average of 300 m3 s1 is taken off the river: its natural flow in the early 20th century (1914–1935) was around 590 m3 s1, whereas in the last few decades (1960–1990) it was 430 m3 s1, i.e. a 28% reduction.9 A summary of the present and potential future drivers and pressures acting upon different economic areas in the Ebro basin (agriculture, industry and tourism) is shown in Table 2.2.3.

Table 2.2.3

Present Agriculture Industry

Tourism Other Future Agriculture

Summary of main drivers and pressures in the Ebro basin Drivers

Pressures

Increased productivity Abandonment of agricultural land in high altitude zones Increased use of hydroelectricity

Water supplies Soil quality

Chemicals, metals and manufacturing industries Influx of tourists during summer months Recreational use of waters Population concentration in urban areas/ depopulation in rural areas Increased production of non-food crops Increase in agrochemical applications

Industry Tourism Other

Increase in irrigated area Growth in metals, chemicals, construction and manufacturing industries Potential increase in tourists Population concentration in urban areas/ depopulation in rural areas

Water supply downstream Sediment loads Water supplies Water quality Water supplies Water quality Water supply Land use Water supply Soil quality Water quality Water supply Water quality Water supply Water supply Water quality Water supply

40

Chapter 2.2

2.2.4.1.2

Management Structure at the River Basin Level

Overall responsibility for water resources in Spain belongs to the Ministry for the Environment. Individual river basins are controlled by the relevant river basin authority, the Confederacion Hidrografica del Ebro, created in 1926 (website http://www.chebro.es/). Under the Water Law dated 1985, the responsibilities of the local water authorities are as follows: Planning: – implementation, monitoring and revision of the Hydrological Plan for the basin as appropriate. Management: – administration and control of public works and water supply, – administration and control of uses which are in the general interest or which affect more than one Communinade Automona. Investment: – planning and construction of works appropriate to the organisation and approved by the state.

2.2.4.1.3

Problems Encountered in the Management of Groundwater Resources

The key issues in this river basin are the following. Water scarcity. The quantity of water available is insufficient to meet the agricultural, industrial, commercial and domestic needs of the population within the basin. Many irrigation facilities in the basin are old and inefficient, but the cost of improving them would not be offset by increased yields and reduced water demand. The construction of large dams to retain limited water resources has altered the hydrological regime of the basin dramatically, which has resulted in many ecological impacts. Salinisation. The abstraction of ground and surface water and widespread irrigation has increased the deterioration of water and soil quality. High levels of salts in both the soil and underlying geology have resulted in high levels of salts in the available water supply. This is compounded by groundwater abstraction, which concentrates salts and increases invasion of aquifers by marine water. Water, soil and sediment quality. The surface waters of the Ebro basin are of poor quality due to high levels of industrial and agricultural chemicals from both point and diffuse sources. High levels of mercury and DDT are a particular problem, as are levels of nitrates in soil and groundwater. Soil erosion. Intense agriculture on saline soils leads to a deterioration in soil structure and quality, which in turn leaves the soil more susceptible to erosion. The areas most susceptible to soil erosion in the basin are in the Catalan Chain to the south of Fraga, Lleida, Alcaniz and Caspe.


41

Ecological issues. The ecology of the delta is under pressure from two key areas. First is the poor water quality as described above, which leads to a deterioration of habitats, with consequences for the wildlife which depends upon it for survival. Second are the reductions in water flow and sediment transport caused by dam construction, which results in erosion of the Ebro delta, a nature reserve of European importance. This is of less importance for groundwater issues.

2.2.4.1.4

Main Contributions of AQUATERRA in the Ebro Basin

A workshop managed by AQUATERRA was held on 7 April 2006 in Barcelona with stakeholders from representatives of unions and water authorities, through academics, industry representatives and students, agricultural community and water distribution companies involved in the Ebro river basin. Aims of the workshop were (i) to discuss the key environmental issues affecting the water–sediment–soil–groundwater system in the Ebro river basin; (ii) to discuss the evolution of these issues in the context of the DPSIR model used throughout AQUATERRA, with a view to establishing a suitable area for an economic case study; (iii) to inform stakeholders from the Ebro river basin about the work being undertaken by AQUATERRA within the basin area and to engage stakeholders in the work of AQUATERRA; and (iv) to interest them in further activities aimed at interaction with the project. Having identified the key issues for the basin, the concept of DPSIR was thus applied to assess what the drivers, pressures, impacts and responses to key environmental issues were and suggesting how they might change in the future. For the drivers (e.g. the factors of change), the main confirmed ones were the WFD and other water-related policies, the planning of water management by the Confederation Hidrogra´fica del Ebro and the demographic developments within and beyond the basin, especially on the coast, including the concentration of the population in the urban areas and the development of industry in zones with a limited water supply. For the pressures, agriculture is the main one within the basin, as it affects both salinity and water quantity and quality problems. Some isolated pressures from local industries are also significant in terms of water quality. A less important, but nonetheless significant pressure is exerted by large urban areas and tourist zones on the coast, both within and beyond the basin due to water transfers. In this basin, a large range of contaminants are encountered and should be monitored at different scales. For the impacts, those on flora and fauna and ecology are the most significant. For the responses, e.g. the measures taken or to be taken in order to respond to impacts, the reduction of water consumption in general would be the only efficient, long-term, sustainable solution to the current situation. The activity most directly and immediately affected by this change would be agriculture; industry and urban areas would also be affected by water economy. Within AQUATERRA, the sub-project MONITOR aims to provide, develop and validate analytical tools necessary for the monitoring of organic and

42

Chapter 2.2

inorganic pollutants in water/sediment/soil compartments with the final objective to identify those present at relevant concentrations. In the Ebro basin, studies started on the occurrence and the distribution of contaminants in the river basin. Numerous compounds have been analysed (trace metals, pesticides, PAHs, BTEX, PCBs, etc.). High concentrations were detected in some areas describing situations where the environmental impact is high (i.e. with fluoranthene, benzo[a]pyrene, benzo[b]fluoranthene). But in the field of environmental monitoring, the information of interest is spread over time and space. Therefore the choice of monitoring locations can be a key issue in terms of global costs. Optimisation of the sampling network is currently done in an effective and economic way (by improving on time intervals for sample retrieval and analysis of selected compounds). The environmental assessment of the Ebro river basin will be improved by the coupled use of chemometrics modelling of monitoring data, various sampler devices, Geographical Information System (GIS) and remote sensing information systems.

2.2.4.1.5

Problems to be Solved in the Future

The identification and validation of major issues within the Ebro basin as a whole lead to the problems that should be solved in the future in the context of progress resulting from knowledge. One of the most important problems to be solved is that of water abstraction. A reduction of water consumption in agriculture would have positive consequence on both the quantity of water consumed and soil quality, as an excess of stagnant water left on fields after irrigation increases soil salinity through water evaporation. For agriculture the improvement of the water efficiency of irrigation systems should be urgently solved. One solution would be the reuse of affected groundwater. Reuse of water would also improve the perception of agriculture by the population in general. A further alternative solution to the lack of water is desalinisation of seawater. Although this was only considered as a temporary solution, it was nevertheless considered preferable to water transfers from the Ebro to southern regions outside the basin. In the case of very acute localised problems, the compulsory purchase of agricultural land in order to avoid cultivation has been suggested, although this would clearly have a significant economic impact for those directly affected and is not a widely supported proposal for the long term. Changing to less water-intensive crops could also be undertaken, potentially at a large scale, although the economic viability of producing a different crop must be considered. At a very large scale this measure should be carefully evaluated and would only be possible under technical and market conditions.

2.2.4.1.6

The Lessons Learned

Discussions have been conducted with the major stakeholders with a common agreement on the fact that drinking water is the priority use in the basin and that future planning should aim to maintain the quantity and quality of the


43

supply. More general problems arising in the management of the basin are (i) the fact that water is free or not charged at it true value, (ii) the implementation of water management policy is not optimal across the basin, (iii) the extent of irrigated areas is continuously increasing requiring an evaluation of the impact of these new areas before improving the irrigation systems of existing areas (to be done at the river basin level to be efficient) and (iv) there is an imbalance in the quality of water treatment between that which is treated and exported to users outside the basin and that which is produced for use within the basin; consequently there is less investment in water within the basin. AQUATERRA and future research projects should investigate some of these areas in the future, like defining decision-making tools for solving identified specific important problems relevant to the river–sediment–soil–groundwater system and socioeconomic analysis of the impacts and responses (like in the Coordination Action Risk-Base of the 6th FP). These tools could be relevant for a policy directive like the WFD. But in many ways, the River Ebro is unique, with its specific management administration system, its combination of distinct climatic issues and some specific problems (such as salinisation). Therefore the Ebro does not easily fit into a standard ‘‘model’’ of a European catchment.

2.2.4.2 2.2.4.2.1

The Meuse Case Study Context

The Meuse river basin is one of the smallest international districts in Europe. The Meuse has its source in the Langres Plateau in France and flows from Belgium and the Netherlands to the North Sea. A part of the Meuse river basin belongs to Germany and Luxembourg even though the river itself does not cross these countries. The Meuse is about 950 km long and its basin represents between 30 000 and 35 500 km2 according to the various information sources (Figure 2.2.4). The rivers of the Meuse basin are mainly plain rivers, characterised by broad valleys and weak medium slopes. Belgium is a central country for the Meuse management since it contains 41% of the total area of Meuse basin (5% in Flanders and 36% in Wallonia) and the basin represent 46% of the whole country area (Figure 2.2.4). The remaining area belongs to the Scheldt river basin. France is the second country, representing 26% of the Meuse river basin (9% in Champagne-Ardennes region, 3% in Nord-Pas-de-Calais and 14% in Lorraine region) whereas the Meuse basin only represents less than 1% of French territory. One-fifth of the basin is located in the southern part of the Netherlands (14% in NoordBrabant region, 6% in Limburg and less than 1% in Zuid-Holland) and the Meuse basin represents 22% of the country. The German Nordrhein-Westfalen (NRW) land covers 11% of the basin (6% of Du¨sseldorf region and 5% of Ko¨ln) and the Meuse basin represents only 1% of the German area. Around 5 to 10% of the Luxembourg area is also included in the basin but it represents less than 300 km2.

44

Figure 2.2.4

Chapter 2.2

Meuse basin (source: IMC11,12).

The Meuse river basin can be characterised by a low runoff coefficient. Rivers are dominated by a rainfall–evaporation regime and the annual runoff is generally about 258 mm per year in the River Meuse. The long-term average discharge is 250 m3 s1 and total volume per year is 7.6 km3 on average. However, since the Meuse is a rain-fed river, the maximum flow can be as


45

high as 3000 m3 s1 in the winter and the minimum flow as low as 10 m3 s1 in the summer. The main land uses in the Meuse river basin are distributed as shown in Figure 2.2.5. The most industrialised regions and those having the most intensive agriculture are also those having the highest population density. A summary of the present and potential future drivers and pressures acting upon different economic areas in the Meuse basin (agriculture, forestry, industry and urban development) is shown in Table 2.2.4.

2.2.4.2.2

Management Structure at the River Basin Level

In the 15th and 16th centuries, the present Dutch and Belgian territories and some parts of France formed one country, the 17 Dutch provinces. But until the independence of Belgium in 1839, both countries began to formulate their own policies, particularly about transport, industrial and agricultural activities in the Scheldt and Meuse basin.10 Diverging interests arise concerning waterways and sharing availability of water (quality, quantity) for drinking water production. In the 20th century there were several attempts to have an international agreement concerning the Meuse basin and finally an agreement was signed in 1994. Thus, collaboration between the basin states of the Meuse is in its infancy compared to that of the Rhine.11 The main agreements concerning the Meuse basin signed to date are: 1992 1994

1995

2002

w

Agreement about the protection and use of transboundary watercourses and international lakes signed 17 March 1992. Agreement on the protection of the Meuse (France, the Walloon region, the Flemish region, the Brussels Region and the Netherlands) signed 26 April 1994 in Charleville-Me´zie`res. Meuse Discharge Treaty (Netherlands and the Flanders region). Equal sharing of water by both partners during low-water periods and a common responsibility for the Border Meuse.w Creation of the International Commission for the Protection of the Meuse (ICPM). The agreement on the flood protection action plan for the Meuse. The ministers of the riparian states declared in Arles (France) the necessity to reduce the flood risk and harmonise flood-reducing measures. They particularly stressed the harmonisation of spatial planning, land use and water management in the river basin. International agreement on the Meuse (Germany, Belgium, the Brussels, Walloon and Flanders regions, France, Luxembourg and the Netherlands) signed 3 December 2002 in Gand.

For further information about the Border Meuse management, see ‘‘River management and low flows in the river Meuse in the Netherlands’’.

46

Chapter 2.2

59% 28% 11%

1%

Figure 2.2.5

Table 2.2.4

Urban areas

Agricultural areas

Others

Water areas

1%

Forest areas

Land use in the Meuse basin (adapted from9).

Summary of main drivers and pressures in the Meuse basin.

Present Agriculture

Forestry Industry

Urban development

Others

Future Agriculture Industry Urban development Other

Drivers

Pressures

Increased productivity: cereals, maize and meadows Intensive farming systems like horticulture, glasshouse cultivation and breeding Increased sylviculture activities Increased industrial discharges Accidental and hazardous pollution Increased water abstraction volumes

Water supplies Water quality Soil quality

Energy, mining, chemistry, metals, construction and agro-food industries Population concentration in urban areas Increased water abstraction Water sanitation (discharge of wastewater) Fluvial transport (numerous navigable rivers and canals) Leisure and tourism versus discharge of wastewater Increased of agrochemical applications Increased in glasshouse cultivation Growth in some of the main industrial activities Population growth and concentrations, increased density Increased leisure demands

Water quality Sediment loads Water supply downstream Water supplies Water quality Water supplies Water supply Water quality Water quantity Water quantity Water quality Soil quality Water quality Water quantity Soil quality Water quantity and quality Water Supply Water sanitation Water quantity and quality


47

Actually, the International Meuse Commission (IMC) plays an important role in international consultations. Its purpose is to obtain sustainable integrated water management in the international basin of the Meuse. From its creation, the IMC has established three action programmes: the preliminary programme (1995–1997), the short-term programme (1998–2003) and the longterm programme (2003–2010). The important targets of the Meuse short-term programme included:12

reduction of urban and industrial discharges; reduction of diffuse pollution; prevention form accidental and hazardous pollution; monitoring water quality; sediment management; ecological improvement of water quality; and knowledge exchange between countries.

The present action program (2003–2010) also includes flood prevention and protection and has a main target to contribute to the implementation of the WFD.

2.2.4.2.3

Problems Encountered in the Management of Groundwater Resources

The key issues in this river basin are the following. Water scarcity. The quantity of water available could be insufficient to meet the needs of the different activities within the basin (agriculture, industries and domestic uses). In some areas, water exploitation is concentrated in specific surface or ground waterbodies, causing large problems for water availability and localised environmental problems to the water ecosystems. A significant example is the Tournaisis aquifer (Hainaut, Wallonia), where water is extracted at rate higher than the aquifer recharge capability. The Netherlands is also strongly dependent on the surface water of the Meuse for its drinking water production. Groundwater quality. Even if the groundwater system does not have a homogeneous network of measures all along the Meuse river basin, its quality seems to become degraded from the upper to the lower stretch of the river basin, in particular with nitrates and pesticides. Other contaminants are also brought by point-source pollution from industry (hydrocarbon fuels, solvents, but also chloride, iron bore and heavy metals from mines in France). Bad quality of the past mining reservoirs may affect the groundwater quality of the Meuse Dogger aquifer: – by the rising of saline water from the mines reservoirs towards the bottom of the aquifer, and/or – by the infiltration of surface water contaminated by the waters out flowing from the flooded mines.

48

Chapter 2.2

The hydromorphology modification of the Meuse. Since 1883, the Meuse has been largely managed by humans to support navigation, economic development and flood protection. The major interventions involved the building of river control structures (weirs, dams and sluices, particularly in Walloon and Dutch parts) and banks (training walls, quays, etc.) and modifying the river bed by regular dredging of sediment. Such modifications to river imposed changes to the morphology (river fragmentation) and to the hydrological regime which have an impact on other factors including: – Flooding: the extent of zone liable to extreme flooding (over the protection designed) is limited and hence the water volume that may be naturally stored in floodplains is also limited. – Sedimentation and transport: when the level discharge is low, the sedimentation transport is greatly reduced behind weirs. As the discharge increases, the deposited sediments are carried again and the volume load is greater than without weirs. According to IMC,11 the amount of sediments transported have tripled because of the hydromorphological modification. – Ecological function: as a consequence, biodiversity became poor in certain stretches and migration routes for fish have been disturbed. Diffuse pollution. Agriculture, industrial activities and household discharges all contribute to diffuse pollution on soil and water bodies. The main pollutants are nitrates, fertilisers, manure and pesticides (like diuron and atrazine) which may induce the eutrophication of the river and thus that of the North Sea. The total concentration of diffuse pollutant in the water system is very important from a drinking water perspective as the River Meuse is a source of drinking water for more than 6 million inhabitants. Some new dangerous chemicals like endocrine disruptors, and new medicines from pharmaceutical industry have recently appeared. When looking at effects on water quality, all pollutants should be considered together for understanding their impacts on the ecology of the system.

2.2.4.2.4

Main Contributions of AQUATERRA in the Meuse Basin

The main study objects in the Meuse basin are: (i) contaminated floodplain sediments; (ii) soil–sediment–groundwater–river interaction in the catchment of the Dommel (a tributary to the Meuse); (iii) groundwater quality effects by river–groundwater interaction in the Belgian part of the Meuse system, more specifically near Lie`ge (Fle´malle cookery site) and in the Geer basin; (iv) quantification of coupled ecotoxicological effects of contaminants from sediment, suspended solids and freshwater on aquatic organisms at a number of Walloon locations; and (v) socioeconomic aspects of global change. This needs


49

integration of multiple disciplines, from geosciences, environmental engineering and chemistry to socioeconomic sciences, from the catchment to the regional scale, as is done in AQUATERRA. The main AQUATERRA contributions in the Meuse presented during the 2nd International Meuse Symposium in Sedan (18–19 May 2006) organised under the auspices of the IMC13 (http://www.meuse-maas.be/) concerned: metal speciation and bioaccumulation in floodplain soil and sediment; nitrate trends in the chalk aquifer of the Geer sub-basin in Belgium; breakdown of leaves as an indicator for sediment quality and ecosystem functioning; ecotoxicological effects of contaminants on aquatic organisms in the River Meuse; metal availability in the Dommel sub-catchment; and measurement of metal concentrations (Cd, Zn, Cu, Fe, Ni) in soil and earthworms in the Dommel sub-catchment. In parallel with the work done in BASIN, the sub-project FLUX investigates the temporal and spatial trends of suspended matter pollution in the Meuse River. Spatial and temporal differences in suspended matter (SPM) quality in the Dutch part of the Meuse can be attributed to point sources contribution, variation in river discharge and sedimentation processes. The large fluctuations in metal content during summer indicate an irregular contribution of point sources from the upstream area, most likely from the Liege industries; to control the high pollution levels of suspended matter, this contribution from Belgian industrial areas should be reduced. The spots of contaminated suspended matter deposition in the summer should be traced back to reduce the risks of higher contamination levels on floodplain areas by re-suspension of contaminated river bed sediment and possible long-term exposure of contaminated bed sediment in the river environment during dry periods.

2.2.4.2.5

Problems to be Solved in the Future

Among the major issues for the Meuse River, the sanitation in the Walloon part of the Meuse basin is one of the most important. Although 95% of the population is connected to a collective sewage system, collective treatment capacity (1.1 million IE) only represents 37% of the urban and industrial wastewater volume (respectively 2 and 1 million IE) that had to be treated in 2002. Figure 2.2.6 shows that numbers of treatment plants are under construction, granted or under study, but 18% remain non-existent. Moreover the whole Walloon stretch is considered as a ‘‘sensitive area’’ under the Council Directive 91/271/EECz: all treatment plants with capacity exceeding 10 000 IE have to ensure a treatment against phosphorus and nitrates (tertiary treatment). z

Council Directive 91/271/EEC of 21 May 1991 concerning urban wastewater treatment.

50

Chapter 2.2 non existent plants 18% plants in study 9%

existing plants 37%

projected plants 1% granted plants 24%

Figure 2.2.6

plants in construction 11%

Treatment plants in the Walloon stretch in 2002: distribution of IE to be treated.

Treatment capacity rates vary within the Walloon stretch from 13% in the ‘‘Meuse aval’’ sub-basin to 73% in the ‘‘Vesdre’’ sub-basin.

2.2.4.2.6

The Lessons Learned

Discussions have been conducted with the major stakeholders of the Meuse river basin and have shown that building a global scenario for managing the basin at the river basin scale remains a challenge for several reasons: The stakeholders have limited knowledge and information on possible future changes in the key drivers and their trends. The identified drivers, the economic activities, are really comprehensive and need to be disaggregated to be more issue-specific (e.g. for agriculture, irrigation drainage, fertilisers and pesticide uses) to be sufficiently analysed and used for decision-making. The complexity of the management system needs to be simplified by developing accurate indicators based on basic simple figures, still to be identified. There are different decision-making levels (local, regional, national and international in the case of the Meuse river basin) and the important issues at one of these levels are not necessarily viewed as important at another, in particular at the river basin scale. Stakeholders concerned by environmental problems have a perception coming from societal pressures that differs from the scientifically based approach based on detailed factual information. Even the time horizon taken into consideration by both is different, being shorter for the stakeholder issues. Developing scenarios for the sediment–soil–water system is very complex, with multiple possible entries: (i) system components (soil, sediment, groundwater and surface water); (ii) drivers and factors of change (climate, economic and policy changes); (iii) spatial scale of analysis (local/small area, regional, national, sub-basin and basin levels);


51

(iv) temporal scale (short, medium and long term); (v) potential issues at stake (organic pollution, diffuse pollution by nitrate, sediment contamination, water pollution by heavy metals, etc.). Therefore, defining longterm scenarios for some specific issue at case study scale should be more appropriate than global management scenarios.

2.2.5

Discussions of the Contributions of AQUATERRA to Societal Use Related to Groundwater Resources Management

As shown in the case studies discussed above, human activities have greatly modified the water resources systems through climate change, land use changes, water engineering and releases of wastes and pollutants in the aquatic systems (Table 2.2.5). Nevertheless, aquatic systems have also high natural spatial heterogeneity. Therefore, understanding and modelling the degree of heterogeneity of water systems will be different and will depend on the scale of interest (Table 2.2.6). The sediment–soil–water issues have to be addressed at different levels, i.e. local, regional, national or international. An important issue at the regional/ national level is not necessarily viewed as an important issue at the river basin scale. The response and management concerns may also vary between the different levels, especially for a transboundary river basins such as that of the Meuse. This complexity of the studied system induces some key elements for the new modern ways of knowledge production. Knowledge production should be socially accountable and should acknowledge the multiple rationalities and different viewpoints that are brought in by the variety of stakeholders that are involved. This means that the research methods should contain the following key elements. Multi- and/or transdisciplinary research methods – Acknowledge the disciplines that should be involved in the research to give more insight in the policy issue, the problem framing and the solutions. – Work together in multidisciplinary teams and emphasise the process of interaction between the disciplines and evaluate these processes. – Take time to develop new theoretical multidisciplinary scientific frameworks. Involvement of the stakeholders in the research process – Develop methods to involve stakeholders and to deal with their viewpoints and values in the research process. – Appreciation of the values, interests and viewpoints of involved stakeholders ask for development of new research methods to develop

x x x x

x

x X

x x

x

x

A

X x x x

x x x

x

x

x x

x

x X X x

x x

x

X

x x x

x

X

C

x

B

x

x

x x

x

x

x

X X x

X

D

x

x

x

x

x

x x x x

E

A: human health; B: hydrological cycle balance; C: water quality; D: global carbon balance; E: fluvial morphology; F: aquatic biodiversity; G: coastal zone impacts.

6. Irrigation / water transfer

5. Urban waters

4. Industrialisation and mining

3. River damming and channelling

2. Land use changes

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

1. Climate variability and changes

development of non-perennial rivers segmentation of river networks changes in flow regimes development of extreme flow events changes in wetlands distribution/function changes in chemical weathering changes in soil erosion salt water intrusion in coastal groundwater salinisation through evaporation wetland filling or draining changes in water pathways changes in sediment transport urbanisation alteration of first-order streams nitrate and phosphate increase pesticides increase nutrient and carbon retention retention of particulates loss of longitudinal and lateral connectivity creation of new wetlands increases of heavy metals and POPs Acidification of surface waters salinisation sediment sources nitrate and phosphate increase enhancement of waterborne diseases organic pollution heavy metals and POPs increase partial to complete decrease of river fluxes salinisation (evaporation and percolation)

Local to regional changes of environmental states

Global impacts

x

x

X

X x

X x

x x

x x

x x x x x

F

Major global pressures on continental aquatic systems and the mapping of local- to regional-scale impacts.1

Pressures

Table 2.2.5

x x

x x

X

x x x x x X x

x x x x x

x

G

52 Chapter 2.2

53


Table 2.2.6 Scale level Plot scale

Heterogeneity of the water systems, natural and anthropogenic issues according to the scale level. Natural issues

distribution of root sys

Small catchment scale

tems lateral transfer of water

groundwater aquifer upstream–downstream flow structure

biogeochemistry parameters of environment compartments

Anthropogenic issues

land cover changes diffuse atmospheric pollution

land

River basin scale

river flow regimes, links

between groundwater and surface waters sediment supply and transfer water quality linked to climate

cover and use changes land management micro-climate changes artificial modifications of water bodies diffuse and local pollutions

water uses and demands impacts of urbanisation impacts of human activi-

ties such as agriculture, industries, mining

and share knowledge. The instrument of joint fact finding14 is already available, but it should be developed further. – Testing of scientific results with stakeholder panels is another direction of development.

Emphasis on learning in policy – Learning between stakeholders, between scientists, between stakeholders and scientists, etc., and on different levels (individual, team and organisation) have a central role in the fourth generation and deserve special arrangements. – Organisation of reflection, feedback and evaluation of the goal achievement of the policy is needed.

As ground water policy is getting involved in a much broader policy field, one can expect that these trends will also relate to groundwater policy. In effect, the AQUATERRA research project is an example of the above mentioned trend of multidisciplinary research. As stakeholder involvement and policy learning are other policy trends, the next sections will go into these topics.

54

Chapter 2.2

2.2.5.1

Stakeholder Involvementy

For stakeholder involvement with respect to groundwater policy there are several arguments. Apart from the basic fact that stakeholders have an impact on the quality and quantity of groundwater, the main arguments can be grouped into three themes: obstructive power, enrichment and fairness. The early involvement of stakeholders reduces the risk of the use of obstructive power by them, and thus the policy not being carried out. On the other hand, stakeholders possess resources, like money or knowledge required for the design planning and implementation of sophisticated policies, which governments do not possess. Knowledge is distributed among several stakeholders and government. So stakeholders should be involved to get all the (pieces of the puzzle) together. Stakeholder involvement can provide good ecological practices in this way. The last argument for stakeholder involvement is fairness. It is fair to involve actors affected by a certain policy, and give them a say in the decision-making process. But who are the stakeholders and how can one involve them? Derived from Slob et al.,15 we can make this definition: (stakeholders are all those people or organisations who have an effect on or are affected by groundwater policies.) We can make a distinction between two types of stakeholders: Organisations and people that have a direct impact on groundwater quality or are directly affected by the relevant policies. This group includes: industries using groundwater, farmers, water authorities, regulators on the local, regional, national and international level and citizens that are directly affected by the measures planned or taken. Organisations and people that have an impact on the relevant decisionmaking. This group covers citizens, landowners, homeowners, insurance companies, NGOs such as Greenpeace and the WWF, scientists and drinking water companies. A process of stakeholder involvement requires an independent chairperson or process manager. The first step for the process manager in the organisation of stakeholder involvement is to find out which stakeholders should be involved. Next, it is vital to collect information about the goals, ambitions and problem definitions (from the various perspectives) of the stakeholders. The process manager should ensure that all these interests are heard and acknowledged in the course of the process. The mobilisation of the stakeholders is an important issue. It is the duty of the manager to let stakeholders realise what the benefits are. Why should they join the process? A sound and deliberate consideration of interests might persuade less interested parties to join and will be a signal to dominant forces not to overreact. To create more certainty into the process, the process requires ‘‘rules of the game,’’ that contain rules for entering the process in later stages, y

This section is an excerpt of a chapter about stakeholder involvement in one of the (SedNet) books.15


55

how decisions are made, how information is brought into the process, etc. These ‘‘rules of the game’’ should be discussed and should be approved by the involved stakeholders. The process of involvement can be arranged with different goals: information: providing information to the stakeholders; consultation: ascertaining what stakeholders think must be done; advising: letting stakeholders advise on the policy and taking their recommendations into account; co-producing: stakeholders are regarded as equal policy-makers but decision-making remains in the political domain; and co-deciding: decision-making power is handed over to stakeholders. Every situation is unique and therefore the level should be chosen that fits the specific situation.

2.2.5.2

Learning Approaches in Policy-making

With the whimsicality of the fourth generation of environmental policies, the policy approach that fits best for the (local) groundwater issue is not ‘‘standard’’ and not known beforehand. This means that the policy approach required should be developed with the involved stakeholders specific for the local groundwater issue, and should be ‘‘tailor made.’’ To do this effectively an adaptive approach should be followed, in which actions are deliberated with the stakeholders, and will be followed up by their implementation. Then, to see the results of the actions, it is needed to follow quite closely how the groundwater system responds to these actions with all stakeholders involved. If the actions do not give the desired results, new or extra actions should be deliberated and undertaken. In this manner the way the system responds will be followed and the understanding of the system will grow. Continuously monitoring of the groundwater system plays a key role in sustaining the ‘‘learning approach’’ of groundwater policy and to find the best local approach to the groundwater issue. A summary of this learning approach could be as follows. Discussion on the groundwater issue with involved stakeholders, the problem (for instance with the DPSIR model) and the shared ambitions to solve the problem. Discussions on the actions that should be undertaken to reach the ambitions. Monitoring of the results and the way the system responds to the actions. Discussion on the results: does the system respond to the actions as was foreseen? Are new or extra actions needed? And so on. Implementation of these (extra) actions and monitoring of these results. Then the cycle starts all over again. In this way actions will be found that match best with the articulated ambitions.

56

2.2.6

Chapter 2.2

Conclusions

In this chapter, we present the preliminary results of the Integrated Project AQUATERRA in the framework of policy and societal use. AQUATERRA fits fully with part of the general conclusions given at the end of the World Science Forum16 (http://www.sciforum.hu/index.php). For example, conclusion 1 stated that ‘‘Scientific research is having a more immediate societal influence and facing an increasing set of requirements on the part of the public. As a result of internal scientific development and societal need, new research priorities emerge, requiring the cooperation of various disciplines. Therefore an integration of the natural and social sciences is taking place for problem-solving. Such an integration reinforces the need for establishing interdisciplinary frameworks. This is to be reflected in the institutional structures of science and of science policy as well.’’ This was clearly evidenced throughout this chapter. In the not too distant future, the main developments in AQUATERRA will certainly agree with another conclusion of the World Science Forum (2003) that stated that ‘‘the development of science and the demands of society will remove the rigid boundaries between theoretical and applied research, between the academic and innovation sector, . . . and scientific communities shall communicate the achievements of science and shall support decision-making . . . ,’’ especially within the COMPUTE sub-project (by setting a modelling tool box to assess impacts on water quantity and quality at different investigation scales: bench, catchment, river basins) and INTEGRATOR sub-project (by elaborating operational tools for the stakeholders, including a first assessment of social and economic impacts of policies in a scientific basis decision-making system). Finally, we showed that such an Integrated Project is in full agreement with the most important conclusion of the World Science Forum16 stating that ‘‘There exist appropriate scientific guidelines for improving the quality of life. Solutions offered by science are generally delayed by socio-economic patterns and inappropriate information transfer. It is the common responsibility of politicians, scientists and decision-makers to ensure the proper implementation of knowledge to improve the quality of life.’’

References 1. M. Meybeck, Hydrol. Proc., 2005, 19, 331. 2. M. H. Gerzabek, D. Barcelo´, A. Bellin, H. H. M. Rijnaarts, A. Slob, D. Darmendrail, H. J. Fowler, Ph. Negrel, E. Frank, P. Grathwohl, D. Kuntz and J. A. C. Barth, J. Environ. Manag., 2007, 84, 237–243. 3. F. Boons, L. Baas, J. J. Bouma, A. de Groene and K. LeBlansch, The Changing Nature of Business, International Books, Utrecht, The Netherlands, 2000. 4. H. Spliethof and J. van der Kolk, Environmental management: developments and interactions between technology and organisation, in Technology and Environmental Policy, [Technologie en milieubeheer],


5. 6. 7.

8. 9.

10. 11. 12.

13.

14.

15.

16.

57

ed. A. P. J. Mol, G. Spaargaren and A. Klapwijk, SDU, The Hague, 1991 [in Dutch]. G. Keijzers, J. Clean. Prod., 2000, 8, 179. L. Simons, A. Slob, H. Holswilder and A. Tukker, Environ. Qual. Manag., 2001, 11, 51. M. Gibbons, C. Limoges, H. Nowotny, S. Schwartzman, P. Scott and M. Trow, The New Production of Knowledge, Sage Publications, London, 1994. H. Nowotny, P. Scott and M. Gibbons, Minerva, 2003, 41, 179–194. N. Geilen, B. Pedroli, K. Van Looy, L. Krebs, H. Jochems, S. Van Rooij and T. Van Der Sluis, Final Report of the IRMA/SPONGE Project, no 9, 2001. R. J. Batalla, C. M. Gomez and G. M. Kondolf, J. Hydrol., 2004, 290, 117–136. IMC, Proceedings of the First International Scientific Symposium on the River Meuse, Maastricht, The Netherlands, 2002. IMC, Rapport interme´diaire du programme d’action ‘‘Meuse’’: Mise en oeuvre de la 1e`re phase et pre´paration de la 2e`me phase. Commission Internationale de la Meuse, Lie`ge, 2001. J. Joziasse, J. Vink, D. Slijkerman, E. Foekema, J. Brils, J. Batlle Aguilar, P. Orban, S. Brouyere, A. Poot, E. Bleeker, C. van der Wielen and M. He´mart, AquaTerra research activities in the Meuse river basin, 2nd International Meuse Symposium, Sedan, 18–19 May 2006. J. R. Ehrmann and B. L. Stinson, Joint fact-finding and the use of technical expertise, in: The Consensus Building Handbook, ed. L. Susskind, S. McKearnan and J. Thomas Larmer, Sage Publications, London, 1999. A.F.L. Slob, L. Gerrits and G. J. Ellen, Sediment management and stakeholder involvement, in Sediment Management on the River Basin Scale, ed. Ph. Owens, Elsevier, in press. World Science Forum, Budapest, 8–10 November 2003; available at http:// www.sciforum.hu/index.php2003.

CHAPTER 2.3

Groundwater Management and Planning: How Can Economics Help? JEAN-DANIEL RINAUDOa AND PIERRE STROSSERb a

Bureau de Recherches Geólogiques et Minie`res (BRGM), 1034 rue de Pinville, FR-34000 Montpellier, France; b ACTeon, Le Chalimont, BP Ferme du Pre´ du Bois, FR-68370 Orbey, France

2.3.1

Introduction

Public water managers and stakeholders involved in the design of groundwater management are increasingly aware of, and concerned by, the economic implications of the technical choices they promote. The definition of sustainable groundwater quality objectives as well as the identification of technical actions needed to achieve these objectives require investigating trade-offs between economic and environmental considerations. It also requires balancing interests between economic sectors through proportioning constraints imposed to sectors generating pollution or depletion with costs incurred by sectors depending on groundwater quality. Although the economic (and political) dimension of groundwater management might be taken into consideration implicitly by water planners in the large majority of European countries, it is rarely explicitly considered and quantitatively assessed. The promulgation in 2000 of the European Union (EU) Water Framework Directive (WFD) has drastically improved this situation by clearly integrating economics into water management and policy. This directive promotes the application of economic principles such as the polluter pays principle. It recommends applying economic methods such as cost-effectiveness analysis to support the identification of measures to achieve the environmental objectives of the directive. And it calls for a wider consideration of economic instruments (e.g. water pricing, charges and taxes) to provide adequate (financial) incentives for reducing pressures exerted on water resources.

58

Groundwater Management and Planning: How Can Economics Help?

59

Clearly, the recognition of the need for more systematic and robust economic assessments to support water management programmes is not the sole result of the changing regulatory context. Indeed, it reflects an increasing demand by actors and economic interests for more robust economic justifications to water protection decisions, choices and orientations. More attention is given to the efficient use of limited financial resources allocated to environmental protection. And the assessment of the economic impacts of constraints imposed on economic sectors by environmental regulation (e.g. restriction or ban of dangerous substances in industry) is increasingly called for: as it may result in competition distortion with non-EU producers and may generate losses of competitiveness, income and employment. In this context, the assessment of these costs, the proof that they are minimised and fairly distributed between sectors and, in certain cases, the demonstration that they generate proportionate social benefits, are elements that can help policy makers to justify decisions and gain social acceptance. Groundwater planners can expect three distinct types of contributions from economists. The first one is the description of the hydro-economic system to better capture the interactions between, and dynamics of, economic activities and groundwater resources. This involves the construction of indicators characterising the economic significance of water uses (in terms of employment, turn over, added value, etc.). It also requires understanding plausible futures for economic activities and resulting pressures on groundwater resources. In some cases, this description might benefit from the assessment of financial flows between economic sectors and water users, or the evaluation of the damage costs imposed on third parties because of the degradation of groundwater resources (pollution or depletion). The second type of contribution from economists consists in assessing and comparing the economic cost and/or benefits of different groundwater management options. The role of economics can either be to identify the least costly way to achieve a given environmental objective (cost-effectiveness analysis) or to assess the net benefit (i.e. total benefits minus total costs) of alternative management options (cost–benefit analysis). It is important to point out that the ranking between different policy options based on economic criteria is only one element provided to support a policy decision, not the decision itself. The third type of contribution consists in the design of economic instruments such as prices, charges (abstraction/pollution) or taxes that might influence water users’ behaviour. The need to influence users’ behaviour for enhancing the sustainability of groundwater resources is increasingly recognised by water planners. And they call for the design and implementation of economic instruments aimed at regulating water demand or pollution. The main purpose of this chapter is to show how practical economics can help provide some light on these different demands. In its first section, the chapter demystifies key economic principles, concepts and methods. It then presents four concrete applications of economic methods applied to groundwater management. It then discusses the potential role of economic instruments to enhance the sustainability of groundwater resources. The chapter concludes

60

Chapter 2.3

by discussing the needs for, and challenges in, integrating economics with technical expertise and knowledge, an element that is essential to the relevance of economics for policy-making in the field of (ground-) water management. The conclusion also identifies gaps between policy demand and the economics toolbox where further research is needed to enhance the potential effectiveness of economics in supporting groundwater management and policy decisions.

2.3.2

Economic Methodologies and Tools: Four Possible Ways of Supporting Groundwater Management and Planning

Economics as a science, its underlying theoretical principles as well as its operational tools and methodologies are not always given due considerations in groundwater management discussions and processes. This section presents a short overview of the economic methods and tools that can be mobilised to support groundwater management and to help answer the groundwater policy demand briefly presented above.

2.3.2.1

Economic Characterisation of Water Uses

Characterising the economic dimension of groundwater use has both static and dynamic dimensions. First, assessment can be made to identify how important groundwater use is today for the economy, e.g. what are the economic weights in terms of turnover, value added, employment, economic value, etc., that one can attach to different abstractors and polluters of groundwater resources. Integrated with information on pressures and/or impacts on the state of the groundwater aquifer, it helps obtaining first insights into the trade-offs between economic development and groundwater protection—and into possible economic impacts that might occur as a result of stricter groundwater protection. The information can be used to justify how essential groundwater protection is for key economic sectors for which development is directly linked to groundwater quality—or because groundwater of adequate quality has a high economic value that justifies its protection. Second, economics can help capturing the dynamics of the system and identifying future groundwater protection challenges. Indeed, static situations in terms of pressures and impacts on groundwater are rare. Changes in sector policies, the internal dynamics of economic sectors (influenced, for example, by global changes and world trade) or the implementation of current environmental legislation is expected to impact on economic development and thus on pressures imposed on groundwater resources. In some cases, improvements in groundwater quality might result from the implementation of existing environmental legislation or from changes in economic sector policies favouring less polluting input and production processes. In other cases, urban development and the opening of new markets for industrial products might lead to increased abstraction. And understanding the dynamics of groundwater


61

systems and connected users is essential in the search for solutions and measures aimed at improving the protection of groundwater resources. The characterisation reports prepared by EU member states in line with the requirements of Article 5 of the WFD provide very diverse illustrations of economic characterisation of water uses. Few of these reports, however, clearly link economic indicators to water use or to groundwater use.1 Investigating today’s situation and plausible socioeconomic futures for given groundwater systems requires collating economic information (often available in statistical systems complemented whenever necessary by surveys of citizens or economic operators), developing links between economic and technical (groundwaterrelated) information or applying foresight methods. Because of the scarcity of information directly available, it builds on exchange and interaction with stakeholders of the groundwater system considered.

2.3.2.2

Methods to Assess Environmental Costs

Recent studies have clearly shown that impacts on water quality and quantity of human activities are becoming significant at the European level2,3 as well as elsewhere in the world.4 As a result of increasing nitrates and pesticides concentrations, chlorinated hydrocarbons contamination or salt water intrusion (to quote only selected problems), different segments of society are incurring economic losses, referred to as damage costs hereafter. An assessment of the significance of these damage costs—which could be avoided through appropriate action—is information policy-makers might use to justify the relevance of engaging into costly groundwater protection actions. Four different types of damage costs can be distinguished: (i) cost of disease when the population is exposed to unsafe levels of substances in water; (ii) avoidance costs for water users who have to undertake averting or corrective actions (treatment of water, purchase of bottled water, etc.) in response to groundwater deterioration; (iii) ecological damage of surface ecosystems and subsequent loss of recreational value, when groundwater contamination has an impact on surface ecosystems (rivers, wetlands); and (iv) loss option value (possibility to use groundwater in the future) and non-use value (bequest value). Different economic methods have been developed to assess these costs. The cost of illness method aims at assessing costs generated by exposure to contaminants in drinking water, e.g. the cost of foregone wages for victims of contamination and the cost of treating the illness.5,6 This approach may be relevant in cases where health effects are the major cost component, as is the case with arsenic groundwater contamination in Bangladesh for instance.7 It does not seem relevant to pollution in Europe where concentrations of harmful substances found in drinking water very rarely reach a very high level (apart for some rural areas in Romania and Bulgaria, for example). The illness cost due to pollution with pesticide and some selected dangerous organic compounds could be significant, but the lack of epidemiological studies does not allow any evaluation of these costs.

62

Chapter 2.3

The avoidance cost method, or averting expenditure approach, consists in assessing the costs of actions undertaken to prevent or mitigate the adverse effects of contamination. Rooted in the household production function model, this approach assumes that consumption of goods or services can substitute for groundwater quality change. The implementation of this method offers a means to generate lower-bound estimates of an important component of the cost of groundwater pollution, namely the use of groundwater as a drinking water source. The averting behaviours reported include purchasing bottled water, monitoring of private borehole water quality and installation of filtering devices. Their estimated cost ranges between US$125 and US$330 per year and per household.8 The contingent valuation method aims at assessing the value that households assign to the preservation of groundwater quality. It builds on household surveys during which respondents have to state their willingness to pay (WTP) for hypothetical groundwater protection (or restoration) scenario. The stated WTP contains both use and non-use values, since households may want to preserve groundwater for their present or future consumption, for the consumption of future generations and/or for the resource as such. The estimated WTP for groundwater protection can be considered as an estimate of the cost of groundwater degradation. Contingent valuation is by far the most widely used method for assessing the benefits of water protection. Household WTP values reported in the literature range from less than h20 to h550 per household per year.9,10

2.3.2.3

Economic Methods for Appraisal of Groundwater Projects and Policies: Cost-effectiveness and Cost–Benefit Analysis

Due to the large number of economic sectors having an impact on groundwater, groundwater overexploitation or pollution problems can almost systematically be solved using different technical options or focusing on different sectors generating pressures. Consider for instance an overexploited aquifer where total water abstraction exceeds groundwater recharge, resulting in a sustained drop of the water table. The hydrological balance can be restored by reducing water abstraction, through artificial rechargew or a combination of both. The reduction of water abstraction can be achieved through the development of alternative water resources (inter-basin transfers, creation of reservoir dams, desalination), improvement of irrigation efficiency, the development of recycling technologies in industry or wastewater re-use. To identify the cheapest way to restore the hydrological balance, different combinations of measures can be compared in terms of their costs and their effects: an approach referred to as ‘‘cost-effectiveness analysis’’. A concrete example of cost effectiveness w

Procedure in which water supplies, sometimes using treated waste water, are pooled over highly permeable aquifers for increased infiltration. In some applications, injection wells can be used to recharge water to deeper aquifer levels or into less pervious aquifers.


63

analysis of water scarcity management can be found elsewhere.11 An additional example of cost-effectiveness analysis applied to a water pollution remediation is further detailed in Section 4 below. The problem can be even more complex when the different technical groundwater management options envisaged do not lead to the same environmental outcome. This can be the case, for instance, when different levels of contaminated site remediation are envisaged, or when the targeted pollutant concentration is not fixed by law. In such situations, water planners may be interested to assess not only the costs of alternative groundwater remediation options but also their benefits or net benefits (i.e. the difference between total benefits and total costs of actions). Cost–benefit analysis thus provides a rational and systematic framework for identifying and assessing in monetary terms all positive and negative effects of alternative options. In some cases, it involves the translation into monetary terms of non-marketed environmental, social and other impacts using some of the methods described in Section 2.2. An example of cost–benefit analysis applied to groundwater remediation is presented in Section 5.

2.3.2.4

Economic Behavioural Models

Behavioural models aim at simulating how changes occurring in the economic, regulatory or natural environment affect decisions of economic agents (farmers, households, industrialists) using water (groundwater abstraction) and/or impacting on its quality (pollution). Two broad types of behavioural models can be distinguished: decision-based models and statistical models. Decision-based models aim at representing the decision process that determines water use or pollution emission. Such models have been, for example, extensively applied for simulating farmers’ choice in terms of cropping patterns, water use (irrigation) or fertiliser use. Two types of models can be distinguished: optimisation models and rule-based models. Optimisation models assume that farmers select the combination of crops which maximises their income under a set of technical, regulatory and economic constraints; these models estimate crop choices, input consumption (fertiliser, labour, energy) and farm income for different input parameter values (agricultural prices and subsidies, input prices, minimum set aside constraint, etc). Optimisation models have also been developed for households to investigate households’ groundwater use.12 Rule-based models assume that individual decisions result not only from an economic optimisation process but also from interactions between various categories of actors and social objectives which may overrule the profit maximisation objective. Such models are often formulated using agent-based modelling techniques and implemented using object-oriented programming language,13 for an example applied to groundwater management. Statistical models aim at exploring the relationships between (ground) water use behaviour and socioeconomic characteristics of a panel of water users. Such models have been extensively used following the seminal work of Howes and Linaweaver14 for modelling drinking water demand and assessing the

64

Chapter 2.3

sensitivity of water use to price increase.15 More recently, statistical models have also been applied to the farming sector, using positive mathematical programming which investigates relationships between farm crop choices and factors determining production choices.16,17

2.3.2.5

Selected Illustrations

In the following sections, practical case studies are presented to illustrate how the methods and tools described above can be implemented in practice, and which results they deliver. The first case study focuses on the assessment of pollution damage costs due to agricultural diffuse pollution with nitrates and pesticides (Section 3). The second example, based on a Slovenian case study, illustrates how cost-effectiveness analysis can be used to identify the least costly way to reduce groundwater pollution (Section 4). Section 5 presents the results of assessments of costs and benefits of alternative groundwater protection scenarios from a practical case study conducted in Latvia. In Section 6, we illustrate how economic and groundwater models can be integrated to simulate long-term global change scenarios.

2.3.3

Assessing and Simulating Current and Future Socio-Economic Impact of Groundwater Deterioration

The first example focuses on the assessment of socioeconomic impact of groundwater pollution with nitrates and pesticides in the upper Rhine valley aquifer. The demand for this economic assessment emerged in the early 2000s following publication of the results of a trans-boundary groundwater quality survey showing a steady increase of nitrate concentration in very large areas of the aquifer, combined with a drastic increase in the pesticide detection frequency. Reversing this trend would require a drastic reinforcement of groundwater protection measures, in particular for the agriculture sector. But regional policy-makers feared that this would lead to significant opposition and protest from the farm lobby. To improve the social acceptance and the legitimacy of groundwater protection measures, they decided to launch a research project to quantify in monetary terms past and future socioeconomic impacts of groundwater pollution. The first part of the study consisted in assessing socioeconomic impacts of nitrate and pesticide pollution which had occurred during the last 15 years. The analysis was based on a consultation of experts, review of archives of financial actors and interviews with municipalities and industry representatives. The results showed that one-third of drinking water utilities, representing 177 municipalities and 432 000 inhabitants, had been directly concerned by nitrate or pesticide pollution during the 15 years period. And investments made to respond to this pollution were estimated at h26 million resulting in an average increase by h0.2 per cubic metre of the drinking water price (or h30 increase of


65

the water bill per household per year). This increase would have been twice as high if investments made by drinking water utilities had not been heavily subsidised. Groundwater pollution also contributed to eroding population trust in tap water and to increasing bottled water consumption, generating total additional costs estimated at h165 million over the 15-year period considered.18 The second part of the study consisted in estimating future costs that would occur in the absence of additional groundwater protection measures. It required developing a tool that would enable the simulation of future groundwater quality changes, combined with an economic method for assessing damage costs that would be incurred by water users as a result of groundwater quality changes. The evolution of water quality was simulated using geostatistical methods, assuming trends in water quality observed in the recent past could be extrapolated to the 2015 time horizon.19 The results of the extrapolation (Figure 2.3.1) showed that the estimated average nitrate concentration would increase up to 26.3 mg l1 by 2015 (against 25.7 mg l1 in 2003) with the area where drinking water threshold value for nitrates is exceeded (50 mg l1) being doubled (10% of total area in 2015 versus 5% today). Concerning pesticides, the evolution would differ from one substance to another, with the concentration of atrazine and its metabolites decreasing while metolachlore and alachlore concentrations would increase. Overall, the study showed that 33 drinking water wells belonging to 21 public drinking water utilities would be contaminated by 2015 (mainly by pesticides). Moreover, 68 additional drinking water wells would be at risk as pesticide

2003

2015

Pesticide concentration

Pesticide concentration

2015

Newly affected areas

Sum of all pesticide concentration < 0,05 µg/l Sum of all pesticide concentration > 0,05 µg/l & drinking water threshold values not exceeded Drinking water threshold values exceeded for one individual substance of the sum of substances

Figure 2.3.1

Maps of pesticide concentration for 2003 and 2015.

66

Chapter 2.3

concentration for these wells would be above 0.05 mg l1 in 2015, i.e. 50% of the drinking water threshold value, thus with the possibility that drinking water thresholds are occasionally exceeded. The damage costs associated with the contamination of the 33 wells were estimated assuming that all concerned water utilities would install activated carbon filter treatment units for eliminating pesticides. In addition, collaborative agreement with farmers would be established to further reduce input of pesticides and nitrates in the nearby aquifer wells. Total costs, including treatment cost (h0.07 per m3) and compensations paid to farmers (h230 per hectare and per year), were estimated at h1.75 million per year in 2015.

2.3.4

Cost Effectiveness Analysis of Groundwater Protection Measures: Finding the Least Costly Way to Reduce Nitrate Pollution to Groundwater

The second example, based on a Slovenian case study, focuses on the identification of the least costly way to reduce nitrate pollution in groundwater. The aquifer selected for this case study is the Krsko kotlina aquifer located near the Croatian boarder in Slovenia.20 This aquifer faces increasing nitrate and pesticide pollution originating mainly from agriculture and municipal wastewater collection and treatment. If current pressures remain as they are today, nitrate concentration in groundwater is expected to rise above the 50 mg l1 drinking water threshold value. Pesticides are present at all monitoring locations, with the concentration of some pesticides increasing and being sometimes above threshold values specified in relevant legislation. Potential measures for reducing groundwater pollution and their cost-effectiveness were investigated in the context of a pilot project aimed at testing methodologies for supporting the implementation of the EU WFD.20 For the agriculture sector, measures considered included adopting agri-environmental measures with lower fertiliser use, the replacement of cropped area by meadows, in particular in water resources protection zones, better management of farm yard manure on farm or the shift to organic farming. For the municipal sector, measures identified included the installation of new sewage and wastewater treatment facilities, the renewal of leaking sewage, the connection of disconnected households to public sewage networks or the installation and efficient management of sceptic tanks. The expected effects in terms of reduction in nitrate leaching to the groundwater were estimated for individual measures. Costs that were considered included direct investment costs, operation and maintenance costs and in some cases indirect economic costs imposed on economic sectors (e.g. differences in farm income/value added when shifting from today’s agriculture to organic farming). All costs were annualised based on the expected time life of investments and the distribution of (direct, indirect) costs over time, using a discount rate of 7%. Table 2.3.1 presents estimates obtained for different measures and used for building the cost-effective

70 ha

2163 ha

4800 ha 2088 ha

0 ha

1431 ha

2400 ha

696 ha

40 ha 348 ha

0 ha

Shift to meadows in water protection zone 1 for Drnovo and Brege abstraction wells Limits imposed on fertiliser use (170 kg N ha1) in water protection zone 2 (both wells) Best management practice for agriculture Winter green cover

Buffer zones Ecological farming

Stricter limitation of fertiliser use in the Brege water protection zone 2 (100 kg ha1) 497 ha

139 ha 1044 ha

Maximum coverage

Potential measures 960 084 SIT ha1

797 SIT ha1

897 SIT ha1 28 997 SIT ha1 1764 SIT ha1 45 397 SIT ha1 1 104 097 SIT ha1

0.00652 mg l1 ha1

0.00056 mg l1 ha1

0.00056 mg l1 ha1 0.00119 mg l1 ha1

0.00596 mg l1 ha1

0.00585 mg l1 ha1 0.00176 mg l1 ha1

Annualised costs

Reduction in nitrates (in absolute terms)

0.0054

3.3172 0.0388

0.0410

0.6243

0.7023

0.0068

Cost effectiveness ratio (mg l1 SIT1)

Coverage, costs and effectiveness of potential measures (PE ¼ population equivalent).

Actual coverage (already implemented)

Table 2.3.1

10

1 6

5

4

3

9

Ranking based on costeffectiveness ratio

Groundwater Management and Planning: How Can Economics Help? 67

(continued )

Wastewater treatment and sewage for groups of houses with less than 50 PE Wastewater treatment and sewage for small villages (50 o PE o 2000) Wastewater treatment and sewage for larger settlements (PE 4 2000)

Stricter limitation of fertiliser use in the Drnovo water protection zone 2 (100 kg ha1) Improved septic tanks

Potential measures

Table 2.3.1

50 ha

10 194 PE 297 PE

1019 PE

8115 PE

2548 PE

0 PE

0 PE

0 PE

Maximum coverage

0 ha

Actual coverage (already implemented) 917 SIT ha1

26 042 SIT PE1 19 094 SIT PE1 10 115 SIT PE1 8941 SIT PE1

0.00184 mg l1 ha1

0.00010 mg l1 PE1

0.00017 mg l1 PE1

0.00017 mg l1 PE1

0.00010 mg l1 PE1

Annualised costs

Reduction in nitrates (in absolute terms)

0.0190

0.0168

0.0052

0.0038

0.0067

Cost effectiveness ratio (mg l1 SIT1)

7

8

11

12

2

Ranking based on costeffectiveness ratio

68 Chapter 2.3


69

programme of measures aim at stabilising average nitrate concentration in groundwater at 50 mg l1.20 The activity undertaken as part of the Krka pilot project had a clear methodological focus and was not aimed at delivering results as part of a decision-making process. However, the assessments showed that the most costeffective programme for reaching the 50 mg l1 threshold for nitrate concentration would require changes in agriculture for 5400 hectares, with measures for reducing municipal sector pollution targeting around 9100 Population Equivalent.20 Total costs of such a programme were estimated roughly at 340 million SIT (or h1.5 million), equivalent to an average cost of around h200 per hectare for the entire aquifer area. The analysis emphasised that implementing in priority basic measures required for the implementation of existing legislation (Wastewater Treatment Directive, Nitrate Directive), independently of their cost-effectiveness ratio, would lead to significantly higher costs for groundwater protection up to h370 per hectare.

2.3.5

Cost–Benefit Analysis of Groundwater Protection: Finding the Economically Optimal Level of Groundwater Protection

The fourth example illustrates how cost–benefit analysis can be applied to assess whether proposed quality objectives or threshold values proposed for groundwater protection can be justified from an economic point of view. It is based on the results of a case study developed in Latvia as part of the EUfunded BRIDGE research project (see Ref. 21 for more details). The case study investigated the economic consequences of two alternative groundwater quality objectives for petroleum products in the shallow groundwater aquifer underlying the capital city of Riga. There are two main types of pressures on the aquifer that explain pollution by petroleum products: historical pollution sources linked to former military zones, industries, the past operation of fuel filling stations and tanks or highly contaminated sites that are officially registered; and current sources of pollution such as fuel filling stations and tankers, parking places or car disposal places (car cemeteries) that represent potential sources of pollution. The latter has clearly a marginal impact on the existing pollution by petroleum products, with the former representing around 90–92% of the observed pollution. Some pollution in the shallow aquifer also originates from street runoff and unused wells. While the area polluted with petroleum products might appear rather small (97 polluted sites representing 70 hectares out of a total of 30 500 hectares for the entire city of Riga), extremely high concentrations in petroleum products can be found, exceeding the existing (environmental) water quality threshold value of 0.2 mg l1 set in the legislation by 100 to 1000 times. This does not impose threats on human health as shallow groundwater is rarely used for drinking water purposes (drinking water for the city of Riga comes from an upstream dam on the Daugava River). But it creates a permanent floating layer

70

Chapter 2.3

of petroleum products that, for contaminated sites close to surface waters, imposes a risk to connected surface water ecosystems. Some petroleum product pollution in the Daugava River crossing the city of Riga (close to its outfall into the Gulf of Riga) is, for example, directly coming from the connected shallow aquifer. Highly contaminated sites in the middle of the city also impose constraints on urban and economic development: as new owners of such sites need to clean (groundwater and soil) pollution (from mainly historical sources for which they have no responsibility) whenever new urban developments are proposed. Environmental policies and urban planning strategies are already in place for dealing with petroleum product pollution in shallow aquifers. The review of planning and strategic documents stressed that financial resources have already been secured for cleaning four highly polluted sites in the coming years. In addition, good management practices are compulsory for fuel filling stations, petrol stations and parking places. Clearly, however, these measures will not be sufficient for restoring good shallow groundwater quality in the entire aquifer. The analysis investigated in some details the economic implications of two different objectives or scenarios. Scenario 1 considers cleaning all sites polluted with petroleum products up to the actual 0.2 mg l1 threshold for 30 sites out of 97 that are highly polluted and (i) connected to surface water ecosystems or (ii) located in a current or future residential area. Scenario 2 is more ambitious and considers cleaning all (97) sites polluted with petroleum products. Different actions and measures that would help reducing petroleum product pollution were identified. These include pumping up existing floating layers, treating contaminated soils or directly shallow groundwater, building rain water collectors or installing and operating rain water pre-treatment facilities. Based on individual cost figures (investment, operation and maintenance costs) for each measure, total costs (annualised, using a 4% discount rate) were estimated for each scenario assuming that measures would be implemented by 2015 in both scenarios. Cost ranges were estimated at h26–30.5 million and h50–58.6 million for scenarios 1 and 2, respectively. The benefits accruing to urban and economic development were estimated, assuming these benefits would be equal to the total area cleaned under each scenario multiplied by surrounding land property values existing for new developments. Benefits were estimated by summing up the cadastral value of properties in Riga for each site multiplied by the concerned area of each site. Total benefits (annualised, using a 4% discount rate) accruing to urban and economic development were estimated at h1.9 million and h4.3 million for scenarios 1 and 2, respectively. To estimate total benefits resulting from groundwater quality improvements, a contingent valuation survey was carried out in the case study area. Overall, 510 citizens from the city of Riga and neighbouring areas were interviewed to capture the total value people attach to groundwater quality improvement and to the elimination of petroleum product pollution. Average WTP for removing petroleum products from shallow groundwater was estimated at h24.5 per household per year, with no significance difference between the WTP for


71

different levels of groundwater quality improvement corresponding to scenarios 1 and 2. This WTP value applied to all inhabitants of the city of Riga leads to total benefits at around h68 million (for more details, see Ref. 21), thus significantly higher than the benefits accruing from urban development presented above. This stresses the importance of the non-use value people attach to groundwater. The comparison between total (annualised) costs and benefits showed that both scenarios can be justified from an economic point of view—with the first scenario yielding a better economic outcome (total benefits minus total costs) than the second (more ambitious) scenario. This results mainly from the very high population (density) in the area (Riga is the capital city of Latvia and hosts nearly one-third of the entire population of the country) to which WTP values for groundwater quality improvements are applied. Further analyses investigated the time dimension of groundwater improvement programmes. If proposed measures are delayed and implemented for a longer time period to account for the availability of financial resources and budgetary constraints, annualised costs are lower. As annualised benefits are also lower (because benefits are obtained after a longer time period), net benefits would be higher than those calculated for a faster implementation time. This could be the basis to justify a longer implementation period for measures: a result which clearly is very site and pollutant specific and that cannot be extrapolated!

2.3.6

Integrating Economic and Groundwater Models for Simulating Nitrate Pollution in the Upper Rhine Valley Aquifer

The fourth example illustrates how economic and groundwater models can be integrated to investigate the environmental impact of changes in economic sectors/sector policy. The example is based on the results of the MONIT InterReg project conducted by French, German and Swiss partners in the upper Rhine valley aquifer.22 The objective of the project was to develop a modelling tool capable of simulating future evolution of nitrate concentration corresponding to different global (economic and environmental) change scenarios. The approach developed recognises that the dynamics of groundwater systems mainly depends on economic drivers influencing activities generating pollution (e.g. farming), and, subsequently, that the processes governing this economic activity must also be represented and analysed in an integrated model. A farm sector economic model was developed to assess the extent to which levels of pollution emission generated by the farming sector could change depending on global change scenarios. The economic model is developed using linear programming (LP). LP models assume that farmers select the combination of crops which maximises their income under a set of technical, regulatory and economic constraints.23 The models simulate crop choices, input

72

Chapter 2.3

consumption (fertiliser, labour, energy) and farm income for different input parameter values (agricultural prices and subsidies, regulatory constraints, changes in the price of input such as energy, fertiliser, labour, minimum setaside constraint, etc.). The models incorporate constraints related to crop rotations, labour availability, production quotas (sugar-beet and milk quotas), manure storage and management (for livestock-oriented farms). Risk is also integrated in LP models to account for the variability of output prices and yields. A crop production function (for corn only) is also introduced in the model to capture the crop yield response to nitrogen input. The LP models representing production choices of twenty selected representative farms were calibrated for sample farms by comparing simulation results with current cropping patterns. The output of the economic model (area under each crop, nitrate use) provides input data to a large-scale nitrate balance model which calculates nitrate leaching for the entire aquifer. This nitrate balance model is developed using the STOFFBILANZ software.23 Some parameters of STOFFBILANZ are adjusted through the use in simulation of a process-oriented nitrate input model developed at a plot scale (soil-plant-model STICS) to determine the nitrate input into the groundwater as a function of crop type, management practice, fertilising practice and climate. The output of the nitrate balance model feeds into the groundwater flow model coupled with a transport model, developed using MODFLOW and MT3D (for more details, see Ref. 23). The economic and soil–groundwater models were then used to simulate the impact of three global change scenarios developed by a group of experts. Given the uncertainty associated with future changes, three contrasted scenario were developed, a baseline scenario and two variants inspired from the scenario developed by the International Panel on Climate Change (IPCC).24 The baseline scenario assumes that the corn root worm detected since 2003 in the region extends over large areas, forcing farmers to increase crop rotation and limiting corn area to less than 50% of the total cultivable area. Energy prices (gazoil) are supposed to increase by 6% per year on average (2015 prices are twice those of 2003). And it is assumed that no financial compensation mechanism is implemented by national governments. As a result of energy price growth, fertiliser price increases by 1.5% per year. Due to the European enlargement, temporary labour force increases significantly (+66% in twelve years), reducing the profitability of vine, fruit and vegetable crops (in particular in Germany where foreign labour force is significantly used for harvesting of these crops). In France, it is assumed that the reform of the Water Act establishes a new water abstraction tax of h0.025 per m3. In Germany, the tax called Wasserpfennig is maintained at its 2003 level (i.e. h0.05 per m3). In both countries, farmers are allowed to produce bio-diesel for their own farm use only. And the bio-fuel industry develops but only for producing ethanol using corn (no bio-diesel industry). The second scenario, called ‘‘A1’’ with reference to IPCC emission scenario A1,24 depicts a more liberal future. Agriculture development aims at maximising competitiveness in markets which tend to function without protectionist


73

barriers and with minimal environmental constraints: taxes on water abstraction are eliminated in Germany and not introduced in France. Liberalisation favours the full implementation of the decoupling principle promoted by the Common Agriculture Policy reform in France, which result in a drastic change in crop gross margins and relative profitability. Energy price increase is compensated by a fiscal stabilisation mechanism and limited to 40 and 68% in Germany and France, respectively. Significant technical means are mobilised by government agencies to fight against the corn root worm (pesticides are sprayed by helicopter). And the biofuel industry develops, representing new market opportunities for farmers. The third scenario, inspired from the B2 IPCC emission scenario, corresponds to a vision of the future where agriculture evolved under the double pressure of increasing input prices (energy, fertiliser) and more stringent environmental constraints. Water abstraction taxes are established at the level of the baseline scenario. A tax on fertiliser is introduced at h0.15 per kg in France and h0.26 per kg in Germany. Due to high energy price and an active governmental support to bio-fuel development, crops used for bio-fuel production represent a very attractive market. The proliferation of the corn root worm compels farmers to reduce the area under corn. And financial support is granted to fruit and vegetable farms to invest in machinery and compensate for the increase of temporary labour costs. The consequences of these three global change scenarios were assessed using the chain of models described above. The results show that future nitrate concentration in the aquifer will not significantly differ from one scenario to the other in the short term (estimated at 19 mg l1 for the baseline and A1 scenarios, and 19.5 mg l1 for the B2 scenario). More significant differences are expected in the longer term (average concentration of 16 mg l1 for baseline and A1 scenarios, 18.2 mg l1 for the B2 scenario). The model shows that the area where nitrate concentration exceeds drinking water thresholds (50 mg l1) will drastically fall from 17 000 hectares in 2005 to around 4000 ha for the baseline and A1 scenarios, and to 6000 ha for the B2 scenario (Figure 2.3.2). Surprisingly, scenario B2 which depict a world with more stringent environmental constraints is also the worst scenario in terms of water pollution due to the increase in areas under industrial crops used for producing bio-fuels.

2.3.7

Designing Economic Instruments for Groundwater Management

As indicated above, recent policy developments have also put emphasis on the potential role economic instruments might play in enhancing the sustainability of water resources. In parallel to requesting EU member states to undertake economic assessments for supporting the selection of measures, the EU WFD also promotes a sounder application of the polluter pays principle, an adequate recovery of the costs of water services and water pricing that provides an incentive to more efficient water use. Furthermore, economic instruments such

74

Chapter 2.3 17000 Baseline

Area where [NO3] > 50 mg/l (ha)

15000

A1 B2

13000

11000

9000

7000

5000

3000 2005

Figure 2.3.2

2015

2025

2035

2045

Simulated evolution of the area (in ha) where nitrate concentration exceeds 50 mg l1 (adapted from Ref. 22).

as water charges/taxes and systems of tradable water rights can be considered when developing programmes of measures. Indeed, economic instruments that provide financial incentives to reduce groundwater abstraction might be relevant options when the ecological consequences of groundwater depletion (e.g. sea water intrusion in coastal areas, drying up of wetlands or reduced discharges of river springs) cannot be mitigated by technical solutions alone. The following paragraphs investigate the two main types of economic instruments: environment taxes and tradable groundwater licences/rights.

2.3.7.1

Environment Taxes and Charges

Environmental taxes and charges can be imposed on users who abstract groundwater as an incentive for reducing water use and thus pressures on the aquatic ecosystem. The higher the taxes, the lower abstraction is expected to be. Designing an environment tax system is, however, not trivial. From a theoretical point of view, the tax level should be such that it covers all costs generated by groundwater use, including external costs caused to third parties and to the environmentz. In practice, the design of a tax system consists in finding a z

Environmental costs can be significant where groundwater overexploitation leads to irreversible environmental and economic damages costs (e.g. abandonment of drinking water wells in case of sea water intrusion in coastal aquifer, destruction of natural habitat in wetlands).


Table 2.3.2

Unitary rates for abstraction taxes and charges for selected member states of the European Union (2004 values: in Ref. 27).

Country France (basic rate of the Seine Normandie river basin)

Estonia

75

Sources of water to which the Tax/charge unitary tax/charge is applied rate (h per m3) Surface water: on volume abstracted Surface water: on volume consumed Groundwater: on volume abstracted Groundwater: on volume consumed Surface water Groundwater

Hungary

All sources of water

Slovenia

All sources of water

0.00071 0.04 0.024 0.04 0.013–0.016 0.016–0.048 depending on use 1.147 for mineral water 0.007–0.02 depending on use 0.03

compromise between economic efficiency and social acceptability: the latter criterion being of utmost importance as taxes will generally be resisted by water users who consider water as a free resource.25,26 And the full cost recovery principle promoted by current policies (e.g. the EU WFD of 2000, the 1994 water reform in Australia) is in fact never implemented in practice. As a result, external costs and damages remain borne by economically less sensitive sectors (frequently households). Environmental taxes and charges are more widely used in Europe than in the USA for instancey. Such taxes or charges for water abstraction are rather common as illustrated in Table 2.3.2. When different rates are proposed for surface water and groundwater abstraction, those for groundwater are usually higher emphasising the higher status groundwater protection might have in the field of water policy. However, a closer look at current tax and charge levels shows that they remain very low as compared to other production costs or value added.28 In some cases, tax exemption is also given to economic sectors to limit its expected negative impact on competitiveness. This situation is illustrated by the Netherlands, where a groundwater tax system was introduced in 1995 as part of an effort to broaden the tax base by shifting the emphasis of revenue generation from conventional taxes to environmental taxes other than energy taxes. The amount of the tax payable is based on the volume of groundwater abstracted with a tax rate of h0.1785 per m3 (2004 figures). Companies could, however, obtain a significant tax rebate (h0.1495 per m3) when infiltrating water back y

See Ref. 25 for a discussion of water pollution taxes versus pollution discharge rights.

76

Chapter 2.3

into the aquifer. Furthermore, small groundwater abstractors were exempted. Although small- and medium-sized enterprises (SMEs) and industries faced a water price increase ranging from 40 to 113% when the groundwater tax was first introduced, the revenue collected by the groundwater tax amounted to 0.03 and 0.08% of the industry turnover and value added, respectively. This was equivalent to 0.33% of pre-tax profits in the industrial sector,28 thus unlikely to impair the competitiveness of Dutch industry. Overall, environmental taxes and charges are often justified on environmental grounds, i.e. as an incentive to economic sectors to limit water use or reduce pollution. In practice, however, existing taxes and charges do have limited incentive effects. They mainly play a revenue-raising role, revenues that in some cases (when they are earmarked to environmental protection) can help support financially the implementation of groundwater protection programmes and projects.

2.3.7.2

Tradable Groundwater Licences and Rights

The establishment of tradable groundwater abstraction licences or rights is another type of economic instrument that might play a role in groundwater management. Based on hydrogeological knowledge, the total maximum volume that can be abstracted from an aquifer (or sustainable yieldz) can be estimated and converted into a number of individual water abstraction licences or property rights which can be allocated to individual water users. Individual rights can be expressed in volume or in percentage of a sustainable aquifer yield, which can be reassessed every year to account for climatic variability and the effect of past abstractions. Whereas licences are issued for a specific period and can be revised or cancelled, water (property) rights are granted on a permanent basis. The initial allocation of licence or rights can be based on historical consideration (first in time, first in right) or on economic consideration (through an auction mechanism, for example, to allocate groundwater to its highest marginal values). Following the initial allocation, water licences or rights can be leased or sold among users. The underlying assumption is that water users will accept to sell part or whole of their licence/right if the financial compensation (price) they can obtain for the transfer exceeds the benefits they would derive from using their water directly. In theory, market forces ensure that water is (re-)allocated to the most efficient users, that waste is minimised and that the total economic value of water use is maximised. Water markets have been established in the USA, leading to the reallocation of surface and groundwater resources from one region to another, and from one economic sector to another.29,30 For a detailed description of groundwater markets in Texas and Arizona, see Ref. 31. Similar markets also function in well-regulated river basins in arid area of Chile.32,33 Following the successful z

Sustainable aquifer yield is defined as the groundwater extraction regime, measured over a specified planning time frame that allows acceptable levels of environmental impact on associated ecosystems and protects the higher value uses that have a dependency on the water.


77

development of water market in the Murray–Darling basin in the mid1980s,33,34 Australian authorities have also promoted the development of groundwater licence trading as an alternative to the ‘‘command and control’’ approach to moving water use to areas with higher economic value and efficiency.35 In Europe, although trading of water use rights has locally been functioning for centuries within irrigation systems in Spain,30 they are rarely considered as measures in water management and policy. As an exception, tradable water permits (quotas) were considered as possible measure for reducing overabstraction for the Beauce aquifer (central France) at the end of the 1990s.36 However, they were never implemented because of the drastic legislation change they would have required. The EU WFD refers also to tradable water rights as possible measure for reaching its environmental objectives. But no EU member state is expected to make use of this type of economic instrument at least in the short and medium term.

2.3.8

Conclusions

This chapter has provided examples of the role economics can play in supporting decisions in the field of groundwater management and planning. At the same time, the illustrations presented above clearly show that economics cannot answer policy issues and questions alone. Indeed, supporting groundwater management and planning decision requires multidisciplinary approaches where economics and technical knowledge must be well integrated. Integration between economics with hydrogeology is required for the development of scenarios of plausible futures described for example in Sections 3 and 6. In developing scenarios, economic assessment or modelling can either be used ‘‘upstream’’ or ‘‘downstream’’ from hydrogeological assessment and modelling. An ‘‘upstream’’ use of economics is important when a significant evolution of economic activities generating pressures on groundwater (abstraction or pollution) is expected to occur during the time horizon considered. As illustrated in Section 6, an economic model can then help in simulating likely future evolution of economic activities and resulting pressures affecting the state of groundwater resources, under different assumption of global economic change (e.g. changes in agriculture policy, demographic growth, changes in energy prices), environmental change (e.g. climate change) or policy action scenario (e.g. implementation of an environmental tax). A ‘‘downstream’’ use of economics is relevant to situations with significant economic interests depending on the state of groundwater resources. As illustrated in Section 3, this can help assess the economic consequences (e.g. in terms of economic sector production and value added, employment, competitiveness) of expected future groundwater status changes obtained from hydrogeological model simulation. Multidisciplinary approaches which integrate economics with hydrogeology are also essential to the evaluation of alternative groundwater management

78

Chapter 2.3

options, in line with the requirements of the EU WFD. Whether the evaluation requires a cost-effectiveness analysis (as illustrated in Section 4) or a cost– benefit analysis (presented in Section 5), economic assessments must build on a robust understanding of the impact of possible management actions on groundwater status. And such understanding requires hydrogeological knowledge or modelling. However, studies where specific groundwater models and economic assessment are conducted together remain rare.37,38 This leads to a relatively high uncertainty associated with cost-effectiveness or cost–benefit estimates: an issue that requires further investigation by the research community. Although the need for multidisciplinary research is increasingly recognised by scientific and policy-making communities,39 the actual development of multidisciplinary research work in that field is somewhat impaired by the fact that academic evaluation (in both the hydrogeology and economics fields) rarely gives credit to such multidisciplinary approaches. The search for more multidisciplinary work reinforces the issues of scale. On the one hand, it questions the extrapolation to larger scales of results from innovative economic methods developed and applied at local case study level, in particular in the domain of environmental cost and benefit valuation (e.g. contingent valuation). The issue of aggregation of values obtained from contingent valuation studies for estimating total benefits of a given management option and environmental quality change at the river basin scale remains a challenge. Whether the same protocol should apply for building contingent valuation surveys at local or large scale is also a research question that might require further attention. On the other hand, economists are faced with the need to link global economic models representing entire economies (partial equilibrium models or general equilibrium models) to impacts on water and groundwater resources at the macroscale. This implies linking macroeconomic models with groundwater models at regional or national level, a scale at which hydrological functioning is difficult to simplify (and might not have much meaning). Another key challenge for economists working on groundwater management is the need to enhance the appropriation of economic principles, tools and results by policy-makers and stakeholders involved in the preparation of groundwater management plans and programmes. Whereas policy actors might be, by tradition, more familiar with the knowledge mobilised by hydrogeologists, there is still a limited understanding of the role economics can play in supporting their decision—although it is clear that policy-makers are very sensitive to economic arguments. In Europe, this situation is slowly changing with the implementation of the EU WFD gaining momentum and more economic analyses being undertaken and discussed among policy-makers and stakeholders at the national and river basin district scales. Also, one can observe an emerging demand for economic assessments of water management options at the local level (for instance in the Arde`che river basin in France). But we are still at the infancy of more systematic applications of economic assessments. Also, how the results of economic analyses will be used for supporting policy decisions (if used at all) remains to be seen.


79

A better appropriation of economic concepts and methods by policy-makers and stakeholders will require specific efforts by the economic community in awareness raising and information, including on the strengths and weaknesses of assessments, on underlying uncertainties and on the demonstration of the added value of the integration between economics and technical expertise. In some cases, it might also require a drastic change in the economists’ thinking and discourse, away from the assumption that individuals are acting purely along the simplified concepts and principles that are the basis of the microeconomic theory. The real-life application of economic assessments that will be produced in the coming years in the context of the implementation of the WFD and of its daughter groundwater directive will be instrumental in enhancing the shared knowledge between economists, hydrogeologists, policy-makers and stakeholders.

References 1. World Wide Fund for Nature and European Environment Bureau, EU Water Policy: Making Economics Work for the Environment. Survey of the Economic Elements of the Article 5 Report of the EU Water Framework Directive, report prepared by the World Wide Fund for Nature and the European Environment Bureau, Brussels, 2006. 2. A. Scheidleder, J. Grath and G. Winkler, Groundwater Quality and Quantity in Europe, European Environmental Agency, Copenhagen, 1999. 3. S. Nixon, Z. Trent and C. Marcuello, Europe’s Water. An Indicator Based Assessment, Topic Report 1/2003, European Environmental Agency, Copenhagen, 2003. 4. T. Shah and D. Molden D, The Global Groundwater Situation: Overview of Opportunities and Challenges, International Water Management Institute, Colombo, 2000. 5. W. Harrington, J. Krupnick and W. A. Spofford, J. Urban Econ., 1989, 25, 116–137. 6. P. Dasgupta, Environ. Develop. Econ., 2004, 9, 83–106. 7. A. H. Smith, E. O. Lingas and M. Rahman, Bull. World Health Organ., 2000, 78, 9. 8. C. W. Abdalla, Am. J. Agricul. Econ., 1994, 76, 1062–1067. 9. A. Stenger and M. Willinger, J. Environ. Manag., 1998, 53, 177–193. 10. B. Go¨rlach and E. Intervies, Economic Assessment of Groundwater Protection: A Survey of the Literature, Ecologic, Berlin, 2003. 11. A. Gerarsidi, P. Katsiardi et al., Cost-effectiveness analysis for water management in the island of Paros, Greece, International Conference on Environmental Science and Technology, Lemnos Island, Greece, 2003. 12. M. Montginoul, J.-D. Rinaudo, P. Garin, Y. Lunet delajoncquie`re and J. P. Marchal, Water Policy, 2005, 7, 523–541. 13. S. Feuillette and F. Bousquet et al., Environ. Model. Software, 2003, 18(5), 413–427.

80

Chapter 2.3

14. C. W. Howes and F. P. Linaweaver, Water Resour. Res., 1967, 3(1), 13–32. 15. F. Arbue´s and M. A. Garcia-Valinas et al., J. Socio-Econ., 2003, 32(1), 81–102. 16. R. E. Howitt, Am. J. Agricul. Econ., 1995, 77, 329–342. 17. L. Judez and J. M. de Miguel et al., Math. Comput. Modell., 2002, 35(1–2), 77–86. 18. J.-D. Rinaudo, C. Arnal, R. Blanchin, P. Elsass, Meilhac and Loubier, Water Sci. Technol., 2005, 52(9), 153–162. 19. J.-D. Rinaudo, P. Elsass, R. Blanchin, C. Arnal and Meilhac, Assessing the Social and Economic Impact of Nitrate and Pesticide Contamination of the Alsatian Aquifer: Synthesis Report [in French], report Brgm/RP-53172FR.Orleáns, BRGM, 2006. 20. Selecting Measures to Improve Water Status in the Krka River Sub-Basin, technical report of the Krka Pilot Project, Ljubljana, 2006. 21. K. Pakalniete, H. Bouscasse and P. Strosser, Assessing Socio-Economic Impacts of Different Groundwater Protection Regimes. Latvian Case Study Report, research report developed under the EU-funded BRIDGE research project, ACTeon, Orbey, 2006. 22. LUBW, Simulating Nitrate Groundwater Pollution in the Upper Rhine valley [in French and German], Monit Intergreg IIIA project, final report, LUBW, Karlsruhe, 2006. 23. P. B. R. Hazell and R. D. Norton, Mathematical Programming for Economic Analysis in Agriculture, Macmillan, New York, 1986. 24. Intergovernmental Panel on Climate Change, Emissions Scenarios: Summary for Policy Makers, IPCC special report, 2000. 25. C. W. Howes, Environ. Res. Econ., 1994, 4, 151–169. 26. A. Dinar, The political economy context of water pricing reforms, in The Economics of Water Management in Developing Countries, ed. P. Koundouri, P. Pashardes, T. Swanson and A. Xepapadeas, Edgard Edwards, Cheltenham, UK, 2003. 27. P. Strosser and S. Speck, Environmental Taxes and Charges in the Water Sector. A Review of Experience In Europe, technical report, Catalan Water Agency, Barcelona, 2004. 28. Ecotec et al., Study on the Economic and Environmental Implications of the Use of Environmental Taxes and Charges in the European Union and Member States, report for the European Union, European Commission, Brussels, 2001 (http://europa.eu.int/comm/environment/enveco/taxation/ environmental_taxes.htm). 29. T. Anderson and P. J. Hill, Water Marketing: The Next Generation, Rowman and Littlefiel, New York/London, 1997. 30. P. Strosser, Analysing alternative policy instruments for the irrigation sector: an assessment of the potential for water market development in the Chishtian sub-division, Pakistan, PhD thesis, Wageningen Agricultural University, 1997. 31. C. W. Howes, Environ. Develop. Econ., 2002, 7(2), 605–616.


81

32. C. J. Bauer, Against the Current: Privatization, Water Markets and the State in Chile, Kluwer, Boston MA, 1998. 33. H. Bjornlund and J. McKay, Environ. Develop. Econ., 2002, 7(2), 769–795. 34. S. Thoyer, Int. J. Sust. Develop., 2006, 9, 2. 35. ARMCANZ, Allocation and Use of Groundwater: A National Strategy for Improved Groundwater Management in Australia, policy position paper for advice to states and territories, Agriculture and Resource Management Council of Australia and New Zealand, Task Force on COAG Water Reform, Sustainable Land Water Resources Management Committee, 1996. 36. N. Kosciusko-Morizet, H. Lamotte and V. Richard, What can we expect from the establishment of transferable water entitlements in France: the case of irrigated agriculture [in French], Conference on Irrigation and Collective Water Resource Management in France and in the World, SFER, Montpellier, 19–20 November 1998. 37. J.-D. Rinaudo and S. Loubier, Cost benefit analysis of large scale groundwater remediation in France, in Cost Benefit Analysis and Water Resources Management, ed. R. Brouwer and D. Pearce, Edgard Edwards, Cheltenham, UK/Northampton, MA, 2005, pp. 290–314. 38. R. Brouwer, S. Hess, M. Bevaart and K. Meinardi, The Socio-Economic Costs and Benefits of Environmental Groundwater Threshold Values in the Scheldt Basin in the Netherlands, case study report produced under the BRIDGE research project, IVM report R06-05, Institute for Environmental Studies (IVM), Vrije Universiteit Amsterdam, 2006. 39. Ph. Quevauviller, P. Balabanis, C. Fragakis, M. Weydert, M. Oliver, A. Kaschl, G. Arnold, A. Kroll, L. Galbiati, J. M. Zaldivar and G. Bidoglio, Environ. Sci. Pol., 2005, 8, 203–211.

3. Groundwater Regulatory Framework

CHAPTER 3.1

European Union Groundwater Policyw PHILIPPE QUEVAUVILLER European Commission, DG Environment (BU9 3/142), Rue de la Loi 200, BE-1049 Brussels, Belgium

3.1.1

Introduction

The European Union (EU) regulatory groundwater framework was developed at the end of the 1970s with the adoption of Directive 80/68/EEC on the protection of groundwater against pollution caused by certain dangerous substances. This directive provides a groundwater protection framework by preventing the (direct or indirect) introduction of high-priority pollutants into groundwater and limiting the introduction into groundwater of other pollutants so as to avoid pollution of this water by these substances. In 1982, a major assessment of groundwater resources was carried out in Europe, which consisted of a general survey (Groundwater Resources of the European Community: Synthetical Report) of groundwater quantity. Since it was published, attention has turned in Europe (and the USA) to quality, and not only have groundwater quality monitoring programmes been greatly expanded but many groundwater protection schemes have been put into place. The declaration of the ministerial seminar on groundwater held at The Hague in 1991 recognised the need for further action to avoid long-term deterioration of the quality and quantity of freshwater resources and called for a programme of actions to be implemented by the year 2000, aiming at sustainable management and protection of freshwater resources. This was followed by the elaboration of an action programme for integrated protection and management of groundwater in 1996, in which the Commission pointed to the need to establish procedures for the regulation of abstraction of freshwater and for the monitoring of freshwater quality and quantity. These w

The views expressed in this chapter are purely those of the author and may not in any circumstances be regarded as stating an official position of the European Commission.

85

86

Chapter 3.1

considerations coincided with the request made by the European institutions to the Commission to come forward with a proposal for a directive establishing a framework for a European water policy. They were hence naturally embedded into the development of this large policy framework development, which resulted in the adoption of the Water Framework Directive (WFD) 2000/60/ EC on 23 October 2000. This chapter summarises the different elements described above, and provides background information about parent legislation which has also to be considered within an enlarged groundwater regulatory framework.

3.1.2

The 1980 Groundwater Policy Framework

In the context of Directive 80/68/EEC on the protection of groundwater against pollution caused by certain dangerous substances,1 groundwater is defined as ‘‘all water which is below the surface of the ground in the saturation zone and in direct contact with the ground or subsoil.’’ This directive provides a groundwater protection framework by preventing the (direct or indirect) introduction of high-priority pollutants (List I) into groundwater and limiting the introduction into groundwater of other pollutants (List II) so as to avoid pollution of this water by these substances (Table 3.1.1). Indirect discharges have to be understood as ‘‘the introduction into groundwater of substances in lists I or II after percolation through the ground or subsoil’’ while direct Table 3.1.1 List I

List II

Lists of substances regulated under Directive 80/68/EEC.

This list contains eight groups of substances, exception being made of substances which are considered inappropriate on the basis of low risk of toxicity, persistence and bioaccumulation: (1) organohalogen compounds and substances which may form such compounds in aquatic environment; (2) organophosphorus compounds; (3) organotin compounds; (4) substances which posses carcinogenic, mutagenic or teratogenic properties in or via the aquatic environment (if this is the case for certain substances of List II, they are included under this category); (5) mercury and its compounds; (6) cadmium and its compounds; (7) mineral oils and hydrocarbons; and (8) cyanides. This list contains individual or groups of substances which could have a harmful effect on groundwater, in particular: (1) metalloids and metals and their compounds such as zinc, copper, nickel, chrome, lead, selenium, arsenic, antimony, molybdenum, titanium, tin, barium, beryllium, boron, uranium, vanadium, cobalt, thallium, tellurium, silver; (2) biocides and their derivatives not appearing in List I; (3) substances which have a deleterious effect on the taste and/or odour of groundwater, and compounds liable to cause formation of such substances so as to render water unfit for human consumption; (4) toxic or persistent organic compounds of silicon, and substances which may cause the formation of such compounds in water, excluding those which are biologically harmless or are rapidly converted in water into harmless substances; (5) inorganic compounds of phosphorus and elemental phosphorus; (6) fluorides; and (7) ammonia and nitrites.

European Union Groundwater Policy

87

discharges correspond to ‘‘an introduction without percolation.’’ Pollution is defined as ‘‘The discharge by man, directly or indirectly, of substances or energy into groundwater, the results of which are such as to endanger human health or water supplies, harm living resources and the aquatic ecosystems or interfere with other legitimate use of water.’’ In this framework, consequences of pollution that has already occurred have to be checked or eliminated as far as possible (Article 1). This implies the following. With regard to List I substances, direct discharges are prohibited, whereas indirect discharges (due to disposal or tipping for the purpose of disposal) of these substances are prevented, which is linked to an authorisation procedure preceded by a thorough investigation on a caseby-case basis. In this respect, all appropriate measures have to be taken to prevent any indirect discharges due to either disposal or other activities on or in the ground other than disposal. With regard to List II substances, direct discharges have to be limited and appropriate measures have to be taken to limit any indirect discharges of these substances due to either disposal or other activities on or in the ground other than disposal. An authorisation procedure preceded by a thorough investigation is required in the case of direct discharge or disposal or tipping for the purpose of disposal of these substances. The authorisation is only granted if all the technical precautions for preventing groundwater pollution by these substances are observed. It should be noted that this directive does not apply to discharges of domestic effluents from isolated dwellings not connected to a sewerage system and situated outside areas protected for the abstraction of water for human consumption. In addition, it does not apply to discharges of List I and II substances which are found in a quantity and concentration so small as to obviate any present or future danger of deterioration in the quality of the receiving groundwater, nor does it apply to discharges of matter containing radioactive substances. Another derogation clause concerns the authorisation of discharge of List I substances in groundwater which has been revealed as being permanently unsuitable for other uses (especially domestic or agricultural), providing that their presence does not impede exploitation of ground resources. These authorisations can only be granted if all technical precautions have been taken to ensure that these substances cannot reach other aquatic systems or harm other ecosystems. In addition, authorisation (after prior investigation) may be granted for discharges due to re-injection into the same aquifer of water used for geothermal purposes, water pumped out of mines and quarries or water pumped out for civil engineering works. Finally, artificial recharges for the purpose of groundwater management are subject to a special authorisation on a case-by-case basis, which may only be granted if there is no risk of groundwater pollution. The directive provides specific requirements regarding

88

Chapter 3.1

the authorisation procedures, distinguishing direct discharge and indirect discharge. In the above context, monitoring is required only for those specific cases of authorisation for the purpose of compliance checking and for assessing the effects of discharges on groundwater. Application of measures relevant to this directive may on no account lead, either directly or indirectly, to pollution of groundwater. Finally, where appropriate, one or more member states may individually or jointly take more stringent measures than those provided for under this directive. From the above description, it can be concluded at first sight that Directive 80/68/EEC ensures a stringent groundwater protection regime against pollution for all the activities that present a risk of groundwater deterioration through direct or indirect discharges of a wide range of pollutants. The implementation of this directive is, however, sometimes faced with the difficulties of a lack of groundwater quality data and objectives. In other words, infringement cases may be difficult to judge in some instances in the absence of clear information on background groundwater quality levels in the zone affected by discharges, and of quality objectives on the basis of which deterioration may unambiguously be identified. This directive will be repealed in 2013 under the WFD (2000/60/EC),2 after which the protection regime should be continued through the WFD and the new Groundwater Daughter Directive,3 which are further discussed below.

3.1.3

Preliminary Assessment (1982)

In 1982, the Directorate-General for the Environment, Consumer Protection and Nuclear Safety of the European Community carried out a major assessment of groundwater resources within its (then) nine member states. It consisted of a general survey (Groundwater Resources of the European Community: Synthetical Report) and individual reports from each member state.4 This report dealt with four major themes:

the aquifer inventory: location and type; groundwater hydrology: flows within these aquifers; groundwater abstraction; and groundwater availability by area.

The study concluded that the Community had enough groundwater to meet most of its needs but that in most countries abstraction rates were already high: Belgium was abstracting some 70% of its available groundwater resources, Denmark 40%, France 25–50%, Italy 50%, Luxemburg 37%, the Netherlands 62% and the UK at least 25%. Only Ireland, which was abstracting some 3% of its groundwater resources, had major untapped resources. This assessment was mainly concerned with groundwater quantity. Since it was published, attention has turned in Europe (and the USA) to quality, and not only have groundwater quality monitoring programmes been greatly


89

expanded but many groundwater protection schemes have been put into place. These considerations have progressively led authorities to reflect on the need to develop a more integrated protection regime for groundwater, both quantitatively and qualitatively. These steps are summarised below.

3.1.4

The Groundwater Action Programme (1996)

The declaration of the ministerial seminar on groundwater held at The Hague in 1991 recognised the need for further action to avoid long-term deterioration of the quality and quantity of freshwater resources and called for a programme of actions to be implemented by the year 2000, aiming at sustainable management and protection of freshwater resources. The final declaration stated that: groundwater is a natural resource with both ecological and economic value, which is of vital importance for sustaining life, health, agriculture and the integrity of ecosystems; groundwater resources are limited and should be protected on a sustainable basis; and it is essential to protect groundwater resources from over exploitation, adverse changes in hydrological systems resulting from human activities, and pollution, many forms of which can produce irreversible damage. The declaration stressed that the objective of sustainability should be implemented through an integrated approach, meaning that: surface water and groundwater should be managed as a whole, paying equal attention to both quality and quantity aspects; all interaction with soil and atmosphere should be taken into account; and water management policies should be integrated within the wider environmental framework as well as with other policies dealing with human activities such as agriculture, industry, energy, transport and tourism. Requests made by the Council in 1992 and 1995 in the form of resolutions recommended an action programme and a revision of Directive 80/68/EEC to be undertaken. This was followed up by the presentation by the Commission of a proposal for a Decision of the European Parliament and of the Council on an action programme for Integrated Protection and Management of Groundwater,5 which was adopted on 25 November 1996, and in which the Commission pointed to the need to establish procedures for the regulation of abstraction of freshwater and for the monitoring of freshwater quality and quantity. The action programme was designed along four main lines of action which are summarised below. (i) Action line 1. Planning and management principles: protection and use of groundwater is conceived here in the context of integrated planning and management of fresh water resources, i.e. groundwater is seen as an

90

Chapter 3.1

integral part of the hydrological cycle dynamically interacting with surface water in terms of quantity as well as quality aspects. Protection measures related to (over)exploitation, pollution prevention and integrated planning are described at member state and at Community levels. (ii) Action line 2. Abstraction of freshwater: in this action line, a rational regulatory framework for the abstraction of freshwater in large urban, industrial and agricultural areas, and tourist centres is discussed. Recommendations are expressed on how to secure an appropriate quantity management of groundwater and surface water within each river basin. Here again, actions are defined at member state and Community levels. (iii) Action line 3. Diffuse sources of pollution: environmental challenges from diffuse sources of pollution are discussed, highlighting the difficulty of identifying individual polluters, in particular for groundwater pollution where the time-lag between application or release of polluting substances and the possibility for detection of their presence in the groundwater may span up to several decades. The action distinguishes diffuse sources from agricultural and industrial activities, traffic and urbanisation either through local impacts or long distance via atmospheric deposition. It provides recommendations to reduce and where possible avoid threats to groundwater from diffuse sources in the form of policy recommendations linked to environmental sustainability of agriculture, environmental challenges from nitrates and other mineral emissions and from the use of sewage sludge. (iv) Action line 4. Control of point-source pollution from activities and facilities which may affect groundwater quality: completing action line 3, the recommendations focus on protection from activities and installations producing liquid and solid effluent and/or representing a potential risk of accidental pollution of groundwater resources. Point sources are in principle traceable to specific activities and may be prone to measures at the source to avert or limit spreading of polluting substances. The action line refers to member states’ responsibilities with regard to authorisation regimes in place in parent legislation (e.g. the Integrated Pollution Prevention and Control Directive) and actions at Community level. The action programme clarifies the role of the Commission, and provides a framework for national action programmes, implementation and review. Considerations expressed in this communication coincided with the request made by the European institutions to the Commission to come forward with a proposal for a directive establishing a framework for a European water policy. They were hence naturally taken into account in the development of this large policy framework development, which resulted in the adoption of the WFD 2000/60/EC on 23 October 2000 (see Section 4). Some of them, however, have still not been taken on board, and are still to be considered in the integrated groundwater management under development within the WFD and the new Groundwater Directive.


3.1.5

91

The Groundwater Policy Framework Under the WFD

The WFD (Directive 2000/60/EC) took the 1996 action programme into account and was set to complement the Directive 80/68/EEC by stipulating that member states should implement the measures necessary to prevent or limit the input of pollutants into groundwater and to prevent the deterioration of the status of all bodies of groundwater. In this context, member states have to protect, enhance and restore all bodies of groundwater, ensure a balance between abstraction and recharge, with the aim to achieve good groundwater (chemical and quantitative) status by 2015, following the definitions given in Table 3.1.2. These requirements include a range of derogation clauses which are summarised in Table 3.1.3. An overview of the different articles and annexes of the directive is given in Appendix I. The WFD also requires the implementation of measures necessary to reverse any significant and sustained upward trend in the concentration of any

Table 3.1.2

Definitions of good quantitative and chemical status.

Ref. WFD

Good status

Good quantitative status (Annex V.2.1.2)

The level of groundwater in the groundwater body is such that the available groundwater resource is not exceeded by the long-term annual average rate of abstraction. Accordingly, the level of groundwater is not subject to anthropogenic alteration such as would result in: (a) failure to achieve the WFD environmental objectives for associated surface waters, (b) any significant diminution in the status of such waters, and (c) any significant damage to terrestrial ecosystems which depend directly on the groundwater body. Alterations to flow direction resulting from level changes may occur temporarily, or continuously in a spatially limited area, but such reversals do not cause saltwater or other intrusion, and do not indicate a sustained and clearly identified anthropogenically induced trend in flow direction likely to result in such intrusions. The chemical composition of the groundwater body is such that the concentrations of pollutants do not exhibit the effects of saline or other intrusions (as determined by changes in conductivity) into the groundwater body, do not exceed the quality standards applicable under other relevant Community legislation in accordance with Article 17 of the WFD and are not such as would result in failure to achieve the WFD environmental objectives for associated surface waters nor any significant diminution of the ecological or chemical quality of such bodies nor in any significant damage to terrestrial ecosystems which depend directly on the groundwater body.

Good chemical status (Annex V.2.3.2)

92

Table 3.1.3

Chapter 3.1

Derogation clauses.

Article (WFD)

Derogation

4.4

Extensions may be granted when improvements of status cannot be reasonably achieved within the timescales for reasons of technical feasibility, disproportionate costs or natural conditions. The extension request has to be explained in the river basin management plan under Article 13, as well as a summary of measures required under Article 11 and the reason for the delay in making these measures operational. Extensions are limited to a maximum of two further updates of the RBMP (2027) except where the natural conditions are such that the objectives cannot be achieved within this period. Less stringent environmental objectives may be set out for specific bodies of water when they are so affected by human activity (as determined by the analysis of pressure and impact under Article 5), or their natural condition is such that the achievement of these objectives would be infeasible or disproportionately expensive and that (a) the environmental and socioeconomic needs served by such human activity cannot be achieved by other means; (b) member states ensure the least possible changes to good groundwater status considering that impacts could not have reasonably been avoided due to the nature of the human activity of pollution; (c) no further deterioration of the affected body of water occurs; (d) the establishment of less stringent objectives and the reasons for it are specified in the RBMP and this is reviewed every six years. Derogation also concerns temporary deterioration due to natural causes or force majeure which are exceptional or not foreseeable (e.g. extreme floods or droughts, accidents), providing that (a) all practicable steps to prevent further deterioration are taken, (b) the circumstances are declared in the RBMP, (c) measures are included in the programme of measures, (d) an annual review is undertaken and all practical restoration measures are taken in order to recover the initial status, and (e) a summary of effects of the circumstances and measures are included in the next update of the RBMP. Member states will not be in breach of the directive when failure to achieve good groundwater status or to prevent deterioration in the status of a body of groundwater is the result of alterations to the level of bodies of groundwater, providing that (a) all practical steps are taken to mitigate the adverse impact on the status of the body of water, (b) the reasons for those alterations are set out and explained in the RBMP, (c) the reasons for those alterations are of overriding public interest and/or the benefits of the alterations outweigh those of achieving the WFD environmental objectives, and (d) the beneficial objectives served by these alterations cannot be achieved by other reasons for reasons of technical feasibility or disproportionate costs. The application of the above derogation clauses should not exclude or compromise the achievement of the directive objectives in other bodies of water within the same river basin district, and is consistent with the implementation of other Community environmental legislation.

4.5

4.6

4.7

4.8


93

pollutant resulting from the impact of human activity in order to progressively reduce groundwater pollution. Under this directive, the framework for groundwater protection imposes the following on member states. Delineate groundwater bodies within river basin districts to be designed and reported to the European Commission by member states, and characterise them through an analysis of pressures and impacts of human activity on the status of groundwater in order to identify groundwater bodies presenting a risk of not achieving WFD environmental objectives. This characterisation work had to be carried out in 2004–2005 and reported to the European Commission following requirements summarised in Tables 3.1.4 and 3.1.5. A report giving a synthesis of member states’ reports has been prepared by the European Commission and made available on the europa website in March 2007. Establish registers of protected areas within each river basin districts for those groundwater areas or habitats and species directly depending on water, which had to be carried out in 2004–2005. The registers have to include all bodies of water used for the abstraction of water intended for human consumption6 and all protected areas covered by the Bathing Water Directive 76/160/EEC,7 vulnerable zones under the Nitrates Directive 91/676/EEC8 and sensitive areas under the Urban Wastewater Directive 91/271/EEC,9 as well as areas designated for the protection of habitats and species including relevant Natura 2000 sites designated under Directives 92/43/EEC10 and 79/409/EEC.11 Registers should be reviewed under the River Basin Management Plan (RBMP; see below) updates. In this context, vulnerable zones are defined as ‘‘all known areas of land in member states territories which drain into the waters affected by pollution and waters which could be affected by pollution and which contribute to pollution.’’ For these vulnerable zones, action programmes are required under the Nitrates Directive to reduce pollution caused or induced by nitrates and prevent further pollution. Based on the results of the characterisation phase, establish a groundwater monitoring network providing a comprehensive overview of groundwater chemical and quantitative status, and design a monitoring programme that had to be operational by the end of 2006. Monitoring will have to be reported, following requirements summarised below and detailed in Chapter 6.1. Set up a RBMP for each river basin district which will include a summary of pressures and impact of human activity on the groundwater status, a presentation in map form of monitoring results, a summary of the economic analysis of water use, a summary of the programme(s) of protection, control or remediation measures, etc. The first RBMP is scheduled to be published at the end of 2009. A review is then planned by the end of 2015, and every six years thereafter. By 2010, take account of the principle of recovery of costs for water services, including environmental and resource costs, having regard to

94

Chapter 3.1

Table 3.1.4

Characterisation.

Annex II.2 (WFD)

Characterisation

Initial characterisation (par. 2.1)

The initial characterisation concerns all groundwater bodies, assessing their uses and the degree at which they are at risk to meet WFD environmental objectives. This analysis may use existing hydrological, geological, pedological, land use, discharge, abstraction and other data, identifying: the location and boundaries of the groundwater body or groups of bodies, the pressures to which the groundwater is subject to (diffuse and point sources of pollution, abstraction, artificial recharge), the general character of the overlying strata in the catchment area from which the groundwater body receives its recharge, and those groundwater bodies for which there are directly dependent surface water ecosystems or terrestrial ecosystems. It concerns the groundwater (or groups of) bodies which have been identified as being at risk, and aims to establish a more precise assessment of the significance of such risks and the identification of any measures top be required under the WFD Article 11. This characterisation has to include relevant information on the impact of human activity and, where relevant, on geological characteristics of the groundwater body (including the extent and type of geological units), hydrogeological characteristics (including hydraulic conductivity, porosity and confinement), characteristics of the superficial deposits and soils in the catchment from which the groundwater body receives its recharge (including the thickness, porosity, hydraulic conductivity, and adsorptive properties of the deposits and soils), stratification characteristics of the groundwater, an inventory of associated surface systems (including terrestrial ecosystems and bodies of surface water, with which the groundwater body is dynamically linked), estimates of the direction and rates of exchange of water between the groundwater body and associated water systems, sufficient data to calculate the long-term annual average rate of overall recharge, and characterisation of the chemical composition of the groundwater (including specification of the contribution from human activity—member states may use typologies for groundwater characterisation when establishing natural background levels for these bodies of groundwater).

Further characterisation (par. 2.2)

the economic analysis conducted under Article 5 of the WFD, and in accordance with the polluter pays principle. Establish a programme of measures for achieving WFD environmental objectives (e.g. abstraction control, prevent or control pollution measures) by the end of 2009, to be operational by the end of 2012. Basic


Table 3.1.5

95

Review of impacts on groundwater and authorisations.

Annex II.2 (WFD)

Reviews of impacts

Impact of human activity (par. 2.3)

For transboundary groundwater bodies (crossing the borders of two or more member states) or bodies identified at risk following the initial characterisation, additional information, where relevant, will have to be collected and maintained for each groundwater body: (a) location of points in the groundwater body used for the abstraction of water (with the exception of points providing less that 10 m3/day or points for abstraction of water intended for human consumption providing less than 10 m3/day or serving less than 50 persons); (b) the annual average rates of abstraction from such points; (c) the chemical composition of water abstracted from the groundwater body; (d) the location of points in the groundwater body into which water is directly discharged; (e) the rates of discharges at such points; (f) the chemical composition of discharges to the groundwater body; and (g) land use in the catchment (or catchments) from which the groundwater body receives its recharge, including pollutant inputs and anthropogenic alterations to the recharge characteristics such as rainwater and run-off diversion through land sealing, artificial recharge, damming or drainage. Bodies for which lower objectives are to be specified (see Table 3.1.2) have to be identified by member states, including consideration of the effects of the status of the body on (i) surface water and associated terrestrial ecosystems, (ii) water regulation, flood protection and land drainage, and (iii) human development. Similarly, bodies of groundwater for which lower objectives are to be specified under Article 4.5 of the WFD (see Table 3.1.2) have to be identified as a result of the analysis of impact of human activity (Article 5.1).

Impacts of change in groundwater levels (par. 2.4)

Impact of pollution on groundwater quality (par. 2.5)

Article (WFD)

Authorisations

11.3( j)

Authorisations concern: (a) reinjection into the same aquifer of water used for geothermal purposes; (b) injection of water resulting from hydrocarbon extraction or mining activities into geological formations which for natural reasons are permanently unsuitable for other purposes; (c) reinjection of pumped groundwater from mines and quarries or associated with the construction or maintenance of civil engineering works; (d) injection of gas or liquefied petroleum for storage purposes into geological formations which for natural reasons are permanently unsuitable for other purposes, or where there is an overriding need for security of gas supply and where the injection is such as to prevent future deterioration of the receiving groundwater; (e) construction, civil and building works or similar activities on or in the ground which come into contact

96 Table 3.1.5

Chapter 3.1 (continued )

Article (WFD)

Authorisations with groundwater, in accordance with general binding rules developed by the member states; (f) discharges of small quantities of substances for scientific purposes, providing that such discharges do not compromise the achievement of environmental objectives established for that body of groundwater.

measures include, in particular, controls over the abstraction of groundwater, controls (with prior authorisation) of artificial recharge or augmentation of groundwater bodies (providing that it does not compromise the achievement of environmental objectives). Point-source discharges and diffuse sources liable to cause pollution are also regulated under the basic measures. Direct discharges of pollutants into groundwater are prohibited subject to a range of provisions summarised in Table 3.1.5. The programme of measures has to be reviewed and if necessary updated by 2015 and every six years thereafter. Strategies to prevent and control pollution of groundwater are covered by Article 17 of the WFD, which requires the establishment of criteria for assessing good groundwater chemical status and for the identification of significant and sustained upward trends and for the definition of starting points for trend reversals, considering the following. The characterisation of bodies of groundwater as detailed in Annex II.2 of the WFD (see Table 3.1.4). Good status definitions as detailed in Table 3.1.2, which are based on groundwater level regime (quantitative status) and conductivity and concentrations of pollutants (chemical status). Monitoring requirements to respond to the needs of obtaining a comprehensive overview of groundwater status and to detect the presence of long-term anthropogenically induced upward trends in pollutants. In this respect, surveillance monitoring is aimed at supplementing and validating the impact assessment procedure (carried out under Article 5 of the WFD) and provide information for use in the assessment of long-term trends both as a result of changes in natural conditions and through anthropogenic activity, while operational monitoring should be undertaken in the periods between surveillance monitoring programmes in order to establish the chemical status of all groundwater bodies or groups of bodies determined as being at risk and to establish the presence of any long-term anthropogenically induced upward trend in the concentration of any pollutant. Further details are given in Section 5. Monitoring results should be used to identify long-term anthropogenically induced upward trends in pollutant concentrations and to set up starting points for reversing these trends.


97

Article 17 requests the European Commission to present a proposal based on the above requirements. This proposal of a new groundwater directive was issued in 20033 and resulted in the new Groundwater Directive of which the main orientations are described in the next section.

3.1.6

The New Groundwater Directive 2006/118/EC

While quantitative status requirements are clearly covered by the WFD, it does not include, however, specific provisions on chemical status, i.e. the different conceptual approaches to groundwater protection did not allow achieving an agreement on detailed provisions within the WFD at the conciliation. As mentioned in Section 5, this justified including a provision, Article 17, requesting the Commission to come forward with a proposal of specific measures to prevent and control groundwater pollution. This proposal was adopted by the Commission on 19 September 2003 (COM(2003)550 final) and was adopted after a conciliation phase among the European Parliament and the Council on 12 December 2006.12 Directive 2006/118/EC is based on the following three main pillars. (i) Criteria linked to good chemical status evaluation, which are based on compliance with EU existing environmental quality standards (nitrates, plant protection products and biocides) and to ‘‘threshold values’’ (playing the same role as EQS) for pollutants representing a risk to groundwater bodies. The latter category of standards has to be established by member states, using common methodological criteria (which were the basis of research developments under the BRIDGE project; see Chapter 9.1), at the most appropriate scale (national, regional or local), taking account hydrogeological conditions, soil vulnerability, types of pressures, etc. They will have to be reported to the Commission by the end of 2008, and will be used as quality objectives for further compliance checking. (ii) Criteria for the identification of sustained upward trends of pollutants in groundwater bodies characterised as being at risk. These include measurement principles and requirements regarding trend reversals. (iii) Requirements on the prevention/limitation of pollutant inputs to groundwater, which will ensure a continuity of Directive 80/68/EEC after its repeal in 2013, i.e. the same principle of prevention of hazardous substances introduction and limitation of other pollutants so as to avoid pollution will apply. Other elements concern clarifications about the groundwater use as drinking water (albeit this is well covered by Article 7 of the WFD) and its relation with the present directive, which relates to WFD environmental objectives. Recommendations to undertake research on groundwater ecosystems are also expressed in a recital, which illustrates the awareness for required scientific integration. Finally, review of technical annexes of the directive (in particular concerning the establishment of groundwater threshold values and methods for

98

Chapter 3.1

identifying and reversing pollution trends) is requested, taking into account scientific progress, before the end of 2012 and every six years thereafter. This review will have to be carried out following ‘‘comitology’’ rules, i.e. adoption of possible decisions by a regulatory committee composed of member states. Since 2006, these rules imply that the European Parliament will have right of scrutiny on adopted decisions. An evaluation of the functioning of the directive in the light of consistency with parent legislation (see Section 7) is also foreseen. In summary of Sections 2, 5 and 6, the WFD2 (including the new ‘‘daughter’’ groundwater directive3) will complement and ensure a continuity of the Directive 80/68/EEC protection regime.1 This will be achieved through a systematic analysis of pressures and impacts (not done under Directive 80/68/EEC), and requirements related to good chemical status and pollutant trend identification/ reversal backed up by surveillance and monitoring programmes. The programme of measures also sets out provisions that are aimed at replacing the existing protection regime. The new groundwater directive aims to provide the necessary common criteria regarding chemical status evaluation, identification and reversal of significant and upward trends in pollutant concentrations, as well as specific clauses regarding direct and indirect inputs of pollutants into groundwater to make sure that the existing protection regime will be appropriately strengthened. An overview of the different articles and annexes of the directive is given in Appendix II.

3.1.7

Policy Integration

One of the key aspects of environmental integration in the light of the EU Sustainable Development Strategy is linked to policy coordination and integration. With regard to the groundwater policy framework, this policy integration appears to be quite complex since it concerns a range of various directives as illustrated in Figure 3.1.1. This section examines how various relevant directives interact with the groundwater policy under the WFD and Directive 80/68/EEC.

3.1.7.1

Nitrates Directive

The Nitrates Directive8 aims to reduce water pollution caused or induced by nitrates from agricultural sources and to prevent further such pollution. It obliges member states to designate vulnerable zones which correspond to all known areas of land in member state territories which drain into the waters (including groundwater) affected by pollution and waters which could be affected by pollution and which contribute to pollution. A reference is made to action programmes to reduce pollution caused or induced by nitrates and to prevent further pollution, and to requirements for identifying groundwater vulnerable zones as ‘‘those waters which contain more than 50 mg l 1 or could contain more than 50 mg l 1 nitrates if an action programme is not undertaken.’’ The link with groundwater policy is clear in that respect, i.e. nitrate

99

European Union Groundwater Policy Policy integration Seveso, IPPC

Env. Impact assessment Birds, Habitats

URBAN SECTOR

DRINKING WATER

AGRICULTURE

INDUSTRY

Drinking water

Sewage sludge landfill Urban waste water, Construction Products Bathing water

Pesticides, Nitrates, biocides

groundwater

WFD

Figure 3.1.1

The overall groundwater policy framework: integration needs.

contamination levels should not be over the trigger value set at 50 mg l 1 (this argument has been used for proposing this value as an EU groundwater quality standard for groundwater in the new Groundwater Directive (see Section 6). The Nitrates Directive8 requires the implementation of suitable monitoring programmes to assess the effectiveness of action programmes at selected measuring points, making it possible to establish the extent of nitrate pollution in the waters from agricultural sources. The designation and monitoring of vulnerable zones is to be carried out at regular intervals at sampling stations which are representative of groundwater aquifers, taking into account the provisions of the Drinking Water Directive.6 The monitoring has to be repeated at least every four years, except for those sampling stations where the nitrate concentration in all previous samples has been below 25 mg l 1 and no new factor likely to increase the nitrate content has appeared (in which case the monitoring programme needs to be repeated only every eight years). The directive also stipulates that reference methods of measurement have to be used. This, however, concerns freshwaters, coastal waters and marine waters (i.e. no specific mention is made of groundwater).

3.1.7.2

Urban Wastewater Treatment Directive

The Urban Wastewater Directive9 aims to protect the environment from the adverse effects of discharges of urban wastewater and wastewater from certain industrial sectors. In this context, the identification of ‘‘sensitive areas’’ relates essentially to freshwaters, estuaries or coastal waters which are found to be eutrophic, lakes and streams reaching lakes/reservoirs/closed bats with poor water exchange and surface freshwaters intended for the abstraction of drinking water which could contain more than 50 mg l 1 nitrates. This directive is

100

Chapter 3.1

indirectly relevant to groundwater (protection of receiving groundwaters from possibly contaminated wastewaters originating from freshwater sources). The Urban Wastewater Directive9 monitoring obligations are directly related to verifications of appropriate treatment, prior regulations and/or specific authorisations of discharges from urban waste treatment plants to freshwaters and estuaries. In the framework of this directive, monitoring will have to focus on discharges from urban wastewater treatment plants to verify compliance with requirements set out in the directive (corresponding to criteria concerning different types of discharges) and following control procedures laid down in the annex (reference monitoring methods and evaluation of results). These requirements focus on flow-proportional or time-based 24-hour sample collection at welldefined points in the wastewater treatment plant outlet and if necessary in the inlet in order to monitor compliance with the directive’s requirements for discharged wastewater. They include an obligation to apply good international laboratory practices in order to minimise the degradation of samples between collection and analysis. Note that these monitoring obligations do not concern groundwater.

3.1.7.3

Plant Protection Products Directive

The Plant Protection Products Directive13 concerns the authorisation, placing on the market, use and control within the Community of plant protection products in commercial form. Regarding groundwater, authorisations are only granted when plant protection products have no harmful effect on humans or human health, directly or indirectly, or on groundwater, and they have no unacceptable influence on the environment, particularly contamination of water including drinking water and groundwater. The ‘‘uniform principles’’ set out in the directive specify that no authorisation shall be granted if the concentration of the active substance or of relevant metabolites, degradation or reaction products in groundwater, may be expected to exceed, as a result of use of the plant protection product under the proposed conditions of use, the lower of (i) the maximum permissible concentration laid down by Directive 80/778/ EEC,7 or (ii) the maximum concentration laid down by the Commission when including the substance listed in the directive, on the basis of appropriate data (in particular toxicological data), or where that concentration has not been laid down, the concentration corresponding to one tenth of the ADI (acceptable daily intake) laid down when the active substance was included in the directive. The monitoring obligations concern the authorisation regime imposed by the member states according to the directive’s provisions. Decision-making provisions are included in the annex to the directive. The granting of authorisations has to take account of the agricultural, plant health or environmental (including climatic) conditions in the areas of envisaged use (this implicitly concerns groundwater, even if this is not specifically mentioned). These considerations may result in specific conditions and restrictions of use and, where necessary, in authorisation being granted for some but not other areas within the member state. The control measures are obviously linked to the current analytical knowledge (and authorisation may be limited to a limited period if limitations


101

in analytical science and technology are recognised), with requirements regarding the method’s reproducibility. As mentioned in Section 6, the directive makes a direct reference to groundwater contamination (with drinking water standards not allowed to be exceeded), which therefore requires monitoring. There are no specific monitoring criteria in this respect other than the mention that analytical methods must reflect the state of the art, and analytical criteria on the method’s performance as set out in the annex.

3.1.7.4

Biocides Directive

The Biocides Directive14 concerns the authorisation and the placing on the market for use of biocidal products. Similarly to Directive 91/414/EEC,13 authorisation of biocidal products may only be granted if the products have no harmful effect on humans or human health, directly or indirectly, or on groundwater, and they have no unacceptable influence on the environment, particularly contamination of water including drinking water and groundwater. Similar principles as the ‘‘uniform principles’’ of Directive 91/414/EEC are set out, which means that the 0.1 mg l 1 quality standard of 80/778/EC6 plays a role of maximum concentration for all groundwater, but that lower standards may be established following the procedure for including the active substance in Annex I of the directive. The decision-making provisions of the annex to the Biocides Directive14 follow the same lines as that described above (related to the Plant Protection Products Directive) with respect to groundwater. Monitoring obligations are closely linked to the authorisation regime which requests a prior risk assessment for which criteria are defined in the evaluation provisions of the same annex. This risk assessment has to take into account any adverse effects arising in any of the three environmental compartments—air, soil and water (including sediment)—and of the biota. The analytical work has, therefore, to focus on the properties and potential adverse effects of the active substances present in the biocidal product for its classification. In case this classification is not possible, information on bioaccumulation potential, persistence characteristics, information from toxicity studies, etc., have to be taken into account. If appropriate, adequately measured exposure data, likely pathways to environmental compartments, potential for adsoption/desorption and degration, etc., have to be evaluated. This obviously includes effects on groundwater. Specific monitoring requirements are, however, not included, except the mention that testing should be carried out according to Community guidelines if these are available and applicable. Where appropriate, other methods can be used (e.g. ISO, CEN or other international standard method, national standard method or other methods accepted by the member state) and if relevant field data exist, these can also be used.

3.1.7.5

IPPC Directive

The Integrated Pollution Prevention and Control (IPPC) Directive15 concerns integrated pollution prevention and control, which lays down measures designed to prevent or reduce emissions in the air, water and land from a range of

102

Chapter 3.1

activities listed in Annex I of the directive. It establishes provisions for issuing permits for existing and new installations, and makes a specific reference to groundwater, indicating that the permits shall include appropriate requirements ensuring protection of the soil and groundwater on the basis of emission limit values for pollutants which may be supplemented or replaced by equivalent parameters or technical measures based on best available techniques. The permit procedure under the IPPC Directive15 includes a provision for suitable release monitoring, specifying measurement methodology and frequency, evaluation procedure and obligation to supply data required for checking compliance with the permit. The directive includes a provision for installations that may have significant negative effects on the environment of another member state. Monitoring is focused on the releases, the results of which have to be regularly submitted by the operator to the competent authority (and without delay in the case of any incident or accident significantly affecting the environment). There are no specific monitoring requirements for groundwater, but the directive’s provisions obviously imply that risks to groundwater be appropriately monitored.

3.1.7.6

Landfill Directive

The Landfill Directive16 concerns the landfill of waste, which aims to provide for measures, procedures and guidance to prevent or reduce as far as possible negative effects on the environment, including groundwater. Similarly to the IPPC Directive,15 the directive establishes provisions for issuing permits based on a range of conditions including impact assessment studies. Regarding groundwater, site characteristics have to locate groundwater and geological and hydrogeological conditions in the area, prevent groundwater from entering into the landfilled waste, take appropriate measures to collect/treat contaminated water and leachate and prevent pollution of the soil, groundwater or surface water using appropriate technical precautions (e.g. combination of geological barrier and bottom liner). The directive establishes criteria for waste testing and acceptance, taking due consideration on the protection of the surrounding environment, including groundwater. The Landfill Directive16 imposes control and monitoring procedures with a frequency which is to be defined by the competent authority (and in any event at least once a year) and on the basis of aggregated data, in order to demonstrate compliance with permit conditions. The corresponding article notifies that the quality control of the analytical operations of the control and monitoring procedures are carried out by competent laboratories. Further requirements are provided in the annex. They include reporting obligations for meteorological data (volume of precipitation, temperature, wind, evaporation, atmospheric humidity) to check whether leachate is building up in the landfill body or whether the site is leaking. Sampling of leachate and surface water is also required to be collected at representative points (for surface water, no less than two points, i.e. one upstream from the landfill and one downstream). A separate section on the protection of groundwater is included, which requests the provision of information on groundwater likely to be affected by the discharging


103

of waste, with at least one measuring point in the groundwater inflow region and two in the outflow region (this number can be increased on the basis of a specific hydrogeological survey and the need for an early identification of accidental leachate release in the groundwater). Sampling has to be carried out in at least three locations before the filling operations in order to establish reference values for future sampling (following the requirement of the ISO 5667 standard Sampling Groundwaters, Part 11, 1993). The parameters to be analysed in the samples taken must be derived from the expected composition of the leachate and the groundwater quality in the area, with account of mobility in the groundwater zone and a frequency adapted to the local conditions. Adverse effects are considered to have occurred when an analysis of the groundwater sample shows a significant change in water quality as defined by a trigger level which should be determined by the competent authority (taking account of the specific hydrogeological formations in the location of the landfill and groundwater quality) and laid down in the permit whenever possible.

3.1.7.7

Sewage Sludge Directive

The Sewage Sludge Directive17 seeks to encourage the use of sewage sludge in agriculture and to regulate its use in such a way as to prevent harmful effects on soil, vegetation, animals and humans. To this end, it prohibits the use of untreated sludge on agricultural land unless it is injected or incorporated into the soil. Treated sludge is defined as having undergone ‘‘biological, chemical or heat treatment, long-term storage or any other appropriate process so as significantly to reduce its fermentability and the health hazards resulting from its use.’’ The directive also requires that sludge should be used in such a way that account is taken of the nutrient requirements of plants and that the quality of the soil and of the surface and groundwater is not impaired. It sets out requirements for the keeping of detailed records of the quantities of sludge produced, the quantities used in agriculture, the composition and properties of the sludge, the type of treatment and the sites where the sludge is used. Limit values for concentrations of heavy metals in sewage sludge intended for agricultural use and in sludge-treated soils are given in annexes to the directive. In the framework of the Sewage Sludge Directive,17 monitoring requirements are focused on specifies rules for the sampling and analysis of sludges and soils, i.e. there are no specific requirements concerning groundwater.

3.1.7.8

Other Directives

Article 4 of the WFD18 also requires that waste be recovered or disposed of without endangering the environment, which may have an (indirect) effect on protecting groundwater. Finally, the Construction Product Directive19 concerns regulatory provisions for construction products. It indirectly concerns groundwater in that construction products for construction works have to be fit for their intended

104

Chapter 3.1

use and respond to requirements regarding hygiene, health and the environment; in particular there should be no threat to the hygiene or health of occupants or neighbours as a result of pollution or poisoning of the water or soil. The directive focuses on conformity aspects of construction products, taking into account possible risk to water environments (in particular release of dangerous substances to water). As such, the directive does not provide for specific groundwater monitoring other than the requirement than a verification that the construction work is designed and built in such a way that it will not generate pollution of the water or soil. Conformity testing of construction products is generally carried out at the factory or on site from a batch which is ready for delivery. The surveillance concerns the factory production and product testing rather than monitoring possible effects on the environment.

3.1.8

Conclusions: The Need for Worldwide Cooperation

Groundwater protection regimes against pollution and overexploitation are established worldwide either in the form of national laws or regional, national or international conventions, as illustrated by a recent compilation of treaties and other legal instruments produced by FAO/UNESCO.20 These regulatory frameworks should ensure that appropriate actions are being undertaken to protect, enhance and restore the quantitative and quality status of groundwaters. However, much remains to be done to actually effectively implement the different legal instruments. The UN International Law Commission (ILC), which is the UN body in charge of the progressive development and the codification of international law, adopted at first reading in June 2006 a set of draft articles on the law of transboundary aquifers.21 The ILC has sent the draft articles to states, requesting them to provide comments and observations by 1 January 2008. In Europe, efforts linked to the WFD Common Implementation Strategy22 enable exchange of experiences, discussion of difficulties and finding possible solutions in the framework of an international participatory approach (see Chapter 4.1). This certainly constitutes an excellent platform for implementing groundwater protection under the WFD, the new Groundwater Directive and parent legislation at the EU scale, but this will require a constant coordination and accompanying efforts to make this happen. The example could be extended to international cooperation, and initiatives are already under way, e.g. in the Mediterranean basin, leading to shared recommendations.23 Finally, international programmes such as the UNESCO International Hydrological Programme24 offer excellent opportunities to disseminate expertise and experiences, and participate in worldwide harmonisation of groundwater protection regimes and education. UNESCO-IHP has contributed to the work of ILC mentioned above on the draft articles on transboundary aquifers by providing the Special Rapporteur with scientific and technical support and assistance on the science of hydrogeology (more details can be found at www.isarm.net).


105

References 1. Council Directive 80/68/EEC of 17 December 1979 on the protection of groundwater against pollution, Official Journal of the European Communities L 20, 26.1.1980, p. 43. 2. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy, Official Journal of the European Communities L 327, 22.12.2000, p. 1. 3. Proposal for a Directive of the European Parliament and of the Council on the protection of groundwater against pollution, COM(2003)550 final. 4. Groundwater Resources of the European Community: Synthetical Report, European Commission, 1982. 5. Proposal for a European Parliament and Council Decision on an action programme for integrated groundwater protection and management, Official Journal of the European Communities C 355, 25.11.1996, p. 1. 6. Council Directive 80/778/EEC of 15 July 1980 relating to the quality of water intended for human consumption, Official Journal of the European Communities L 229, 5.12.1998, p. 32. 7. Council Directive 76/160/EEC of 8 December 1975 concerning the quality of bathing water, Official Journal of the European Communities L 31, 5.2.1976, p. 1. 8. Council Directive 91/676/EEC of 12 December 1991 concerning the protection of waters against pollution caused by nitrates from agricultural sources, Official Journal of the European Communities L 375, 31.12.1991, p. 1. 9. Council Directive 91/271/EEC of 21 May 1991 concerning urban waste treatment, Official Journal of the European Communities L 135, 30.5.1991, p. 40. 10. Habitats Directive 92/43/EEC, Official Journal of the European Communities L 206, 22.7.1992, p. 7. 11. Birds Directive 79/409/EEC, Official Journal of the European Communities L 103, 25.4.1979, p. 1. 12. Directive of the European Parliament and of the Council of 12 December 2006 on the protection of groundwater against pollution and deterioration, Official Journal of the European Communities, L 372, 12.12.2006, p. 19. 13. Council Directive of 15 July 1991 concerning the placing of plant protection products on the market, Official Journal of the European Communities L 230, 19.8.1991, p. 1. 14. Directive 98/8/EC of the European Parliament and of the Council of 16 February 1998 concerning the placing of biocidal products on the market, Official Journal of the European Communities L 123, 24.4. 1998, p. 1. 15. Council Directive of 24 September 1996 concerning integrated pollution prevention and control, Official Journal of the European Communities L 257, 10.10.1996, p. 26.

106

Chapter 3.1

16. Council Directive 99/31/EC of 26 April 1999 on the landfill of waste, Official Journal of the European Communities L 182, 16.7.1999, p. 1. 17. Sewage Sludge Directive 86/278/EEC, Official Journal of the European Communities L 181, 8.7.1986, p. 6. 18. Waste framework directive 75/442/EEC amended by Directive 91/156/ EEC. 19. Council Directive of 21 December 1988 on the approximation of laws, regulations and administrative provisions of the member states relating to construction products, Official Journal of the European Communities L 040, 11.12.1989, p. 12. 20. S. Burchi and K. Mechlem, Groundwater in International Law, FAO Legislative Study, FA0/UNESCO, 86, 2005. 21. Report of the International Law Commission, 58th session, A/61/10, 2006, p. 185 (http://www.un.org/law/ilc/). 22. Common Implementation Strategy for the Water Framework Directive, European Communities, 2003 (ISBN 92-894-2040-5). Final CIS document available at: http://europa.eu.int/comm/environment/water/waterframework/implementation.html. 23. Technical Report on Groundwater Management in the Mediterranean and the Water Framework Directive (http://www.emwis.org/GroundwaterHome. htm). 24. UNESCO, International Hydrological Programme (http://www.unesco. org/water).

CHAPTER 3.2

US Drinking Water Regulation: The Ground Water Rulew CRYSTAL RODGERS-JENKINS US EPA, 1201 Constitution Avenue NW, MC-4607M, Washington, DC 20460, USA

3.2.1

Introduction

The United States Environmental Protection Agency (EPA) finalised the Ground Water Rule (GWR) in 2006.1 The EPA established the GWR due to concerns related to faecal contamination in groundwater systems (GWSs). The occurrence of faecal contamination in groundwater sources is an indication that viral and bacterial pathogens may also be present. The goal of the GWR is to provide Americans that drink water from public groundwater sources with water that is safe for human consumption. The GWR is the first federal drinking water regulation in the USA that requires GWSs, not under the influence of surface water, to monitor their groundwater sources for faecal indicators. In addition to source water monitoring, the GWR requires periodic sanitary surveys of GWSs to inspect for significant deficiencies that may lead to contamination. The rule requires GWSs that have an indication of faecal contamination and GWSs with significant deficiencies to take corrective action to ensure that the drinking water from groundwater sources is safe for human consumption. Prior to the GWR, the only microbial US drinking water regulations that applied to GWSs were the Total Coliform Rule (TCR) and, in part, the Surface Water Treatment Rule (SWTR). The TCR applied to all public water systems and the SWTR applied to public water systems using surface waters or groundwaters under the direct influence of surface water. The TCR and SWTR were published in 1989.2,3 The TCR requires that each GWS monitor its distribution system for total coliforms and faecal coliforms/E. coli; no monitoring w

The views expressed in this chapter are those of the individual author and do not necessarily reflect the views and policies of the US Environmental Protection Agency.

107

108

Chapter 3.2

is required at the groundwater source. All GWSs are required to comply with the TCR provisions, although for some GWSs with pristine groundwater sources—for which the system can demonstrate it has a protected groundwater source and has no history of total coliform presence—the State may allow a reduced distribution system monitoring frequency. Under the TCR, periodic sanitary surveys must also be performed at small GWSs;2 however, the frequency and scope of the TCR sanitary surveys differ from the GWR requirements. The sanitary survey frequency and scope are discussed in Section 3.2.4. The SWTR requires that all GWSs under the direct influence of surface water to disinfect and achieve at least 3- and 4-log removal and/or inactivation of Giardia and viruses, respectively. Since 1989 the EPA has finalised five other microbial drinking water rules for surface water systems: the Information Collection Rule,4 the Interim Enhanced Surface Water Treatment Rule,5 the Filter Backwash Recycling Rule,6 the Long Term 1 Enhanced Surface Water Treatment Rule7 and the Long Term 2 Enhanced Surface Water Treatment Rule.8

3.2.2

Federal Statutory Authority

The GWR was established pursuant to the 1996 amendments to the Safe Drinking Water Act (SDWA). In the 1996 SDWA amendments, US Congress authorised the EPA to require disinfection, if necessary, as a treatment technique for GWSs and to establish criteria for determining when disinfection is needed. The GWR applies to all public water systemsz that use groundwater sources, in whole or in part (including consecutive systems that receive finished groundwater from another public water system). Public water systems that combine all of their groundwater with surface water prior to treatment at a surface water treatment plant or groundwater under the direct influence of surface water are not required to comply with GWR. These systems must comply with the Surface Water Treatment Rule.3 Private wells are not regulated under the GWR or by the EPA.

3.2.3

Challenges in Developing the GWR

In 2000, the EPA published the proposed GWR9 for public review. The EPA received several thousand comments on the proposed regulation. The EPA, water industry and States shared a common interest: public health protection. However, the three entities had various perspectives on which requirements would be necessary to protect public health. In general, the water industry focused on minimal capital and operational and maintenance costs which translates to fewer rule requirements; and States—oftentimes struggling with implementing existing regulations due to limited personnel and financial z

Public water systems are systems that provide drinking water to the public through pipes or other constructed conveyances, if such system has at least 15 service connections or regularly serves an average of at least 25 people per day at least 60 days out of the year.

US Drinking Water Regulation: The Ground Water Rule

109

constraints—emphasised the need for greater flexibility. The EPA’s goal in developing the GWR was to meet the federal statutory requirement through a scientifically sound, flexible and cost-effective regulation. The EPA met numerous challenges in achieving this goal. Some of the main challenges were: the number of GWSs: approximately 150 000 systems; the number of very small GWSs: more than 60% of the GWSs serve r500 people; the lack of a single, comprehensive national microbial occurrence study of GWSs in the USA; and limited data on public health risks related to exposure to microbial pathogens in groundwater.

3.2.3.1

US Groundwater System Demographics

A total of approximately 150 000 GWSs are unevenly spread across the country’s extensive (over 3.7 million square miles (over 9.6 million km2))10 and extremely varied landscape. The water quality of the public wells that provide groundwater to GWSs also differs throughout the US regions. In some geographical areas in the USA, the raw groundwaters from public wells are clean and safe to drink or the GWSs may need to provide very little treatment prior to human consumption. US groundwater occurrence data and information on the waterborne disease outbreak in groundwater systems suggest this is not the case for other areas in the USA. The nearly 150 000 GWSs supply drinking water to approximately 114 million people. The EPA divides public water systems into three categories: community water systems (CWS), non-transient non-community water systems (NTNCWS) and transient non-community water systems (TNCWS).y Nontransient non-community water systems and transient non-community water systems account for about 70% of the total number of GWSs in the USA. A relatively small percentage of the population (13%) consumes groundwater supplied by NTNCWSs or TNCWSs. About 70% of NTNCWS and TNCWSs serve o100 people and over 60% of CWSs serve o500 people. During implementation of the GWR, GWSs that serve very small populations (e.g. o500 people) will have the greatest economic impact compared to larger GWSs due to economies of scale. The GWR requires systems of all sizes to adhere to rule requirements intended to ensure the same level of public health protection. The EPA addressed the challenges of the large number of GWSs and the substantial percentage of very small GWSs through GWR’s risk-targeted strategy. This strategy targets the subset of the total number of GWSs that are susceptible to y

Community water systems serve at least 25 people year round. Non-transient non-community water systems provide water to the same 25 people for at least six months per year (e.g. schools). Transient non-community water systems serve 25 persons per day for 60 days per year (e.g. restaurants).

110

Chapter 3.2

Table 3.2.1

Ground Water Rule population and system baselines.

System size (population served) CWS r100 101–500 501–1000 1001–3300 3301–10 000 10 001–50 000 50 001–100 000 100 001–1 million 41 million National total NTNCWS r100 101–500 501–1000 1001–3300 3301–10 000 10 001–50 000 50 001–100 000 100 001–1 million 41 million National total TNCWSs r100 101–500 501–1000 1001–3300 3301–10 000 10 001–50 000 50 001–100 000 100 001–1 million 41 million National total Grand total

Total population

Total number of ground water systems

749 084 3 377 075 3 310 229 10 867 911 16 242 873 29 704 225 10 889 872 21 334 033 3 933 533 100 408 836

12 843 14 358 4649 5910 2884 1444 167 103 3 42 361

476 998 1 527 684 1 274 145 1 109 731 376 195 207 644 66 000 110 000 – 5 148 396

9456 6758 1894 715 73 10 1 1 – 18 908

2 461 310 3 360 731 1 242 779 828 502 371 291 299 629 51 850 125 000 – 8 741 092 114 298 324

64 448 18 993 1940 585 74 19 1 1 – 86 061 147 330

Note: detail may not add to totals due to independent rounding. Source: Ref. 14.

faecal contamination and requires these at-risk systems to take corrective action. The strategy is discussed in greater detail later in Section 3.2.4. The population and system baselines of the GWR are summarised in Table 3.2.1.

3.2.3.2

Occurrence Data

Occurrence data play a major role in risk assessments conducted for drinking water regulations. Some regulations, for example the Long Term 2 Enhanced Surface Water Treatment Rule, are based on comprehensive occurrence


111

studies—the Information Collection Rule (ICR) and the ICR Supplemental Surveys—conducted specifically for that regulation.11 Groundwater is a vital natural resource in the USA; however, there are a limited number of studies that assess the groundwater quality. The final GWR risk assessment relied on occurrence data from multiple studies due to lack of a single study on groundwater systems across the USA. The EPA’s GWR risk assessment analyses data from 15 occurrence studies to obtain an estimate of the national occurrence of viral pathogens and faecal indicators, specifically evaluating the occurrence of enteroviruses based on culture methods, and E. coli.12 When the EPA proposed the GWR, it based the estimate of the national occurrence of pathogenic viruses and faecal indicators on two occurrence studies although 16 studies were evaluated. The two studies were more comprehensive than the other 14 studies and were selected because they provided a relatively large amount of occurrence data among different geographical areas: one of the studies included data from 17 States and 2 US territories, the other study analysed groundwater sources in 35 States and 2 US territories. The remaining 14 studies were more limited in scope and mostly individual State occurrence studies. The EPA intended to use the two datasets selected for the proposed GWR risk analysis to estimate the national occurrence of pathogenic viruses and faecal indicators. The EPA did not include bacterial pathogen data in the risk analysis due to the lack of available occurrence data. Based on the objectives and scope of the studies, the EPA preliminarily concluded that the two sets of data represent the national pathogenic virus and faecal indicator occurrence and concentrations in US groundwater sources.9 The EPA received adverse comments during the public comment period for the proposed GWR related to the EPA’s reliance on two studies to represent national pathogenic virus and faecal indicator occurrence in US groundwater sources. Therefore, the EPA re-evaluated the 16 occurrence studies cited in the proposed GWR and several other studies that became available after the GWR was proposed.13 The EPA faced the challenge of using multiple studies with varying objectives and scopes to estimate that national pathogenic virus and faecal indicator occurrence. To address this challenge, the EPA convened a two-day statistical workshop in 2005 to obtain advice from expert statisticians on a scientifically sound and meaningful way to select relevant studies and combine data from multiple studies to model the occurrence of pathogenic viruses and faecal indicators in public water wells.12 The statisticians recommended that EPA use all available data unless it was aware of data quality assurance problems or the well contamination situation was outside the realm of normal operation of public wells in the USA. The EPA followed the statisticians’ recommendations and used all available and relevant US groundwater studies that had enterovirus cell culture and E. coli data. The modelling approach used to combine data from the studies used in the GWR risk analysis is outside the scope of this chapter; however, detailed information about the modelling approach can be found in Economic Analysis for the Final Ground Water Rule.14

112

3.2.3.3

Chapter 3.2

Public Health Risks

Some individuals may be at risk due to exposure to viral and bacterial pathogens because the pathogens and/or faecal indicators may occur in GWSs that provide no treatment, provide less than optimal treatment (o4-log treatment of viruses) or have treatment deficiencies/failures. EPA estimates that GWSs serve untreated groundwater to 20 million people. About half of the 20 million people receive their water from CWSs and the other 50% from NTNCWSs and TNCWSs. Waterborne disease outbreak data (Table 3.2.2) indicate that GWSs that do not disinfect may be susceptible to faecal contamination. The data also show that waterborne disease outbreaks have occurred at disinfecting GWSs due to treatment deficiencies and failures. Waterborne disease outbreaks in GWSs demonstrate that individuals have been exposed to pathogenic viruses and bacteria,1 although the EPA believes that the data underestimates a significant number of waterborne disease outbreaks and cases of illness due to underreporting. Gastrointestinal illness is the commonly reported illness in waterborne disease outbreaks. In healthy individuals the illness may cause mild symptoms like mild diarrhoea, fever and vomiting which is usually treated at home without seeking medical attention. In infants the illness may be more severe and, in some cases, life-threatening.

Table 3.2.2

Sources of waterborne disease outbreaks in groundwater systems, 1991–2000.

Cause of contamination Community water systems Untreated groundwater Treatment deficiency Distribution system deficiency Miscellaneous/unknown Total Non-community water systems Untreated groundwater Treatment deficiency Distribution system deficiency Miscellaneous/unknown Total Combined Untreated groundwater Treatment deficiency Distribution system deficiency Miscellaneous/unknown Total Sources: Refs. 15–19.

Number of outbreaks

Outbreaks (%)

Cases of illness

Illness (%)

Cases per outbreak

5 7 5

26 37 26

167 1,624 803

6 58 29

33 232 161

2 19

11 100

183 2777

7 100

92 146

23 19 6

47 39 12

4057 3264 442

50 40 5

176 172 74

1 49

2 100

386 8149

5 100

386 166

28 26 11

41 38 16

4224 4888 1245

39 45 11

151 188 113

3 68

4 100

569 10 926

5 100

190 161


113

Prospective epidemiological studies and pathogen dose–response information are needed to assess the endemic ‘‘baseline’’ level of health risk related to human consumption of contaminated water from GWSs. The actual endemic risk level associated with consumption of contaminated groundwaters is difficult to ascertain because in many situations the illnesses that may result do not rise to the level of medical attention. In addition, there are very few epidemiological studies conducted in communities where the people drink water solely from GWSs. In addition, dose–response data for pathogenic viruses are available but are also limited. Pathogenic bacteria dose–response data are not considered due to lack of bacterial pathogen occurrence information. Therefore, the quantitative risk analysis is underestimated because it is based solely on viruses. The reader is referred to the Economic Analysis for the Final Ground Water Rule14 for more information on the health effect predictions for the GWR. Because of the limited epidemiological studies and dose–response information, the EPA believes that the GWR risk analysis understates the viral illnesses and deaths that are related to GWSs. The estimated endemic level is also underestimated because illness and deaths due to exposure to bacterial pathogens, e.g. E. coli O157:H7, in groundwater are not quantified in the GWR risk analysis due to data limitations. E. coli O157:H7 causes kidney failure and possibly other severe illnesses. The EPA dealt with the challenges related to estimating the viral baseline illness and risk reductions (benefits due to the number of avoided illness and deaths that would result from the GWR) by quantifying the risk using dose– response information for rotavirus and echovirus in the GWR quantitative risk analysis. Also, the Economic Analysis for the Final Ground Water Rule14 includes a detailed discussion on unquantified benefits, for example benefits due to avoided bacterial illness and deaths.

3.2.4

The Risk-Targeted Approach

In the final GWR, EPA established the risk-targeted approach to focus the GWR requirements on GWSs that are susceptible for faecal contamination. The intent of the risk-targeted approach is to target a subset of the 150 000 GWSs that have groundwater sources that may be vulnerable to faecal contamination and to require those GWSs to take corrective action. The proposed GWR established a multiple barrier approach. The multiple barrier approach and the final GWR risk-targeted strategy involve the same concept: identification and corrective measures for GWSs at risk of faecal contamination. The multiple barrier approach included five provisions: sanitary surveys, triggered source water monitoring, hydrogeological sensitivity assessments/routine source water monitoring, treatment technique requirements and compliance monitoring. The main difference between the two strategies is that the multiple barrier approach included a hydrogeological sensitivity assessment (HSA) requirement

114

Chapter 3.2

to determine if the GWSs obtain water from sensitive aquifers (e.g. karst, gravel or fractured bedrock aquifers). The multiple barrier approach also included a routine source water monitoring requirement for GWSs that obtain water from sensitive aquifers. In developing the final GWR, the EPA removed the HSA requirement and the accompanying routine source water monitoring requirement. The EPA believes these changes alone resulted in a more flexible and more targeted approach since it provides the State with the option to conduct more comprehensive monitoring (assessment source water monitoring) at GWSs that may have a higher likelihood of faecal contamination. The State may choose to use HSA along with other information to determine GWSs most vulnerable to faecal contamination. Below is a summary of the final GWR provisions. The reader is referred to the GWR regulatory text and preamble1 for the detailed rule requirements and a comprehensive discussion of the rationale for the requirements. Sanitary surveys. Sanitary surveys must be conducted at all GWSs. The purpose of the sanitary survey is to identify significant deficiencies that may result in contamination of the public water supply. The State is responsible for conducting the on-site assessment every 3 years for CWSs and every 5 years for non-community water systems (NCWSs) after the initial sanitary survey cycle. The rule includes conditions for the State to conduct sanitary surveys every 5 years (in lieu of every 3 years) for CWSs. Under the TCR, sanitary surveys are required only for small systems that routinely collect less than 5 TCR distribution system samples per month. Sanitary surveys must be conducted every 5 years for small CWSs and NCWSs (serving populations of 4100 or less) after the initial sanitary survey cycle. For NCWSs using only protected and disinfected groundwater, the system must undergo subsequent sanitary surveys at least every 10 years. The scope of the on-site assessments required under the GWR includes evaluations of the following 8 critical elements: (1) source; (2) treatment; (3) distribution system; (4) finished water storage; (5) pumps, pump facilities and controls; (6) monitoring, reporting and data verification; (7) system management and operation; and (8) operator compliance with state requirements. The TCR does not specify the scope of the sanitary surveys. Under the GWR, if a significant deficiencyz is identified during the sanitary survey then the GWS must take corrective action. The GWR requires the GWS to inform its customers of an uncorrected significant deficiency. The GWS must continue to inform the public annually until the significant deficiency is corrected. The flexibility of the rule provides the State with the option of requiring GWSs to inform customers of corrected significant deficiencies. z

In the GWR, significant deficiencies include, but are not limited to, defects in design, operation or maintenance, or a failure or malfunction of the sources, treatment, storage or distribution system that the State determines to be causing, or have potential for causing, the introduction of contamination into the water delivered to consumers.


115

Triggered source water monitoring. The triggered source water monitoring requirement builds on the existing TCR. The monitoring requirement applies to GWSs that do not provide at least 4-log treatment of viruses and have a TCR distribution system sample that has tested total coliformpositive8. A GWS that is subject to the triggered monitoring provision must monitor its groundwater source(s) for one of three State-specified faecal indicators (E. coli, enterococci, coliphage). If the groundwater source sample is faecal indicator-positive, the GWR requires the GWS to notify the State and the public within 24 hours. Unless directed by the State to take immediate corrective action following the initial faecal indicator-positive sample, the GWS must collect and test 5 additional groundwater source samples for the presence of the same Statespecified faecal indicator within 24 hours. If any one of the 5 additional groundwater source samples tests positive for the State-specified faecal indicator, the GWR requires the GWS to notify the State and public and to take corrective action. As a complement to the triggered source water monitoring provision, States have the discretion to require GWSs to conduct assessment source water monitoring on a case-by-case basis. This optional provision provides States with flexibility and the opportunity to require additional source water monitoring and to further evaluate GWSs that may have a higher likelihood of faecal contamination. Treatment technique requirements. The treatment technique requirements, also called corrective action requirements, apply to GWSs that have received written notice from the State of a significant deficiency and GWSs that have received written notice from the laboratory of a faecal indicator-positive sample. The GWR requires that within 120 days of receiving the notification from the State of a significant deficiency the GWSs must take corrective action. Under the GWR, GWSs that have received notice from the laboratory that one of the 5 additional groundwater source samples has tested faecal indicatorpositive must take corrective action within 120 days. In some cases, the State may require corrective action to be taken within 120 days following the initial faecal indicator-positive sample. The GWR treatment technique provision requires that GWSs implement at least one of the following corrective action options when taking corrective measures: correct all significant deficiencies; provide an alternative source of water; 8

A GWS is not required to collect a faecal indicator sample following a TCR total coliform-positive distribution system if (a) the State determines that the total coliform-positive sample is caused by a distribution systems deficiency, or (b) the TCR total coliform-positive distribution system sample is collected at a location that meets State criteria for distribution system conditions that will cause total coliform-positive samples.

116

Chapter 3.2

eliminate the source of contamination; and provide treatment that reliably achieves 4-log treatment of viruses. Compliance monitoring. The compliance monitoring requirement applies to GWSs that at anytime during implementation of the GWR provide at least 4-log treatment of viruses using chemical disinfection, membrane filtration or a State-approved alternative treatment technology. The purpose of the provision is to ensure that treatment effectiveness is maintained.

3.2.5

Conclusions

Federal drinking water regulations developed to protect consumers from microbial pathogens that may occur in surface water systems have existed for many years. The finalisation of the GWR has made a significant contribution to US public health protection objectives aimed at providing adequate protection against pathogenic viruses and bacteria for many Americans that obtain drinking water from GWSs. The EPA faced many changes in developing the regulation. However, through the EPA’s careful consideration of the public comments received on the proposed GWR rule requirements, incorporation of expert recommendations on meaningful ways to use multiple occurrence studies and recognition of the limitations of the GWR risk assessment, the EPA has met the challenges and believes that the GWR is a scientifically sound, flexible and cost-effective regulation.

Acknowledgements I would like to thank the following individuals for their review and contributions to this chapter: Philip Berger, Michael Finn, Jennifer McLain, Michael Messner and Stig Regli.

References 1. EPA, National Primary Drinking Water Regulations; Ground Water Rule; Final Rule, Federal Register, 8 November 2006, vol. 71, p. 65574 (http:// www.epa.gov/fedrgstr/EPA-WATER/2006/November/Day-08/w8763.htm). 2. EPA, Drinking Water; National Primary Drinking Water Regulations: Total Coliforms (Including Fecal Coliforms and E. coli); Final Rule, Federal Register, 54(124): 27544-27568, 29 June 1989 (http://www. epa.gov/safewater/disinfection/tcr/regulation.html#1989rule). 3. EPA, National Primary Drinking Water Regulations; Filtration, Disinfection; Turbidity, Giardia lamblia, Viruses, Legionella, and Heterotrophic Bacteria; Final Rule, Federal Register 54(124): 27486, 29 June 1989. 4. EPA, National Primary Drinking Water Regulations: Monitoring Requirements for Public Drinking Water Supplies; Final Rule, Federal Register


5.

6.

7.

8.

9.

10. 11.

12.

13.

14.

15.

16.

17.

117

61(94): 24353-24388, 14 May 1996 (http://www.epa.gov/fedrgstr/EPAWATER/1996/May/Day-14/pr-20972DIR/pr-20972.txt.html). EPA, Interim Enhanced Surface Water Treatment Rule; Final. Federal Register 63(241): 69477-69521, 16 December 1998 (http://www.epa.gov/ fedrgstr/EPA-WATER/1998/December/Day-16/w32888.htm). EPA, National Primary Drinking Water; Filter Backwash Recycling Rule; Final Rule; Federal Register 66(111): 31085-31105, 8 June 2001 (http:// www.epa.gov/fedrgstr/EPA-GENERAL/2001/June/Day-08/g13776.htm). EPA, Long Term 1 Enhanced Surface Water Treatment Rule; Final Rule, Federal Register 67(8): 1812-1844, 14 January 2002 (http://www.epa.gov/ fedrgstr/EPA-WATER/2002/January/Day-14/w409.htm). EPA, Long Term 2 Enhanced Surface Water Treatment Rule; Final Rule, Federal Register 71(3): 653-702, 5 January 2006 (http://www.epa.gov/ fedrgstr/EPA-WATER/2006/January/Day-05/w04a.htm). EPA, National Primary Drinking Water Regulations: Ground Water Rule; Proposed Rule, Federal Register, 10 May 2000, vol. 65, p. 30194 (http:// www.epa.gov/fedrgstr/EPA-WATER/2000/May/Day-10/w10763.htm). US Geological Survey, Materials in Use in U.S. Interstate Highways, fact sheet 2006-3127, US Department of Interior, October 2006. EPA, Occurrence and Exposure Assessment for the Long Term 2 Enhanced Surface Water Treatment Rule, EPA–821–R–06–002, US Environmental Protection Agency, Office of Water, Washington, DC, 2005 (http:// www.epa.gov/safewater/disinfection/gwr/regulation.html). EPA, Occurrence and Monitoring Document for the Final Ground Water Rule, EPA–815–R–06–012, US Environmental Protection Agency, Office of Water, Washington, DC, 2006 (http://www.epa.gov/safewater/disinfection/ gwr/regulation.html). EPA, National Primary Drinking Water Regulations; Ground Water Rule; Notice of Data Availability, Federal Register, 27 March 2006, vol. 71, p. 15105 (http://www.epa.gov/fedrgstr/EPA-WATER/2006/March/Day-27/ w2931.htm). EPA, Economic Analysis for the Final Ground Water Rule, EPA–815–R– 06–014, US Environmental Protection Agency, Office of Water, Washington, DC, 2006 (http://www.epa.gov/safewater/disinfection/gwr/regulation.html). R. S. Barwick, D. A. Levy, G. F. Craun, M. J. Beach and R. L. Calderson, Surveillance for waterborne-disease outbreaks: United States, 1997–1998, Morbid. Mortal. Weekly Rep., 2000, 49(SS-04), 1–35 (http://www.cdc.gov/ mmwr/preview/mmwrhtml/ss4904a1.htm). M. H. Kramer, B. L. Herwaldt, R. L. Calderson and D. D. Juranek, Surveillance for waterborne-disease outbreaks: United States 1993–1994, Morbid. Mortal. Weekly Rep., 1996, 45(SS-1), 1–33 (http://www.cdc.gov/ mmwr/preview/mmwrhtml/00040818.htm). S H. Lee, D. A. Levy, G. F. Craun, M. J. Beach and R. L. Calderon, Surveillance for waterborne disease outbreaks: United States, 1999–2000, Morbid. Mortal. Weekly Rep., 2002, 51(SS-08), 1–28 (http://www.cdc.gov/ mmwr/preview/mmwrhtml/ss5108a1.htm).

118

Chapter 3.2

18. D. A. Levy, M. S. Bens, G. F. Craun, R. L. Calderon and B. L. Herwaldt, Surveillance for waterborne disease outbreaks: United States, 1995–1996, Morbid. Mortal. Weekly Rep., 1998, 47(SS-05), 1–34 (http://www.cdc.gov/ mmwr/preview/mmwrhtml/00055820.htm). 19. A. C. Moore, B. L. Herwaldt, G. F. Craun, R. L. Calderon, A. K. Highsmith and D. D. Juranek, Surveillance for waterborne disease outbreaks: United States, 1991–1992, Morbid. Mortal. Weekly Rep., 2003, 42(SS-05), 1–22 (http://www.cdc.gov/mmwr/preview/mmwrhtml/00025893.htm).

4. Stakeholder Interactions

CHAPTER 4.1

Principles of the Common Implementation Strategy of the WFD: The Groundwater Working Groupw PHILIPPE QUEVAUVILLER,a JOHANNES GRATHb AND ANDREAS SCHEIDLEDERb a

European Commission, DG Environment (BU9 3/142), Rue de la Loi 200, BE-1049 Brussels, Belgium; b Umweltbundesamt GmbH, Spittelauer Laende 5, AT-1090 Wien, Austria

4.1.1

The Need for Multi-stakeholder Involvement in the Environmental Policy Development and Implementation Process

The development of environmental policies is a complex process, which mixes legal requirements with issues of technical feasibility, scientific knowledge and socioeconomic aspects, and which requires intensive multi-stakeholder consultations. In this context, the consideration of scientific progress and access to technical information represent key aspects for the design of new policies and the review of existing ones.1 This is discussed in detail in Chapter 2.1. Within the European Union (EU), this consideration is fully embedded in the Sixth Environment Action Programme which stipulates that ‘‘sound scientific knowledge and economic assessments, reliable and up-to-date environmental data and information, and the use of indicators will underpin the drawing-up, implementation and evaluation of environmental policy.’’2 This requires, therefore, that scientific inputs should constantly feed the environmental policy process. This integration also involves various players, namely the scientific and w

The views expressed in this chapter are purely those of the authors and may not in any circumstances be regarded as stating an official position of the European Commission.

121

122

Chapter 4.1

Figure 4.1.1

Consultation process involving stakeholders.

policy-making communities, but also representatives from industry, agriculture, non-governmental organisations (NGOs), etc. (Figure 4.1.1). This aspect is described in depth in Chapter 4.4. An example of successful multi-stakeholder consultation and participation concerns the implementation of the Water Framework Directive (WFD; 2000/ 60/EC). In this context, a Common Implementation Strategy (CIS) has been agreed with the EU member states, candidate and associate countries and stakeholder organisations and has been operational since 2001.3 In this framework, various topics are under discussion by experts from EU member states, industry, agriculture, scientists, etc., with the aim to gather and share knowledge and concern on WFD relevant issues, as examined from different perspectives. This approach, albeit time-consuming, has considerably enhanced the knowledge and common interpretation of the key provisions of the WFD, and it has been considered as a very powerful tool for sharing good practices and an example of good governance.

4.1.2

The WFD Common Implementation Strategy

4.1.2.1

General Principles

As already mentioned above, it has become clear, soon after the WFD adoption, that the successful implementation of the directive will be, at the least, equally as challenging and ambitious for all countries, institutions and stakeholders involved. Therefore, a strategic document establishing a CIS for the WFD has been developed and finally agreed by the EU’s Water Directors under the Swedish presidency in 2001.3 Despite the fact that the full responsibility of the individual member states for implementing the WFD was recognised, a broad consensus existed among the Water Directors of the

Principles of the Common Implementation Strategy of the WFD

123

member states, Norway and the Commission that a European joint partnership was necessary in order to:

develop a common understanding and approaches; elaborate informal technical guidance including best practice examples; share experiences and resources; avoid duplication of efforts; and limit the risk of bad application.

Furthermore, the Water Directors stressed the necessity to involve stake holders, NGOs and the research community in this joint process as well as to enable the participation of candidate countries in order to facilitate the cohesion process. Following the decision of the Water Directors, a comprehensive and ambitious work programme was started of which the first phase, including ten Working Groups and three Expert Advisory Forum (EAF) groups, was completed at the end of 2003 and led to the availability of fourteen guidance documents which are publicly available (in the form of CD-ROM and on the internet on the WFD europa website). The second phase of the CIS (2003–2004) involved four Working Groups, namely on Ecological Status (WG A), Economics and Pilot River Basins (WG B), Groundwater Body Characterisation and Monitoring (WG C) and Reporting (WG D), as well as two EAF groups, of which the discussions focused on developing policies linked to the WFD (i.e. Priority Substances Directive, and revision of the Reporting Directive). These groups were re-conducted in the third phase (2005–2006), and this process is now continued under new mandates for the period 2007–2009 (see Figure 4.1.2), which is detailed with regard to groundwater in the section below.

4.1.2.2

Supporting Activity: The Pilot River Basin Network

In the context of the CIS, a network of Pilot River Basins (PRBs) has been established to test and validate guidance documents developed under the CIS of the WFD. The network covers 15 PRBs in 12 countries. Activity reports are published on a yearly basis. Besides the CIS process, a range of PRBs are linked to research and demonstration projects. They indeed represent an opportunity for researchers to test new developed techniques or methodologies (e.g. risk assessment methods, monitoring tools, modelling) in well-characterised areas which have direct links to WFD implementers. This network is described in detail in Chapter 4.2.

4.1.3

The CIS Working Group on Groundwater

4.1.3.1

Objectives

The CIS Groundwater Working Group (WG C) aims both to clarify groundwater issues that are covered by the WFD and to prepare the development of technical guidance documents and exchange best practices on several issues in

124

Chapter 4.1

Figure 4.1.2

CIS operational diagram for the period 2007–2009.

the light of the orientations of the newly adopted Groundwater Directive (see Chapter 3.1).

4.1.3.2

Leadership and Network

The Commission/DG ENV chairs the WG C which is co-chaired by Austria. The Working Group is composed of representatives of EU member states, associated and candidate countries, industrial and scientific stakeholders and NGO representatives (around 80 members in total). Plenary meetings are open to all participants, while ad hoc activities are operated by groups of a maximum of 15–20 participants which develop documents that are scrutinised by the plenary group.

4.1.3.3

Achievements from 2003 to 2006

The focus in the period 2003–2006 was on the development of technical reports and guidance documents primarily focusing on the issues covered by the WFD, namely monitoring, prevent/limit measures and groundwater protected areas. In addition, a specific activity concerned exchange of views on groundwater management in the Mediterranean area (linked to the EU Water Initiative). Activities of the WG were conceived with the view of collecting targeted data


125

and information, avoiding duplication with existing guidance documents and ensuring an efficient use of available data and information. Series of workshops were held in 2003–2004, which led to three technical reports gathering member states’ practices in the field of groundwater risk assessment, monitoring and programmes of measures.4–6 The orientations in 2005–2006 were concerned with the drafting of guidance documents on groundwater monitoring, protected areas and measures to prevent/limit pollutant introduction into groundwater. The Monitoring Guidance for Groundwater document was finalised and endorsed by the EU Water Directors on 30 November 2006 at the meeting under the Finnish presidency in Inari (Lapland).7 The two other guidance documents were slightly delayed owing to the negotiation of the new Groundwater Directive, and were rescheduled for the 2007–2009 work programme of WG C (see Section 3.4).

4.1.3.4

Perspectives for 2007–2009

The main orientations of the 2007–2009 mandate of WG C were discussed at the occasion of the Groundwater Conference in Vienna on 22–23 June 20068 and through an e-mail consultation of all WG C members. The main aims and objectives of WG C for the period 2007–2009 are to pursue exchanges in support of the implementation of the new Groundwater Directive along the CIS principles,9 focusing in particular on: best practices related to groundwater programmes of measures, including measures related to diffuse sources of pollution and megasites; common methodology for the establishment of groundwater threshold values; compliance, status and trend assessment; and recommendations for integrated risk assessment, including conceptual modelling. The WG C work programme 2007–2009 consists of three core activities led or co-led by member states or stakeholder organisations, which will develop their work programme as described in activity sheets. The activities (drafting or exchanges of good practices) will be undertaken with selected WG participants (groups of ideally 15–20 participants) willing to actively contribute to the drafting of documents and to participate in ad hoc meetings (possibly organised by the activity leaders). The progress of the activities will be reported and discussed at plenary meetings of WG C held twice a year and organised under the EU presidency umbrella. The objectives of WG C for the period 2007–2009 are separated into three core activities. 1. Activity 1: Programmes of Measures (PoM). Discussions will focus on exchanges on best practices and recommendations needed by member states in the context of the identification of measures related to

126

Chapter 4.1

groundwater that will have to be included in the First River Basin Management Plan. The activity will cover, in particular, the following. The finalisation of the ‘‘prevent/limit’’ guidance initiated in 2006, which aims to support the implementation of Article 6 of the new Groundwater Directive. Exchanges on best available technologies related to groundwater measures, taking into account programmes of measures required under other EU directives. These will include guidance on both point sources of pollution (including historical contaminated sites) and on diffuse sources, including agricultural diffuse pollution and megasites (large polluted areas, e.g. harbour areas and industrially contaminated sites). This item will be closely linked to a CIS expert group on ‘‘WFD & Agriculture’’ with regard to diffuse agricultural sources. 2. Activity 2: Compliance and Trends. The work will be directed toward the development of a guidance document on compliance and trends, as well as on recommendations establishing on groundwater threshold values, considering the following. Adoption of a common methodology for the establishment of groundwater threshold values based on the outcome of the methodology developed by the BRIDGE project (see Chapter 9.1), and exchanges of experiences among the member states in support of the new Groundwater Directive (see Chapter 3.1). Development of the status compliance and trends guidance document, concerning both quantitative and chemical status issues. This had been planned in the former work programme but could not be initiated owing to delays in the adoption of the new Groundwater Directive. With respect to trend assessment, the document will be largely based on a technical report developed in 2002,10 and will provide recommendations to member states on how to undertake and interpret trend studies (including considerations on lag time of groundwater systems and how to integrate this in trend assessment). 3. Activity 3: Integrated Risk Assessment. Discussions will focus on recommendations for improving risk assessment for groundwater at river basin level in an integrated way, in view of the preparation of the First River Basin Management Plan. The activity will cover, in particular, the following. Discussions on how to improve risk assessment and recommendations on conceptual modelling for water systems, including (publicly available) databases, mapping (e.g. vulnerability, hydrogeology) and visualisation of subsurface processes. Good management practices, including issues such as artificial recharge and transboundary aquifer management. This item will be operated in close connection with a CIS expert group on water scarcity and drought.


4.1.4

127

Perspectives

The successful implementation of the new Groundwater Directive will closely depend upon an efficient participatory approach and harmonised groundwater risk assessment, monitoring and programmes of measures throughout the EU. The CIS Working Group on Groundwater will be an indispensable element supporting this implementation, in particular in view of the preparation of the First River Basin Management Plan expected for publication at the end of 2009. The working group will also help in gathering the necessary scientific and technical knowledge which will enable a scientifically sound review of the new Groundwater Directive required under Article 7, i.e. by the end of 2012, which will coincide with the operational start of the WFD Programme of Measures.

References 1. Ph. Quevauviller, P. Balabanis, C. Fragakis, M. Weydert, M. Oliver, A. Kaschl, G. Arnold, A. Kroll, L. Galbiati, J.M. Zaldivar and G. Bidoglio, Environ. Sci. Pol., 2005, 8, 203. 2. European Commission, 6th Environment Action Programme 2002–2012, 2002. 3. Common Implementation Strategy for the Water Framework Directive, European Communities, 2003 (ISBN 92-894-2040-5). Final CIS document available at: http://cc.europa.eu/environment/water/water-framework/ objectives/implementation-en.htm. 4. European Commission, Groundwater Body Characterisation, technical report, 2004. 5. European Commission, Groundwater Risk Assessment, technical report, 2004. 6. European Commission, Groundwater Monitoring, technical report, 2004. 7. European Commission Monitoring Guidance for Groundwater, CIS Guidance no. 15, Common Implementation Strategy of the WFD, European Commission, 2007. 8. Proceedings of the European Groundwater Conference, Vienna, 22–23 June 2006. 9. Mandate of Working Group C ‘‘Groundwater,’’ Common Implementation Strategy of the WFD, European Commission, 2007. 10. Technical report on Statistical aspects of the identification of groundwater pollution trends, and aggregation of monitoring results, Common Implementation Strategy, 2001.

CHAPTER 4.2

The Pilot River Basin Network: Examples of Groundwaterrelated Activitiesw LORENZO GALBIATIa AND GIOVANNI BIDOGLIOb a

Age`ncia Catalana de l’Aigua, Provença 204-208, ES-08036 Barcelona, Spain; b European Commission, Joint Research Centre, Via E. Fermi 1, TP 460, IT-21020 Ispra (VA), Italy

4.2.1

Introduction

The Water Framework Directive1 (WFD) sets the guidelines for sustainable water management in Europe. Its aim is to achieve ‘‘good status’’ for all environmental waters by 2015 at the latest. This is to be achieved through programmes of measures in which each river basin is treated as a coherent unit (see Chapter 3.1). To help the compliance of this ambitious objective the European Union (EU) water directors have agreed to establish a Common Implementation Strategy (CIS) of the WFD (see Chapter 4.1). During the 2001–2002 CIS period, a series of guidance documents (GDs) concerning all major aspects the WFD implementation were developed by working groups (WGs) including representatives of EU member states, accession countries, national experts and the European Commission. In order to test and cross-validate these GDs, a Pilot River Basin (PRB) network has been established. It was foreseen that such a network would act as an interface between the Commission and member state authorities and thus contribute to the implementation of the WFD. The PRB exercise was structured into two phases. Phase I focused on testing and reporting on coherence amongst the different GDs, leading to the longterm development of River Basin Management Plans and preparation of w

The views expressed in this chapter are purely those of the authors and may not in any circumstances be regarded as stating an official position of the Catalan Water Agency or the European Commission.

128

The Pilot River Basin Network: Examples of Groundwater-related Activities

129

programmes of measures. This first phase was finalised at the end of 2004, the main deliverables of which were: the PRB outcome report on the testing of WFD GDs;2 the Article 5 reports produced by some PRBs (available on the PRB’s website and on the PRB section of CIRCA); and the PRB thematic workshops, which were organised on the topics of groundwater, water body delineation, economy, Mediterranean dimension, and research and technology integration in support of the WFD. Phase II mainly consisted of gaining experience and developing methods necessary to draw up the monitoring programmes and programmes of measures for the PRBs according to the WFD deadlines. In this phase, PRBs have been used as tools to reach the goals of the WGs in the CIS project. Phase II of the exercise started at the beginning of 2005 and was finalised at the end of the 2006.

4.2.2

Science–Policy Integration in the PRB Exercise Linked to Groundwater Management

Under Article 5 of the WFD, member states had to identify water bodies by end of December 2004. In this context, member states carried out an initial characterisation of all groundwater bodies including their location and boundaries as well as the identification of pressures and groundwater bodies at risk of failing to meet the objectives of the WFD. In a forefront initiative to support this specific aspect of the implementation process, the PRBs tested the applicability of the GD on groundwater. Generally in this phase the PRBs perceived that the definition of the river basin district boundaries is the most important and complex issue which is not only affecting the activities underlined by the Planning GD but is strongly affecting the activities related with the Groundwater GD, especially in those cases where the groundwater body definition concerns shared aquifers. These criteria were chosen in most countries at national level, and then adopted with the necessary adaptation by all basin districts. Criteria were also set for assignment of each shared groundwater body to all the pertaining river basins, based on available information about hydrogeology (bedrock geology, tracing study results, groundwater flow regime and direction) and the presence of dependent ecosystems (groundwater-fed lakes, rising from underground streams, groundwater-dependent terrestrial ecosystems). In the case of shared river basins or groundwater bodies, the issue of cooperation becomes essential, particularly in the development of the programme of measures and river basin management plans, to ensure that such interconnected water bodies and associated ecosystems are adequately protected. Most of the activities related to groundwater management undertaken by the PRBs have been focused on the different criteria applied to define the groundwater bodies. Section 3 presents a methodology proposed by the Shannon PRB for the managing of groundwater bodies.

130

Chapter 4.2

During phase II of the exercise, the PRBs developed their activities under the umbrella of different CIS WGs contributing to the testing of different methodologies to tackle and solve problems related with the WFD implementation. For example, Article 17 of the directive stipulates that criteria for establishment of the groundwater chemical status should be developed in a proposal made by the European Commission, i.e. the ‘‘Groundwater Daughter Directive’’ adopted in December 2006 (see Chapter 3.1). In this directive, the Commission specifies that the good chemical status of groundwater should be partly defined by the establishment of groundwater standards (‘‘threshold values’’) by member states themselves. The idea is that the chemical status of groundwater will be based on existing community quality standards and on the requirements for member states to identify pollutants and related threshold values that are representative of groundwater bodies found being at risk, in accordance with the analysis of pressures and impacts carried out under the WFD. The interface role of the PRBs qualified them as appropriate places for researchers to test new methodologies (e.g. risk assessment and monitoring approaches, monitoring tools).3 An example is the role of the PRBs in the development of pollutant threshold values in the context of the BRIDGE project,4 which involved the following PRBs: Tevere (Italy), Pinios (Greece), Scheldt (Belgium) and Odense (Denmark). Section 4 gives a brief overview of the case study developed by the Tevere PRBs for the identification of the groundwater natural background levels and the definition of threshold for an aquifer in a Mediterranean catchment.5

4.2.3

Managing Groundwater Bodies in the Shannon PRB for the Implementation of the WFD

The Shannon PRB is the largest river basin in Ireland draining a land area of some 18 000 km2 in central Ireland. It includes part of 18 local authorities in the Republic of Ireland and has a small transboundary component of approximately 6 km2 in County Fermanagh, Northern Ireland. Carboniferous rocks dominate the bedrock geology of the Shannon PRB. Of these, highly karstified pure bedded limestones predominate in the upper reaches of the basin. Groundwater flow in these rocks is dominated by conduit flow. In contrast, in most of the rest of the basin, groundwater flows through fissures and faults in relatively low transmissivity aquifers. In the west, on either side of the Shannon estuary, bedded shales and sandstones of Namurian age dominate. Between the upper and lower reaches of the basin, embedded pure limestone and impure limestone are folded around cores of older rocks. Agriculture is the principal activity in the river basin (73% of total area); the dominant land use being pasture. There are some significant areas of wetland (12%), mainly peatland. The catchment is not notably industrialised and agri-industries, such as milk and meat processing, are the most prominent.


4.2.3.1

131

Groundwater Body Delineation

The Geological Survey of Ireland (GSI) has carried out the delineation of groundwater bodies in Ireland, including the Shannon PRB. The delineation process involved several stages. Mapped rock units were assigned an aquifer class based on the existing GSI aquifer classification system. These aquifer classes were then grouped into four aquifer types based on groundwater flow regime, i.e. karst aquifers, gravel aquifers, productive fissured bedrock aquifers and poorly productive bedrock aquifers. Preliminary groundwater bodies were then delineated using no-flow geological boundaries, as well as boundaries based on groundwater highs, differing flows and flow lines. Final delineation incorporated major surface water catchment boundaries except in areas where the influence of topography is diminished (e.g. karstic or confined aquifers). This process resulted in the delineation of 97 bedrock groundwater bodies with a median size of 53 km2.

4.2.3.2

Groundwater Management for the WFD

The first requirement of the WFD is to identify groundwater bodies at risk of failing to meet the environmental objectives set out in Article 4. To achieve these objectives requires making operational the programme of measures specified in the River Basin Management Plan. A proposed risk assessment methodology to identify groundwater bodies (GWBs) at risk is presented. This process will allow for the prioritisation of resources in the River Basin Management Plan. The focus of the programme of measures should be on the high impact potential areas of ‘‘at risk’’ GWBs. Different aquifer types will require different management responses appropriate to their spatial extent, flow regime, degree of groundwater–surface water interaction and connectivity with groundwater-dependent terrestrial ecosystems. This approach will require a detailed conceptual understanding of each GWB to ensure that the most suitable programmes of measures are applied and the use of limited resources is optimised.

4.2.3.3

Example of Risk Assessment Methodology for Diffuse Groundwater Pollution in the Shannon PRB

The following approach is a screening exercise using available GIS layers and follows the ‘‘source-pathway-receptor’’ model. The objective is to identify groundwater bodies at risk and allow for prioritisation in the programme of measures and river basin management plan. To reach this goal a five-step procedure has been applied. The first and second steps are to develop of a good conceptual understanding of each groundwater body and combine information on groundwater vulnerability with aquifer flow regime characteristics using risk matrices to identify the degree of pathway susceptibility to diffuse pollution (Figure 4.2.1). Then pressure magnitude thresholds, e.g. for stocking density, are set up (Figure 4.2.2). Thresholds will

132

Chapter 4.2

Figure 4.2.1

Step 1 and step 2: develop a good conceptual understanding and combine information on groundwater vulnerability.

Figure 4.2.2

Step 3: set pressure magnitude thresholds.

need to be developed for all pollutant types. The next step is the combination of the pathway susceptibility and pressure magnitude using risk matrices to produce an impact potential map (Figure 4.2.3). As a last step, the combination of all these steps produces the final risk designation (Figure 4.2.4). The percentage area affected by pollution combined with a verification using monitoring data will determine the identification of whether a groundwater body is ‘‘at risk’’ or not. Lack of monitoring data and pressure layer information will affect the confidence in the risk designation. Further assessment may be required to determine whether associated surface waters or groundwaterdependent terrestrial ecosystems are adversely impacted.


Figure 4.2.3

Step 4: pathway susceptibility and pressure magnitude.

Figure 4.2.4

Step 5: final risk designation.

4.2.4

133

Groundwater Natural Background Levels and Threshold Definition in the Tevere PRBs Under the BRIDGE Project

The BRIDGE project has been structured with different working packages (WPs), each of them being a different step in the definition of the background criteria for the identification of groundwater threshold (see further details in Chapter 9.1): work package 1: survey of representative groundwater pollutants project launching and co-ordination; work package 2: study of groundwater characteristics; work package 3: criteria for environmental thresholds and methodology to define a good status; work package 4: representative sites/water body studies and compliance testing; work package 5: economic and social costs linked to the establishment of groundwater threshold values; and work package 6: information and dissemination.

134

Chapter 4.2

The example of the Tevere and the PRBs presented in this section are related to the developed in the framework of the WP4. The objective of WP4 was to evaluate the developed approach and application of environmental thresholds recommended in BRIDGE WP3 at selected European representative sites. Results of this phase were then transferred to WP5 and WP6 for economic assessment and communication, respectively. The aim of the BRIDGE WP3 was to propose a practical approach for a methodology to define the good status for groundwater bodies and a general structure for developing criteria related to this status definition, providing guidance on the derivation process for environmental thresholds. Following the structure of the project, the activities developed in this WP referred to data and information coming from WP1 and WP2. In this section a brief summary of the case study developed by the Tevere in the framework of the BRIDGE project is given. A complete description of the area and the groundwater characterisation are available as a full report at the BRIDGE web page (www.wfd-bridge.net). The section focuses on the description of a methodology applied to obtain the groundwater status evaluation by using threshold values.

4.2.4.1

Tevere PRB: The Colli Albani Case Study

The Colli Albani volcanic area has a surface of about 1950 km2. The area is characterised by four WBs for the surface aquifers on the basis of groundwater watersheds and flow direction. Overexploitation is the main problem in all the four WBs. The impacts produced by this pressure on the aquifer affect different receptors, among them groundwater itself, terrestrial ecosystems, aquatic ecosystems and drinking water. All of the receptors considered in BRIDGE are present in the Colli Albani aquifer, the following in particular: watercourses extending in a radial direction along the slopes of the volcanic structure fed by point and linear sources; lakes in hydraulic continuity with groundwater; wetlands associated with terrestrial ecosystems dependent on groundwater; coastal areas subject to saline intrusion; areas destined for water abstraction for drinking water use; and groundwater itself. The main pressure in the area is groundwater abstraction, coming from about 33 000 points legally authorised and from an estimated equivalent number of illegal wells. Diffuse and point sources of pollution, coming from agricultural areas (38% of the surface) is also an important pressure in the area. Concerning the determination of a natural background level (NBL) of pollutants no existing methodologies are available neither at a national nor regional level. However, many studies on the characterisation of the hydrochemical facies of the Colli Albani aquifers are available, based on a monitoring network developed in the area since the 1970s. It is possible to observe several important effects caused by the main pressure on groundwater. With respect to the quantitative aspect these are: substantial variation of the piezometric level (especially in the intensely exploited areas), lowering of the level of the lakes, 60% reduction of the total base flow in the


135

watercourses and salty water intrusion in the coastal areas. Looking at the groundwater quality as affected by groundwater abstraction, no significant values for anthropogenic hazardous substances are visible. The abstraction is also affecting the water-dependent ecosystems. In this case the main impact is related to the variations in the extension of wetlands and disappearance of springs, which also led to a reduction of the base flow in surface watercourses. As a consequence, the dilution capacity of watercourses is substantially reduced and under certain conditions discharges completely replace base flow. Consequently, groundwater itself becomes the receptor of the water recharge coming from the watercourses.

4.2.4.2

Groundwater Status Evaluation by Threshold Values

The application of WP3 methodologies was preceded by a study of the hydrogeochemical characteristics of the Colli Albani aquifer. For this purpose chemical analyses carried out on 15 springs characterised by marked hydrogeochemical anomalies were used. About 100 samples from the period 1970– 1980 were selected. They were characterised by low anthropogenic pressure. Furthermore, chemical analyses from a measurement campaign carried out in 2005 by the Lazio Regional Environment Agency on about 50 wells used for drinking water abstraction were considered, especially focusing on natural elements such as arsenic, vanadium and fluorides. In some cases these natural elements exceeded drinking water standard values. For example, high values of arsenic, vanadium and fluorides seem to be located in measuring points in the southeastern sector of the volcanic structure. The high variability of the chemical characteristics of water and consequently of the natural background levels of volcanic aquifers suggested that WP3 methodologies can be applied successfully only if the water families that interact with receptors are identified first. The following steps were carried out in this case study: mapping of the hydrochemical characteristics of water in the entire aquifer; mapping of the receptors; description of receptor–groundwater body interaction; selection of Environmental Quality Standards (EQS) values that can be associated with the receptors; identification of NBL values of groundwater that interact with the receptor; identification of significant parameters that may cause variations in the receptor; and application of WP3 methodologies for threshold identification. This approach requires that for each receptor the type of interaction with groundwater, the EQS values to take into account and the natural background levels of local groundwater that interact with the receptor are known. Threshold values were calculated only for parameters that cause negative impacts on

136

Chapter 4.2

the receptor and show growing trends in respect to the natural background values. At present, there are not sufficient data to produce a detailed map of the chemical characteristics of the entire Colli Albani aquifer. Therefore, the identification of threshold values was carried out starting from each receptor in relation to available data. In particular, WP3 methodologies were applied to two specific areas: groundwater abstraction area destined for drinking water use (SaloneAcqua Vergine area); and salty water intrusion near the protected area of Castelporziano.

4.2.4.3

Case Study of Salone-Acque Vergini System

The study area is a protected zone known as the Salone-Acqua Vergine system. The main pressure on the area is groundwater abstraction for drinking water use, with four main abstraction points, which supply a total of 600 l s 1. Monitoring activities, are carried out mainly by water supply agencies and consist of about 2000 samples collected from 1992 to 2005. In 2001 a field campaign was undertaken to analyse the hydrogeochemical composition of the water which is feeding the springs and wells in water abstraction points for drinking water supply. Eighteen regularly sampled monitoring points were identified. The considered EQS values were drinking water standards according to those proposed by Directive 1998/83/EC.6 Time-series analyses of parameters such as Ca, Mg, SO4, Na, Cl, Si, Fe, V, Li, B and NH41 have shown fluctuations around a central value that remains constant. This can be ascribed to seasonal fluctuations or to analytical variability. The NO3 parameter was considered significant in respect to the impact on the receptor due to its significant upward trend, probably due to anthropogenic pressures (Figure 4.2.5). The trend analysis from 1997 showed an increase in some wells of the nitrates value, which reached 30 mg l 1 in this specific area, probably due to anthropogenic factors related to agricultural activities and/or wastewater point sources coming from urban areas. NBL values of the chemical compounds were derived from 90th percentile calculations of analyses carried out on six monitoring points in 1997. Arsenic is another significant parameter that characterises the receptor. It varied between 5 and 10 mg l 1 and did not show significant trends. The background levels derived from the statistical analysis of NO3 and as are presented in Table 4.2.1.

4.2.4.4

Case Study of the Protected Area of Castelporziano

The national protected area of Castelporziano was included in the ‘‘Natura 2000’’ ecological network as a Special Conservation Zone. It is a strip of Mediterranean forest, which has remained unaltered for many centuries, with terrestrial ecosystems depending on water and wetlands. Reclamation works carried out in the surrounding marshland during the 1930s as well as groundwater overexploitation for agricultural, household and industrial uses modified


137

Figure 4.2.5

Nitrates concentration trend in two wells located in protected area for drinking water uses.

Table 4.2.1

Nitrates and arsenic threshold values in the drinking water of the Salone-Acqua Vergine area.

Parameter 1

Nitrate (mg l ) Arsenic (mg l 1)

NBL BRIDGE

TV

Drinking water standard

24 10

37 10

50 10

Case 2 Case 3

aquifer recharge causing progressive infiltration of saline water from the sea and the final stretch of the Tevere River. Phenomena linked to mainly gaseous endogenous fluids take place in the inland area of Castelporziano, in the Malafede valley, characterised by the presence of mineral and thermomineral water. Since salinity variations may cause adaptive changes in the flora and fauna of the protected area’s terrestrial ecosystem, conductivity, salinity and piezometric levels are monitored in the area. Methodologies for the identification of salinity levels that may induce these adaptive changes still need to be consolidated. Currently, EQS values to use for salinity and conductivity are being assessed by expert judgement, mainly considering trend analysis rather than threshold values. Tables 4.2.2 and 4.2.3 show the results of two measurement and chemical field analyses, carried out in the area during 1999 and 2004.7 A more detailed reconstruction of the phenomenon is shown in Figure 4.2.6, where also sampling points external to the protected area were taken into account. The map shows that higher conductivity levels are located in the most intensively urbanised areas, while lowest conductivity levels are located at the centre of the protected area.

T (1C)

18.0 17.7 17.6 16.6 18.1 15.9 16.8 17.2 16.0 15.5 17.2 17.9 15.5 18.1 17.0 15.9 17.2 18.0 18.1

F2 F3 E1 E2 E3 E4 E7 E8 E11 E26 E27 C1 min max med P10 P50 P90 P97.7

6.7 6.6 7.3 7.2 7 6.9 6.6 6.1 6.8 6.6 7.7 6.9 6.1 7.7 6.9 6.6 6.9 7.3 7.6

pH

1.95 0.87 1.14 0.79 1.04 0.99 0.34 1.26 0.98 1.44 1.05 1.12 0.34 1.95 1.08 0.80 1.05 1.28 1.40

Cond. (mS cm 1) 144.0 87.0 54.0 40.0 108.0 40.0 38.0 67.0 54.0 76.0 75.0 56.0 38.0 144.0 69.9 40.0 61.5 79.2 101.4

Ca (mg l 1) 14.8 9.2 3.7 5.4 4.3 4.4 8.8 5.6 2.0 6.6 32.4 3.4 2.0 32.4 8.4 3.4 5.5 11.2 27.5

K (mg l 1) 194.0 39.0 123.0 92.0 90.0 113.0 16.0 135.0 124.0 149.0 69.0 100.0 16.0 194.0 103.7 42.0 106.5 136.4 146.1

Mg (mg l 1) 54.5 24.8 36.5 12.3 10.8 25.0 3.7 31.0 22.7 35.0 39.3 34.7 3.7 54.5 27.5 11.0 28.0 36.8 38.7

Na (mg l 1) 888.0 403.0 481.0 422.0 407.0 505.0 169.0 533.0 565.0 537.0 470.0 497.0 169.0 888.0 489.8 403.4 489.0 539.8 559.2

HCO3 (mg l 1) 204.0 85.1 76.4 45.1 114.0 66.7 17.9 146.0 48.0 186.4 86.0 86.3 17.9 204.0 96.8 45.4 85.6 150.0 178.0

Cl (mg l 1)

40.7 0.2 64.5 29.2 40.1 34.0 13.2 37.0 30.4 33.6 54.0 52.4 0.2 64.5 35.8 14.8 35.5 55.1 62.3

SO4 (mg l 1)

Chemical analyses and physical parameters of 12 samples in the protected area of Castelporziano, 1999.

Sample

Table 4.2.2

1540.0 648.0 839.0 646.0 775.0 788.0 267.0 955.0 846.0 1023.0 825.0 829.0 267.0 1540.0 831.8 646.2 827.0 961.8 1008.9

TDS (mg l 1)

138 Chapter 4.2

T (1C)

17.6 18.5 18.1 18.9 17.4 19.3 17.8 18.7 16.9 18.5 18.6 18.6 16.9 19.3 18.2 17.4 18.5 18.9 19.2

F2 F3 E1 E2 E3 E4 E7 E8 E11 E26 E27 C1 min max med P10 P50 P90 P97.7

7.0 7.2 6.9 6.8 6.8 7.1 6.5 6.8 6.8 6.9 7.7 7.1 6.5 7.7 7.0 6.8 6.9 7.2 7.6

pH

1.34 0.91 1.74 0.91 1.25 0.65 0.78 1.55 0.41 1.40 0.99 1.11 0.41 1.74 1.09 0.66 1.05 1.57 1.70

Cond. (mS cm 1) 110.0 40.0 184.0 104.0 119.0 69.2 69.3 92.4 28.7 159.0 52.0 131.0 28.7 184.0 96.6 41.2 98.2 161.5 178.8

Ca (mg l 1) 10.9 10.8 6.8 7.3 6.5 5.4 6.8 9.1 4.0 7.7 19.2 6.1 4.0 19.2 8.4 5.5 7.1 10.1 17.1

K (mg l 1) 27.8 26.4 46.4 14.4 16.4 11.2 13.0 42.7 5.9 34.5 24.5 35.8 5.9 46.4 24.9 11.4 25.5 43.1 45.6

Mg (mg l 1) 140.0 101.0 94.2 69.5 113.0 46.0 48.3 168.0 35.3 72.5 93.0 54.4 35.3 168.0 86.3 46.2 82.8 118.5 156.6

Na (mg l 1) 713.7 433.1 701.5 420.9 390.4 298.9 311.1 488.0 164.7 610.0 359.9 481.9 164.7 713.7 447.8 300.1 427.0 619.2 682.6

HCO3 (mg l 1) 137.0 89.0 182.0 96.0 183.0 58.0 60.0 238.0 41.0 172.0 76.0 92.0 41.0 238.0 118.7 58.2 94.0 188.5 226.6

Cl (mg l 1)

35.0 0.0 105.0 36.0 81.0 21.0 21.0 86.0 16.0 26.0 22.0 48.0 0.0 105.0 41.4 16.5 30.5 87.9 101.1

SO4 (mg l 1)

Chemical analyses and physical parameters of 12 samples in the protected area of Castelporziano, 2004.

Parameter Sample

Table 4.2.3

1174.0 700.0 1320.0 748.0 909.0 510.0 529.0 1124.0 295.0 1082.0 647.0 849.0 295.0 1320.0 823.9 511.9 798.5 1143.6 1279.4

TDS (mg l 1)

The Pilot River Basin Network: Examples of Groundwater-related Activities 139

140

Chapter 4.2

Figure 4.2.6

Delimitation of conductivity levels in the protected area of Castelporziano as measured in 2004.

Historical data on groundwater conductivity before that saline intrusion occurred are not available; therefore sampling points located in the protected area of Castelporziano characterised by the lowest salinity values were selected as NBL value identification. In this case the NBL value is about 1000 mS cm 1 lower than the P90 values in the areas of about 1300 mS cm 1, as shown in Tables 4.2.2 and 4.2.3. The threshold value was calculated as 2000 mS cm 1, since an EQS value derived from a consolidated methodology is not given. The EQS value will be set by means of expert judgement and probably it will be inferior to the threshold value, considering the particular naturalistic value of the area. The application of the methodology in the protected area of Castelporziano showed that the threshold value for parameters measuring marine water intrusion were exceeded in sample coastal areas, showing a general upward trend of the values from 1999 to 2004.

4.2.5

Conclusions

The chemical composition of the water in the Colli Albani aquifer system is very variable due to the nature of the rocks and the interaction with gasses and


141

fluids originating at depth. The statistical treatment (P90, P97.7) of a large number of samples from different parts of the aquifer can lead to assessment errors of NBL values even though they are not subject to anthropogenic pressure. The application of WP3-BRIDGE methodologies under these conditions is useful for the purpose of threshold identification and trend reversal measures only if the geochemical groundwater families that interact with the receptors are known. WP3 methodologies were applied in two areas of the hydrostructure where there were sufficient data to characterise groundwater and to identify significant parameters of receptor interaction. A general lowering of the piezometric level is visible due to intense water abstraction. Therefore, quantitative recovery should be the first objective of the programme of measures to be adopted by the management plan of the WFD. The Colli Albani aquifer is currently subject to safeguard measures aimed at limiting abstraction permits in critical areas where there are concentrated withdrawals.

References 1. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy, Official Journal of the European Communities L 327, 22.12.2000. 2. L. Galbiati, J. M. Zaldivar, F. Somma, F. Bouraoui, M. C. Moren Abat, G. Bidoglio and J. D’Eugenio, Pilot River Basin Outcome Report: Testing of the WFD Guidance Documents, EUR Report 21518, 2005. 3. Ph. Quevauviller, P. Balabanis, C. Fragakis, M. Weydert, M. Oliver, A. Kaschl, G. Arnold, A. Kroll, L. Galbiati, J. M. Zaldivar and G. Bidoglio, Environ. Sci. Pol., 2005, 8, 203. 4. Bridge general application and evaluation of a proposed methodology for derivation of groundwater threshold values: a case study summary report. 5. A. Di Domenicantonio, M. Ruisi and P. Traversa, Groundwater natural background levels and threshold definition in the Colli Albani volcanic aquifers in central Italy, Autorita` di Bacino del Fiume Tevere BRIDGE Project, 2007. 6. Directive 1998/83/EC of the Council of 3 November 1998 on the quality of water intended for human consumption, Official Journal of the European Communities L 330, 05.12.1998. 7. M. Bucci, Stato delle risorse idriche, Il sistema ambientale della tenuta presidenziale di Castelporziano, Accademia Nazionale delle Scienze detta dei Quaranta, Scritti e documenti XXVII, 2006, pp. 327–387.

CHAPTER 4.3

The Harmoni-CA Initiative GEO E. ARNOLD, WIM J. DE LANGE AND MICHIEL W. BLIND RIZA, PO Box 17, NL-8200 AA Lelystad, The Netherlands

4.3.1

Introduction

The need to improve the integration of research into the policy-making process is one of the major challenges in managing complex environmental problems like water resources management. For many years, research and technology development (RTD) activities have paid more and more attention to incorporating policy-relevant topics in their research agendas. Current RTD projects have established operational links with practitioners. However, the objective of transferring newly developed tools from the research community to operational use by water managers has not been achieved and the efforts of different projects have not been co-ordinated. One of the actions for enhancing the use of tools and closer cooperation between the RTD and the European Union (EU) Water Framework Directive (WFD) worlds was the establishment of the Harmoni-CA (Harmonised Modelling Tools for Integrated Basin Management, EVK1-2001-00192) project in October 2002. Harmoni-CA is a concerted action, supported by the European Commission (DG RTD) under the 5th Framework Programme, which aims to facilitate the specific clustering activities. Harmoni-CA aims to facilitate the dialogue and help bridge the gap between research and policy, by synthesising the available knowledge produced by the various RTD (CatchModw) projects and facilitating the development and the use of these methodologies and tools to support the use of information, communication and technology (ICT) tools in implementing the WFD. Harmoni-CA started in October 2002 and will finalise in September 2007. This chapter gives a brief description of the way Harmoni-CA started the process of bridging the gap between research and policy. The process as initiated by the Harmoni-CA project and the lessons learned will be of interest w

CatchMod is a group of EC FP5- and FP6-funded projects aiming at development of ICT tools and supporting methodologies for integrated river basin management (IRBM).

142

143

The Harmoni-CA Initiative

for the implementation of the new Groundwater Directive, adopted by the European Parliament in December 2006.

4.3.2


The concerted action Harmoni-CA, supported by the European Commission (EC) under the 5th Framework Programme,1 aims to facilitate the dialogue and to help bridge the gap between research, consultancy, operational management and policy. Harmoni-CA therefore synthesises the available knowledge produced by the various RTD (CatchMod) projects and facilitates the development and the use of these methodologies and tools to support the use of ICT tools in implementing the WFD (Figure 4.3.1). The long-term objective of Harmoni-CA is to set up a forum for communication, information exchange and the harmonisation of the use and development of ICT tools that goes beyond the duration of Harmoni-CA. Harmoni-CA therefore formulates a set of specifications for the continuation and extension of the communication forum.

4.3.3

Process of Bridging the Gap Between Research and Policy/Water Management

As described above, Harmoni-CA acts along two tracks. Firstly, Harmoni-CA facilitates activities within the CatchMod cluster, like the identification and enhancement of complementarities between different research projects, and disseminates research results focusing on the projects in the EC-supported CatchMod modelling cluster. Secondly, Harmoni-CA brings together the demand and support for ICT tools and methodologies for the implementation of the WFD. The activities can be divided into a process-related activity and products/tools that support this process. Establishing a dialogue among the scientific and policy-making communities is one of the most important achievements of Harmoni-CA. For establishing a dialogue among the scientific and policy-making communities Harmoni-CA organised yearly forums and conferences. The forums and conferences were supported by workshops on specific topics.

Harmoni-CA

Fundamental Research

Figure 4.3.1

Applied research (e.g. Catch Mod)

mediation

WFD implementation

The role of Harmoni-CA between WFD and applied research.

144

4.3.3.1

Chapter 4.3

Harmoni-CA Forums and Conferences

As stated above, the aim of organising forums and conferences was to improve the relation between research and policy. The first step in bringing policymakers and researchers together was to understand the demands and needs from the policy and water management side for the implementation of the WFD and hence the support to be given from the research and development communities. The ‘‘demand ’’ from the policy-makers and the operational managers was based on the experiences in the WFD pilots and in the implementation of the WFD. The ‘‘support’’ from the methodology and technology providers was given by the EC-supported research (mainly CatchMod) projects, national initiatives and other sources. The main objective of the forums and conferences was to facilitate the dialogue between these groups (Figure 4.3.2). The forums and conferences were targeting activities following the time schedule of the WFD implementation, such as the characterisation reports (WFD, Article 5), the monitoring activities and the preparation of integrated river basin management plans. The focus of the conferences was on stimulating the networking between the four target groups involved. During the first conference (February 2004) it became clear that a large gap existed between the ‘‘demand’’ of the policy-makers and the operational managers and the ‘‘support’’ offered by the scientists. This gap is for different reasons, among others: scientists and policy-makers speak a different language and they have different interests and targets, different agendas and different timetables. Discussions between scientists, policy-makers and stakeholders showed that knowledge generated by many research and demonstration

Policy makers

Harmoni-CA Operational managers

Forums and Conferences

Methodology providers

Technology providers

Figure 4.3.2

Target groups for Harmoni-CA (forums and conferences).


145

projects does not reach policy-makers in an efficient way. On the other hand, the policy-making community does not consider research results, as it should do, mainly for political reasons and difficulties in integrating research developments in legislation. This is further discussed in Chapter 11.3. An important conclusion of this conference was the need to improve and to make operational a ‘‘science–policy interface’’ linked to the implementation of the WFD. As a result a common ‘‘scope paper’’2 has been drafted to strengthen the cooperation between CatchMod/Harmoni-CA (under the umbrella of DG RTD) and those responsible for implementing the WFD (under the umbrella of DG Environment). In this scope paper the three following activities were defined: linking WFD requirements and RTD products; building of a web portal; and close co-operation within Pilot River Basins (PRBs). Since 2004 these actions have played a central role in the activities of HarmoniCA.3 At the same time the EC started a discussion on the integration of scientific and technological progress into the policy-making and implementation process.4 A need was felt for science–policy integration in the implementation process of the EU WFD.

4.3.3.2

CatchMod/Harmoni-CA Workshops

In combination with and additional to the conferences Harmoni-CA supported the organisation of CatchMod/Harmoni-CA workshops aiming to facilitate activities within the CatchMod cluster, like the identification and enhancement of complementarities between different research projects and to strengthen the discussion and contacts between researchers and operational managers. The workshops were organised around specific topics, in relation to ICT tools and focused on the tasks and activities of the different Harmoni-CA work packages (toolboxes, planning methodology, joint use of monitoring and modelling and public participation). Since the end of 2003, but in particular since the 1st Harmoni-CA Forum and Conference in April 2004, the CatchMod/Harmoni-CA consortium started to link its activities with some Common Implementation Strategy (CIS) activities initiated by DG Environment, policymakers and water managers.

4.3.3.3

Conclusions and Lessons Learned from the Conferences and Workshops

Some conclusions and lessons from the conferences and workshops organised by Harmoni-CA/CatchMod are the following. Conferences, workshops, action plans, reports, websites and newsletters are good communication tools but should be ‘‘tailor-made’’ for each user category (policy-makers, operational managers, stakeholders, etc.).

146

Chapter 4.3

New research is too late for actual use. This is often the case for new research projects under the 6th EU Framework Programme and certainly for new research programmes under the 7th Framework Programme. Exceptions are those projects that are truly linked to a WFD activity (e.g. projects for policy support, such as REBECCA (http://www.environment.fi/syke/rebecca) and BRIDGE (see Chapter 9.1)). Many research projects provide results and tools that do not yet meet the requirements for broad application by water managers. Some additional investments are required to make operational tools and create supporting organisations. In addition to the yearly conferences and workshops, it is of utmost importance to participate in regional meetings and river basin meetings to listen to and to discuss the experiences of water managers and to demonstrate and to bring locally relevant research to water authorities. There is a need for a closer co-operation between DG ENV and DG RTD, including the development of funding mechanisms for joint research projects in which the role of water managers is strengthened.

4.3.4

Products/Tools of Harmoni-CA

Beside the process as described above, Harmoni-CA is delivering products and tools to support the communication process. Important products are the WISE-RTD web portal, guidance documents, synthesis reports and summaries.

4.3.4.1

WISE-RTD Web Portal

In close co-operation with Commission services Harmoni-CA developed a web portal which aims to provide direct access to scientific information supporting water policy implementation (see also Chapter 11.3). This portal is under development and will be linked to the official launch of WISE (March 2007). The WISE-RTD portal provides an intelligent search of information on tools and experiences issued from RTD projects, as well as guidance documents in support of the implementation of the WFD. The search is based on WFDspecific issues (e.g. WFD milestones, WFD terminology) and serves multiple user groups (policy-makers, water managers, stakeholders, modellers, etc.). Hence the portal links ‘‘demands and support/offers’’, while taking the different languages/terminologies of the target groups into account. The users can get support from a ‘‘Communication Service Center’’ (CSC). Figure 4.3.3 shows the main idea behind the web portal. WISE-RTD links to websites that contain a wide range of information such as CIS guidance documents, reports from PRBs, reviews and selections of ICT tools, or results of national and EC-funded projects (e.g. the CatchMod cluster). The system already links to more than 60 research and application projects, over 100 tools and provides direct entry inside CIS guidance and


Figure 4.3.3

147

Information flow.

technical guidance documents (January 2007 status). The portal prototype can be viewed at www.wise-rtd.info.

4.3.4.2

Guidance Documents, Synthesis Reports and Summaries

Another activity to support the process of matching the demand and support for knowledge and ICT tools for implementing the WFD is the preparation of guidance documents and synthesis reports. Guidance documents are meant to give guidance to the implementation of activities or tasks. Target groups are operational managers or more general, implementers of a certain task within the WFD implementation. Examples of guidance documents in preparation are:

Uncertainty Calibration Sensitivity analysis Environmental economics Quality assurance in modelling Public participation Monitoring network design Planning methodology for the WFD.

148

Chapter 4.3

Besides guidance documents activities (synthesis and summaries) are carried out within Harmoni-CA, which bring results of projects forward, provide support to new research, etc. Synthesis reports give an evaluation of results on a study or a methodology, etc. Synthesis reports describe amongst others the area of application, gaps, etc. Target groups can be researchers or at a more strategic level policy-makers, water managers and also research funding boards. Examples of these activities are: improving the toolbox on nutrient emission tools to include additional tools; developing a portal for useful standards relating to software in water management; improving quality assurance support tools to make them better adaptable to new (modelling) processes; analysis of decisions support systems, leading to advice on how to develop such systems successfully; and analysis of end-user involvement in European-funded projects leading to advice as to how this can be done more effectively, and thus should lead to improved outputs of research with respect to operational water management. The guidance documents and synthesis reports will also be incorporated in the web portal.

4.3.5

SPI-Water

Four years into the Harmoni-CA project it can be stated that a fruitful communication between research and policy has been established. However, the duration of the concerted action Harmoni-CA is limited. The project started in October 2002 and will end in October 2007. As stated before, the long-term objective of Harmoni-CA is setting up a forum for communication, information exchange and the harmonisation of the use and development of ICT tools that will go beyond the duration of Harmoni-CA. This objective will partly be carried out by SPI-Waterz, a new project that is funded by the 6th Framework Programme, as part of Priority 8.1: policy-oriented researchscientific support to policies. SPI-Water elaborates on a theme already started by Harmoni-CA and proposes a number of concrete actions to bridge the gap in communication by developing and implementing a ‘‘science–policy interface,’’ focussing on setting up a mechanism enhancing the use of RTD results in the WFD implementation. As a first action, existing science–policy links will be investigated. RTD and LIFE projects with direct relevance for the implementation of the WFD will be identified and analysed and their results will be extracted, ‘‘translated’’ and z

Science–Policy Interfacing in support of the WFD implementation.


149

synthesised in a way that can efficiently feed the WFD implementation. Secondly, the information system (WISE-RTD web portal) will be further developed to cater for an efficient and easy-to-use tool for dissemination as well as retrieval of RTD results.5 The web portal will be tested in four selected river basins to better tune the ‘‘product’’ to the needs of WFD stakeholders, policymakers and scientists. In parallel, the web portal will be disseminated to WFD stakeholders. As a third and last action, this science–policy interfacing of WFD-related topics will be extended to non-EU countries, taking their specific needs into account. An assessment of recent practices and needs of non-EU countries, together with an in-depth analysis of the operational needs in two Mediterranean pilot river basins, will allow the preparation of recommendations for an efficient transfer of knowledge.

4.3.6

Groundwater Directive

The initiatives started by Harmoni-CA and the lessons learned during the Harmoni-CA project will be very important for the implementation of the Groundwater Directive (see Chapter 3.1). Indeed, groundwater and surface waters are strongly connected and a lot of the tools already developed and experiences gained can be used for groundwater. Communication between water managers, policy-makers and surface water specialists is not easy. For groundwater, an invisible and ‘‘hidden’’ water resource with quite different characteristics, this is even more difficult. In this case public participation, one of the topics within the Harmoni-CA project, can play an important role. The WISE-RTD web portal, already developed by Harmoni-CA and continued and extended by the new SPI-Water project, will be a challenge for the implementation of the water policies in general, and the new Groundwater Directive in particular. All groundwater activities and projects are invited to make use of this tool.

References 1. Description of Work HarmoniCA (Harmonised Modelling Tools for Integrated Basin Management), EVK1-2001-00192, 2002. 2. G. Arnold and Drafting Group, Research supporting the WFD implementation, Mutual gains from cooperation (scope paper), Lelystad, The Netherlands, 2004. 3. G. E. Arnold, W. J. De Lange and M. W. Blind, Environ. Sci. Pol., 2005, 8, 213–218. 4. Ph. Quevauviller, P. Balabanis, C. Fragakis, M. Weydert, M. Oliver, A. Kaschl, G. Arnold, A. Kroll, L. Galbiati, J. M. Zaldivar and G. Bidoglio, Environ. Sci. Pol., 2005, 8, 203–211. 5. P. Willems and W. J. De Lange, Environ. Sci. Pol., in press.

CHAPTER 4.4

Linking Public Participation to Adaptive Management CLAUDIA PAHL-WOSTL, JENS NEWIG AND DAGMAR RIDDER Institute of Environmental Systems Research, University of Osnabru¨ck, Barbarastrasse 12, DE-49069 Osnabru¨ck, Germany

4.4.1

Introduction

In recent years the awareness that sustainable and integrated water resource management cannot be realised based on expert knowledge and technical solutions alone has increased. Participatory approaches are required in which stakeholders are involved in developing, implementing and monitoring management plans to cope with the complexity of issues to be tackled and the ensuing conflicts of interest.1 Such is also the spirit of the novel European Water Policy aiming at the integration of the hitherto existing fragmented regulatory framework. The European Water Framework Directive (WFD; Directive 2000/60/EC) can be labelled as an example of a new generation of European Union (EU) directives with which the EU seeks to partly overcome the established technocratic, top-down method of European policy-making. The EU member states have more freedom to develop an implementation plan targeted towards their needs, taking into account national and regional conditions.2 Organised interest groups and the public at large are to be involved in the process. The upcoming Groundwater Directive should be implemented in the same spirit and public involvement can be expected to encounter similar challenges in implementing participatory approaches as experienced during implementation of the WFD. The recognition of uncertainties such as climate change or changes in administrative systems underline the need for new, more adaptive management styles in water management in general and groundwater management in particular. Participatory processes are expected to foster the learning of individuals and groups. Eventually, social learning processes are expected to support adaptive management. One precondition for achieving this is 150

Linking Public Participation to Adaptive Management

151

participatory processes that go beyond informing or consulting the public by trying to actively involve the relevant stakeholders in decision-making.

4.4.2

Adaptive Management and Public Participation

Water management in Europe largely lacks experience and expertise in involving the public. At the operational level of water management, public participation is often perceived as being too resource intensive and is seen as a barrier to efficient management rather than an opportunity for developing and implementing innovative management approaches. However, innovation towards more flexible governance systems and management strategies that take different kinds of uncertainties into account are urgently neededw. We argue that stakeholder participation is required to develop, implement and sustain such management approaches, taking into account that: ambiguity exists when defining operational targets for the different management goals to be achieved, and that conflicts of interest require participatory goal setting (not by experts alone) and a clear recognition of uncertainties in this process; the outcomes of management measures are uncertain due to the complexity of the system to be managed and to uncertainties in environmental and socioeconomic developments influencing the performance of implemented management strategies; new knowledge about system behaviour may suggest options for change in management strategies; and changes in environmental and/or socioeconomic conditions may demand changes in management strategies. Given current water management practice, one can identify a clear need for a more coherent and comprehensive approach, an approach based on sound conceptual foundations to deal with uncertainties in water management in general and groundwater management in particular. The idea of adaptive management has already been the subject of ecosystem management discussions for some time.3–6 It is based on the insight that the ability to predict future key drivers influencing an ecosystem, as well as a system’s behaviour and its responses, is inherently limited. Hence management must be adaptive and must include the ability to change managerial practices based on new insights. One form of adaptive management uses management programmes that are designed to experimentally compare selected policies or practices by evaluating alternative hypotheses about the system being managed (e.g. Refs. 7–9). However, adaptive management is not limited to an experimental approach but can more generally be defined as a systematic process for improving management policies and practices by learning from the outcomes of w

The development of such strategies is the main objective of the EU-funded project NeWater ‘‘New approaches to adaptive water management under uncertainty’’ (www.newater.info).

152

Chapter 4.4

implemented management strategies. As Bormann et al.10 pointed out, ‘‘Adaptive management is learning to manage by managing to learn.’’ Learning may encompass a wide range of processes that span the ecological, economic and sociopolitical domains in the testing of hard and soft approaches.11,12 Adaptive management emphasises the importance of the process nature of management without claiming that the process is an end in itself but by explicitly recognising that management strategies and even goals may have to be adapted during the process. We argue that this can be carried out most effectively if a learning cycle unites all the relevant actors in the different phases of policy development, implementation and monitoring. The actors to be involved can be represented by the public at large. Nonetheless, the more common and realistic form of participation in adaptive management is stakeholder participation: organised groups are represented by one or more persons. Organised groups can be very diverse, e.g. environmental NGOs, groups representing social classes, age or gender groups, lobby groups of industrialists, or they can even be another state or regional authority that has traditionally not been involved in the decisionmaking process. The whole adaptive management process (see Figure 4.4.1) requires a number of steps that are part of an iterative cycle. All steps should be participatory (0, 1, 2, 3, 4). In a participatory process different perspectives need to be taken into account in the definition of the problem (0). The design of policies should include scenario analyses to identify key uncertainties and to find strategies that perform well under different possible but initially uncertain future developments rather than searching for a strategy that performs optimally under very specific

Figure 4.4.1

Iterative cycle of policy development and implementation in adaptive management.


153

conditions (e.g. climate) but performs purely if these conditions are not met (1). Policies must be understood as semi-open experiments that require the careful evaluation of potential positive or negative feedback mechanisms by planning and implementing other related policies (1, 2). Decisions should be evaluated by the costs of reversing them. Large-scale infrastructure or rigid regulatory frameworks increase the costs of change. But costs may also be related to a loss of trust and credibility if uncertainties and the possible need for changes are not addressed by the competent authority during policy development (3). Monitoring programmes should include processes to highlight undesirable developments at an early stage. This might imply different kinds of knowledge, including community-based monitoring systems13 (3). The policy cycle must include institutional settings in which actors assess the performance of management strategies and implement change if needed (4). Continuous replanning and reprogramming based on the results of monitoring and evaluation should be institutionalised (4).

In principle, the WFD is compatible with such an iterative and adaptive approach. However, implementation seems to follow largely established management practice based on a more linear and optimisational approach. Uncertainties are taken into account only to a limited extent.14 The implementation of an adaptive management approach is only possible if certain structural conditions are fulfilled. Hence the implementation of adaptive management needs an integrated system linked to a new understanding of management as learning rather than a control process. Response to new insights is only possible if the measures implemented can be changed. The transition to adaptive management relies on increasing the adaptive capacity of the (water) system. It aims at an integrated system based on the understanding of the interdependence between technologies, economic and environmental factors and the formal and informal institutional context. The aim is to increase the ability of the entire system to respond to change rather than to simply react to undesirable impacts of change. Institutionalising this learning capability in the long run will secure the adaptive foundation of management. What are the current requirements for the adaptive management of groundwater resources? New information must be made available and/or consciously collected (e.g. indicators of the performance of management regimes like the relation of groundwater retrieval to groundwater recharge) and monitored over appropriate time scales (given the often very slow percolation of pollutants through the soil and the long groundwater travel times, monitoring time scales need to be considerably longer than those mandated by short-term political objectives).

154

Chapter 4.4

The actors within the management system must be able to process this information and draw meaningful conclusions from it. This can be best achieved if the learning process unites the actors in all the phases of assessment, policy implementation and monitoring. If the information collected provides answers to the questions (hypotheses) they posed, then management is transparent to all those involved. Change must be possible in ways that are open and understandable to all actors. Management must have the ability to implement change, based on processing new information in a learning process where it is clear as to who decides how and when to change management practices, based on which evidence, and why.

Such conditions are not met in many countries in Europe, in particular in groundwater management where long-lasting conflicts may prevail. In the upper Guadiana basin, for example, groundwater use exceeds capacity and the groundwater table drops due to the existence of numerous illegal wells. Representatives from the competent authority argue for the need to enforce legal regulations and for the wells to be closed down by governmental intervention (personal communication). However, given the failure of such attempts in the past one may need to reconsider whether participatory approaches based on learning rather than on control may have a greater chance of triggering change. It can be concluded that adaptive management requires the incorporation of a learning element in its management process. This learning component encompasses social learning: in which the actors involved learn about their different perspectives with regard to groundwater management and their motives behind opting for certain management strategies and actions. Only then can it be guaranteed that the planned activities respect the process nature of groundwater management and that management strategies leave sufficient room for reaction to these changes. Active participation is the foundation for achieving social learning and adaptive management. In order to guarantee that new information is freely distributed and equally understood, the learning cycle must unite all the relevant actors in the different phases of policy development, implementation and monitoring. The actors to be involved can be represented by the public at large or by stakeholders. The more common and realistic form of participation in adaptive management at this stage is stakeholder participation: organised groups are represented by one or more persons. Organised groups can be very diverse including, e.g. environmental NGOs, groups representing social classes, age or gender groups, lobby groups of industrialists, or they can even be another state or regional authority that was traditionally not involved in the decision-making process. In addition to the current forms of participation such as informing and consulting other people or groups, new, more active forms of participation will be required to optimise the learning process and eventually the adaptability of management.

155


4.4.3

Rationales and Requirements for Effective Participation in Groundwater Management

4.4.3.1

Rationales and Goals for Public Participation

‘‘Public participation is not an end in itself but a tool to achieve the environmental objectives of the Water Framework Directive.’’15 This quote from the CIS guidance document on public participation according to the WFD perhaps best describes the dominant rationale for public participation in the current EU water policy. Thus, the current emphasis on participation seems to be predominantly rooted in a certain disillusionment in the effectiveness of governmental steering efforts in the face of the continuing implementation deficits of state environmental policy.16–18 It expresses both a hope and an expectation that participatory processes will lead to an improved compliance and implementation (measured by the agreed environmental goals) due to a more sound knowledge base and an improved acceptance of decisions: in short, an enhanced effectiveness of the pursued policy.19,20 Moreover, some observers generally expect that an increasing societal complexity requires poly-centric and participatory modes of governance.1,16,21,22 Taking a closer look at the CIS guidance document on public participation, two main strands of arguments become apparent (see Table 4.4.1). One is that the quality of decisions is expected to be better: In the course of the participatory process, information is generated or made available that would not have been so otherwise; furthermore, the decision benefits from this information, i.e.

Table 4.4.1

Different rationales for public participation as they appear in the CIS Public Participation Guidance Document relative to the Water Framework Directive.15 (Table after Ref. 60.)

Rationales for public participation

CIS PP Guidance

Quality of decision

pp. 24, 26, 41

Quality of implementation

Make available lay local knowledge to the authority Make available knowledge regarding attitudes and acceptance on the part of the non-state actors to the authority Improve environmental quality, reach environmental goals Increase environmental awareness, education, information on the part of the non-state actors Build acceptance of and identification with a decision on the part of the nonstate actors Build trust among non-state actors and between these and the authority Alleviate conflicts by mediation of interests

p. 24 pp. 7, 26 pp. 4, 26 pp. 4, 26, 41 pp. 26, 41 pp. 26, 41

156

Chapter 4.4

the information is actually incorporated into the decision. Thus, water policy and management decisions can benefit from the factual knowledge of involved actors about their (local) conditions,23–25 assuming that those who are closest to a problem develop the best understanding of it.26,27 Other authors, however, contest this claim and hold that it is rather the authorities who have different and usually more reliable means of information provision at their disposal,28 especially regarding highly technical issues, and the corresponding need for specialised expert knowledge.29 Then again, there may be information that ‘‘emerges’’ from the close interaction of actors in a group process. Many authors stress the positive effects of social learning, the plurality of perspectives and thus the more creative decision-making as characteristics of participatory decision-making.30 Yet group processes also have the potential to create adverse effects. For instance, Cooke31 points out problematic findings from social psychology regarding consensus-oriented group processes, such as the tendency towards taking risky decisions or an immunisation towards independent and critical arguments. Another type of information from which decisions could profit is information regarding the extent to which planned measures will be accepted by the addressees. In this respect, participation becomes an ‘‘instrument for the anticipation of resistance to planning and implementation.’’32 Generally, participation is expected to prevent implementation problems from occurring.33 Quite plausibly, the addressees of a decision must know of it in order to be able to implement it: obey rules, comply with requirements. If future addressees are involved in decision-making, they can be assumed to be thoroughly informed about these decisions, and a higher rate of compliance can reasonably be expected, as the possibly necessary measures of reorganisation and adaptation to new (regulatory) conditions, which usually take some time, can duly be taken. Furthermore, compliance with a decision is expected to depend positively on the degree of acceptance, or even identification, on the part of the addressees (e.g. Refs. 33, 34). Acceptance may, firstly, be supported by providing the interested actors with early and comprehensive information. This may prevent actors from feeling left out or ignored, and create a sense of involvement and belonging. Also, certain educational effects, e.g. in the sense of an improved environmental awareness, can play a role.35 Moreover, an intensive involvement of the concerned actors in a decision process that is perceived as fair and based on mutual communication is expected to enhance the acceptance of the decision. This even holds when the result does not correspond to the actors’ expectations,27,36 as procedural justice research has found that the acceptance of a decision crucially depends on aspects of fairness of the decision procedure.33,37–39 Furthermore, a decision that involves conflicting interests is more likely to be accepted by the different parties if it is based on either a consensus or at least a compromise to which most of the parties agree. This in turn most likely requires an intensive participatory process that allows the concerned actors to effectively claim their stakes, but also a spectrum of interests that does not fundamentally rule out any consensual solutions.


157

Furthermore, in the medium and long term, the building of trust relationships among the non-state actors, the state actors and also between non-state and state actors through participation33 can lead to an increased regional collective social capital, and can thus influence the context of future decision processes. In particular, the building of trust can improve acceptance of and thus the willingness to comply with measures, as empirical studies in other contexts have shown.39

4.4.3.2

Public Participation Provisions in the WFD

The WFD demands different forms of involving the public in decision-making, which are explicated in further detail in the PP Guidance Document, although of course not in a legally binding manner.

4.4.3.2.1

Legally Binding Three-stage Consultation Process for the River Basin Management Plans

Most stringently regulated is the consultation for the river basin management plans (RBMPs). RBMPs form the principal instrument of transparency and communication of the current status of waters and planned measures.40 From the end of 2006 onwards, the public must be informed, and can voice its concerns, at three annual intervals regarding the working programme (2006), the most important water management issues (2007) and the draft management plans (2008), leaving the public 6 months each time to produce written statements (Art. 14 (1.2) and (2) WFD). Since this formal consultation procedure only has to be implemented at the level of the rather large river basin districts, the impact of public participation on decision-making can be expected to remain rather poor. This is even more the case since the current cooperation in the river basin districts in federative EU member states such as Germany is restricted to exchange between authorities, while the actual planning process takes place at the level of the federal states. While consultation is only required for the RBMP and not for the more important programmes of measures, the latter is required by the SEA Directive. In practice, both consultations will most probably be combined into one single procedure (see Ref. 2).

4.4.3.2.2

Free Access to Background Information

According to Art. 14 (1.3) WFD, the public is to be granted free access to the documents used for preparing the RBMP on request.

4.4.3.2.3

Encouragement of ‘‘Active Involvement’’

Probably the most important provision regarding public participation is the required encouragement of the ‘‘active involvement’’ of all interested parties

158

Chapter 4.4

(Art. 14 (1.1) WFD). Active involvement differs from the consultation process in three respects. First, relevant involvement relates to the ‘‘interested parties’’ and thus to a smaller circle of the public than in consultation. Second, ‘‘active involvement’’ implies a much stronger involvement of actors than in consultation. Third, ‘‘active involvement’’ relates to the implementation of the whole directive and not exclusively to the RBMP. Although this requirement is less legally binding than the consultation process and, moreover, not at all operationalised by the directive, regional authorities in the member states have already taken great pains to do this requirement justice (see the examples below).

4.4.3.2.4

Adaptive Regulatory Impact Assessment

As a learning instrument and to systematically monitor its success, the WFD requires reports to be drawn up about the public participation conducted. According to Art. 13 (4), RBMPs must also include information on how public participation has affected or changed the plan. The Guidance Document stresses that this instrument serves not only ex-post control by the Commission but, predominantly, to improve public participation in the following planning cycle. While this instrument appears to have been underestimated until now, it allows the successive improvement of public participation from one planning cycle to another. The collection and systematic analysis of experiences enables the adaptation of public participation in the following cycle. Thus, for the first time, legal evaluation is not only institutionalised as a retrospective regulatory impact assessment41 but, moreover, as an adaptive management procedure.

4.4.3.2.5

Imperative to Conduct Actor Analyses?

When the competent authorities determine the relevant public to be involved, differentiating according to the different phases and demands of implementation, and also distinguish different forms and degrees of participation—in short, when they tailor participation instruments to their target groups—this will in many cases require a systematic identification, analysis and classification of (potentially) relevant actors.15,42–44

4.4.3.3

German Experiences with Public Participation

Although the WFD commands implementation at the level of the river basin districts, the federal organisation in Germany has led to implementation structures at the level of the federal states. Thus, the most important public participation instruments are organised at state level. These differ considerably from state to state. As Figure 4.4.2 illustrates, different forms of participation have been institutionalised. Some are located at the state level (such as councils or a steering group) but most are situated at more local scales. And while some instruments are targeted at the public at large, others include only selected

159


state level

Steering Group (1 state)

Council (7 states)

Road Shows (3 states) Regional Conferences (4 states) working level the public at large

Councils (1 state)

Area Cooperations (1 state)

Working Groups (2 states)

selected stakeholders

degree of participation and of decreasing publicness

Figure 4.4.2

Overview of public participation institutions (‘‘active participation’’) in German federal states (adapted from Ref. 2).

stakeholders, thus allowing for more intensive cooperation in the WFD’s implementation. On the whole, a surprising multitude of participation activities is being undertaken, covering information, consultation and active participation, even before the official consultation process regarding the RBMPs has even started. The Groundwater Directive is not new in terms of substantive provisions but rather in that it expands the procedural law in terms of instruments in order to attain the water quality goals.45

4.4.3.4

Example: Regional Participation in Groundwater Protection from Agricultural Nitrate

This example focuses on the participation activities in Lower Saxony, Germany’s ‘‘principal agricultural state.’’ Agriculture is one main, if not the main, addressee of groundwater regulations because of its high potential to affect groundwater quality due to its area-wide operation. Diffuse pollution due to nitrate is still one of the most important unresolved problems of groundwater quality. Experiences made during the implementation process of the WFD provide valuable insights for the groundwater directive, too. Information is provided by the authorities through a leaflet and two internet pages. Roadshows in different places and for different catchment areas (‘‘regional’’ and ‘‘area conferences’’) have been organised to inform

160

Chapter 4.4

stakeholders and the interested public about the first inventories that were developed for the sub-basins within the administrative boundaries of Lower Saxony. The number of participants ranged between 70 and 200 persons. In January 2003, the Lower Saxon Ministry of Environment (MELS) established a council for the implementation of the WFD in Lower Saxony, aiming to inform the most important stakeholders at the state level, but also to allow some room for discussion as regards the current and future implementation steps. The council meets once or twice a year and is made up of about 50 representatives from different sectors. At the ministerial level, a technical expert group involving about a dozen stakeholders has been established to support the development of the methodological basis for the implementation of the WFD, with a specific emphasis on the objectives for groundwater bodies. Although the official process for informing and consulting the public has not yet begun, the variety of the more or less institutionalised participation efforts illustrated above show that some of the provisions of the WFD have already been surpassed by the water authorities in Lower Saxony. On the other hand, these institutions do not allow non-state actors to participate in planning concrete measures at the regional and local scales and therefore hardly meet the needs of regional stakeholders, since the meetings are organised at higher levels, making the translation of partly abstract and technical decisions into the lower levels difficult. Moreover, the number of participants, e.g. in the council, is too large for a constructive working atmosphere. Consequently, the MELS established a more local and direct form of active involvement. In autumn 2005, 30 so-called ‘‘areas of cooperation’’ were initiated at the sub-sub-basin level covering the whole of Lower Saxony. They were designed as long-term institutions that would contribute to the formulation of the RBMPs, while leaving the final decision competence with the state authorities (MU Niedersachsen 2005, p. 2). Although the official consultation process at the level of the whole river basin districts, starting by the end of 2006, could also influence the implementation of the WFD, the most important discussions, and perhaps decisions, will take place within these areas of cooperation. They hold a great potential to break up the existing alliances of agriculture and state actors and allow for a true consideration of the environmental goals provided by the WFD. Whether or not, and to what extent, the area of cooperation will succeed in terms of the stringent implementation of WFD demands and whether the measures decided upon by the area of cooperation will in fact lead to substantially reduced nutrient intakes into the region’s waters depends on a series of factors that will be analysed in the next section. The successful implementation of the WFD in a region with intensive agriculture surely depends on many aspects. First of all, the historical and economic role of agriculture, the regional experiences in water management and the different resources and interests of actors are decisive. A crucial aspect for Lower Saxony in general will be the specific motivation of stakeholders as to why and how the WFD should be implemented. Is the WFD only seen in terms of fulfilling European legislation or is there a true desire to improve the status of water bodies? Where the latter is the case—which will be true if a clear


161

benefit for the person or group in question becomes obvious—the area of cooperation can be of great assistance in achieving ecological targets, without completely dismissing economic interests. Exploring all possible measures and instruments that can contribute to good water status and negotiating the opportunities is a great chance for stakeholder involvement and also gives weak actors the chance to state their interests.46 Integrating new actors and perspectives and developing the capacity for effective and efficient communication require processes of social learning.

4.4.4

Social Learning in Public Participation: Support for Adaptive Management

What is social learning? As already explained above, adaptive management places an emphasis on the improvement of management processes by learning. Here, it becomes obvious how the concept of social learning has greatly influenced the development of the meaning of ‘‘adaptive management.’’ Learning at its best should be an active process, and social learning with respect to sustainable development is based on the participatory processes of social change and societal transformations.16 The necessity of participatory approaches is also considered as crucial in adaptive management processes. Therefore both concepts, social learning and adaptive management, cannot be applied without active stakeholder involvement in planning and decisionmaking. In a very broad sense, social learning is referred to as building knowledge within groups, organisations or societies. Mostert47 explains it as ‘‘the growing capacity of social entities to perform common tasks, such as the management of a water resource.’’ But it should not be forgotten that it is often also an individual’s gain in knowledge in a well-managed stakeholder process that can make the difference. The concept of social learning can therefore very generally support the traditional use of economic and hydrological information as well as general expert knowledge in making water management decisions. In bringing together a large spectrum of relevant actors for groundwater management not only communication will be improved and thus information better exchanged, but also new information will be gained. A better understanding of feedback mechanisms and actors’ dependencies will evolve. This newly acquired knowledge facilitates better cooperation and eventually solution-finding in a team. The HarmoniCOP projectz resulted in a handbook on improving social learning and participation in water management. This process was briefly summarised as ‘‘learning together to manage together.’’48 The major difference of this interpretation of social learning from Bormann’s13 ‘‘adaptive management is learning to manage by managing to learn’’ is its focus on learning as a collective experience. z

Harmonising Collaborative Planning (HarmoniCOP), an EU project funded within the 5th framework programme (www.harmonicop.info).

162

Chapter 4.4

The following aspects should be distinguished in order to create conditions conducive to achieving social learning:

appropriate framing conditions (legal framework/authorities); well-designed process management; well-selected methods and tools; and leadership issues.

If water managers manage to use active participation and social learning to result in collective decision-making, it feeds back to the water managers and finally to policy-makers. Successful social learning therefore amplifies its impact by continuously improving its positive policy environment, and thereby supports the idea of the iterative policy cycle (see Figure 4.4.1).

4.4.4.1

Appropriate Framing Conditions

Social learning not only benefits the creation of new joint ideas and common visions to realise bottom-up planning of societal relevance, but it can and should also support institutional change. In practise, this could be realised, for example, by the improved cooperation of agencies of agriculture with agencies of nature and water protection. During the discussions on how to implement the participatory aspects of the European WFD it was requested at an early stage that the environmental measures in agriculture should become an integral part of implementing the WFD.49 Corresponding to the requests of the WFD to implement water protection measures in a cross-sectoral manner, it will also require a change in behaviour of the relevant actors coming from different fields of policy.50 Here, the above-mentioned change of relations and the knowledge built between individuals, groups and organisations helps change the common practise by increasing the understanding of the complexity of the problem. Compared to merely having information, social learning is therefore the more sustainable form of participation: facilitating adaptive management and supporting institutional change into a desired direction determined in a transparent process of negotiation. Social learning therefore accordingly supports the demand of the WFD (‘‘active involvement of all interested parties’’ according to Art. 14 WFD and of the guidance document on ‘‘Public participation in relation to the WFD,’’ Section 3: ‘‘Active involvement of all interested parties in the planning process of the directive’’15 to involve a large spectrum of actors. Aspects such as trust, social learning and the building of networks are nowadays considered to be the key elements of sustainable water management.12,14 Conducive framing conditions for social learning and adaptive management are characterised by a policy framework that, among other things, allows the following conditions to evolve: decentralisation of decision-making to make it as local as appropriate (e.g. if the topics to be discussed and decided upon become very abstract for stakeholders, their motivation to participate decreases because it


163

becomes difficult for them to integrate their knowledge into the overall context; but the right level of decentralised decision-making implies also that factors like the NIMBY effect are taken into consideration which, if it occurs, would call for a higher level of decision-making); high flexibility in planning groundwater management as well as flexible and adaptive working structures to better react to uncertainties; and efficient management of relations among relevant actors as well as their roles in the network to guarantee long-term cooperation with clear-cut defined roles (including requirements and benefits) for the stakeholders. Stakeholder participation and social learning in groundwater management must develop within a solid legal and institutional framework. Otherwise there is a great risk that the stakeholders are reluctant to participate. It is especially important to define:51 the rights and duties of representatives; and procedures for those who are continually reluctant to participate (this may imply personal visits and additional information for the stakeholders concerned or even special incentives to participate but also a clear working modus with deadlines for continuation). Although social learning may initiate institutional change, institutional factorsy can also become a barrier to social learning.52 Bureaucratic working procedures, centralised structures of organisations and corresponding centralised decision-making reduce the necessary flexibility of participatory processes. Moreover, it increases the risk of ‘‘solving’’ problems that are not the actors’ real problems. Centralised structures and rigid, over-regulated bureaucracy hinder the creation of a conducive, consensus-oriented culture of discussion from the very beginning.

4.4.4.2

Well-designed Process Management

At the beginning of a stakeholder process in groundwater management it is important to define a common problem. For example, a water supplier is interested in abstracting groundwater of a sufficient quality and quantity to reduce production costs. Agricultural associations, for example, will defend the interests of farmers to continue their current practise of fertilisation if compensation for losses in productivity is not guaranteed. In such a case, starting out with a problematic statement like ‘‘reducing the nitrate levels in groundwater to potable water level’’ would be too one-sided. A more holistic statement allowing give-and-take in terms of measures and benefits should be preferred. y

The term ‘‘institutions’’ is used in a sociological sense where formal institutions such as laws, acts and policies are distinguished from informal institutions such as general unwritten rules and norms. Institutions in general guarantee the organisation of individuals and groups in communities (cf. Ref. 59).

164

Chapter 4.4

One simple example could be ‘‘improving the groundwater status,’’ rather than already focusing on one pollutant and giving indirect thresholds. Examples of important aspects in successful process management are: making use of external facilitation for setting the rules of the process as well as having a ‘‘neutral mirror’’ for reflecting upon the process; making clear how the process influences decisions and results; and the rather informally designed participatory process should increasingly result in formal agreements and responsibilities. Well-designed process management will lead to social learning among participating actors by sensitising to for the problems and needs of others. This new perspective for individuals and groups is based on their interaction leading to not only improved but also new relations. The following paragraph will demonstrate how selected methods and instruments support this learning process over a required time span. The results of learning are measurable in terms of the quality of the relations among actors and in terms of the quality of the technical outcomes of measures and interventions. Unfortunately in water management, many results only become measurable after a longer time span. This is especially true for groundwater. In the preceding paragraph the importance of an efficient management of network relations was already emphasised. This is partly guaranteed by the right framework. But the process management in itself can also contribute to the good management of actor relations within the network. Within the process of participation it is first necessary to sensitise actors to the importance of such functional network management; help set up or improve the network; and train the relevant actor(s) in managing and maintaining the network. To improve communication and relations among stakeholders, it is even recommended to engage social scientists to map the existing communication networks amongst the various ‘‘message senders’’ and ‘‘message receivers’’ involved in the management and to use a specific aquifer.51 The organisation in charge of the participation process must take these results into account for their process design.

4.4.4.3

Well-selected Methods and Tools

Selecting the appropriate method or the right information and communication (IC) support tool can determine the success of participation. IC methods and tools are defined as ‘‘material artefacts, devices or software, that can be seen and/or touched, and which are used in a participatory process to support the interaction between stakeholders (including scientists) and with the public through two-way communication processes.’’53 Examples include maps and Geographic Information Systems (GIS), as well as group model building and


165

role playing games. The precondition is that the method or tool supports the dialogue: information should be dissipated equally in all directions. Since social learning requires a multitude of well-set conditions to evolve and also defines a multitude of goals in itself, it is unrealistic to believe that one stakeholder meeting, one workshop or one method will suffice to guarantee its development. For this reason, social learning is explicitly defined as a process. This means that the IC tools are not always appropriate in each phase of the participation process (in the starting, managing and improving phases; compare Table 4.4.2) and that not all conditions and goals of social learning are equally supported. Table 4.4.2 shows the assessment of tools and methods with regard to their applicability in different participation phases (good to low applicability), depicting their basic effect with regard to selected conditions and/or objectives of social learning.54 The applicability and benefits of IC methods and tools in participatory processes can be summarised as follows: different methods support the various phases of participation with different intensity; different goals of the process require different methods; trust-building methods are a precondition for creating the transparency of the process; methods must be culturally adapted so that participants are encouraged to formulate their interests; interest groups must be integrated into the development of new methods of participation; and additional information and expertise will finally be gathered.

4.4.4.4

Leadership Issues

The importance of good and neutral facilitation has already been highlighted under the issue of process management. In many cases governmental authorities will not only take the lead in the participation process but will also, to a certain extent, facilitate working sessions of stakeholder groups in which they represent a stakeholder themselves. During the implementation of the WFD it is and has often been the authority that maintains the final power to decide which other stakeholders to invite to a meeting in which all were supposed to have a say. The fact that one authority or representative may invite, moderate and contribute to a participatory process can certainly raise difficulties in creating an egalitarian working atmosphere. It is an important topic when planning the overall moderation and process management of participation. Additionally, it may be the case that other technical authorities are invited to these meetings that may compete with each other for governmental budgets. This would be an additional reason to call for professional moderation, adding to the credibility of the process.

’

’ ’

’

K

K

K

’, Good applicability; K, medium applicability; m, low applicability.

Website

Maps Spatial mental models and maps

Facilitated session in which participants build a model to improve their understanding of the issue Game situation in which players act out roles in a real or imaginary context Facilitated and reported open discussion between participants System used for the storage, mapping and analysis of geographic data Graphic scale models Geographic representation and structuring of perceptions about issues Computer-based collection of information accessible on the internet, sometimes including a forum

Short description

Phase: starting

’

’ ’

’

’

’

’

Phase: managing

’

m

m

’

K K

K

’ ’

’

’

Phase: improving

m

Knowing about system complexity

’

’

Fairness of the process

’

’

Learning about other perspectives

’

’

’

Distribution of information

’

’

Common problem definition

Tools and methods and their applicability in different participation phases, and their effect with regard to selected conditions and/or objectives of social learning.

Geographic Information System (GIS)

Round table conference

Role playing game

Group model building

Name of tool or method

Table 4.4.2

166 Chapter 4.4


167

Experiences with participation and social learning are currently largely based on the interaction of the representatives of interest or stakeholder groups in the form of informal to formal working groups, forums and workshops. Unfortunately, there is a large knowledge gap on the mechanisms that guarantee that individual stakeholders correctly represent the interest of their group in the participation process and that information gathered during the participation process is correctly fed back to their group. One precondition for a successful participation process is certainly that representatives should have the clear mandate of their group. Here, it is also the structure of the organisation they represent that influences the kind of mandate provided and the way in which information and decisions are fed back into the organisation. In additionally to the provided mandate, the ‘‘quality’’ of a stakeholder in a process depends on his or her leadership qualities. On one hand the individual qualities and skills of the representatives largely determine the satisfaction of their represented group with the outcomes of participation. On the other hand these qualities and skills determine the acceptance and individual success of the representative in the participation process. It therefore becomes clear that the two concepts of social learning and leadership influence the success or failure of the participation process. Examples of individuals’ high leadership qualities have already been described by Maslow.55 Among the characteristics of people with high leadership qualities Maslow mentioned ‘‘they focus on problems outside themselves.’’ This attribute is at the same time one inherent goal of social learning: becoming aware of other perspectives of a problem and better understanding why other people hold a contrasting position. The importance of leadership issues was underlined by the outcomes of the HarmoniCOP project, where it was recognised that ‘‘the success of social learning is dependent on the participation of key individuals and their attitude.’’56 Not only organisations and facilitators should become increasingly aware of the required leadership skills but also research should be conducted to investigate which kinds of leadership styles and competencies may be most needed at different scales of participation in water management. Finally, another risk to the often used stakeholder participation must be mentioned. Groups that are composed of different representatives are often not really representative, despite all efforts to involve all groups or people concerned. Groups that meet on a regular basis may tend towards corporatism, which involves the risk that decisions are taken that may have a negative impact on those actors who did not participate. This risk underlines the necessity of a careful process management that may include a stakeholder or actor analysis at the beginning of the process (cf. Section 3.2). All steps of the process require an intermediary evaluation to notice when corporatism with its negative effects is developing.

4.4.4.5

An Example of Participation-based Measures in Groundwater Management

Experiences from an EU-funded project on sustainable groundwater management show good results with voluntary agreements (in the form of contracts)

168

Chapter 4.4

between land managers and the water supply companies and/or the federal state of Lower Saxony in Germany.57 Under the terms of the contracts, the enterprises (principally farmers, but also forestry and horticultural enterprises) are obliged to observe particular restrictions and conditions which go beyond good practice. Farmers are paid compensation for economic losses that may arise. Thus, the farmers make a contribution towards groundwater protection which they would otherwise not be required to do, in return for payment. The money required comes from water abstraction charges and, in some cases, from the EU. The basis for developing these voluntary agreements and implementing a variety of measures therein was the development of the cooperation model in Lower Saxony. Cooperation committees were formed for all of the waterworks where representatives from agriculture and forestry operations, the water companies and the authorities involved, namely the Chamber of Agriculture and the Water Authority, sit at one table. The objective of this cooperation was to find a common solution to lower the nitrate emissions into groundwater for drinking water protection. The preconditions for the successful implementation of these voluntary agreements are the following. Local water advisors work in the field to supervise/aid farmers. Measures are jointly developed by local and regional water management authorities, water suppliers and farmers. Economic disadvantages of farmers are compensated. Random control of measures: farmers are directly approached if they do not comply with contractual agreements. In worst cases, farmers can be excluded from further voluntary agreements after the payment of compensation has stopped. Since money for compensation is limited, this instrument has only been implemented for water protection areas until now. Available measures are restricted to changes in agricultural production. But restrictions of this instrument exist, too: e.g. even a high economic incentive cannot work for farmers with high production levels of liquid manure because they are physically forced to get rid of the manure if they do not completely change their farming system. What makes it difficult to judge success is that the expected benefit of reduced nitrate levels in groundwater—depending on the measure in detail and the natural factors—are only visible after several years. One advantage of voluntary agreements lies in their self-control mechanism. Groups that join for a common goal or common intervention tend to socially control each other so that no ‘‘free-riders’’ benefit from the system. The given example shows what kind of steps can be taken towards achieving more adaptive groundwater management in practise. If this new instrument in the form of voluntary agreements becomes more effective in the sense of adaptive groundwater management, special attention should be paid to monitoring and evaluating the process and to the learning that has taken place among the actors involved. Naturally, more adaptive groundwater management requires more


169

than just new instruments. The requested new management approaches may include different working attitudes, new institutions and the previously mentioned innovative instruments for implementing the management strategies. In the case of Germany, groundwater protection presents the most important drinking water resource. Accordingly, water suppliers and relevant authorities are urgently awaiting a rigid daughter directive of the WFD on groundwater.58 But since large implementation gaps already exist for policies, laws and ordinances in water protection, it is unlikely that augmented regulation will necessarily aid groundwater protection. Cross-sectoral collaboration and stakeholder participation become increasingly important because of the currently acknowledged complexity of problems. Also, in an era of job cuts in the relevant authorities, it is also an economic necessity. This development underlines the necessity of making use of new instruments. This is because these participatory-based voluntary agreements may demand less effort in controlling their successful implementation than ‘‘traditional’’ instruments in water protection based on mechanisms of command and control. Besides the economic needs that become drivers of change in water management, the previously mentioned factors such as climatic change, including the increasing uncertainty in water scenarios, also make adaptive management an appropriate solution. Adaptive management will consist of many small experiments trying to come up with new ideas and instruments like the one described. All of these steps, whether successful or not, should be planned, directed and certainly documented and well reflected upon to prepare the floor towards a more open, transparent and participatory management style.

4.4.5

Conclusions

Adaptive water management and participatory approaches are not advocated as a panacea to solve all kinds of water problems. Different management traditions, cultures and styles make it necessary to carefully explore what is possible and to develop approaches adapted to the socioeconomic and environmental context. The acceptance of this new approach by water managers and other actors is a precondition for its success. If adaptive management cannot be meaningfully embedded in its policy environment, it would also be useless to force its implementation. When can adaptive groundwater management been considered especially useful? new, more creative solutions are required due to new, unpredicted problems; major uncertainties about the effect of measures prevail; slow processes of change with unreliable data and information; problems and their solutions have and will have an impact on many actors; and there is a general need to adapt traditional instruments of groundwater management.

170

Chapter 4.4

When can adaptive groundwater management been considered too resourceintensive to be implemented on a routine basis? Decisions are routine and very clear: the processes are technical; and there is an agreement among actors in groundwater management that techniques, methods and instruments are most appropriate and functional to solve a problem. Eventually, it is necessary for water managers to understand that adaptive water and adaptive groundwater management are composed of many measures and activities that can be placed on a continuum from ‘‘hardly’’ adaptive to ‘‘very’’ adaptive. It is the final selection in a participatory process carefully evaluating a whole range of possible management approaches that makes it the best option for a particular problem in a specific area.

References 1. C. Pahl-Wostl, Aquat. Sci., 2002, 64, 394–411. 2. J. Newig, Z. Umweltpolitik Umweltrecht, 2005, 28(4), 469–512. 3. C. S. Holling (ed). Adaptive Environmental Assessment and Management, John Wiley, New York, 1978. 4. C. J. Walters, Adaptive Management of Renewable Resources, McGraw Hill, New York, 1986. 5. C. Pahl-Wostl, The Dynamic Nature of Ecosystems: Chaos and Order Entwined, Wiley, Chichester, 1995. 6. K. N. Lee, Conserv. Ecol. 1999, 3(2) (online at: http://www.consecol.org/ vol3/iss2/art3). 7. L. Gunderson, Conserv. Ecol. 1999, 3(1) (online at: http://www.consecol.org/vol3/iss1/art7). 8. C. F. Kiker, J. W. Milon and A. W. Hodges, Ecol. Econ., 2001, 37, 403–416. 9. B. D. Richter, R. Mathews, D. L. Harrison and R. Wigington, Ecol. Appl., 2003, 13, 206–224. 10. B. T. Bormann, P. G. Cunningham, M. H. Brookes, V. W. Manning and M. W. Collopy, Adaptive Ecosystem Management in the Pacific Northwest, USDA Forest Service General Technical Report PNWGTR-341, 1994. 11. P. Gleick, Science, 2003, 302, 1524–1528. 12. C. Pahl-Wostl, Transitions towards adaptive management of water facing climate and global change, Water Resour. Manag., 2007, 21(1), 49–62. 13. D. Ridder, E. Mostert and H. A. Wolters (eds), Learning Together to Manage Together: Improving Participation in Water Management, Handbook of the HarmoniCOP project, 2005 (ISBN 3-00-016970-9). 14. V. Galaz, Does the Water Framework Build Resilience? Harnessing SocioEcological Complexity in European Water Management, policy paper 1, Swedish Water House/The Resilience and Freshwater Initiative, 2005


15.

16.

17. 18. 19. 20.

21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

31.

32.

33. 34.

171

(http://www.swedishwaterhouse.se/images/partners/20050810093304WFD BuildResiliencReport.pdf/). European Union, Guidance on Public Participation in Relation to the Water Framework Directive. Active Involvement, Consultation, and Public Access to Information, Common Implementation Strategy, Guidance Document no. 8, Luxemburg, 2002. J. Minsch, P. H. Feindt, H.-P. Meister, U. Schneidewind, T. Schulz and J. Tscheulin, Institutionelle Reformen fu¨r eine Politik der Nachhaltigkeit. Studie im Auftrag der Enquete-Kommission ‘‘Schutz des Menschen und der Umwelt’’ des Deutschen Bundestages, Springer, Heidelberg/New York/Berlin, 1998. A. Voßkuhle, Verwaltungsarchiv, 2001, 92, 184–215. M. Lee and C. Abbot, Modern Law Rev., 2003, 66(1), 80–108. C. Coglianese, Duke Law J., 1997, 46, 1255–1346. F. H. J. M. Coenen, D. Huitema and L. J. O’Toole, Participation and environmental decision quality: an assessment, in Participation and the Quality of Environmental Decision Making, ed. F. H. J. M. Coenen, D. Huitema and L. J. O’Toole, Kluwer Academic, Dordrecht/Boston/ London, 1998, pp. 307–324. V. Ostrom, C. M. Tiebout and R. Warren, Am. Polit. Sci. Rev., 1961, 55(4), 831–842. C. Pahl-Wostl, Environ. Modell. Software, 2007, 22(5), 561–569. J. A. Lo´pez Cerezo and M. Gonza´lez Garcı´ a, Phil. Tech., 1996, 2, 1. L. Pellizzoni, Environ. Values, 2003, 12(2), 195–224. S. Yearley, S. Cinderby, J. Forrester, P. Bailey and P. Rosen, Environ. Values, 2003, 12(2), 247–262. J. Steele, Oxford J. Legal Studies, 2001, 21(3), 415–442. J. C. Thomas, Public Participation in Public Decisions. New Skills and Strategies for Public Managers, Jossey-Bass, San Francisco, CA, 1995. A. Fisahn, Demokratie und O¨ffentlichkeitsbeteiligung, Mohr Siebeck, Tu¨bingen, 2002. G. Munnichs, J. Agric. Environ. Ethics, 2004, 17, 113–130. J. Doak, Changing the world through participative action: the dynamics and potential of local agenda 21, in Participation and the Quality of Environmental Decision Making, ed. Frans H. J. M. Coenen, Dave Huitema and Laurence J. O’Toole, Kluwer, London, 1998, pp. 77–91. B. Cooke, The social psychological limits of participation? in Participation: The New Tyranny? ed. B. Cooke and U. Kothari, Zed Books, London/ New York, 2001, pp. 102–121. W. Linder and A. Vatter, Kriterien zur Evaluation von Partizipationsverfahren, in Planung und Kommunikation, ed. K. Selle, Bauverlag, Wiesbaden/Berlin, 1996, pp. 181–188. H. Bulkeley and A. P. J. Mol, Environ. Values, 2003, 12(2), 143–154. T. Webler and O. Renn, A brief primer on participation: philosophy and practice, in Fairness and Competence in Citizen Participation: Evaluating

172

35.

36. 37. 38. 39. 40.

41.

42. 43.

44. 45. 46.

47.

48. 49.

50.

51.

Chapter 4.4

Models for Environmental Discourse, ed. O. Renn, T. Webler and P. Wiedemann, Kluwer Academic, Dordrecht, 1995, pp. 17–33. H. Ryffel, Bedingende Faktoren der Effektivita¨t des Rechts, in Zur Effektivita¨t des Rechts, ed. M. Rehbinder and H. Schelsky, Bertelsmann Universita¨tsverlag, Du¨sseldorf, 1972, pp. 225–246. J. L. Creighton, The Public Involvement Manual, Abt Books, Cambridge, MA, 1981. E. A. Lind and T. R. Tyler, The Social Psychology of Procedural Justice, Plenum, New York/London, 1988. T. R. Tyler, Why People Obey the Law, Yale University Press, New Haven, CT, 1990. K. Murphy, Law Human Behav., 2004, 28(2), 187–209. European Union, Common Implementation Strategy for the Water Framework Directive (2000/60/EC), Guidance Document no. 11 on the Planning Processes, Luxemburg. 2003. C. Bo¨hret and G. Konzendorf, Guidelines on Regulatory Impact Assessment (RAI), prepared for the Federal Ministry of the Interior and for the Ministry of the Interior of Baden-Wuerttemberg, Speyerer Forschungsberichte, Speyer, 2004, vol. 234. W. M. Babiuch and B. C. Farhar, Stakeholder Analysis Methodologies Resource Book, National Renewable Energy Laboratory, Colorado, 1994. A. Harrison, G. Schmidt, C. Avis and R. Hauser, WWF’s preliminary comments on public participation in the context of the Water Framework Directive and Integrated River Basin Management, Copenhagen, 2001. E. Mostert, Water Policy, 2003, 5(2), 179–197. V. Nies, B. Rocha and A. Skupien, Agrar- und Umweltrecht, 2006, 4, 120–126. B. Kastens and J. Newig, Will participation foster the successful implementation of the WFD? The case of agricultural groundwater protection in North-West Germany, Local Environ., accepted for publication. E. Mostert, Participation for sustainable water management, in Sustainable Management of Water Resources: An Integrated Approach, ed. G. Giupponi, A. J. Jakeman, D. Karssenberg and M. Hare, The Fondazione Eni Enrico Mattei Series on Economics, the Environment and Sustainable Development, Cheltenham, UK, 2006, p. 171. D. Ridder and C. Pahl-Wostl, Regional Environ. Change, 2005, 5(4), 188–196. A. Ro¨hring, Die Umsetzung der EU-Wasserrahmenrichtlinie: ‘‘problems of interplay’’, in Das Flussgebiet als Handlungsraum. Institutionenwandel durch die EU-Wasserrahmenrichtlinie aus raumwissenschaftlichen Perspektiven, ed. T. Moss, Mu¨nster, 2003, p. 256. T. Moss, Induzierter Institutionenwandel ‘‘von oben’’ und die Anpassungsfa¨higkeit regionaler Institutionen. Zur Umsetzung der EUWasserrahmenrichtlinie in Deutschland, in Das Flussgebiet als Handlungsraum. Institutionenwandel durch die EU-Wasserrahmenrichtlinie aus raumwissenschaftlichen Perspektiven, ed. T. Moss, Mu¨nster, 2003, p. 158. H. Gardunõ, M. Nanni, A. Tuinhof, K. Kemper and C. Dumars, Stakeholder participation in groundwater management: mobilizing and


52. 53.

54.

55. 56.

57.

58. 59. 60.

173

sustaining aquifer management organization, in Sustainable Groundwater Management: Concepts and Tool, Briefing Note 6, World Bank, Washington, DC, 2003. J. Tippet, B. Searle, C. Pahl-Wostl and Y. Rees, Environ. Sci. Pol., 2005, 8(3), 287–299. P. Maurel (ed.), Public Participation and the European Water Framework Directive. Role of Information and Communication Tools, Work Package 3, Report of the HarmoniCOP Project, 2003 (www.harmonicop.info). N. Kranz, M. Patel and D. Ridder, Public Participation in European River Basin Management. Lessons from the HarmoniCOP Project, Ecologic, Briefs Ecologic, Berlin, 2006. A. Maslow, Motivation and Personality, Harper & Row, New York, 1954. Y. Rees, B. Searle, T. Tippett and A. Johannessen, Good European Practices for Stakeholder Involvement: Lessons from Real Planning Processes, Work Package 5, report of the HarmoniCOP project, 2005 (www.harmonicop.info). C. Aue and K. Klaassen (coord.), Sustainable Groundwater Management: Handbook of Best Practice to Reduce Agricultural Impacts on Groundwater Quality (Water4All-Project), Aalborg, Denmark, 2005. F. Haakh, Anforderungen an den Gewa¨sserschutz aus Sicht der Wasserversorgung, Wat-Berlin Tagungsband, DVGW, 2006, pp. 18–19. H. Esser, Soziologie. Spezielle Grundlagen, Bd 5, Institutionen, Frankfurt aM/New York, 2000. J. Newig, Does public participation in environmental decisions lead to improved environmental quality? Towards an analytical framework, Communication, Cooperation, Participation. Research and Practice for a Sustainable Future, 2007, 1(1), 51–71.

5. Groundwater Characterization and Risk Assessment

CHAPTER 5.1

Groundwater Characterisation and Risk Assessment in the Context of the EU Water Framework Directivew ANDREAS SCHEIDLEDER,a JOHANNES GRATHa AND PHILIPPE QUEVAUVILLERb a

Umweltbundesamt GmbH, Spittelauer Laende 5, AT-1090 Wien., Austria; European Commission, DG Environment (BU9 3/142), Rue de la Loi 200, BE-1049 Brussels, Belgium

b

5.1.1

Legal Background

One of the primary and key steps within the implementation of the European Union (EU) Water Framework Directive (WFD) is the analysis of the characteristics of groundwater bodies, the review of the environmental impact of human activity and the economic analysis of water use as it is laid down in Article 5 and specified in Annex II in order to identify groundwater bodies presenting a risk of not achieving WFD environmental objectives laid down in Article 4 of the WFD. The first analysis and review was due on 22 December 2004 and had to be reported by EU member states to the European Commission in March 2005. These analyses have to be reviewed, and, if necessary, updated at the latest in 2013 and every six years thereafter. The content of this chapter is very much inspired by the discussion within and the outcome of technical workshops of WG C ‘‘Groundwater’’ (see Chapter 4.1) on groundwater body characterisation and risk assessment1–3 during 2004–2005 which were already based on guidance documents elaborated during the first phase of the Common Implementation Strategy.4–7 w

The views expressed in this chapter are purely those of the authors and may not in any circumstances be regarded as stating an official position of the European Commission.

177

178

Chapter 5.1

The identification and delineation of groundwater bodies is the first part of the analysis of the characteristics of the river basin districts under the WFD (Article 5). Each groundwater body has to be characterised in order to assess the uses, the degree to which they are at risk of failing to meet the environmental objectives and to identify any measures to be required under Article 11. The specification for this impact review for groundwater is laid down in WFD Annex II and includes five parts (see Figure 5.1.1): initial characterisation, including identification of use, pressures and risk of failing to achieve objectives; further characterisation of groundwater bodies identified as being at risk; review of the impact of changes in groundwater levels for groundwater bodies for which lower objectives are to be set according to Article 4(5); review of the impact of human activity on groundwaters for transboundary and at risk groundwater bodies; and review of the impact of pollution on groundwater quality for which lower objectives are to be set. The most important goal of this first review was to understand the significant water management issues within each river basin and how they affect each individual water body. The timetable for completing the first pressures and impacts analyses and reporting their results was very short. The first analyses therefore relied heavily on existing information on pressures and impacts and existing assessment methods.

Figure 5.1.1

The WFD specifies requirements for impact analysis.6

Groundwater Characterisation and Risk Assessment

179

The groundwater risk assessment is part of the characterisation and the review of the environmental impacts of human activity already introduced. For each groundwater body the degree to which it is at risk of failing to meet the objectives under Article 4 has to be assessed. Ideally, a pressures and impacts assessment will be a four-step process. identifying the driving forces (especially land use, urban development, industry, agriculture and other activities which lead to pressures) without regard to their actual impacts and identifying pressures with possible impacts on the water body and on water uses; identifying the significant pressures, by considering the magnitude of the pressures and the susceptibility of the water body; assessing the impacts resulting from the pressure; and evaluating the likelihood of failing to meet the objective. To undertake the four key stages, three supporting elements must be considered (shown on the left of Figure 5.1.2). The description of a water body and its catchment area will underpin the pressures and impacts analysis. During the process, monitoring data relevant to the water body may be introduced. A comparison of monitoring data with driving forces may help to screen where pressures are likely to cause a failure in meeting objectives. It is also necessary to understand the objectives against which the actual state is compared. In many cases these key stages need not be undertaken as a linear sequence, but in general all key stages are to be addressed.

Figure 5.1.2

Key components in the analysis of pressures and impacts.6

180

5.1.2

Chapter 5.1

Groundwater Body Identification and Delineation

As the environmental objectives of the WFD must be applied to ‘‘water bodies’’ it is important to have a closer look at the identification and delineation of groundwater bodies. The application of the term ‘‘body of groundwater’’ must be understood in the context of the hierarchy of relevant definitions provided under Article 2 of the WFD wherein a ‘‘body of groundwater’’ means a distinct volume of groundwater within an aquifer or aquifers; ‘‘groundwater’’ means all water that is below the surface of the ground in the saturated zone and in direct contact with the ground or subsoil; and ‘‘aquifer’’ means a subsurface layer or layers of rock or other geological strata of sufficient porosity and permeability to allow either a significant flow of groundwater or the abstraction of significant quantities of groundwater. Based on the definitions a body of groundwater must be within an aquifer or aquifers. However, not all groundwater is necessarily within an aquifer. The WFD’s definition of aquifer requires two criteria to be considered in determining whether geological strata qualify as aquifers: a significant flow of groundwater or the abstraction of significant quantities of groundwater. If either of the criteria is met, the strata will constitute an aquifer or aquifers. The significance of groundwater flow should be understood in the context of the purpose and provisions of the WFD. A key purpose of the WFD is to prevent further deterioration of and protect and enhance the status of aquatic ecosystems, and with regard to their water needs, terrestrial ecosystems and wetlands directly depending on groundwater. The objective of protecting and restoring good groundwater status is designed to help achieve this purpose. It applies to all bodies of groundwater. Consequently, to ensure that the purpose of the directive can be achieved, the definition of significant flow must encompass all groundwater flow that is important to aquatic and terrestrial ecosystems. Geological strata that permit such flow should therefore qualify as aquifers. In practice, the criteria mean that nearly all groundwater in the Community would be expected to be within aquifers (Figure 5.1.3). The WFD leaves flexibility to the member states for the delineation of groundwater bodies to adopt the most effective means of achieving the directive’s objectives. The delineation should take into account that the groundwater bodies can be accurately described and take regard of major differences in the status of the groundwater at different depths. This does not mean that a body of groundwater must be delineated so that it is homogeneous in terms of its natural characteristics, or the concentrations of pollutants or level alterations within it. Groundwater bodies should be delineated in three dimensions and the depth of groundwater considered should depend on the risk to fail the directive’s objectives. It seems appropriate to differentiate between shallow areas which


Figure 5.1.3

181

The WFD’s definition of aquifers.5

are more immediately affected by pressures on the surface and deeper groundwater bodies which might be affected as well but react after a certain time-lag. Nearly all member states started with the identification of geological and hydrogeological boundaries and applied a comprehensive, additional set of further criteria like vulnerability maps, subsoil properties, risk potential, utilisation and protection need, economic importance and water management aspects. The most important aim of the member states was to achieve efficient and practical inventory and management units and to keep the administrative burden and the financial efforts within practicable dimensions. Each groundwater body has to be assigned to a river basin district. The identification of groundwater bodies must be consistent and coordinated within a river basin district. In particular, the international river basin districts need to develop common approaches for the whole river basin. Groundwater bodies may be grouped for the purposes of the risk assessment, for monitoring, reporting and management purposes where monitoring of sufficient indicative or representative water bodies in the subgroups of groundwater bodies provides for an acceptable level of confidence and precision in the results of monitoring, and in particular the classification of water body status. The ability to group bodies will depend on the characteristics of the river basin district and the type and extent of pressures on it and can contribute to a cost efficient and pragmatic implementation of the directive and management of groundwater. Especially the Nordic countries Finland and Norway are

182

Chapter 5.1

confronted with a huge number of groundwater bodies due to the specific geological characteristics. However, such grouping must be undertaken on a scientific basis so that monitoring information obtained for the group provides for a suitably reliable assessment that is valid for each body in the group and for an acceptable level of confidence and precision in the results of monitoring. It should be emphasised that the identification of ‘‘groundwater bodies’’ is a tool and not an objective in itself. Groundwater bodies are the units which are used for reporting and assessing compliance with the directive’s principal environmental objectives. Member states had to identify such bodies by 22 December 2009 and where necessary verify and refine the body identification in the period before the publication of each river basin management plan (RBMP) by 22 December 2009 and then every six years. As commonly information from the characterisation process and monitoring was not available before 2004 it is most likely that member states will need to update the delineation of groundwater bodies, and therefore verification and refinement steps of groundwater body identification should be foreseen in the implementation process. However, all groundwater bodies must at least be fixed for each planning period. A key descriptor in this context is the ‘‘status’’ of those bodies. If water bodies are identified that do not permit an accurate description of their status, member states will be unable to apply the directive’s objectives correctly. At the same time, an endless subdivision of water bodies should be avoided in order to reduce administrative burden if it does not fulfil any purpose as regards the proper implementation of the directive. In addition, the aggregation of water bodies may, under certain circumstances, also help to reduce meaningless administrative burden, in particular for smaller water bodies. Finally, it has to be recognised that the objectives of preventing or limiting inputs of pollutants (Article 4.1 of WFD) and reversing any significant and sustained upward trend in the concentration of any pollutant are subject of all groundwater and not only groundwater bodies.

5.1.3

Initial Characterisation

The need for an initial characterisation of all groundwater bodies is laid down in Article 5 of the WFD and specified in Annex II. It covers the analysis of the river basin district characteristics, the identification of uses and pressures and a review of the environmental impact of human activity for assessing the degree to which the groundwater bodies are at risk of failing to meet the objectives of Article 4 of the WFD, namely the achievement of good (quantitative and chemical) status of groundwater at the latest by the end of 2015. Groundwater bodies may be grouped for the purposes of this initial characterisation. Based on the delineation of groundwater bodies, pressures to which the groundwater bodies or groups of bodies are liable to be subjected need to be identified (including diffuse and point sources of pollution, abstraction and


183

artificial recharge). In addition, the general character of the overlying strata in the catchment from which the groundwater body receives its recharge shall be described, as well as the groundwater bodies for which there are directly dependent surface water ecosystems or terrestrial ecosystems. The initial characterisation requires a general analysis of pressures corresponding to that described above, but set in the context of evaluating the risk of failing to meet the objectives. This requires an understanding of the nature of the impact that may result from a pressure, and appropriate methods to monitor or assess the relationship between impact and pressure as it is described by the conceptual model. Where no monitoring data for a groundwater body are available, the likely presence or absence of pressures and impacts should be considered when making a decision of the likely status of the groundwater body. Where it is clear from monitoring data that the groundwater body is ‘‘at risk’’, or where there are inadequate data to make a decision with reasonable confidence that a groundwater body is ‘‘at risk’’, the process should continue to further characterisation.

5.1.4

Further Characterisation

Following the initial characterisation, a further characterisation has to be carried out for those groundwater bodies or groups of bodies which have been identified as being at risk in order to establish a more precise assessment of the significance of such risk and identify any measures to be required under Article 11 of the WFD. The approach recommended follows that outlined for the initial characterisation, but requires the collection of more detailed information and data, such as that detailed in Annex II 2.3, e.g. geological and hydrogeological characteristics, the characteristics of the superficial deposits and soils, stratification characteristics in the groundwater, an inventory of associated surface systems including terrestrial ecosystems and bodies of surface water, with which the groundwater body is dynamically linked, estimates of the directions and rates of exchanges of water between the groundwater body and associated surface systems, long-term annual average rate of overall recharge and characterisation of the chemical composition of the groundwater, including specification of the contributions from human activity. The wording of Annex II suggests that the information specified shall be included ‘‘where relevant’’. In this context ‘‘relevant’’ is taken to mean relevant to the assessment of risk of failure to meet the environmental objectives of the WFD. This does not give license to neglect collecting information. ‘‘Relevance’’ also involves questions of the level of detail that should be sought and, for human activities, the timescale over which the effects of the activity may be deemed relevant. In deciding these matters it is important to refer back to the purpose of further characterisation: to improve the assessment of risk and identify any measures to be required under Article 11.

184

Chapter 5.1

5.1.5

Additional Requirements of the WFD

5.1.5.1

Transboundary Groundwater Bodies

Within the characterisation process specific provisions concern those groundwater bodies that are transboundary and cross the boundary between two or more member states, focusing mainly on quantitative aspects. Information is requested on, for example, the location of groundwater abstraction points serving more than 10 m3 a day or more than 50 persons, the abstraction rates and direct discharges to groundwater and the chemical composition of water abstracted and recharged. First experiences show that the way to deal with transboundary groundwater bodies is not yet fully clarified. In many cases considerable, time-consuming discussion, cooperation and commitment between member states sharing a common transboundary groundwater body are still needed in order to characterise the overall groundwater body, review the impacts and assess the risk of failing to achieve the objectives of the WFD in a harmonised way.

5.1.5.2

Groundwater Bodies with Lower Objectives

Connected to the further characterisation, the WFD also requires the identification of those bodies of groundwater for which lower objectives are to be specified under Article 4 where, as a result of the impact of human activity, and as determined in accordance with the analysis of pressures and impacts, the body of groundwater is so polluted that achieving good groundwater chemical status is infeasible or disproportionately expensive.

5.1.5.3

Interaction with Aquatic and Terrestrial Ecosystems

Important aspects to be considered for the characterisation of groundwater bodies are the interactions with associated surface waters and terrestrial ecosystems. Indeed, the definition of good groundwater status implies that the concentrations of pollutants and directions and rates of water exchange in a defined groundwater body should not result in failure to achieve the environmental objectives under Article 4 of the WFD for associated surface waters nor any significant diminution of the ecological or chemical quality of such bodies nor in any significant damage to terrestrial ecosystems which depend directly on the groundwater body.

5.1.6

Conceptual Model/Understanding

Assessing the impacts on a water body requires some quantitative information to describe the state of the water body itself and/or the pressures acting on it. This assessment requires a conceptual understanding of what happens in a groundwater body and what causes impacts.


185

Conceptual understandings or conceptual models are simplified representations, or working descriptions, of the hydrogeological system being investigated and reflect the understanding of how the real hydrogeological system of a groundwater body, its processes and reactions are believed to behave. It is very much a set of working hypotheses and assumptions, based on evidence and should be written down so that it can be tested. The development of a conceptual understanding is mainly based on the work carried out as part of the characterisation process and refined by additional information and monitoring data gathered during the following implementation procedure of the WFD. As the amount of, and confidence in, available environmental information increases, the accuracy and complexity of the model/understanding improves, so that it becomes a more effective and reliable description of the system. The level of refinement needed in a model is proportionate to (a) the difficulty in making the assessments or predictions required, and (b) the potential consequences of errors in those assessments. A conceptual model/understanding is furthermore necessary to design monitoring programmes and it is necessary to interpret the data provided by those programmes, and hence assess the achievement of the directive’s objectives. The testing of conceptual models/understandings is important to ensure they provide for acceptable levels of confidence in the assessments they enable. A successful pressure and impact assessment will not be one that follows prescriptive guidance. It will be a procedure in which there is proper understanding of the objectives, a good description of the water body and its catchment area (including monitoring data) and a knowledge of how the catchment system functions. A conceptual understanding/model is dynamic, evolving with time as new data and information are obtained and as the model is tested. Its development and refinement should adopt an iterative approach (see Figure 5.1.4). The approach therefore fits in well with the various levels of knowledge required at different stages of the WFD. For example a basic model will be appropriate for initial characterisation; this (if appropriate) will be refined and improved during further characterisation, and again during the review cycle of the RBMP. The

Figure 5.1.4

Refinement of conceptual model/understanding.7

186

Chapter 5.1

drafting of basic conceptual models of groundwater flow and chemical systems and then of groundwater bodies should be undertaken early in the process of initial groundwater characterisation. This will include the delineation of the groundwater body boundaries and an initial understanding of the nature of the flow and geochemical system, interaction with surface water bodies and terrestrial ecosystems and an early assessment of pressures.

5.1.7

Identification of Driving Forces and Pressures

A common understanding of terms and the most effective approach for groundwater risk assessment was developed within the IMPRESS working group during the first phase of the Common Implementation Strategy.6 The widely used driver, pressure, state, impact, response (DPSIR) analytical framework had been adopted with definitions as in Table 5.1.1. It is worth noting in the context of the DPSIR framework as described above that objectives defined by the WFD relate to both the state and the impact, since standards from other European water quality objective legislation relate to the concentration of pollutants in the water body (i.e. its state), while the biological elements of the WFD clearly indicate impacts. Driving forces are sectors of activities that may produce a series of pressures. A pressure results from an activity that may directly cause deterioration in the status of a water body. In most cases a pollution pressure relates to the addition or release of substances into the environment. This can be the discharge of a waste product, but may also be the side effect or by-product of other activities, such as leaching of nutrients from agricultural land. A pollution pressure may also be caused by an action such as a change in land use. The most usual categorisation of pollution pressures is to distinguish between diffuse and point sources. However, the distinction between point and diffuse sources is not always clear, and may again relate to spatial scale. For example, areas of contaminated land might be considered as either diffuse or point sources of pollution. A quantitative pressure relates to the change of Table 5.1.1

The DPSIR framework as used in the pressures and impacts analysis.

Term

Definition

Driver

An anthropogenic activity that may have an environmental effect (e.g. agriculture, industry) The direct effect of the driver (e.g. an effect that causes a change in flow or a change in the water chemistry) The condition of the water body resulting from both natural and anthropogenic factors (i.e. physical and chemical characteristics) The environmental effect of the pressure (e.g. ecosystem modified) The measures taken to improve the state of the water body (e.g. restricting abstraction, limiting point-source discharges, developing best practice guidance for agriculture)

Pressure State Impact Response


187

groundwater levels or the modification of flow directions but also to the intrusion of salinity, the reduced dilution of chemical fluxes or the modification of dependent aquatic or terrestrial ecosystems. Such pressures can be changes in land use like land sealing, water abstraction or artificial recharge. The pressures and impacts analyses should be focused in such a way that the effort involved in assessing whether any groundwater body, or group of bodies, is at risk of failing to achieve its environmental objectives is proportionate to the difficulties involved in making that judgement. A screening approach helps simplifying the tasks prior to additional description and analysis at a later stage, as it means focusing on the search for pressures on those areas and pressure types that are likely to prevent meeting the objectives and pointing out with simple assessments those water bodies that are clearly ‘‘at risk’’ or not at risk in failing to achieve good status in 2015. The screening approach may be carried out using driving force assessment as substitute of pressures. Driving forces are quantified by aggregated data, simple to obtain, e.g. hectares of arable land, population density per area. This screening should identify issues to be addressed in the drawing up of the RBMP, and it may also reveal a number of gaps in data or knowledge that should be filled during the process of drawing up the RBMP and the monitoring programme. A list of pressures and the assessment of impacts on a water body shall ensure the identification of all of the potentially important problems. Assessing the likely impacts arising from each of the pressures will produce a list that can be used to identify points where monitoring is necessary to better understand if the water body is at risk of failing to achieve good status. The screening procedure is not only a way to accelerate data collection by focusing on those pressures that are reasonably expected, it provides an independent assessment of pressures and impact relationships, which is valuable especially if emission and abstraction registers are poorly populated. Clearly the use of GIS facilitates this process.

5.1.8

Identification of Significant Pressures

The inventory of pressures is likely to contain many that have no or little impact on the groundwater body. The assessment of whether a pressure on a water body is significant must be based on the knowledge of the pressures within the catchment area together with a conceptual understanding of the groundwater bodies in the context of the receptors. This understanding coupled with the list of all pressures and the particular characteristics of the catchment makes it possible to identify the significant pressures. However this approach often requires two steps. In the first step, correlation assessment can be carried out. This has the advantage of using monitored data and does not require complex hypotheses. In the second step, the conceptual understanding is embodied in a set of simple rules indicating directly whether a pressure is significant or not.

188

Chapter 5.1

One approach is to compare the magnitude of the pressure with a criterion or threshold relevant to the water body type but this must take account of the particular characteristics of each water body and its vulnerability to the pressure. This approach effectively combines the pressure identification with the impact analysis, since, if any threshold is exceeded, the water body is assessed as likely to fail its objectives. To improve confidence, the estimates of the type and magnitude of pressures should be crosschecked, where possible, with monitoring data and with information on the key drivers for the pressures.

5.1.9

Assessing the Impacts of Pressures

Within the initial characterisation the concept of ‘‘potential impact’’ could be introduced to describe the effects a pressure is likely to have on a groundwater body, and that ‘‘potential impact’’ is used in the evaluation of whether the body is ‘‘at risk’’. This concept recognises that, with the constraints on the characterisation process, it will not always be possible to accurately measure the impact by monitoring groundwater levels and quality. For pollution pressures the potential impact is judged by considering the pollution pressure (where this occurs at the surface) in combination with the vulnerability of the groundwater body to pollution. Thus, for example, a high pollution pressure caused by anthropogenic activities at the surface may have little impact on a groundwater body if that body is protected by a significant thickness of low permeable layers. Within the further characterisation a review of the impact of human activity for groundwater bodies characterised to be ‘‘at risk’’ and for those crossing member state boundaries is required. Assessing the impacts on a water body requires some quantitative information to describe the risk status of the groundwater body and/or the pressures acting on it. The type of analysis will depend on what data are available. Regardless of the particular process to be adopted, and as with the identification of significant pressures described above, the assessment requires a conceptual understanding of what causes impacts. In many cases a simple approach might be absolutely suitable for assessing the impact of a pressure. However, there will be a vast range of catchment types, aquifer types, interacting pressures, process conceptualisations, data requirements and possible impacts, and adopting such a simple model for all cases might not be appropriate. Tools which might assist assessing the impacts comprise the use of observed data to assess and to refine the assessment, a conceptual model, the use of analogue water bodies and the use of numeric models.

5.1.9.1

Tools to Assist

A pressure checklist can be considered as a reminder of the driving forces and the pressures that should be considered and therefore represents a precursor to the actual pressures and impacts analysis.


189

If monitoring data are available for the groundwater body, it might be possible to perform a direct assessment of the impact. However, it must be kept in mind that most pressures do not create a clear-cut impact. Monitoring data may indicate that there are no current impacts. This information itself reveals that none of the pressures identified in the initial screening process is significant, or that the time lag required for a pressure to give rise to an impact has not yet passed. The latter is likely to be of particular importance when assessing groundwater bodies in which pollutants travel very slowly. A conceptual model/understanding of the flow system, chemical variations and the interaction between groundwater and surface ecosystems is essential for characterisation and for assessing the impacts of pressures. A significant strength of the approach is that it allows a wide variety of data types (including, for example, physical, biological and chemical data) to be integrated into a coherent understanding of the system. As new data are obtained they help to refine or adjust the model; conversely the model may indicate errors and inadequacies in the data. A further step could be a shift to mathematical models. In general the more complex the model, the greater the data requirements and the greater the time and costs needed to improve it; therefore such an additional effort seems to be appropriate where water bodies appear to be at risk, or where a detailed programme of measures needs to be developed only. However, in the context of groundwater body characterisation under the WFD there are many questions that may be answered adequately with a simple model. In situations with no observed data, a possible tool to evaluate status is to compare with a similar analogous groundwater body for which data are available, and to assume that the assessment made from the observed data can be applied validly to both bodies. Furthermore, the WFD offers the possibility of grouping water bodies for the purpose of pressure and impact analysis and monitoring. Groundwater vulnerability maps or indices are useful tools for assessing the likely impact of pollution pressures during the characterisation process. By taking account of a range of factors the susceptibility or vulnerability of groundwater to pollution from pollution pressure on the land surface can be ranked. Groundwater vulnerability maps can be used as a screening tool to rapidly assess the relative scale of impacts arising from pressures. They may be useful for assessing whether groundwater bodies are ‘‘at risk’’ from pollution sources during the initial characterisation. Groundwater vulnerability assessments may be combined with models of diffuse pollution source behaviour, to consider the overall risks to water quality on a groundwater body scale.

5.1.9.2

Scaling Issues

Different kinds of pressures do not impact the different water bodies at the same time and space scales. It is important to adopt appropriate temporal scales in the pressures and impacts analysis since some pressures may result in impacts many years in the future, and some future impacts will relate to past

190

Chapter 5.1

pressures that no longer exist. For example, pesticide application may lead to increased concentrations of the pesticide in the groundwater many years after it was released. When a groundwater body currently has good status but it is thought that pressures may cause its status to be rendered poor by 2015, then the body is ‘‘at risk’’ and will require further characterisation. It should be noted that a body currently determined to have poor status will automatically be ‘‘at risk’’. Regarding spatial scales, it is important to consider the location of where the data are gathered from and where the pressures occur, especially if the groundwater body consists of, for example, a separate recharge area and an area of confined groundwater that responds differently to the pressure. For example, considering confined groundwater, the relevant pressure data are those on the recharge area only, not over the total extent of the water body. The WFD does not differentiate between groundwater in different strata: all groundwater requires the same degree of protection from pollution. However, the impact that a pollution pressure is likely to have on groundwater varies from site to site, depending on the hydrogeological properties of the aquifer and geological strata. Consequently, for a given pollution pressure, the impact on the status of a groundwater body, and the potential programme of measures will vary in different aquifers.

5.1.10

Evaluating the Likelihood of Failing to Meet the Objectives

Evaluating the risk of failing to meet the objectives should have been a straightforward comparison of monitored data with the provisions of good status as laid down in the WFD. The first pressure and impact assessment had to be completed by the end of 2004. However, specifications required to meet most of the objectives of the WFD had not been firmly defined by this date as Article 17 requested the European Commission to come up with a ‘‘daughter’’ directive laying down a criteria for assessing good groundwater chemical status and criteria for identifying significant sustained upward trends and for the definition of starting points of trend reversal. After four years of discussion, the new Groundwater Daughter Directive was finally adopted in December 2006 (see Chapter 3.1). The criteria for good chemical status are based on EU-wide quality standards, groundwater threshold values and WFD criteria, wherein the groundwater threshold values have to be developed by member states on a national, regional or local level for those substances which are causing risk of failing to meet the environmental objectives of the WFD. The groundwater threshold values have to be reported together with the draft RBMPs in 2008. The confidence and precision of the estimated environmental effects of different pressure types will also be very variable, depending to a great extent on the quality of national and local information and assessment expertise. This is because consideration of many of the pressures and impacts relevant under

191


Annex V and GWDD

Annex II requirements

Evaluating risk of

requirements

failing to meet Characterisation,

Figure 5.1.5

Defining thresholds for

The iterative evaluation of the risk of failing objectives (based on Ref. 6).

the WFD has not previously been required by other Community water legislation. Member states completed the first analyses using appropriate estimates for pressures and impacts but they had to be aware of, and had to take account of, the uncertainties in the environmental conditions required to meet the WFD’s objectives and the uncertainties in the estimated impacts. The consequence of these uncertainties is that member states’ judgements on which bodies are at risk, and which are not, are likely to contain more errors in the first report than will be the case in subsequent planning cycles. It is clear that the process of evaluating the risk of failure is to some degree an iterative collaboration between those undertaking the pressures and impact analysis, and those defining thresholds for the as yet undefined elements of status (Figure 5.1.5). Furthermore, it will be important for member states to be aware of the uncertainties so that their monitoring programmes can be designed and targeted properly in order to provide the information needed to improve the confidence in the assessments. Where the assessment contains significant uncertainty, those water bodies should be categorised as at risk of failing to meet their objectives. Obvious failing of pressures is not an uncertainty.

5.1.11

Reporting on the Characterisation and Risk Assessment

Article 15 requires member states to submit a summary report of the pressures and impact analyses to the Commission within three months of their completion (i.e. the first report had to be submitted by March 2005). This analysis has to be reviewed, and if necessary updated at the latest in 2013 and every six years thereafter. The summary reports sent to the Commission should be concise and give an overview of relevant characteristics and main water bodies within a river basin district, tables and maps showing significant pressures and water bodies at risk. Furthermore, the summary report should include methodologies, tools, threshold values, environmental quality objectives, classification schemes, etc., used

192

Chapter 5.1

within the risk assessment and an indication of the amount of uncertainty of the pressure analysis and results. All these reports provided to the Commission are publicly available on the internet. More detailed information should be available on demand for public and stakeholder consultation. The European Commission is currently developing guidance on reporting to the Commission in the form of reporting sheets for the single reporting obligations of the WFD contributing to a comparable basis for harmonisation of water management on a river basin scale between member states within international river basin districts and to provide a transparent overview of the analysis and results to communicate with government, stakeholders and the public. Moreover, the development of WISE (Water Information System for Europe) will enable electronic reporting and visualisation of reported data. The system has been publicly opened on 22 March 2007.

References 1. European Commission, Groundwater Summary Report, Technical Report on Groundwater Body Characterisation, Monitoring and Risk Assessment, 2005. 2. European Commission, Groundwater Body Characterisation, technical report, 2004. 3. European Commission, Groundwater Risk Assessment, technical report, 2004. 4. Common Implementation Strategy for the Water Framework Directive, European Communities, 2003 (ISBN 92-894-2040-5). 5. European Commission, Guidance Document no 2, Identification of Water Bodies, 2003 (ISBN 92-894-5122-X). 6. European Commission, Guidance Document no 3, Analysis of Pressures and Impacts, 2003 (ISBN 92- 894-5123-8). 7. European Commission, Guidance Document no 7, Monitoring under the Water Framework Directive, 2003 (ISBN 92-894-5127-0).

CHAPTER 5.2

Groundwater Quality Background Levels EMILIO CUSTODIOa AND MARISOL MANZANOb a

Technical University of Catalonia, Department of Geotechnical Engineering, Gran Capita`, s/n Ed D-2, ES-08034 Barcelona, Spain; b Technical University of Cartagena, Paseo Alfonso XIII, 52, ES-30203 Cartagena, Spain

5.2.1

Introduction

Groundwater forms complex, three-dimensional bodies in which recharge, flow conditions and interaction with the solid matrix are point dependent. This means that, in a given groundwater body, the chemical, radiochemical and biochemical characteristics of water vary both in space (horizontally and vertically) and slowly with time. When anthropogenic effects are added, variations may be intensified with respect to pristine conditions. Thus, the water quality of a given groundwater body cannot be represented by any set of single analytical values, and the degree of human influence cannot be established by a simple comparison to a reference list. The terms background, threshold and baseline quality values have been classically used in many scientific disciplines to try to identify anomalous concentrations with respect to what are considered as ‘‘typical’’ values. These values are critical to define water quality for a given use, and have to be defined for groundwater as a guide for protection and remediation programmes. The baseline chemical composition or baseline quality of a groundwater body may be defined, quite instinctively, as the physicochemical conditions due only to natural processes during recharge, flow and water–rock interaction. Should this be possible, any impact on groundwater quality could easily be shown by comparing actual values to baseline values. Nevertheless in practice, problems appear when defining baseline due to the common variability of the significant different chemical parameters. It is necessary to know whether a given concentration is natural, the result of hydrologic changes due to human activities or the result of introducing substances from outside into the groundwater body. 193

194

Chapter 5.2

Any adverse change in physicochemical properties is called contamination, and the substances producing this change are called contaminants. If these substances are artificially introduced in some way they can be called pollutants, and the result is pollution (see Chapter 1). These are the definitions that will be used in this chapter, although they are not universally admitted. Natural variability can be taken into account by describing the statistical distribution of values for any given parameter of interest reflecting the water’s natural quality of a groundwater body. This implies that a large enough number of unbiased measurements were made, which is not an easy task in a large, threedimensional groundwater body. For a given chemical or physical parameter, the range and distribution of values can be described by a set of statistical magnitudes like the mean, median, standard deviation, percentiles, maximum, minimum, etc., or by its full statistical distribution. In practice, this can be done only for a limited number of samples and for a small number of characteristic parameters, which may vary from case to case. Although this is not a new problem, existing experience is currently limited. However, statistical values are badly needed to correctly and effectively apply the European Water Framework Directive (WFD),1 and especially the recently adopted Groundwater Directive.2 The situation explained above was in mind when the European Union BaSeLiNe (Natural Baseline Quality in European Aquifers, EVK1-CT199900032 and EVK1-CT-2002-00527) project was elaborated. The project started in 1999 and closed in 2003.3 In this project the following definition of baseline was adopted: ‘‘groundwater quality baseline is the concentration range in water of a given present element, species or substance, derived from natural geological, biogenic or atmospheric sources.’’ Thus, chemical concentrations are considered, taking into account water– rock interaction and the natural behaviour of chemical compounds along groundwater flow lines. Both atmospheric contributions and chemical reactions are time dependent, and not all components have the same residence time in the system. As a consequence, the quality baseline of a given aquifer shows a range—sometimes a wide one—of values that varies in space and evolves slowly with time. For a given groundwater body, lithological heterogeneity and the fact that groundwater moves following more or less well-defined flow lines are the main contributors to baseline spatial variability. The main actions controlling baseline temporal variability are chemical reactions (redox, mineral solubility and surface processes such as adsorption and ionic exchange) and recharge conditions. This one involves evapo-concentration of airborne and soil-released salts, as well as solutes contributed by surface water. Groundwater pollution may clearly appear when looking at specific substances introduced as pollutants (tracers) and non-existing (to some extent) from baseline, especially those of fully artificial origin. Substances such as NO3, NH4, F, As, heavy metals, some radioisotopes (and isotope changes) may not be suitable tracers for pollution, since they may also appear naturally in groundwater under some hydrogeological conditions. The objective of the BaSeLiNe project was twofold: (i) setting scientific criteria to define groundwater quality baseline, and (ii) developing standard methods to

Groundwater Quality Background Levels

195

be applied in the European Union territory in order to guide in the fulfilment of the WFD requirements. The project, coordinated by researchers of the British Geological Survey, initially consisted of a consortium of 11 research groups of 9 European countries: Denmark, Spain, Portugal, Belgium, France, Estonia and Poland, United Kingdom and Switzerland as an associate country; later on, in 2002, they were joined by three other countries: Malta, Bulgaria and Czech Republic, which at the time were in the process of joining the European Union. The main objectives of the project were attained by means of theoretical and practical conceptual approaches, based on experience and real data from 21 aquifers of the consortium countries. It is clear that each aquifer is unique and its baseline quality depends on a particular combination of geological, hydrodynamical, climatic and soil cover characteristics. This means that results from a given aquifer cannot be used to accurately characterise any other aquifer. Nevertheless when climatic conditions, including distance to the coast, are not too different and of secondary importance, it is possible to establish some typologies, based essentially on the dominant lithology, in order to help defining baseline quality and planning analysis sampling, monitoring and study. But lithology is only one of the many factors influencing baseline quality. It becomes dominant in well-recharged aquifers by continental rainfall, but climatic conditions may dominate in semi-arid and arid areas, especially near the seashore, reducing and even overcoming the importance of lithological influence. This is a common situation in many peri-Mediterranean areas. Furthermore the residual influence of the sea on sediments may exert an important influence on groundwater quality, as happens in numerous coastal areas. Figure 5.2.1 shows how the airborne chloride deposition rate varies on continental Spain and how this is translated into groundwater chloride content due to water evapo-concentration in the soil. In this chapter some of the main methodological and conceptual results of the BaSeLiNe project are mentioned to illustrate how to determine the natural quality baseline of groundwater. This includes both the items to be considered and the preventive measures to be taken into account. The final report of the project was made available via the internet.3

5.2.2

Rationale to Establish the Groundwater Quality Baseline

The need to define and establish baseline values of groundwater quality, and the criteria to set them, are a consequence of the enforcement, over the years, of different acts and laws. In the USA, the Environmental Protection Agency (EPA) relies on the Federal Water Pollution Control Act Amendments of 1972, amended in 1977 and revised several time through the years. It is known as the Clean Water Act. Although dealing mostly with surface water, this act considers cleaning polluted groundwater in disposal sites and recognises that planning is needed to address the critical problems posed by non-point-source pollution. In the European Union, the WFD (2000) has been enacted, the main objective of which is setting the framework to protect continental, surface,

196

Chapter 5.2 10

France

3

3

1

1

0.5 10

3

1

1

1 0.5

Portugal

3

0.5

3 Airborne salts -2 -1 g·m ·yr Cl

1

3

0.5

N

1

1

3

3

3

3

0

100

200 km

France 20

10

10

5

5

10 5

20

Portugal

5

10 10

10 5

10 10

50 20

Figure 5.2.1

20

5

20 50 100

5

5

10

100

5 10

Recharge, mg/LCl-

20 20 50

N

5 100 0

100

200 km

Change of chloride content in groundwater in continental Spain. The upper panel shows atmospheric airborne chloride contribution in mg m2 yr1 and the lower panel shows the result of climatic/pedologic evapo-concentration, once discounted runoff, as reflected in chloride content (mg l1) in the upper part of the groundwater table. Chloride baseline changes conspicuously throughout the territory. (Modified from Ref. 4.)


197

transitional, coastal and ground water, in order to prevent any further environmental deterioration and to improve current status. The Groundwater Directive,2 which was adopted at the end of 2006, is aimed at establishing strategies to prevent, control and correct groundwater contamination, as stated in Article 17 of the WFD. It neither modifies nor enlarges the WFD objectives, but complements them in a way that presents some difficulties, since it is still insufficiently known by many of the officials that have to apply the provisions. The European directives are obligatory in the European Union territory, and have to be incorporated into national legislations in accordance with the subsidiarity principle. In fact, the WFD is now incorporated into countries’ national laws and water acts, and related legislation and rules have been or will be correspondingly adapted (see Chapter 3.1 for further details). In North America (USA and Canada), the term baseline applied to natural groundwater quality appears often in documents of the EPA and the United States Geological Survey (USGS), since at least the mid-1990s. But it seems that there is neither an official document giving a definition nor the criteria to establish it. In Europe, the WFD does not use explicitly the term baseline, but the expression ‘‘background levels.’’ Moreover, it mentions repeatedly what is called ‘‘quality of surface water and groundwater bodies of the different countries.’’ A ‘‘groundwater body’’ is defined as a clearly differentiated volume of groundwater inside a given aquifer or aquifer system. Furthermore, the new Groundwater Directive2 incorporates the term ‘‘threshold values,’’ meaning ‘‘a concentration limit for a pollutant in groundwater, the exceedance of which would cause a body of groundwater to be characterised as having poor chemical status.’’2 The baseline or ‘‘threshold’’ should be established in waters participating actively in the hydrological cycle, except if they are already modified by contamination. This participation may be due to natural conditions or created by human intervention, such as pumping or deep drainage. Taking into account the spatial and temporal variability of natural quality baseline composition and the current scarcity of monitoring data and knowledge on the functioning of many aquifers, the Groundwater Directive does not provide a list of quality standards to be uniformly applied in the whole European Union, though it prescribes the application of existing nitrate and pesticide norms. In this respect, norms on drinking water, which are useful to protect human health, are not necessarily adequate as environmental guidelines. However, from this derives that groundwater bodies in which the limits are overcome must be classified as water bodies in ‘‘poor chemical status.’’ Thus, a body or group of bodies of groundwater shall be considered as having good groundwater chemical status when, according to the WFD and new Groundwater Directive,1,2 the nitrate concentration does not exceed 50 mg l1 NO3 (or a lower one if established for a nitrate vulnerable zone, following Directive 91/676/EEC), and the total content of active ingredients in pesticides, including their relevant metabolites (degradation and reaction products), does not exceed 0.1 mg l1. With regard to any other polluting substances, and especially for NH4, As, Cd, Cl, Pb, Hg, SO4, trichloroethylene and

198

Chapter 5.2

tetrachloroethylene, groundwater status has to be below the threshold value set by each country.2 The member states of the European Union are encouraged to enlarge the list according to identified risks to groundwater. The rationale for the new Groundwater Directive2 mentions that the 2003 meeting of the BaSeLiNe project in Funchal, Madeira Island, stressed the difficulty of setting uniform quality standards for groundwater, and emphasised the need to consider aquifer characteristics and actual pressures from human activity. Moreover, the document establishes criteria to identify significant and sustained upward trends in pollution from human activity and to determine if there is a reversal, calling for a common methodology to test the statistical significance of these trends (Article 1 of the directive). In order to correctly monitor the possible natural baseline deterioration of the different water bodies, following the conclusions of the BaSeLiNe project, some groundwater ‘‘types’’ may be considered, according to the kind of aquifer containing the water and the concentration range of specific baseline indicators. Nevertheless, as already stated, baseline is a complex result and consequently the aquifer type is only one of the factors.

5.2.3

Methods to Establish the Natural Baseline Quality of Groundwater

To establish the baseline quality of an aquifer or groundwater body, the ideal situation is when available chemical data correspond reasonably well to areas unaffected by human activity. Generally this means pre-industrial age water. But this is not always easy or possible to get. Shallow levels of water table aquifers often contain anthropogenic components of diverse origin (acid rain, airborne pollutants, agrochemicals) and must be discarded. Multilayer aquifers present, in theory, some ideal conditions to obtain non-impacted waters from their deepest levels, provided they are not stagnant or contain saline water. However, often wells and boreholes are poorly constructed and grouted, and may produce a by-pass between exploited deep and shallow contaminated levels. This means that, sooner or later, young contaminated water may penetrate pre-industrial age water levels. Nevertheless, in some cases, there are aquifers containing pollution-free, young water, which are fully acceptable to characterise the reference natural baseline quality. In order to distinguish water of natural origin from anthropogenically impacted water, the BaSeLiNe project3 recommended the following approaches to be adopted, if applicable: (1) looking for the evidence that water age (or mean residence time) exceeds 50 to 100 years; (2) extrapolating available chemical data time series backwards, until it reaches a (theoretical) initial time in which there was no anthropogenic activity in the area; and (3) looking for substances that are clear indicators of human activity. These substances may be agrochemicals and their degradation compounds (including metabolites of pesticides), industrial products or an increase of dissolved nitrogen species or of total dissolved organic carbon. In order to identify the existence of a fraction


199

of young water with an anthropogenic influence, fully artificial substances are especially useful, such as many organic solvents, SF6, CFSs, etc. Even though the presence and the impact intensity of anthropogenic contaminants in water can be currently easily identified through modelling, this approach allows taking reasonable initial decisions on the way, provided that the significant compounds are analysed with the needed analytical sensitivity, at least in the survey stages, independently of being part of the monitoring program. One of the more frequent contamination forms is the arrival to the recharge area of a given aquifer or water body of airborne contaminants external to the zone. This complicates the determination of natural quality baseline in small, intensively exploited aquifers, which are often an essential local water resource characterised by the short turnover time. The BaSeLiNe project suggests using, as a reference for these aquifers, the natural baseline established for other aquifers under similar geological, hydrogeological, climatic, etc., characteristics, reinforced when needed with the help of hydrogeochemical modelling, and the drilling of new monitoring boreholes for sampling, when some parts under natural conditions can be expected to be found. Since the same non-impacted aquifer may contain groundwater bodies of different chemical composition (e.g. due to the presence of redox fronts, ion exchange gradients, waters of diverse marine continental origin), in practice the natural baseline quality of specific water bodies and their characteristic values should be explained by means of some main geochemical processes and the heterogeneities existing in the aquifer. In order to explain correctly a given baseline quality composition of a water body inside an aquifer, it is convenient to use ambient descriptor properties (and terms) such as ‘‘confined,’’ ‘‘water table,’’ ‘‘oxidant,’’ ‘‘reducing.’’ The different integrated tools applied in the BaSeLiNe project are commented on below. They constitute a proposal of methodology to establish the natural groundwater baseline quality, and contain the main conceptual and applied results of the project. These tools are: to study the major and trace inorganic components chemical data, in order to establish the variation range of natural baseline quality; to study the organic carbon data in order to establish the variability range of baseline quality and its usefulness as a contamination indicator; to carry out hydrogeochemical modelling, in order to identify and establish the types and characteristic times of the basic reactions controlling the baseline quality of the different aquifer types; to use tracers and dating techniques to know the time scales that control the variation ranges of the different components under consideration; to study baseline trends to know their causes and how to discriminate between those due to natural processes and those due to contamination. Adequate sampling, which is a key issue to know aquifer water quality, has to address the three-dimensional character of groundwater flow and the importance of surface processes.5,6

200

5.2.3.1

Chapter 5.2

Study of Major and Trace Inorganic Component Chemistry

Natural water quality depends on characteristic concentrations of different components, represented by the statistical mean or median of a set of values and their distribution (dispersion) around these central values. If the distribution is normal or log–normal, it is possible to define their dispersion by means of the standard deviation. However often this is not the case due to the simultaneous or correlative presence of more than one physicochemical process. In the BaSeLiNe project, in order to define the baseline quality, and after a first evaluation of available data, it was decided to adopt the median as the most characteristic value for a parameter or component, and the 2.3% and 97.7% percentiles to show the variation range. Thus, most of the studied population (95.4% for a normal distribution) is inside the range. In order to reasonably describe the possible spatial variation of baseline quality, the chemical study of an aquifer or groundwater body should be carried out by using a large enough number of groundwater samples. This implies that new samplings may be needed. Moreover, this means that historical data prior to 1985 should be evaluated before being integrated to the younger ones, since many analytical techniques were less accurate than the current ones, and detection limits were too high for some components. An approach based on simple statistical, univariant techniques is proposed to establish characteristic values and variation ranges. In the BaSeLiNe project the cumulative frequency representation of data was selected in order to identify the main processes controlling the observed distributions (Figure 5.2.2). This type of representation was already used by Davis and de Wiest7 to show the distribution of elements in fresh groundwater. Figure 5.2.3 is a simple case. Other graphical representation types, which were considered adequate to show the natural baseline, are the box or whiskey plots (Figures 5.2.4 and 5.2.5) and that of time/space evolution of concentrations. Some of the main methodological remarks are the following. For populations not following the normal or log–normal distribution, outliers may be part of groundwater composition natural baseline quality and not necessarily the result of contamination. Such are the appearance of dissolved Fe21 and Mn21 in reducing groundwater ambients, the disappearance of NO3 and SO42 in similar environments or the high concentrations of sulfate when sulfide-rich sediments are supplied with dissolved oxygen or become desaturated. Thus, a careful study of these data is needed. The baseline quality variability may be of the same order of magnitude or even larger than that produced by contamination. The natural baseline quality being established depends on available data, and it is often biased due to diverse circumstances, such as taking samples by pumping (selection of the most permeable layers) or carrying out samplings with preference for some depths, or by mixing different flow lines (Figure 5.2.6).

201


6- Part of values are below analytical sensitivity (no data) 5- Small contents are discarded (less than limit) or have disappeared 1- Normal distribution 2- Concentrations are close to chemical equilibrium with minerals 3- Upper values are solubility controlled 4- Contamination or saline water admixture 1’- Bimodal distribution

Cumulative frequency, %

90 6

2 4 5

1

3 1’

50

Median

10

Log concentration

Figure 5.2.2

Plot of cumulative frequencies of chemical parameters, showing some typical circumstances. The median is used as the regional reference level, or the value to compare different parameters. Type 1, normal distribution; type 1 0 , multimodal (bimodal as shown) distribution, both reflecting variability of recharge, water–rock interaction and turnover (residence) time in natural flow systems. Type 2, small variability due to closeness to chemical equilibria with relevant minerals (for Si, Ca, Mg, etc.). Type 3, small variability at high concentrations reflecting that mineral solubility exerts control (e.g., fluorine content due to fluorite dissolution). Type 4, large variability of high concentrations resulting from addition of contaminated or saline water to a small fraction of samples. Type 5, fast decrease of low values pointing to the preferential reduction or elimination of a component by a geochemical reaction (e.g. nitrate reduction or sulfate reduction). Type 6, low values are below a threshold due to analytical conditions. (After Ref. 3.)

As a consequence, the following guidelines are proposed to determine the baseline quality of an aquifer or water body. To exclude samples known to be contaminated (information provided by some components). To carry out samplings along a flow line and normal to it. The data used must take into account the three-dimensional distribution of water characteristics in the aquifers. This means considering the sample position with respect to the groundwater flow network. In many cases what matters is the flow configuration existing under prevalent aquifer conditions, due to the slow movement and replacement of groundwater, and not the current flow regime under disturbed conditions. This may be a major handicap for studies lacking scientific support.

202

Chapter 5.2

Figure 5.2.3

Plot of cumulative frequencies for major ions in Madrid’s basin aquifer. Fresh recharge water from the basin sides mixes with saline remnants in the central basin sediments, derived from old playa lake situations. The higher values are controlled by reactions with sedimentary silicates. (After Ref. 8.)

To compare local data to information from other areas that are surely not affected by human activity. To use times series, when existing, to detect early time conditions. To use hydrogeochemical modelling as a tool to know if the assumed and deduced processes are natural or need artificial conditions to be active. To make a limited use of statistical techniques, since hydrologic and hydrogeochemical considerations are generally not taken into account and thus not enabling key processes to be clearly identified.

5.2.3.2

Organic Component Chemistry

Organic carbon dissolved in groundwater is an important reactant in natural geochemical processes. Furthermore, it may be useful to determine the natural or uncontaminated status of a given water as a contamination indicator, e.g. of disposal sites and used waters, and may be a potential contaminant as well. It is a source of energy and food to bacterial populations, both in aquifers and in distribution networks, and also plays an important role in the mobilisation of trace metals, radionuclides and biogenic components coming from outside.


Figure 5.2.4

203

Box plot (whiskey plot) of major ion concentrations in groundwater samples from Madrid Tertiary Detrital aquifer. Groundwater is predominantly of the sodium bicarbonate type, in agreement with the arcosic nature of sediments, but there is also some saline groundwater formed in old evaporating playa lakes, of mostly the sodium–calcium sulfate type. Excess sodium is part of baseline. (After Ref. 8.)

Additionally, its presence and concentration is important for the evolution of underground redox fronts. Organic carbon is part of the organic matter present in groundwater, which is of two types: humic and non-humic substances. The latter group includes decomposable plant material, living biomass and woody plant material. Both types can be present as solved species or as particles. All natural waters contain some dissolved organic carbon (DOC), here considered—from a practical approach—as that remaining in water filtered through a 0.45 mm sieve. The total organic carbon (TOC) is that measured in unfiltered waters and includes both the dissolved and the particulate fraction (larger that 0.45 mm), such as bacteria and phytoplankton. For correctly sampled groundwater, in most cases DOC is about the same as TOC since the sieving effect takes place in the aquifer, except for coarse formations, in which even some micro-organisms could be transported. Work carried out mostly in the last decade recognises the significant role of TOC in some important hydrogeochemical processes, such as natural weathering or redox reactions, which control the evolution of some hydrogeochemical environments in aquifers. Furthermore, the particulate matter, especially that of colloidal size (most of it smaller than 0.45 mm),

Figure 5.2.5

Whiskey plots (box plots) to show median and range variation of different major and trace components and physicochemical parameters of interest for baseline quality in fresh water (A) and saline water (B) of the same aquifer (Donãna, southwestern Spain). The aquifer system has at least two hydrogeologically different water bodies. (After Ref. 9.)

204 Chapter 5.2


Figure 5.2.6

205

Schematic representation of aquifer behaviour in El Abalario, Donãna Natural Park (Huelva, Spain), showing a cross-section between El Abalario dome and La Rocina creek. The water table aquifer consists of fine–medium sands with a thin coarse layer of a much more permeable gravely formation, where wells have their screens. The flow lines (lines with arrows) have an important vertical component, which is downwards in most of the area. Nitrate contamination in irrigated fields (mostly fruit trees and strawberries under plastic cover) moves downwards in the sands, at about 0.5 m per year. Currently a large part of the sands is contaminated by high concentration of nitrate, but this still does not appear in many wells that are deep screened nor in deep discharges into the main water course, though it does in the shallow local creeks. Well water still shows baseline values but the water body is seriously damaged, and it will progressively worsen. The medium is oxidant, so nitrate is not reduced. In other areas, in which the aquifer is thinner, nitrate pollution attains the full thickness.

plays an important role in the transport, mobilisation and degradation of contaminants. Natural groundwater often has TOC concentrations less than a few milligrams per litre, although values higher than 50 mg l1 can be occasionally found. Data studies carried out within the BaSeLiNe project show that TOC concentrations decrease with increasing depth, and that differences between median values of natural water from siliceous and carbonate aquifers containing young and old water are small. The measured values vary between 0.7 and 1.8 mg l1, which are significantly smaller than those measured in clearly

206

Chapter 5.2

contaminated aquifers, in which the values may be of the order of several tens to some hundred milligrams per litre. A clear relationship between TOC and other similar parameters such as assimilable organic carbon, halogenated organic compounds or the bacterial counting has not yet been found, and the scientific knowledge about the different organic molecules and their reactivity, toxicity and ability to mobilise various contaminants is still being developed. Though early work was done more than 40 years ago,10 research intensification has occurred during the last ten years, producing excellent manuals11 and a good number of scientific papers. Published work is mostly related to biodegradation in natural remediation, paying attention to some compounds (plaguicides, halogenated organic carbons, hydrocarbons) and their metabolites, as well as selective behaviour through 13C isotope evolution.12,13 The knowledge gained in upcoming years could be relevant.

5.2.3.3

Hydrogeochemical Modelling

As already stated, groundwater natural quality is the result of complex interactions between the solid, gaseous and liquid phases. The resulting composition may be in some places of the same or of a higher order of magnitude than that due to contamination in other places. Hydrogeochemical modelling coupled to water flow is a necessary tool to get a qualitative and quantitative knowledge of the main and more frequent processes controlling groundwater quality (dissolution/precipitation, ion exchange, redox reactions, adsorption; Figures 5.2.7 and 5.2.8). Once the significant processes are known and quantified, modelling can be used to predict future water quality changes, both due to ambient

Figure 5.2.7

Interpretation of chemical logs along boreholes to show hydrogeochemical changes when going through a redox front. Data points are experimental values and the continuous curves are the result of flow and transport modelling. Values are in mmole l1 (mM). (After Ref. 3.)

207


Interpretation of measures along a flow line to show hydrogeochemical changes when there is water–rock interaction. Only Ca21 evolution is shown. Data points are experimental values and the curves are the result of flow and transport modelling, with only one process (insufficient to explain results) and with two processes. (After Ref. 3.)

Concentration

Figure 5.2.8

Extrapolated trends Modelled trend

Baseline data period

First available data

Figure 5.2.9

Today

Time

Schematic representation of trend modelling using the data period. Extrapolation backwards until anthropogenic processes are not present or a steady state is observed allows the approximate knowledge of baseline quality for the chemical parameter being considered. Extrapolation forwards allows visualisation of trends under given scenarios or hypothesis. (After Ref. 3.)

natural variations (e.g. natural climatic changes, tectonically induced flow modification, subsidence, sedimentation, erosion) and to the impact of different human activities, including the future removal of some of the current ones (Figure 5.2.9). Modelling facilitates the study of spatial and temporal trends, and helps in the correct design of expensive monitoring programmes.

208

Chapter 5.2

In the BaSeLiNe project, programme PHREEQC was used to study the processes responsible for the natural composition observed and its evolution, especially the changes occurring along a flow line in three aquifer types that illustrate frequent situations in Europe: silica-dominated sedimentary aquifers with rare carbonates; carbonate aquifers or aquifers containing abundant carbonate; and aquifers with old saline water being displaced by younger freshwater. More sophisticated approaches, although not necessarily more effective, are those simulating flow and reactive mass transport. Nevertheless a detailed insight to processes allows one to define more accurately what is happening14 or to identify controlling facts, e.g. in seawater intrusion into aquifers.15 The most significant conclusions related to modelling are the following. Many patterns and trends appearing often in natural waters, such as ion exchange gradients, are the result of processes occurring at geological time scale and are generally due to flow conditions that preceded those existing today, even if equilibrium is quickly attained. Under such conditions the supply of reactants may be the limiting factor. This means that the ‘‘time’’ parameter must be carefully used in simulation works. Independent dating measures using different tracers are desirable to limit uncertainty. When this information is not available and time is derived from a modelling exercise, the uncertainty associated must be clearly shown. Groundwater development distorts natural chemical gradients, and this is very difficult to correct and simulate. Mixing processes between old and young waters also change the original composition. Therefore, some discontinuities in the natural composition of some aquifers may be due to ‘‘age gradients,’’ while smooth changes following the groundwater head gradient show the existence of continuous processes. When an aquifer does not contain water of natural origin, modelling may yield a realistic estimation of the original concentration of some elements.

5.2.3.4

Tracers and Temporal Scales

In order to interpret water quality changes referring to variation of the natural baseline quality, the knowledge of the age – the turnover time – of groundwater in the flow system is needed, as well as the temporal scale at which the different hydrogeochemical processes explaining natural baseline chemistry occur. Under favourable conditions water age may be reasonably estimated with sufficient accuracy, but this may be often impossible, due to the unavailability of suitable ideal tracers. In order to solve these difficulties, the use of several tracers and the numerical simulation of water flow and solute transport are necessary. The dating principles are the following. It is interesting to measure the components whose concentration in water varies along time, due to known causes distinct from water–rock interaction.

209


These components are like ‘‘clocks.’’ The different known types are: – cumulative processes (3He, 4He, some chemical components, etc.) – radioactive decay processes (39Ar, 14C, 32Si, 3H, 85Kr, etc.) – variable but known incorporation to groundwater (3He, CFCs, SF6, etc.). Within the BaSeLiNe project different tracers (13C, 14C, 3He/3H, 85Kr, 39Kr, CFSs, etc.) and water molecule isotopes (18O, 2H, 3H) are used, mostly to know the groundwater age structure in the studied aquifers (Table 5.2.1). Figure 5.2.10 is an example. Age structure is the result of integrating values derived from different techniques into the hydrogeochemical and hydrodynamical processes responsible for water characteristics, age being one of them. Water age determination implies that processes other than time-dependent ones (e.g. radioactivity) have been adequately corrected to remove other effects such as dilution by dead (non-radioactive) matter or exchange with solid. This is not an easy task. Ages obtained without corrections (or with only partial corrections), namely apparent ages, are only an approximation, and sometimes a crude one. One of the main contributions of time tracers to natural baseline quality determination is the estimation of the residence (turnover) time scale of groundwater in the aquifers being studied. Before carrying out long, extensive and expensive hydrogeochemical studies, this information will help to predict if water will keep in the future its natural composition. This aspect may be shown by determining whether a young water component (3H, 85Kr, 39Ar, CFSs, SF6, etc.) is present or not, taking into account the following. If the presence of some (generally various) of these components shows that water is younger than 50 years, this means that it is potentially impacted by human activity. Therefore, it is probable that the natural background has been changed, and consequently the baseline quality. The lack of any young component guarantees that original natural conditions prevail.

Table 5.2.1

Substances (asterisk indicates radioactive) potentially useful for groundwater to date under favourable circumstances and range of years that can be dated.

Substance

Origin

Range (years)

Application

3

Natural, nuclear bombs, nuclear reactors Nuclear reactors Industrial, domestic Natural Natural, (nuclear bombs) Natural

5–50 (200)

Easy

10–50 10–50 50–2000 1000–20 000

Difficult Medium Difficult Easy

200–10 000 000

Medium

H* (tritium); 3 H*/3He 85 Kr* CFCs, SF6 39 Ar* 14 C* (radiocarbon) 4

He

210

Figure 5.2.10

Chapter 5.2

Simultaneous measurement of three isotopes to date in a north–south cross-section in Donãna National and Natural Parks, southwestern Spain. The figure shows tritium (half-life of 12.43 years) in tritium units (TU), radiocarbon (14C, half-life of 5730 years) in percent modern organic carbon (pmc), and 13C content (stable, indicating origin and behaviour of dissolved carbon in the aquifer) in deviation per mile from the PDB standard. Most samples are mixtures of waters from different depths, where relatively high tritium (young water) may coexist with relatively low radiocarbon (old water). This helps in interpreting chemical data. The measurement of 85Kr and 39Ar at some points allow improved understanding. (After Ref. 16.)

If the lack of young components indicates that water age is greater than 50 years, groundwater quality corresponds to water–rock interactions evolving through time. The knowledge of these reactions needs both hydrogeochemical modelling and dating with adequate tracers. What has been stated allows for the preliminary classification of any chemical parameter, independently of its variation range, into two groups with different residence times: (1) parameters that are measured in water with young components, and (2) parameters that are measured in water free of young components. This simplifies sampling and data treatment, in order to establish the natural baseline quality. The main result of the study of water residence time in the different aquifers considered in the BaSeLiNe project is that a unique universal technique or set of techniques does not exist, but each aquifer needs the application of a series of specific techniques, which mainly depend on the time scale of water residence time, the geochemical ambient (e.g. in redox media many of the available


211

tracers may be not useful), the economic resources and the analytical tools that can be used.

5.2.3.5

Study of Natural Baseline Trends

This activity is a result of what is required in the Groundwater Directive:2 the use of time series in order to observe the possible presence of increasing, sustained and statistically significant trends in the concentration of some chemical components, due to contamination; and monitor and detect the reversal in these trends after implementing the corresponding remediation actions. The use of statistical techniques as a principal indicator is still much discussed in hydrogeological forums. If hydrogeochemical and hydraulic techniques are not considered, there is the risk of mistaking natural increasing trends for contamination. In the framework of the BaSeLiNe project the study of historical data sets has been addressed: to know which type of time series can be expected to be found in different European Union countries; to observe and define natural baseline trends in order to obtain chemical support to understand the natural functioning of aquifers; and based on what has been said above, to discriminate natural changes from those due to anthropogenic activity. Only a few countries have good time quality series on groundwater quality, except for some special aquifers. Length is often less than 15 years, but some Eastern Europe countries have the longest series, up to 70 years. The study of these series has allowed one to distinguish two types of spatial and time trends. 1. Trends of natural origin: Due to processes that cause changes at the aquifer scale. They depend on solute transport velocities through the medium, and therefore are very slow, such as the replacement of saline water by recently recharged freshwater. Due to small-scale space variability caused by aquifer heterogeneity, also of small scale. This causes fluctuations around some level that may be erroneously interpreted, if only a statistical approach is used, as the result of contamination processes (increasing trends) or as reversals of them (decreasing trends following other increasing tendencies). 2. Trends due to aquifer development, which have their only cause in natural processes and not in human contamination. The cause is the flow

212

Chapter 5.2

velocity increase due to groundwater development, thus accelerating the appearance of natural trends that otherwise would require much longer times to be observed. Groundwater development may also induce natural chemical reactions that otherwise would not be produced if abstraction did not exist. This is the case of water mixing by upconing or by changes in the redox state, pH, etc., related to water table oscillation or the displacement of water bodies of different characteristics. The former trends are related to hydrodynamic and hydrogeochemical conditions, and have to be interpreted and studied before trying to interpret the latter ones. Figure 5.2.11 shows a simple example and Figure 5.2.12 shows how different chemical parameters correlate with chloride, indicating processes that may be present in trends. The main methodological and conceptual conclusions of the study of chemical trends are the following. In order to differentiate a natural trend from an evolution due to contamination, the origin of the observed trends has to be understood.

Figure 5.2.11

Chemical trends. The hydrodynamic and hydrogeochemical study is the key to know if there is a trend and if its causes are natural or due to contamination. Time evolution of Ca21 in a pumping well with trend and time series of Mn21 in a pumping well without trend. Central trends are also shown. (After Ref. 3.)


Figure 5.2.12

213

Plot showing correlations of some ions versus chloride content. It shows the existence of geochemical processes that may not be obvious from the time series plot.

The main processes producing changes in groundwater quality trends are – changes in the mixing proportions between different waters; this happens in the salinisation by lateral flow or saline upconing or when substituting connate saline water with freshwater; – changes of redox conditions and fronts, such as the occurrence of pyrite oxidation and the displacement of nitrate, sulfate or iron reduction fronts; and – changes in recharge water composition, such as those due to variations of recharge rate or rainfall composition. In order to confirm the presence of water quality trends, be they natural or of anthropogenic origin, historical series are needed. When they exist, trends have to be interpreted. Going backwards, the original composition of groundwater can sometimes be determined, which is very important for chemical data interpretation. It is recommended that all new drilled wells or boreholes in a given aquifer be adequately sampled, in order to keep this information as an initial reference in databases. In the same way, the measurement of initial chemical characteristics will allow future studies to benefit from a highly valued initial reference

214

Chapter 5.2

information to study the natural trend, and also for numerical modelling and for aquifer management. In order to monitor groundwater quality trends in a given aquifer or groundwater body, firstly the different natural trends have to be identified.

5.2.4

Conclusions

The relatively recent enforcement of the European Union Water Framework Directive,1 and the enforcement of the newly adopted Groundwater Directive,2 with their strict requirements of member states with respect to the quality status characterisation of different groundwater bodies and the implementation of remediation programmes, need a definition of baseline quality with two clear objectives: to distinguish between natural quality and quality modified by the presence of anthropogenic components; and to establish the characteristic natural composition of the different aquifers or groundwater bodies to serve as a reference, in case of implementing remediation activities. In the European Union project BaSeLiNe,3 carried out between 1999 and 2003, a definition of natural baseline was established, with scientific support, being at the same time usable by managers and policy-makers. Furthermore, a methodology to establish the natural baseline quality and its origin was proposed and tested, appearing to be adequate for application to all Europe. This methodology has been summarily presented above, as well as the main applied and conceptual conclusions derived from the application to 21 European aquifers. As a summary, the main conclusions are the following. The interpretation of observed groundwater chemical composition should always be made taking into account the aquifer hydrodynamical functioning. This is the basis to establish natural baseline quality or to deduce which natural or contamination-induced geochemical processes are acting, as well as to detect trends and their possible initial and final time. Since the functioning of most aquifers is poorly known, this introduces the need, supported by water law requirements, of carrying out the studies and undertakings adequate to get a basic knowledge in the upcoming 10–15 years. Among these works are (a) the drilling of some new boreholes and their adequate and complete study by means of geological, geophysical, geochemical and hydrodynamical logs, (b) the start of time quality series to monitor baseline quality and to detect any change since the beginning and (c) the use of numerical models, not necessarily sophisticated ones, in order to (i) complete and reinforce


215

aquifer functioning conceptual models, including water quality origin, (ii) know what is naturally or artificially producing the modifications, (iii) forecast the foreseeable future evolution and (iv) establish the main characteristics of natural baseline quality when it is not reflected in the water currently in the aquifer.

Acknowledgements Most ideas and results reflected in this chapter derive from the joint work of the European Union research project BaSeLiNe (natural baseline quality in European aquifers: a basis for aquifer management, EVK–1–CT1999–0006), in which more than 30 scientists and experts—among them the authors of this chapter—worked during the period 1999–2003. The project results were considered in drafting the text of the European Daughter Directive on Groundwater. A synthesis of the work is available at the project site.3 Due credit should be given to the team responsible for each of the work packages, from whom many of the ideas expressed in this chapter have been borrowed. Some figures come also from the project papers. Besides the documents available through the web, a book is being prepared, and is well advanced, co-edited by Mike Edmunds and Paul Shand, managers of the BaSeLiNe project.17

References 1. Directive 2000/60/EC of the European Parliament and of the Council establishing a framework for the Community action in the field of water policy (EU Water Framework Directive), Official Journal, OJ.L.327, 22-12-2000, 2000. 2. Directive of the European Parliament and of the Council on the protection of groundwater against pollution and deterioration, OJL 372, 12.12.2006. 3. Natural BaSeLiNe quality in European aquifers: a basis for aquifer management, 2003 (www.bgs.ac.uk/hydrogeolgy/baseline/europe/EU_Baseline.pdf ). 4. F. G. Alcala´, Recarga a los acuı´ feros espanõles mediante balance hidrogeoquı´ mico [Recharge to Spanish aquifers by means of hydrogeochemical balance], doctoral thesis, Technical University of Catalonia, Barcelona, 2005. 5. L. G. Everett, Groundwater Monitoring, Genium, Schnectady, NY, 1987. 6. J. R. Boulding, Practical Handbook of Soil Vadose Zone and Ground-water Contamination, Boca Raton FL, Lewis, 1995. 7. S. N. Davis and R. J. M. de Wiest, Hydrogeology, John Wiley, New York, 1966. 8. M. A. Hernańdez-Garcı´ a and E. Custodio, Environ. Geol., 2004, 46, 173–188. 9. M. Manzano and E. Custodio, Groundwater baseline chemistry in the Donãna aquifer (SW Spain) and geochemical controls, 4th Assembleia Luso-Espanhola de Geodesia e Geofı´ sica, Figueira de Foz, Resumos, 2004, S13.7, pp. 729–730.

216

Chapter 5.2

10. M. A. Messineva, in: Geologic Activity of Microorganisms, ed. S. I. Kuznetsov, Trans. Inst. Microbiol. IX. Consultants Bureau, New York, 1962. pp. 6–24. 11. F. H. Chapelle, Groundwater Microbiology and Geochemistry, John Wiley, New York, 2001. 12. J. T. Wilson, R. Kolhatkar, T. Kuder, P. Philp and S. J. Daugherty, Ground Water Monit. Remed., 2005, 25, 108–116. 13. L. G. Kennedy, J. W. Everett and J. Gonzales, J. Cont. Hydrol., 2006, 83, 221–236. 14. T. Xu, J. Samper, C. Ayora, M. Manzano and E. Custodio, J. Hydrol., 1999, 214, 144–164. 15. M. Rezaei, E. Sanz, E. Raeisi, C. Ayora, E. Va´zquez-Sun˜e´ and J. Carrera, J. Hydrol., 2005, 311, 282–298. 16. M. Manzano, E. Custodio and M. Colomines, El fondo hidroquı´ mico natural del acuı´ fero de Donãna (SO Espanã) [Natural hydrochemical baseline of the Donãna aquifer, southwestern Spain], 5th Congreso Ibe´rico de Geoquı´ mica/9th Congreso de Geoquı´ mica de Espanã, Soria, 2005, pp. 1–13. 17. W. M. Edmunds and P. Shand, The Natural Baseline Quality of Groundwater, Blackwell, Oxford, in press.

CHAPTER 5.3

Groundwater Age and Quality KLAUS HINSBY,a ROLAND PURTSCHERTb AND W. MIKE EDMUNDSc a

Geological Survey of Denmark and Greenland, GEUS, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark; b Climate and Environmental Physics, Physics Institute, University of Bern, Sidlerstrasse 5, CH-3012 Bern, Switzerland; c Oxford Centre for Water Research, Oxford University Centre for the Environment, South Parks Road, Oxford OX1 3QY, UK

5.3.1

Introduction

The pressures on groundwater quality and quantity have increased dramatically during the past 50 years due to increasing demands for freshwater and contamination from a wide range of human activities. Before 1950 the human impact on groundwater quality was insignificant or limited, and the groundwater composition was in practice close to the natural background in most aquifers,1,2 Since about 1950 increasing contents of contaminants such as nitrate and pesticides have been found in groundwater and ecosystems globally.2–4 In the same period an increasing number of pollution plumes from point sources appear below urban and industrial areas in most parts of the world. The increasing pressure on groundwater quality and quantity leads to an increasing pressure on both water resources and dependent ecosystems and an increasing need for efficient tools for the development of a sustainable integrated water management and policy. The European water framework and groundwater directives (see Chapter 3.1) provide the framework for developing such tools, e.g. as exemplified by the efforts of estimating the natural background quality of European groundwater and nutrient mobility within European river basins.1,5 Groundwater dating by environmental tracers and numerical groundwater flow models are important tools for understanding the temporal and spatial hydrochemical evolution and the pressures of pollutants on groundwater and dependent ecosystems. Figure 5.3.1 illustrates trends of important environmental tracers in the atmosphere, which can be used to estimate groundwater ages, evaluate the temporal hydrochemical evolution of groundwater and indicate human impact. 217

218

Chapter 5.3 CFC, SF6 (pptv), 14C (pmc)

85

85

600

Kr (mBq m-3)

3

H (TU)

6000

Kr CFC-12 1200

500

5000 3

400

1000

SF6 x100

H

4000 800

300

3000 CFC-11

200

600 400

2000

14

C

100

200

1000

CFC-113 0 1940

Figure 5.3.1

1950

1960

1970

1980 Year

1990

2000

0 2010

Concentration trends of selected environmental tracers in the atmosphere applicable for groundwater dating (modified after Hinsby et al.2).

Important factors controlling both groundwater quality and quantity issues and the effect on dependent ecosystems include how fast groundwater is recharged, how fast it flows and how it interacts with the aquifer sediments and rocks. The computation of these parameters requires detailed geological information, which is most commonly not available for the subsurface. The content of environmental tracers in groundwater, and groundwater age estimation of water sampled from wells or monitoring points provide valuable information on travel times and the risk of contamination at these points. A simple evaluation of the existence of a modern water component in a water sample from a monitoring point or well by measuring the contents of one of more of the tracers shown in Figure 5.3.1 gives a first indication of possible contamination at this point.2,6 Considerable research efforts during the past decade have demonstrated more advanced applications of new environmental tracers and groundwater modelling techniques. The new techniques provide more detailed understanding of subsurface flow systems including the possibility of estimating absolute groundwater ages, history and fate of groundwater contaminants and the interaction with dependent ecosystems (see Chapter 10.2).7–10 The quality and the quantity of the subsurface water resources and their impact on dependent ecosystems are closely linked to the age of groundwater.11,12 The aquifers have a considerable attenuation potential for most contaminants,13,14 and the risk of pollution therefore decreases with increasing groundwater age along flow paths. Sustainable management of the water resources and the dependent ecosystems therefore requires a solid understanding of the groundwater age

Groundwater Age and Quality

219

distribution, the temporal and spatial evolution of the subsurface hydrochemistry and the groundwater/surface water interaction.10–19

5.3.2

Groundwater Age Estimation

There are basically two different ways of estimating groundwater age at a groundwater well or monitoring point: (1) by environmental tracers and geoindicators and (2) by groundwater flow modelling. This chapter focuses on the application of environmental tracers and geoindicators as groundwater dating tools, while groundwater flow modelling is only briefly discussed.

5.3.2.1

The Definition of Groundwater Age

Groundwater age is generally considered as the average travel time for a water parcel from either the surface or from the water table (point of recharge) to a given point in the aquifer. In humid sandy areas with thin (o5–10 m) unsaturated zones the difference between these is generally negligible. In arid areas with thick unsaturated zones the difference may be considerable. For European conditions the difference is generally believed to be of less importance and in this chapter we therefore use the term groundwater age to cover both situations. The term ‘‘residence time’’ was originally defined differently, but we prefer to use it synonymously with ‘‘age’’ as this is commonly done at present, and as this concept is of more use in groundwater studies. Hence, the residence time of groundwater is here defined as the average travel time between the point of recharge and the point of discharge, e.g. to a river or a lake or to any monitoring point in the groundwater zone. The tracer age estimate is normally considered and described as the average age of the water sample. This is a good approximation in cases where the flow system is simple, and can be approximated by a piston flow model (insignificant mixing and dispersion19). However, where significant mixing and dispersion occur, e.g. in long screens in water supply wells, in fractured dual porosity aquifers and in groundwater bodies with significant aquitards, the estimated tracer model age may either underestimate or overestimate the actual mean age of the water parcel.20,21 A sound knowledge of the geological setting and both physical and chemical processes in the aquifers is therefore important for the right interpretation and application of the environmental tracers and computed groundwater ages.

5.3.2.2

Environmental Tracers for Absolute Age Estimation

All environmental tracer dating methods are based on chemical or isotopic concentration variations as function of time. The time resolution of a groundwater dating method depends on two fundamental factors: the accuracy of the functional relation between a measured concentration and the groundwater age; and

220

Chapter 5.3

the temporal gradient of the tracer concentration in relation to the analytical precision of the measurements. Varying tracer concentrations are groundwater is either the result of a changing input concentration or sinks and sources of tracer in the subsurface. For dating the input history and the accumulation or decay rates have to be known. Many natural tracers are produced in the atmosphere due to nuclear reactions with cosmic rays (examples are 39Ar, 14C or 81Kr). A constant atmospheric equilibrium concentration will be achieved if the production rate is not changing. Dissolved in the precipitation these isotopes enter the subsurface and are subject to radioactive decay. The radioactive ‘‘clock’’ is in the ideal case the only process changing the tracer concentration in the subsurface. This is very often true for noble gases and why they are preferred tracers despite the high analytical requirements. On the other hand, chemically reactive tracers are easier to measure, but they can be affected by dissolution or degradation processes, which are difficult to quantify. The use of 14C for example is therefore complicated for groundwater dating in some settings. During the past 50 years, human activities have released a number of chemical and isotopic substances into the atmosphere. In the atmosphere they have mixed and spread worldwide. The monitoring of the atmospheric concentrations of these substances provides an ideal input function for groundwater dating (Figures 5.3.1 and 5.3.2). 3H was introduced in the atmosphere as a result of nuclear bomb tests, chlorofluorocarbons (CFCs) were intensively used in refrigerators and air-conditioning, 85Kr is released during reprocessing of fuel rods from nuclear reactors and SF6 is mainly used as an electrical insulator in high-voltage switches.23 All these tracers are gases, except 3H, which is usually part of a water molecule. The initial amount of tracer dissolved in the water depends therefore also on the recharge conditions namely the recharge temperature and pressure as well as the entrapment of excess air.24 3H and 85Kr decay with half-lives of 12.32 and 10.76 years respectively. CFCs and SF6 are non-radioactive and chemically stable under oxic conditions. However, CFCs can be degraded in anaerobic environments.25–27 resulting in an overestimation of groundwater residence times deduced from CFC measurements. Mixing of younger and older water due to dispersion or extended screen intervals is a factor that has to be considered for all dating tracers. The combination of different tracers or time series can help to identify and to quantify such processes (Figure 5.3.2) The input function and the decay rate in the subsurface are the crucial parameters defining the most sensitive dating range of a tracer. Normalised functional relations of tracer concentrations and groundwater age are shown in Figure 5.3.3(a). Compared to the atmospheric input these concentrations are smoothed out due to dispersion. The analytical uncertainties of the methods divided by the concentration gradient gives an estimate of the time resolution of a dating method (Figure 5.3.3(b)). Other factors affecting the tracer age are not considered here. The high precision of mass spectrometric (MS) measurements result in very accurate dating results for the 3H/3He method28 and age


Figure 5.3.2

221

Dating principle using environmental tracer methods. The tracer concentration in the recharge water is given by the input concentration (1) in the atmosphere and the recharge conditions (2) (temperature, pressure, excess air, etc.). In the aquifer different flow lines with different flow times mix to a whole age distribution, e.g. in a screened borehole (3). The correct age distribution has either to be assumed based on the hydrogeological situation or can be constrained, e.g. by measured time series (4). The admixture of old water (5) (450 years) which is free of 3 H, 85Kr, etc., can be quantified with a two-tracer approach (5). The old water shifts both concentrations towards zero. The dilution ratio indicates the amount of old water whereas the isotope ratio defines the age of the young water (examples: (a) 100% young water with an age of 20 years; (b) 60% young water with an age of 10 years and 40% water older than 50 years). The age of the old water can further be constrained with 39 Ar and 14C measurements.

differences of a few weeks can be measured.29 The uncertainties of 85Kr, SF6 and CFC ages are in the ideal case of the order of months to a few years. Decreasing CFC concentrations since the 1990s reduce the dating resolution for young waters. A similar problem arises for 3H with almost no concentration gradient over the last 10 years. 39Ar dating provides the most reliable dating results in the age range 60–900 years with a dating error of approximately

222

Figure 5.3.3

Chapter 5.3

(a) Normalised tracer concentrations in groundwater in a dispersive aquifer as a function of groundwater age. (b) Dating resolution calculated based on the analytical error of tracer measurements and the concentration gradients shown in the figure above. These errors have to be interpreted as lower limits because of other limiting factors like degassing, unknown recharge conditions, degradation or subsurface production, etc. (c) Most advantageous dating ranges of the different methods.

30–50 years. For older waters 14C is the method of choice. However, because groundwaters are commonly mixtures of waters with different age multi-tracer measurements are required for the characterisation of complex age structures. An example is given in Figure 5.3.2. With the combination of two tracers it is possible to determine, for example, the age and mixing portion of young water components. In the context of this volume it is very often sufficient to investigate whether or not recent water components are present in a sample. For this purpose any of the tracer methods 3H/3He, CFC, SF6 or 85Kr is in principle


223

suitable. The most favourable tracer has to be investigated in the individual case based on the objectives and the hydrological and geochemical conditions.

5.3.2.3

Geoindicators: Estimating Environmental Change and Relative Ages

Indicators of environmental change have been extensively developed, mainly using biological parameters. A series of geoindicators has been proposed,30 which can detect environmental change over a timescale of 100 years. Among these groundwater quality has been proposed as a sensitive indicator of overall change since human impacts are recorded in the groundwater body.31 Two levels of indicators are suggested which monitor physical change: changes in the natural hydrogeochemistry and the main anthropogenic influences. The primary indicators (water level, HCO3, DO, Cl, NO3, SO4, DOC) may be supported where possible by various secondary indicators, which help to characterise the sources of contaminant or the geochemical processes involved. The unsaturated zone in unconsolidated lithologies is also proposed as a target for monitoring. Under favourable circumstances in porous media a decadal record of the recharge rate, recharge history, products of geochemical reaction and records of pollution may be observed since downward rates of transport amount to 0.5–1.0 m yr 1. Chemical information in groundwater may also be used as an indicator of residence time since many processes and reactions are time dependent and these have been used to supplement information obtained from the quantitative radiometric tools.32,33 Chemical tracers are best investigated in downgradient profiles in confined groundwaters where sequential hydrogeochemical changes (including changes in cation ratios, salinity (Cl) changes, build-up of trace elements) may occur along flow lines. Records obtained may represent timescales extending over many millennia (103–105 yr) and give an indication of former climatic or environmental conditions as well as extending age ranges, e.g. above the radiocarbon dating limit.32 Information of relative age may also be obtained from depth profiles, distinguishing modern water (recognisable by various contaminants) from pristine sources at depth. Glynn and Plummer34 provide a comprehensive review of the present knowledge about subsurface geochemistry and the understanding of groundwater systems. Freshwater diagenesis by groundwater moving into sediments formerly occupied by saline waters of marine or continental origins produces sequential changes in the groundwater with distance and depth. These reactions are of two types: gradual dilution of the marine pore water of clay-rich sediments with associated changes in groundwater cation ratios and incongruent reactions between groundwater and carbonate minerals also leading to distinctive cation ratios. In the first case, often termed freshening aquifers,35–37 two end members are concerned: a saline NaCl-type marine water initially filling the aquifer, and a fresh CaHCO3-type recharge water. Cation exchange processes take place after the marine pore water has been replaced by fresh recharge water, causing

224

Chapter 5.3

a disequilibrium between the new pore water and the marine cations that are still adsorbed on the exchange sites of (mainly) the clay minerals. The marine cations Na1, K1 and Mg21 are desorbed and exchanged for Ca21 in the new pore water according to the affinity sequence, leading sequentially to an increase in Na1, followed by a K1 peak, and finally a Mg21 peak, in the upflow direction. Marine carbonates (low- or high-Mg calcites as well as aragonite) contain Mg, Sr and other impurities which help stabilise the mineral lattices under the biogenic conditions of formation. Carbonates under subaerial conditions then form aquifers and undergo freshwater diagenesis, the marine carbonate minerals releasing their impurities to groundwaters, leaving a purer calcite in the process.38 This process of incongruent dissolution is time dependent and Mg and Sr increases (up to limits of dolomite or celestite solubility) can be used to distinguish younger waters from older more evolved waters. This is illustrated for the chalk aquifer of Wessex in southern England in Figure 5.3.4. In the freshwater aquifer, where it can be demonstrated that inert tracers are derived predominantly from atmospheric inputs, it may be possible to infer

Figure 5.3.4

Trilinear diagram showing the evolution of groundwater in the chalk of Wessex, UK. Initial reaction of low-Mg calcite in the unsaturated zone and shallow aquifer produces low Mg/Ca ratios, but increasing Mg/Ca ratios (up to the limiting 1:1 ratio of dolomite saturation) indicate increasing residence time as the marine calcite undergoes recrystallisation. Similar trajectories are shown for the Berkshire area (London Basin) and in both areas the influence of residual connate water is shown by Na+K increase.

225


°

Figure 5.3.5

°

Lithium, strontium and 13C as indicators of groundwater residence time in the East Midlands aquifer, UK. Groundwater temperature has been used as a proxy for distance downgradient.

palaeoclimatic information and, therefore, independently, another indicator of age. This approach has been used in the East Midlands aquifer,32,39 where the Cl concentrations are atmospherically derived and the Br/Cl and the 36Cl/Cl atomic ratios may provide information on past recharge rates and the changes in source areas of precipitation. In turn the chemical data may be used to fingerprint modern and older, pristine water as in the case study below. In a further example from the East Midlands aquifer, reactive tracers such as 13 C, Li and Sr may be used as proxies for residence time.32 In Figure 5.3.5 the groundwater radiocarbon ages are indicated and show a near linear increase with distance from outcrop (here represented by temperature as proxy). There is a gap in ages from approx. 10–20 kyr which is thought to represent an absence of recharge during the LGM (Last Glacial Maximum). Strontium increases and the enrichment in carbon-13 are the result of two processes: the incongruent dissolution of carbonate minerals and the dissolution of trace gypsum with Sr as impurity. Lithium shows a strong linearity and is derived from slow release from silicate minerals (probably K-feldspar) in the aquifer framework. A lithium timescale has been suggested for this aquifer.32 Lithium is unrelated to modern contamination and at outcrop the lithium in young groundwaters is below 10 mg l 1 but concentrations in excess of 20 mg l 1 indicate residence times in excess of 20 kyr.

5.3.2.4

Numerical Modelling of Groundwater Age

Groundwater flow models and environmental tracers have been used to assess travel times and groundwater ages for some decades. The studies have

226

Chapter 5.3

investigated both very old groundwaters in large sedimentary basins40 and young groundwater in shallow aquifers.41,42 Presently the studies are increasingly focusing on young contaminated groundwaters, and the interaction with surface waters (Chapters 9.3 and 10.2),10,42,43 due to increasing environmental impact and the development of more advanced numerical modelling and environmental tracer tools for dating young groundwater. Groundwater ages can be computed by groundwater flow models or integrated hydrological models by various techniques.11,44–46 Many studies have demonstrated that the computation and comparison of groundwater ages estimated by groundwater flow models and environmental tracers are not trivial.21,46–49 Modelling studies have, for example, illustrated that tracer-derived groundwater ages may both underestimate or overestimate the actual average age of groundwater in settings where young waters with human impact are mixed with old waters without human impact20,21,45,50 or where low permeability aquitards or units occur along the flow paths,20,51–53 respectively. While this is a drawback to the application of tracer ages as a calibration tool for groundwater models, unless dispersive processes and mixing in the well can be accounted for in the groundwater flow and transport model, underestimation of groundwater age may actually be considered a benefit when evaluating groundwater vulnerability at water supply wells,6 since soluble pollutants generally behave similar to the environmental tracers. Similarly there may be a considerable uncertainty on results from the numerical modelling as these reflect the uncertainty of the geological model and the physical parameters of the aquifers and aquitards in the subsurface and uncertainty in climatological parameters.45,54,55 However, it is beyond the scope of this chapter to discuss strengths and drawbacks for the different methods. For the purpose of this chapter it suffices to say that both numerical hydrological models and environmental tracers are very strong tools in groundwater research and management, and that the best and most detailed information of the investigated system is obtained by the combined use of both methods. Generally, extensive geological knowledge and quantification of both geochemical and physical subsurface processes are needed to obtain sound estimates of groundwater ages.34

5.3.3

Groundwater Age and Water Quality and Quantity Issues

5.3.3.1

Groundwater Quality as a Function of Age

As mentioned earlier, groundwater quality changes with time due to both changes in anthropogenic activities and water/rock interaction in the surface. Generally the risk of pollution decreases with increasing groundwater age, while the risk of saltwater intrusion and increase in dissolved minerals, trace metals and radionuclides increases. No general rules can be given for the advance of modern contaminated groundwater in European aquifers as it varies considerably with the geological and climatological setting.2 Figure 5.3.6 illustrates the temporal evolution of nitrate concentrations in groundwater in

227

Groundwater Age and Quality 50

30 40

NL g 30

20

20 USf

N in groundwater (mg/L)

N in fertilizer (g/m2/y)

NL f

DK f

10

10 DKg US g 0

0 1940

Figure 5.3.6

1950

1960

1970 1980 Recharge date

1990

2000

Average (smoothed) nitrate concentrations in groundwater in Denmark (for all samples in monitoring database with O2 4 1 mg l 1),68 the Netherlands (Noord-Brabant)12,57 and the USA (watershed in Maryland)7 as a function of recharge year estimated by CFCs (DK and US) and 3H/3He (NL). The NO3-N concentration evolution in groundwater is compared to the fertiliser consumption for the relevant regions in the same period (subscripts g and f indicate groundwater and fertilizer curves, respectively).

Denmark, the Netherlands and the USA compared to the amount of applied fertilisers in the investigated regions. Environmental tracers have been applied widely in evaluation of the transport of nitrate in the subsurface and to dependent ecosystems.7,10,16,56–58

5.3.3.2

Groundwater Age and Monitoring

The location of groundwater monitoring points in relation to the spatial distribution of groundwater age is of major importance. Ideally groundwater monitoring programmes should include also analyses of environmental tracers such as tritium or CFC gases as these provide valuable information on the vulnerability of the monitoring wells and the temporal evolution of the groundwater around them. An example is shown in Figure 5.3.6 were the nitrate contents of oxic groundwater in Denmark and the USA are shown as a function of groundwater ages estimated by CFC gases compared to nitrate contents in groundwater in the Netherlands dated by the 3H/3He method.

228

5.3.3.3

Chapter 5.3

Groundwater Age and Water (Over)exploitation

Generally as the groundwater age and depth increases the risk of overexploitation and mining also increases as the recharge of the groundwater resource may be slow. However, if too high abstraction rates lead to severely lowered water tables and strongly increased hydraulic gradients, the risk of both saline water intrusion and pollution from the surface increases. Furthermore, if groundwater tables are lowered below reduced sediments or minerals this will induce natural geochemical processes, which may result in a groundwater quality breaching quality standards for several elements or substances. Lowered water tables will result in increased oxidation of reduced organic carbon and minerals potentially present in the sediment, and this may lead to increased concentrations of dissolved carbon, bicarbonate, sulfate, arsenic, nickel, etc.,59 Furthermore the high abstraction may affect the quality of the dependent surface waters primarily due to reduced discharge (Chapter 9.3).

5.3.4

Groundwater Age and the Water Framework and Groundwater Directives

The spatial distribution of groundwater ages in the subsurface is an import indicator of vulnerable and less vulnerable parts of the groundwater resources.41,60 and the potential impact from the different parts on dependent aquatic and terrestrial ecosystems.9–11 Hence it is an important parameter for the temporal and spatial evolution of the status of both groundwater itself and the dependent ecosystems.

5.3.4.1

Groundwater Age and Derivation of Natural Background Levels and Threshold Values

It is stipulated in the Water Framework Directive and the daughter Groundwater Directive (see Chapter 3.1) that the water bodies in Europe have to reach good status in 2015. The assessment of the qualitative status for groundwater depends on the natural background of the groundwater quality and the relevant environmental quality standards for groundwater itself and for the dependent ecosystems. The environmental tracers and groundwater dating are the most obvious and important tools for identifying natural background quality groundwater without human impact (see Chapter 5.2).61

5.3.4.2

Groundwater Interaction with Dependent Ecosystems

Shallow young groundwaters affect the dependent ecosystems more than deep groundwaters, quantitatively and qualitatively (Chapters 9.3 and 10.2).10 Hence a sound understanding of the relation between the spatial hydrochemical evolution of groundwater quality and the groundwater age distribution provides important information on the possible impact of groundwater on the

229


dependent ecosystems.7,9 Groundwater/surface water interaction may occur in both directions and may change downstream as some parts of the streams may be gaining, while others may be loosing. Hence a bad status groundwater may affect streams negatively in some parts of a stream, while in other parts the stream water quality may lead to deterioration of the groundwater quality (see Swiss case studies in Section 5.2).

5.3.5

Case Studies

Environmental tracers and groundwater flow modelling are applied for evaluation of groundwater ages in an increasing number of studies as important tools for evaluation of the hydrochemical evolution in both deep and shallow groundwater systems. Both tracers and groundwater flow modelling are generally necessary tools for unravelling groundwater flow and hydrochemical evolution in the usually complex geological structures of the subsurface. Below we briefly describe selected case studies from Denmark, Switzerland, Germany and the UK, in which environmental tracers and groundwater modelling have been cardinal tools.

5.3.5.1

Examples of Danish Case Studies

5.3.5.1.1

Nitrate Reduction in a Pyritic Sandy Aquifer at Rabis Creek

In a classic study, Postma et al. investigate transport and degradation of nitrate in a pyritic sand aquifer.19,56 The study identifies the tritium bomb peak and uses this together with earlier evaluations of the migration of bomb tritium in unsatured and saturated zones at a location close to the Rabis Creek site62 to estimate vertical groundwater flow velocity across the redox boundary, and the progression of the redox front (Figure 5.3.7). Postma et al. demonstrate that the redox front progression is accelerated by a factor of 5 by nitrate pollution to about 2 cm a year as nitrate is reduced and use up pyrite in the sediment.56 Later studies at the Rabis Creek test site investigate the subsurface geology63 Agricultural areas

Forest and Heath T2 T3 T4

T1

60 100 50

T5

T6

T10

100 50

Elevation, m

Water Table 40

? 20

0

Figure 5.3.7

NO3 > 0.1 mM

O3 < 0.05 mM

0.5 km

The advance of the nitrate plume at the Rabis Creek test site. Modified after Postma et al.56 The curves to the left of the multisamplers T1 and T2 indicate the location of the 3H bomb peak (Figure 5.3.1) in the Rabis Creek aquifer in 1988.

230

Chapter 5.3

and model nitrate removal,64 transport of tritium,65 groundwater ages66 and the transport and degradation of CFCs in the saturated25 and unsaturated zone.67 The site is included in the National Groundwater Monitoring Programme and more than 100 monitoring points at the site are sampled and analysed once a year.68

5.3.5.1.2

Natural Background Levels and Hydrochemical Evolution

The hydrochemistry of the deep-lying Ribe Formation aquifer in SW-Jylland, Denmark, was investigated in a European research project ‘‘PALAEAUX’’ on the hydrochemical evolution of European aquifers since the end of the last ice age.51 In this study it was mainly the radioactive isotopes 3H and 14C that were used and proved to be of great value in identifying natural background composition or modern water impacts. After correction for the effects of geochemical and physical (diffusion) processes in the subsurface, the groundwater ages obtained by 14C dating and groundwater flow models compared quite well and showed that the groundwater in the Ribe Formation is a few thousand years old, totally pristine and has a very high quality. The study showed that the Ribe Formation, which is mainly a freshwater sediment deposited in the Miocene period about 20 million years ago, was first salinised during later sea-level high stands and then again (re)freshened during the Pleistocene and Holocene geological periods. The Ribe Formation was studied further in another European Union research project on the estimation of natural baseline (background) levels of hydrochemical parameters in European aquifers.69 The study showed that the aquifer is an excellent example of a groundwater body with a pristine natural background quality that also to some extent can be used to approximate the natural background quality in anoxic parts of carbonaceous Pleistocene sands above.

5.3.5.1.3

Vulnerability of Groundwaters in Buried Quaternary Valleys

Deep buried valleys may contain pristine groundwater of very high quality, but may also create shortcuts or pollution ‘‘highways’’ between the surface and adjacent deep aquifers like the Miocene Ribe Formation described above. Very different scenarios may exist depending on the valley infill, surrounding aquifers and aquitards, and regional climate and exploitation. The environmental tracers are excellent tools for evaluating the vulnerability of the valley aquifer and adjacent aquifers. Selected environmental tracers were applied in a regional European research project on mapping of water resources in buried valleys for evaluating recharge, groundwater age and vulnerability in selected Danish and German Quaternary buried valleys.70 Results demonstrate the value of the environmental tracers in the understanding of groundwater flow systems and in the identification of leakage in some monitoring wells.

231


5.3.5.1.4

Pollutant Breakthrough and Tracer Transport Modelling in the Odense River Basin

In the final Danish example multiple environmental tracers (3H/3He, CFCs, 85 Kr and SF6) were used to constrain and partly calibrate an integrated hydrological model including groundwater/surface water interaction, set up to estimate the breakthrough and future evolution of the pollutant 2,6dichlorobenzamide (BAM) concentrations at water supply wells of the Odense Water Company.45,55,71,72 Figure 5.3.8 shows the simulated evolution of the pollutant concentrations at the well site for the next 50 years.

5.3.5.2

Examples of Swiss and German Case Studies

5.3.5.2.1

Quantification of River Water Infiltration in a Shallow Aquifer in Switzerland

Pollution of surface water can cause degradation of groundwater quality and conversely pollution of groundwater can degrade surface water. Hence, the exchange rate of groundwater with surface water in particular with rivers determines the risk potential for cross contamination between these two systems which have to be regarded as a single resource. In this example the environmental tracers 3H and 85Kr were used in order to estimate the mixing portion of river water in a shallow groundwater body.73 In the area of investigation the river water is enriched in 3H due to emissions from the local BAM root zone (µg/l) 1.50

BAM well field (µg/l) 0.50

0.45 1.25

0.40 0.35

1.00

0.30 0.75

0.25 0.20

0.50

0.15 0.10

0.25

0.05 0.00 0.00 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060

BAM concentration below root zone (input) BAM concentration at well field (fully mixed)

Figure 5.3.8

Simulated future evolution of the pollutant 2,6-dichlorobenzamide (BAM) at a well field.72

232

Chapter 5.3

watch industry compared to 3H in recharge from precipitation. 85Kr on the other hand is exclusively of atmospheric origin. The elevated 3H values found in two groundwater samples (Figure 5.3.9) clearly indicate the presence and the mixing portion of river water in the groundwater. In contrast, 85Kr is a sensitive indicator for the mean residence time of the groundwater.

5.3.5.2.2

Constraining the Age Distribution of a Mineral Water with Multi-Tracer and Time Series Tracer Measurements

The age structure of groundwater is a key parameter for the sustainable exploitation of a resource in terms of water quality and quantity. The accuracy of the dating depends not only on the intrinsic properties of the applied tracer methods (Figure 5.3.3(a)) but also on the understanding of the subsurface flow pattern and the shape of the transit time distribution.74 This information has to be based on the hydrogeological environment but can also be constrained by the tracer data. A sample from a mineral water production well revealed 3H and 85 Kr values of 22.1 TU and 22.3 dpm cm 3 Kr respectively.75,76 The relation between the 3H/85Kr ratio and the mean groundwater residence time depends on the assumed age distribution (Figure 5.3.10). The difference between a piston flow ( pfm: no mixing) and the exponential model (EM: strong mixing),

Figure 5.3.9

Estimation of river water portion and residence time of a shallow aquifer in eastern Switzerland using 3H and 85Kr data. The proportion of river water is about 80% and 30% in samples (a) and (b) respectively. The residence time of groundwater is around 2 years in both cases.

233

Groundwater Age and Quality 140

0.2 a)

PF (Tm = 12 years)

100

EM

0.0

Tritium [TU]

log( 3 H/ 85 Kr)

0.1

b)

120

PM

measured (1996)

-0.1 -0.2 t

-0.3

80 60 40 EM (T =15 years) m 20

-0.4 0 -0.5 0

5

10

15

20

25

mean residence time

Figure 5.3.10

1975 1980 1985 1990 1995 2000

sampling year

3

H and 85Kr concentrations of a mineral water. (a) Tracer ratio H/85Kr as function of mean residence time plotted for two assumed age distributions (PM: piston flow; EM: exponential model); (b) 3H time series. 3

which can be regarded as two extreme scenarios, increases significantly for residence times over 10 years. This ambiguity can be reduced if the 3H time series is taken into account (Figure 5.3.10(b)), which evidently favours the EM age distribution before the PM. Both the 85Kr and 3H tracer ratio and the 3H time series point to an EM age distribution with a mean residence time of 15 years for this sample.

5.3.5.2.3

Inter-Aquifer Leakage

Shallow unconfined Quaternary basins provide the main groundwater resource in many areas in Europe. However, as in southwestern Germany, these basins may be hydraulically interconnected with deeper groundwater systems, e.g. in the Malm-Karst.77 The direction of the hydraulic gradient between these two resources determines whether the shallow systems, which are vulnerable to pollution, are ‘‘fed’’ by older high-quality water from the deep aquifer or whether the deep aquifer is under risk to become polluted from the shallow system. The deep water is often older than 50 years,78 and is therefore free of young residence time indicators (3H, SF6, 85Kr, etc.). Hence, the admixture of deep and old water is manifested by a dilution of the tracer concentrations (Figure 5.3.11). Three groups of waters can be distinguished in Figure 5.3.11. Samples plotting near the suspected model curve (dotted line) represent waters from local recharge. The residence times of these waters range up to 20 years. A group of samples, mainly in the age range 20–25 years and originating essentially from greater depths of the Quaternary basins, become increasingly influenced by the deep water (group 2). The third group of samples is dominated by old water from the deep karst aquifer.

234

Chapter 5.3 45.0

40a

40.0 35.0

Tritium [TU]

30.0

30a 51a

young water age 25a

25.0 20.0

20a 10a

15a

15.0

5a

10.0 80%

60%

5.0

40%

young water portion

20%

0.0 0.0

10.0

20.0

30.0

40.0

50.0

60.0

85Kr[dpm/ccKr]

Figure 5.3.11

5.3.5.3 5.3.5.3.1

Age dating of young water components and estimation of the portion of old water from deeper underlying aquifers using a two-tracer plot.

Example of a British Case Study Chloride as Indicator of Modern and Pristine (Older) Water: East Midlands Aquifer

Chemical tracers can provide diagnostic information of age in many types of aquifer as outlined above: either alone or in conjunction with some other indicator. The East Midlands Triassic sandstone aquifer is well studied and presents a clear example of how tracers32,33,79 behave in groundwater which has no residual salinity over most of its developed section. Chloride acts as an inert tracer and its inputs relate to climatic or anthropogenic factors and geogenic influences are negligible.39 The timescale for groundwaters in the aquifer, obtained from corrected 14C data,80 extends beyond the radiocarbon dating range and several other absolute indicators have now been applied to this aquifer which demonstrate the age profile downgradient, albeit with a degree of mixing. In the example shown (Figure 5.3.12), where groundwater temperature is used as a proxy for distance from outcrop, the timescale is represented qualitatively by d18O. There is a clear separation (about 1.4%) between lighter isotopic signatures representing recharge during the late Pleistocene and recharge from the modern era; the scatter can be related to several factors such as abstraction rates and gives an idea of the amount of mixing in this aquifer which also shows strong vertical age stratification. The extent of invasion by modern groundwater is shown clearly by high Cl (above 20 mg l 1) derived from modern aerosol input with Cl enhanced by industrial emissions, but also locally by agrochemicals and from industrial sources. This ‘‘front’’ is also shown by other chemical indicators such as nitrate.33 The pristine waters are


Figure 5.3.12

235

Modern water with high Cl entering the East Midlands (UK) aquifer in contrast to the low Cl water of Holocene and late Pleistocene age.

remarkably low in Cl and represent the rainfall signature of the early Holocene and Pleistocene; the absence of an increase in Cl with depth over this section contrasts with many aquifers, where residual salinity is still found. In this aquifer salinity does increase in the very deepest groundwater, however, related to formation waters associated with continental evaporites. The main conclusion is that if the aquifer is well understood using a multi-tracer approach, then Cl alone may be used for management purposes to monitor the extent of overprinting of the pollution front.

5.3.6

Conclusions

A sound understanding of groundwater ages and travel times is a prerequisite for the evaluation of the history and fate of contaminants in the subsurface, and hence for evaluating and protecting the quality and quantity of both groundwater itself and its dependent ecosystems. Environmental tracers as well as groundwater or integrated hydrological models are the important and only tools for evaluating travel times in the subsurface, and these should be

236

Chapter 5.3

combined whenever possible to get the best description of groundwater ages and residence times in the hydrological system. There is a relatively large number of environmental tracers that can be applied for evaluating travel times and groundwater ages, as described in this chapter; which ones to use will depend on the hydrogeological setting, etc. Generally, it is recommended to use a combination of multiple tracers as this, for example, provides a possibility of evaluating the relative contribution of young and old waters in mixed water samples. Groundwater age or residence times from properly designed monitoring networks provide valuable information to all assessments of the evolution of groundwater quantity and quality and its effects on dependent ecosystems. Hence, they are important tools for developing a sustainable management policy for the protection of water resources and the aquatic environment.

References 1. W. M. Edmunds and P. Shand, in The Natural Baseline Quality of Groundwater, Blackwell, 1st edn, 2007, in press. 2. K. Hinsby, W. M. Edmunds, H. H. Loosli, M. Manzano, T. Melo and F. Barbecot, Geological Society Special Publications, 2001, 189, 271. 3. W. Alley, in Regional Ground-Water Quality, Van Nostrand Reinhold, 1st edn, 1993. 4. J. N. Galloway, F. J. Dentener, D. G. Capone, E. W. Boyer, R. W. Howarth, S. P. Seitzinger, G. P. Asner, C. C. Cleveland, P. A. Green, E. A. Holland, D. M. Karl, A. F. Michaels, J. H. Porter, A. R. Townsend and C. J. Vorosmarty, Biogeochemistry, 2004, 70, 153. 5. A. L. Heathwaite, G. Billen, C. Gibson, C. Neal, P. Withers and L. Bolton, J. Hydrology, 2005, 304, 1. 6. A. H. Manning, D. K. Solomon and S. A. Thiros, Ground Water, 2005, 43, 353. 7. J. K. Bohlke and J. M. Denver, Water Resour. Res., 1995, 31, 2319. 8. Mattle, W. Kinzelbach, U. Beyerle, P. Huggenberger and H. H. Loosli, J. Hydrology, 2001, 242, 183. 9. P. B. McMahon and J. K. Bohlke, J. Hydrology, 1996, 186, 105. 10. E. Modica, H. T. Burton and L. N. Plummer, Water Resour. Res., 1998, 34, 2797. 11. H. P. Broers, J. Hydrology, 2004, 299, 84. 12. H. P. Broers and B. van der Grift, J. Hydrology, 2004, 296, 192. 13. T. H. Christensen, P. Kjeldsen, P. L. Bjerg, D. L. Jensen, J. B. Christensen, A. Baun, H. J. Albrechtsen and C. Heron, Appl. Geochem., 16, 2001, 659. 14. T. H. Christensen, P. L. Bjerg and P. Kjeldsen, Ground Water Monitor. Remed., 2000, 20, 69. 15. J. K. Bohlke and Denver, Water Resour. Res., 1995, 31, 2319. 16. A. J. Tesoriero, T. B. Spruill, H. E. Mew, K. M. Farrell and S. L. Harden, Water Resour. Res., 2005, 41, 2005.


237

17. R. Kunkel and F. Wendland, J. Hydrology, 2002, 259, 152. 18. F. Wendland, H. Bogena, H. Goemann, J. F. Hake, P. Kreins and R. Kunkel, Phys. Chem. Earth, 2005, 30, 527. 19. C. A. J. Appelo and D. Postma, in Geochemistry, Groundwater and Pollution, A. A. Balkema, Leiden, 2nd edn, 2005, p. 1. 20. P. Maloszewski, W. Stichler and A. Zuber, Isotopes Environ. Health Stud., 2004, 40, 21. 21. E. M. LaBolle, G. E. Fogg and J. B. Eweis, Water Resour. Res., 2006, 42(7). 22. P. G. Cook and A. L. Herczeg, in Environmental Tracers in Subsurface Hydrology, Kluwer Academic, Boston, MA, 1st edn, 2000, p. 1. 23. M. Maiss and I. Levin, Geophys. Res. Lett., 1994, 21, 569. 24. E. Mazor, Geochim. Cosmochim. Acta, 1972, 36, 1321. 25. K. Hinsby, A. L. Højberg, P. Engesgaard, K. H. Jensen, F. Larsen, L. N. Plummer and E. Busenberg, Water Resour. Res., 2007, in press. 26. IAEA, in Use of Chlorofluorocarbons in Hydrology: A Guidebook, International Atomic Energy Agency, Vienna, 1st edn, 2006, p. 1. 27. L. N. Plummer and E. Busenberg, in Environmental Tracers in Subsurface Hydrology, ed., P. G. Cook and A. L. Herczeg, Kluver Academic, Boston, MA, 2000, pp. 441–478. 28. U. Beyerle, W. Eschbach-Hertig, D. M. Imboden, H. Baur, T. Graf and R. Kipfer, Environ. Sci. Technol., 2000, 34, 2042. 29. D. K. Solomon, S. L. Schiff, R. J. Poreda and W. B. Clarke, Water Resour. Res., 1993, 29, 2951. 30. A. R. Berger and W. J. Iams, in Geoindicators: Assessing Rapid Environmental Change in Earth Systems, A. A. Balkema, Rotterdam, 1st edn, 1996. 31. W. M. Edmunds, in Geoindicators: Assessing Rapid Environmental Change in Earth Systems, ed., A. R. Berger and W. J. Iams, A. A. Balkema, Rotterdam, 1st edn, 1996, pp. 135–150. 32. W. M. Edmunds and P. L. Smedley, Appl. Geochem., 2000, 15, 737. 33. P. L. Smedley and W. M. Edmunds, Ground Water, 2002, 40, 44. 34. P. D. Glynn and L. N. Plummer, Hydrogeol. J., 2005, 13, 263. 35. C. A. J. Appelo, Water Resour. Res., 1994, 30, 2793. 36. M. Coetsiers and K. Walraevens, Hydrogeol. J., 2006, 14, 1556. 37. K. Walraevens, M. Van Kamp, J. Lermytte, W. J. M. Van der Kemp and H. J. Loosli, Geological Society Spec. Publ., 2001, 189, 49. 38. W. M. Edmunds, J. M. Cook, W. G. Darling, D. G. Kinniburgh, D. L. Miles, A. H. Bath, M. Morgan-Jones and J. N. Andrews, Appl. Geochem., 1987, 2, 251. 39. J. N. Andrews, W. M. Edmunds, P. L. Smedley, J. C. Fontes, L. K. Fifield and G. L. Allan, Earth Planet. Sci. Lett., 1994, 122, 159. 40. M. C. Castro, P. Goblet, E. Ledoux, S. Violette and G. de Marsily, Water Resour. Res., 1998, 34, 2467. 41. A. Zuber, S. Witczak, K. Rozanski, I. Sliwka, M. Opoka, P. Mochalski, T. Kuc, J. Karlikowska, J. Kania, M. Jackowicz-Korczynski and M. Dulinski, Hydrolog. Processes, 2005, 19, 2247.

238

Chapter 5.3

42. P. G. Cook, S. Lamontagne, D. Berhane and J. F. Clark, Water Resour. Res., 2006, 42, 2006. 43. D. C. Gooddy, W. G. Darling, C. Abesser and D. J. Lapworth, J. Hydrology, 2006, 330, 44. 44. D. J. Goode, Water Resour. Res., 1996, 32, 289. 45. L. Troldborg, Technical University of Denmark/Geological Survey of Denmark and Greenland, 2004. 46. M. Varni and J. Carrera, Water Resour. Res., 1998, 34, 3271. 47. P. Maloszewski, W. Stichler and A. Zuber, Isotopes Environ. Health Stud., 2004, 40, 21. 48. P. Maloszewski and A. Zuber, Water Resour. Res., 1991, 27, 1937. 49. P. Maloszewski and A. Zuber, J. Hydrology, 1982, 57, 207. 50. C. M. Bethke and T. M. Johnson, Geology, 2002, 30, 107. 51. K. Hinsby, W. G. Harrar, P. Nyegaard, P. Konradi, E. S. Rasmussen, T. Bidstrup, U. Gregersen and E. Boaretto, Geological Society, Spec. Publ., 2001, 189, 29. 52. E. M. LaBolle, G. E. Fogg and J. B. Eweis, Water Resour. Res., 2007, 42, 2007. 53. W. E. Sanford, Ground Water, 1997, 35, 357. 54. J. C. Refsgaard, P. van der Keur, B. Nilsson, D. I. Mu¨ller-Wohlfeil and J. Brown, J. Hydrol. Earth Syst. Sci., in press. 55. L. Troldborg, J. C. Refsgaard, K. H. Jensen and P. Engesgaard, Hydrogeol. J., 2007, doi: 10.1007/j.envpol.2007.01.027. 56. D. Postma, C. Boesen, H. Kristiansen and F. Larsen, Water Resour. Res., 1991, 27, 2027. 57. A. Visser, H. P. Broers, B. van der Grift and M. F. P. Bierkens, Environ. Pollut., 2007, doi: 10.1016/j.envpol.2007.01.027. 58. K. Zoellmann, W. Kinzelbach and C. Fulda, J. Hydrology, 2001, 240, 187. 59. F. Larsen and D. Postma, Environ. Sci. Technol., 1997, 31, 2589. 60. A. Zuber, S. M. Weise, J. Motyka, K. Osenbruck and K. Rozanski, J. Hydrology, 2004, 286, 87. 61. W. M. Edmunds, P. Shand, P. Hart and R. S. Ward, Sci. Total Environ., 2003, 310, 25. 62. L. J. Andersen and T. Sevel, in Isotope Techniques in Groundwater Hydrology, IAEA, Vienna, 1974. 63. H. Olsen, C. Ploug, U. Nielsen and K. Sorensen, Ground Water, 1993, 31, 84. 64. P. Engesgaard and K. L. Kipp, Water Resour. Res., 1992, 28, 2829. 65. P. Engesgaard, K. H. Jensen, J. Molson, E. O. Frind and H. Olsen, Water Resour. Res., 1996, 32, 3253. 66. P. Engesgaard and J. Molson, Ground Water, 1998, 36, 577. 67. P. Engesgaard, A. L. Hojberg, K. Hinsby, K. H. Jensen, T. Laier, F. Larsen, E. Busenberg and L. N. Plummer, Vadose Zone J., 2004, 3, 1249. 68. GEUS, Groundwater Monitoring 2005, Geological Survey of Denmark and Greenland (GEUS), 2005 (www.geus.dk). 69. K. Hinsby and E. S. Rasmussen, in The Natural Baseline Quality of Groundwater, ed. W. M. Edmunds and P. Shand, Blackwell, 1st edn, 2007.


239

70. K. Hinsby, in Groundwater Resources in Buried Valleys: A Challenge for Geosciences, ed. BurVal Working Group, Leibniz Institute for Applied Geosciences (GGA-Institute), Hannover, 2006, pp. 141–148. 71. J. A. Corcho Alvarado, R. Purtschert, K. Hinsby, L. Troldborg, M. Hofer, R. Kipfer, W. Aeschbach-Hertig and H.-A. Synal, Appl. Geochem., 2005, 20, 599. 72. K. Hinsby, L. Troldborg, R. Purtschert and J. A. Corcho Alvarado, in Isotopic Assessment of Long Term Groundwater Exploitation, IAEA-TECDOC-CD-1507, 2006, 73–95. 73. K. Osenbruck, Pilot-Isotopen-Studie, Amt fu¨r Umwelt des Kanton Thurgau, 2001. 74. D. W. Waugh, T. M. Hall and T. W. N. Haine, J. Geophys. Res. Oceans, 2003, 108, 2003. 75. P. Hartmann, ETH Zu¨rich, 1998. 76. R. Purtschert, Universita¨t Bern, 1997. 77. B. W. Bertleff, Abh. Geol. Landesamt Baden-Wu¨rttemberg, 1986, 12. 78. M. Heidinger, Grundwasser Bewirtschaftungskonzept Singen, Stadtwerke Singen, Singen, 1996. 79. W. M. Edmunds, A. H. Bath and D. L. Miles, Geochim. Cosmochim. Acta, 1982, 46, 2069. 80. A. H. Bath, W. M. Edmunds and J. N. Andrews, in International Symposium on Isotope Hydrology, IAEA-SM-228/27, Vienna, 1979, vol. II, pp. 545–568.

CHAPTER 5.4

Characterisation of Groundwater Contamination and Natural Attenuation Potential at Multiple Scales THOMAS PTAKa AND JERKER JARSJO¨b a

University of Go¨ttingen, Geosciences Center, Goldschmidtstrasse 3 DE-37077 Go¨ttingen, Germany; b Stockholm University, Department of Physical Geography and Quaternary Geology, SE-106 91 Stockholm, Sweden

5.4.1

Introduction

Groundwater pollution is an important problem at many locations all over Europe. Sources for contaminants in aquifers such as chlorinated compounds, petroleum hydrocarbons, etc., are, for example, leaking underground storage tanks and pipelines, petrol stations, gasworks sites and all types of industries. At these locations, rapid industrial development, missing regulations and/or safety measures, changes in land use and ownership as well as the hydraulic and hydrogeochemical aquifer heterogeneity cause complex and irregular contamination patterns with very often unknown locations of pollutant hot spots. It is generally accepted that present approaches for site investigation and assessment are either not reliable enough or not cost effective, making the development of new approaches for the characterisation of groundwater contamination and natural attenuation potential at multiple scales necessary. The European Union (EU) FP 5 project INCORE (Integrated Concept for Groundwater Remediation, EVK1-1999-00080) and the German BMBFfunded project SAFIRA C2.1 (Erkundung der Schadstofffracht in kontaminierten Aquiferen zur Dimensionierung von in-situ-Sanierungsreaktoren, BMBF 02WT9948/0) are aimed at the development and implementation of a new approaches and methods for contaminated land assessment and revitalisation in urban industrial areas, focusing on groundwater quality and complex 240

Characterisation of Groundwater Contamination

241

contamination patterns at megasites which are typical for many European cities. The principal idea is to start at large scale, e.g. at the scale of an entire industrial site, and, in a first step, to assess groundwater contamination, using an innovative integral investigation method to estimate contaminant concentrations and mass flow rates across control planes as well as the natural attenuation potential. Using backtracking methods, potential contamination source zones can then be delimited, considering parameter uncertainty. Finally, detailed high-resolution investigation methods such as direct push-based profiling or multilevel sampling can be applied for small-scale investigations at the identified hot spot zones. In this approach, a large potentially contaminated area is screened initially at a high level of certainty, but only a small area may be finally considered for further investigations and remediation measures. Consequently, this large- to small-scale screening procedure yields a significant reduction of costs needed for land revitalisation. The new tools were developed, implemented and tested under real-world conditions, considering administrative aspects also. In this way, the project results provide an important basis for the development and implementation of EU directives on contaminated land assessment and revitalisation. This chapter summarises some of the new developed approaches and tools. The principle of the integral groundwater investigation method is described at first. Then a general, integral investigation-based methodology for assessing the effects of aquifer parameter uncertainty on the estimates of mass flow rates and concentrations as well as on delimiting both contaminant source zones and zones absent of source is presented. It is also shown how the integral investigation method can be applied to estimate natural attenuation rates at field scale. Finally, a multilevel version of the integral investigation approach for local-scale investigations is described. In addition, examples of application are given.

5.4.2

The Integral Groundwater Investigation Method

A reliable characterisation of contaminant plumes in groundwater and source zone locations is essential for decisions about future land use at contaminated sites, and for choosing appropriate remediation measures. The basic problem of the characterisation is that at many contaminated sites pollutant hot spots with positions not exactly known, and preferential transport paths and low conductivity zones within the aquifer cause an irregular distribution of contaminants in groundwater (Figure 5.4.1). In such situations, standard subsurface investigation procedures based on interpolation of point-scale concentration measurements are very likely to yield poor results. An effective way of obtaining investigation results with a high level of certainty at large scale is to apply the integral groundwater investigation method,1–4 in which relatively large water volumes are sampled within so-called integral pumping tests (IPTs; Figure 5.4.1). Due to the large sampling volumes the small-scale concentration variance is averaged out, which may otherwise influence and bias point-scale concentration

242

Chapter 5.4 Plot of concentration vs. time during pumping tests (compound specific)

Pumping tests with concentration time series measurements Contaminated Mean groundwater flow direction site Isochrones Source Well 1 Control cross-section of Well 2

C

t1

pollutant Well 3

t2 t 1

Well 2

C

Well 3

t2 t1

t2

Contaminant plume

Contaminant mass fluxes and concentrations at control cross-section

Figure 5.4.1

C Well 1

Transient inversion algorithm based on a numerical flow and transport model of the field site

Concept of the integral investigation method for the quantification of groundwater contamination.3

measurements, and there is no need to interpolate point-scale concentration measurements. Therefore a high level of certainty can be expected for the investigation results.

5.4.2.1

Concepts and Principles

The basic idea of the integral groundwater investigation method1–3 is to cover a whole cross-section of a contaminant plume downstream of a pollutant source, employing pumping tests with multiple contaminant concentration measurements at the pumping wells. Due to the spatial integration of a pumping test, and due to the increasing capture zone with pumping time, both the spatial distribution of the contaminants as well as the total mass flow rate within a contaminant plume can be estimated. To apply the integral investigation method, one or more pumping wells are placed along a control plane (control cross-section) perpendicular to the groundwater flow direction and operated simultaneously, or in subsequent pumping campaigns, downstream of a suspected pollutant source zone. The positions, pumping rates and pumping times are designed in a way so as to allow the well capture zones to cover the overall width of the potentially polluted area (Figure 5.4.1). Typically, the well is operated for a time period of some days, in order to obtain a large enough well capture zone. During pumping, as the capture zones increase, the concentration of groundwater contaminants and/or other groundwater quality parameter values is measured as a function of time at each of the pumping wells. The concentration time series yield information on the position and extent of the contaminant plume(s) as well as on the concentrations of the target substances in the plume(s). In Figure 5.4.2, four typical scenarios of concentration time series are shown as usually observed during IPTs, together with a possible interpretation of the subsurface contaminant plume size and location.4

243

1

Idealized contaminant plume maximum isochrone at end of pumping period pumping well

concentration

concentration


3

time

concentration

concentration

time

2 time

Figure 5.4.2

4

time

Typical scenarios of concentration time series recorded during an integral pumping test and possible interpretation with respect to position and size of the contaminant plume.4

The type 1 scenario represents a site where the pumping well is located outside of a relatively narrow contaminant plume. At the beginning of the pumping, the well capture zone initially covers the uncontaminated aquifer volume, and the concentrations in the well are below the detection limit. As the well capture zone (isochrone) grows into the area of the plume, the concentrations in the well discharge increase. At a later time, when the well capture zone has reached the outer fringe of the plume, the concentrations decrease again, due to dilution with clean water from outside the plume. Note that the measured concentration does not decrease to zero, as always the contaminant plume is within the isochrone and therefore some contaminated water is pumped at the well. It becomes clear from this interpretation, that the shape of the concentration time series is determined by the well location relative to the location of the plume and the plume width. The scenario of type 2 is observed if the pumping well is located within the contaminant plume, which itself is limited in width. In such a case, concentrations in the pumping well are high at the beginning of the pumping, but they decrease as the size of the capture zone and hence the dilution increases with time. This type can also be observed if the fringe of the plume is not sharp but given by decreasing concentrations. In case of the type 3 scenario, the pumping well is located outside of the plume, and the plume width reaches beyond the maximum isochrone extent. In this case, the width of the plume cannot be assessed. Finally, the type 4 scenario is characterised by a more or less constant concentration detected at the pumping well. In this case, the well is located within a wide plume with only slightly varying concentrations. Again, the plume width cannot be assessed. Only for a concentration time series of type 4, the value of the contaminant concentrations measured in the well actually represents the real concentration of the plume within the aquifer. The mass flow rates and mean concentrations within the well capture zones as well as possible contaminant concentration distributions across the control plane can be determined by an inversion procedure.

244

Chapter 5.4

5.4.2.2

The Inversion Problem

5.4.2.2.1

Numerical Solution

The algorithm used for the numerical inversion of the measured concentration time series at the wells is implemented in the newly developed program code CSTREAM,5,6 an improved version of the original code by Schwarz,7 focusing on applications in highly non-uniform groundwater flow systems due to aquifer heterogeneity (hydraulic conductivity distribution) and boundary conditions, and allowing investigations in a broad range of hydrogeological conditions. CSTREAM requires a transient flow model of the field site, which incorporates the pumping rates and pumping times at the abstraction wells, simulates all pumping tests at one control plane in one model run, and which provides the thickness, hydraulic conductivity and porosity of the aquifer as well as the local hydraulic gradient. Using CSTREAM, irregular well capture zones due to aquifer heterogeneity and the effects of the local hydraulic gradient on the development of capture zones are accounted for. Mass flow rates are thus estimated using local groundwater flow terms at the wells. The program operation is briefly described in the following using a two-dimensional, depth-integrated formulation.5 The setting is outlined in Figure 5.4.3, where the mean groundwater flow under natural conditions is along the y-axis. A contaminant plume is assumed to pass a control plane (CP) located at y ¼ 0. The control plane has an area ACP [L2] defined by the thickness of the water body and the maximum capture width LCP [L] during a pumping test performed at a pumping well with the position (0, 0). The distribution of contaminant concentration in the aquifer is denoted by C(x,y,t) [M L3]. C(x,y,0), i.e. the concentration distribution at time t ¼ 0 before start of the pumping, is assumed to be constant in flow direction along the streamlines within the well capture zone. The well is operated for times t 4 0, and the contaminant concentrations at the pumping well are recorded as concentration y pumping well (0,0)

a)

QW(t) CW(t)

b)

b A CP LCP q

q

x

(p)

(n)

LI(t) plume

Figure 5.4.3

b(x,y)

Concept for the inversion of measured concentration time series. (a) The plume and the control plane are located at the well position perpendicular to the mean groundwater flow direction. (b) Convergent flow field with isochrone LI(t) at time t (see text for an explanation of the symbols8).

245


time series CW(t). This measured concentration time series is used in order to estimate the total mass flow rate. The concentration distribution along the CP before pumping, C(x,0,0), can be expressed as a function of the concentration time series CW(t) measured at the pumping well through an integral equation describing the mass balance within the volume of pumped water:5 CW ðtÞ ¼

1 QW

I

Cðx; y; 0Þj~ qðpÞ ðx; yÞjbðx; yÞ dl

ð1Þ

LI ðtÞ !

where jqðpÞ ðx; yÞj ½L T1 denotes the DarcyH velocity during the pumping test, ! b(x, y) [L] is the saturated thickness, QW ¼ LI ðtÞ jqðpÞ ðx; yÞjbðx; yÞ dl ½L3 T1 is the pumping rate at the well and LI(t) [L] is the isochrone corresponding to time t. Equation (1) is solved for each value of the measured concentration time series, taking into account the concentration distribution in the aquifer obtained by the previous CW(t) values. In this way a concentration distribution within the aquifer is obtained, which implicitly yields a concentration distribution across the control plane before start of pumping, C(x,0,0) [M L3]. Figure 5.4.4 summarises the basic principle of the numerical inversion solution of an IPT. The left side of Figure 5.4.4 shows a plan view of the isochrones and the streamlines of the undisturbed groundwater flow field corresponding to their maximum extent. The first concentration measurement shown on the right side of Figure 5.4.4 corresponds to the first, i.e. smallest, isochrone. The sample concentration is the mean (mixed) concentration along the isochrone. Using the assumption that the concentration is practically constant along a streamline at

concentration [c/cmax]

1.0

0.5

0 0

Figure 5.4.4

1

2 3 time [d]

4

Plan view of the isochrones, streamtubes of the undisturbed groundwater flow field and the pumping well (left) as well as measured concentrations at the well (right) for a demonstration run of an integral pumping test.9

246

Chapter 5.4

the scale of the isochrones, the area covered by the streamtube corresponding to the first isochrone can be assigned the mean concentration of the first sample (i.e. zero, in this example). The groundwater of the second sampling is representative of the second isochrone. The concentration of a part of the second isochrone, i.e. the part crossing the streamtube of the first sample, is already known from the first sample. The average concentration of the remaining parts of the second isochrone can then be calculated, yielding a mean concentration along the second isochrone corresponding to the second measured concentration. This procedure is then successively repeated for all measured concentrations. The resulting concentration distribution can then be integrated along the control plane in order to obtain the total mass flow rate under steady state flow conditions across the control plane, MCP [M T1]:5 Z MCP ¼ Cðx; 0; 0Þj~ qðnÞ ðx; 0Þjbðx; 0Þ dx ð2Þ lCP !

wherejqðnÞ ðx; yÞj ½L T1 is the Darcy velocity perpendicular to the CP under natural flow conditions. The numerical program CSTREAM enables the estimation of the total mass flow rate at a CP, MCP, for (known) heterogeneous conditions. The inversion of the measured concentration time series at the pumping well CW(0, 0, t), yielding the concentration distribution along the CP before pumping, is performed based on a transient and heterogeneous numerical flow model of the field site. MODFLOW is used here.10 The code MODPATH11 is used for the definition of the isochrones (transient backtracking during pumping) and of the streamtubes (steady-state forward tracking before pumping, i.e. at time t ¼ 0). Finally, the integral eqn. (1) is solved numerically within the code CSTREAM, and mass flow rates are obtained applying eqn. (2). Verification examples are provided by Bayer-Raich et al.5

5.4.2.2.2

Analytical Solution

If a groundwater flow model of the investigated site is not available, for example in an initial stage of site investigation, and/or if no information on aquifer heterogeneity can be obtained, a simplified analytical solution for eqn. (1) and (2) can be applied for the inversion of the measured concentration time series. Under the assumption of homogeneous aquifer parameters around the pumping well and radially symmetrical flow towards the pumping well, i.e. neglecting the influence of the natural groundwater flow during the pumping period, the following analytical equation may be used for the estimation of mass flow rates:7,12

MCP ¼ 2

n X i¼1

cî Qi

with

1 p iP rk1 rk ci arccos c^k arccos 2 k¼1 r ri i cî ¼ ri1 arccos ri

ð3Þ


247

where MCP [M T1] represents the mass flow rate perpendicular to the control plane, ci [M L3] the concentration measured at the pumping well at time ti, i.e. ci ¼ c(ti), and cî the average of the concentrations of the two streamtubes of the natural groundwater flow field positioned left and right from the pumping well at a distance r [L] (with ri1 o r o ri ). Qi ¼ kjrhjbðri1 ri Þ ½L3 T1 is the discharge under natural, i.e. undisturbed, conditions passing the control plane at both left and right streamtubes. The radius of the isochrone at time t is given by pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ri ¼ QW ti =pbf ð4Þ where QW [L3 T1] is the pumping rate at the well, b [L] the aquifer thickness and f the porosity. For the first time step, c^1 ¼ c1 , and n is the total number of samples. This solution has been applied successfully at several sites (e.g. Refs. 4 and 13). Criteria for a site-specific decision on the application of the simplified analytical solution are discussed elsewhere.2 An expansion of the analytical inversion solution considering non-radial isochrones due to natural groundwater flow is given in Ref. 12.

5.4.2.2.3

Non-uniqueness of Inversion

The inversion of the measured concentration time series is not unique, as many possibilities exist for the distribution of the contaminant in the aquifer to generate the measured concentration time series in the pumping well. Therefore, a symmetrical solution is usually used, where the contaminant mass is distributed equally on both sides of the pumping well, which yields a symmetrical concentration distribution in the aquifer along the control plane. If additional information on concentration in the aquifer is present, i.e. a pointscale measurement within the well capture zone, conditioning of the inversion algorithm is possible to include this a priori information. For aquifers with medium heterogeneity the uncertainty introduced by the non-uniqueness of the concentration distribution can be expected to be smaller than 50% of the estimated mass flow rate.5 If more than one pumping well is used in subsequent pumping campaigns to measure the concentration time series, subsequent overlapping of the individual capture zones allows one to reduce the non-uniqueness of the solution.2 The shifting of the plume position due to pumping in subsequent campaigns is of course considered in the numerical solution.

5.4.2.3

Application of the Integral Investigation Method

Up to now the integral investigation method has been applied under real-world conditions at numerous locations in Europe and North America, covering a large variety of contaminant types and hydrogeological conditions. As an example, some of the results obtained from applications at an industrial site in southwest Germany will be summarised in the following.

248

5.4.2.3.1

Chapter 5.4

Application of the Integral Investigation Method at an Industrial Site in Southwest Germany

5.4.2.3.1.1 Site Description. The investigated site in southwest Germany is a more than 100-year-old industrial area with multiple potential sources of groundwater contamination. Abandoned landfills (with municipal waste, demolition debris and industrial waste) are overlain by former and recent industrial sites. Figure 5.4.5 shows a plan view of the study area, which has an extent of 4.5 km2. The area is contaminated with mineral oil, BTEX compounds, chlorinated hydrocarbons as well as PAHs. The aquifer system under investigation is a Quaternary porous aquifer consisting of poorly sorted sand and gravel deposits. The aquifer thickness shows a high variability between 1 and 10 m mainly due to the irregular surface of the aquifer bottom. The average thickness of the aquifer is 3.5 m, and the mean transmissivity amounts to 6.85 103 m2 s1. The ln K values determined by slug tests and short duration pumping tests range from 9.8 to 3.1 with a variance of 1.9. This indicates the considerable heterogeneity of the aquifer. 5.4.2.3.1.2 Definition of Control Plane Locations, Field Investigations, Design and Performing of Integral Pumping Tests. An extensive compilation of the existing data provided the basis for the planning of the field investigations.13 The relevant data included the characterisation of about 50 potential source zones with groundwater flow directions, aquifer parameters (where available) and the analysis of the existing network of about 300 monitoring wells. Within

source characterization pumping test abandoned landfill

pot. contaminated site

Figure 5.4.5

Plan view of the study area in southwest Germany and position of integral pumping tests.13


249

the first phase of the field investigations starting in April 1997, 59 new monitoring wells were drilled within the study area. The wells for the (source zone characterisation) IPTs were placed along several control planes transverse to the main groundwater flow direction (Figure 5.4.5). The cross-sections are completed by wells which are located downstream of sites with major contaminant sources. In addition, a number of wells were drilled to complete the existing network of monitoring wells. In each of the new wells and in some of the existing wells a short-duration (3–4 h) step drawdown test was conducted providing information about the well capacity and the aquifer parameters. Groundwater samples, obtained at the end of the step drawdown tests were used for some first screening of the kind of contaminants and the range of concentrations to be expected. This information provided the database for the detailed planning of the pumping tests (discharge rates, pumping periods, sampling intervals, type of discharge treatment). The most critical cost factor was how to obtain the maximum possible size of the capture zone within a minimum of pumping time. This in general is achieved through maximum discharge rates. The sampling intervals for concentration measurements were chosen such that every concentration measurement represents equal spatial averaging with respect to the capture zone development. From October 1997 until June 1998, 34 full-scale IPTs were conducted in the study area.13 Due to the high hydraulic conductivity of the aquifer the pumping rates were 5.3 l s1 on average. With an average pumping period of 5.3 days, the resulting capture zone widths range from 30 to 120 m. In general about 10 groundwater samples were obtained during each pumping test and analyzed for the major organic compounds. The numerical model and the numerical evaluation of the IPTs is described in sections below dealing with parameter uncertainty and delimiting of contaminant source zones.

5.4.3

Methodology to Consider Aquifer Parameter Uncertainty and to Delimit Contaminant Source Zones Using Integral Measurements

5.4.3.1

Principles

Although estimations of plume locations form a part of the inversion problem, as outlined in the previous sections, it should be recognised that robust integral estimations of average concentrations and total mass flows may not necessarily require an exact answer regarding the plume location. This is because different plausible plume locations and flow field conceptualisations may yield similar results in terms of concentration and mass flow. Formally, this view is supported by the analytical analyses of inverse problems of Bayer-Raich et al.,12 where it is shown that whereas solutions for exact plume locations are mathematically ill-posed, the corresponding solutions for average concentrations are relatively robust. In the following, we will explore this topic further

250

Chapter 5.4

and systematically review different sources of uncertainty, summarising and discussing current knowledge regarding their influence on integral investigation results. We will also relate this to practical aspects related to IPT investigations of contaminated sites, including discussions and suggestions on how measurements should be performed in the light of existing uncertainties.

5.4.3.1.1

Sources of Uncertainty

One can distinguish between three principal sources of uncertainties in IPT investigations: (i) flow model uncertainties, (ii) transport/reaction model uncertainties and (iii) the non-uniqueness of inversion results, leading to uncertainties of the original contaminant distribution in the aquifer. Regarding (i), we note that flow model uncertainties can be a result of uncertain aquifer properties or uncertain initial or boundary conditions. For instance, considering IPTs performed such that the pumping rate is large relative to the natural flow in the aquifer (a preferable condition for obtaining dependable IPT results), isochrones will be approximately circular around the pumping well in homogeneous aquifers. This is formally the case if the pumping rate Q is greater than or equal to 2pbq02t/ne, where b is the aquifer depth, q0 is the specific discharge, t is the pumping time and ne is the effective porosity, as further outlined in Bayer-Raich et al.12 Then, in homogeneous aquifers, even if the hydraulic conductivity value K is uncertain, the inversion procedure implies that exact estimates of average concentrations can be obtained. The reason is that the circular isochronous positions and linear streamline positions are independent of the uncertain K estimate. However, associated mass flows are in this case associated with uncertainty since they are dependent on K. In contrast, boundary condition uncertainties and uncertain K values may in heterogeneous aquifers result in both uncertain average concentration and mass flow estimates. This is because the isochronous positions and streamlines generally depend on the assumed K distribution and boundary conditions (BCs), which influence inversion results. Jarsjo¨ et al.14 used a MODFLOW site model and a numerical inversion procedure for quantification of the effects on IPT results of a BC uncertainty (that, in turn, was caused by uncertainties in an underlying water balance study). Results showed that the effects were relatively small: a factor two difference in constraining groundwater recharges resulted in relative errors in average concentrations and mass flows that were commonly less than 10%. Additional uncertainties related to spatially variable K values may be quantified using Monte Carlo simulations of the inversion, in which a relatively large number of equally likely aquifer realisations are used for obtaining statistics on the associated variability in IPT results (e.g. Ref. 15). However, for most practical applications, it may be sufficient to consider only some limiting cases of aquifer heterogeneities, such as the fully stratified case (with K variability in the vertical direction only), for which the uncertainty analyses can be considerably simplified by application of numerical or analytical methods to each aquifer layer independently (e.g. Ref. 6).


251

Regarding transport/reaction parameter uncertainties, case (ii) above, Bayer-Raich et al.16 recently investigated the influence of linear sorption on IPT interpretations (during the pumping phase). Results showed that interpretation results are independent of the retardation factor under equilibrium conditions, although linear retardation due to sorption decreases the investigation volume in comparison with the non-retarded case. In other words, whereas this has implications for the technical dimensioning of the tests (e.g. required number of wells, pumping rates and pumping times for investigating a cross-section of certain dimensions), uncertainties in the retardation factor do not influence subsequent average concentration or mass flow results. Furthermore, for obtaining a solution to the inversion problem, it is necessary to make some assumption regarding the contaminant attenuation in the direction of flow. So far, all existing IPT studies have, for simplicity, assumed that concentrations are approximately constant over the well capture zone (in the flow direction). A basic and relevant question posed by Zeru and Scha¨fer17 is how much various violations of this simplifying assumption can bias IPT results, particularly if strong concentration gradients exist within the well capture zone (for instance as a result of dispersion, (bio)degradation and fluctuations of the flow field). In response, Bayer-Raich et al.18 showed that IPT results are unbiased as long as the concentration attenuation along the flow direction is linear, regardless of the concentration gradient. This hence implies that biases in IPT investigations due to the ‘‘constant concentration’’ assumption are only introduced if the concentration attenuation is nonlinear. Jarsjo¨ and Bayer-Raich19 further considered wide contaminant plumes and provided an analytical expression from which the influence of concentration attenuation on IPT results can be quantified. Specifically, effects of exponential first-order decay were investigated, being the most common model for natural attenuation (NA) processes. Results show that the ‘‘constant concentration’’ assumption does not considerably bias predictions even if the investigated contaminant undergoes first-order decay, unless the resulting attenuation is very large. For instance, a 60% concentration decrease over the capture zone extent caused by first-order decay yields a prediction error of 4%. Finally, the uncertainty related to the non-uniqueness of the inversion results, case (iii) above, is illustrated in Figure 5.4.6. The upper part of the figure illustrates a C(t) curve, i.e. a contaminant concentration–time series measured in a pumping well, which is used as input for the IPT inversion analysis. The curve shows that the concentration is zero, or close to zero, in the beginning of the pumping test, implying that the well is located in a relatively clean part of the aquifer. After some time, concentrations in the well increase as one or several contaminant plumes are drawn towards the well due to the pumping. However, the original location of the contaminant plume relative to the well cannot be judged on the basis of the measured C(t) curve, which means that several different interpretations (or conceptual models) regarding the spatial contamination distribution are possible on the basis of the same curve. In analogy with Jarsjo¨ et al.,14 this uncertainty will in the following be denoted contamination model uncertainty.

252

Chapter 5.4

?

concentration in pumping well

Observation:

? time

? Interpretation I: Plume at left hand side

Interpretation III: Plume at right hand side

Interpretation II: Plume at both sides

280

280

280

270

270

270

260

260

260

250

250

250

240

240

240

230

230

230

220 220 220 220 230 240 250 260 270 280 220 230 240 250 260 270 280 220 230 240 250 260 270 280

Figure 5.4.6

Illustration of the non-uniqueness in the inversion results, regarding the plume position relative to the well. The lower part of the figure shows possible hydraulic and contaminant situations in the nearest well vicinity (plan views), with the shaded region representing the plume location, the thin black lines representing isolines of hydraulic head and the closed curves representing isochrones. The well is located within the inner isochrone. The numbers on the axes are spatial coordinates in the x and y directions, in metres.

Figure 5.4.6 illustrates three possible interpretations that can be made regarding the contaminant distribution: the contaminant plume may be located at the left-hand side of the well only, at the right-hand side of the well only, or at both sides. Jarsjo¨ et al.14 quantified this left–right uncertainty considering 19 IPTs performed in a strongly heterogeneous aquifer at an industrial site in southwest Germany (see above). Specifically, for each well, the contaminant mass flow (MF) was quantified for each considered plume position (left, right or both, i.e. symmetrical; see Figure 5.4.6), resulting in three different MF estimations per well and contaminant. The uncertainty in the result was then for each well and contaminant quantified through the relative MF arising from the three considered positions, defined as RMF ¼ (max MF min MF)/(average MF). Results showed that large uncertainties mainly occurred if the aquifer properties were strongly heterogeneous in the nearest vicinity of the pumping well. A regression analysis showed that RMF was correlated to the standard deviation of the log-transmissivity slnT within the capture zone volume of the qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi aquifer model, quantified as sln T 1=ðnCZ 1Þ Sðln T mln T Þ2 , where


253

mlnT is the mean value of ln T, the summation is performed over all the model cells belonging to the well capture zone and nCZ is the total number of model grid cells belonging to the well capture zone. Specifically the correlation was RMF E 0.5slnT + 0.1, with R2 ¼ 0.7. The constant 0.1 in this relation illustrates that as long as the aquifer variability is low (with slnT close to zero) in the nearest well vicinity, the left–right uncertainty is of the order of 10%, even though the considered aquifer as a whole is strongly heterogeneous (in this case, slnT,Tot is 1.6 for the whole aquifer). The relation further shows that the uncertainty increases with a proportionality constant of approximately 0.5 as slnT of the well vicinity increases. In summary, we have here reviewed the main sources of uncertainties in IPT investigations. Some of the uncertainty sources are the same as, or similar to, uncertainties in traditional point measurement interpretations. This is for instance the case for estimations of mass flows on the basis of measured concentrations. However, a main difference between the IPT measurements and the traditional point measurements is that, in the point measurement case, there will for most practical applications always remain an uncertainty as to the contamination situation in between the measurement points. This interpolation uncertainty cannot be quantified unless new measurement points are established. For the case of IPTs, the investigation volume is large and the traditional interpolation uncertainty is transformed into a quantifiable uncertainty related to the (unknown) position of the plume relative to the pumping well. The uncertainty can be quantified if the aquifer property statistics are known or can be estimated at a reasonable level of confidence, as illustrated through the above example. Alternatively, the uncertainties can be reduced by constraining the plume position interpretations (Figure 5.4.6), e.g. on the basis of point concentration data.

5.4.3.1.2

Source Zone Delimiting

The primary outputs of IPT investigations are average concentration values and mass flows of dissolved contaminants over CPs. However, the location of (free phase) sources that give rise to the dissolved contaminant plumes that can be detected at CPs is in many cases unknown. IPT results can aid in delimiting the location of such source zones. In addition, a negative IPT result, i.e. the case that no contaminants are detected in the pumping well, implies that the whole length of the CP is essentially free from contamination. Such a result can be used for delimiting larger upstream areas absent of contaminant sources, which means that these areas can be excluded from further investigation. In the following, we outline and discuss the principles of a technique described in Jarsjo¨ et al.,14 through which source zones and source absence zones can be delimited at a chosen level of confidence, depending on individual or administrative rules and concepts. The source zone delimiting procedure can also be applied to sections of a control plane, defined, for example, by the isochrones corresponding to sampling times within a concentration time series.

254

Chapter 5.4

With respect to delimiting of the position of the source zone, the approach is to use particle backtracking techniques considering different alternative aquifer/boundary condition realisations, thereby accounting for flow model uncertainties (see also the discussion of the previous section and the flow charts below). The principle is illustrated in Figure 5.4.7, in which ‘‘Model 1’’ and ‘‘Model 2’’ represent identified worst-case scenarios with regard to flow directions. Figure 5.4.7(a) shows how particle backtracking then is performed by placing particles along the length of the considered CP and backtracking them to the upstream boundary, hence calculating the upgradient pathway of particles detected at the CP. This yields a distribution of spatial limits in the transverse direction, which is relevant for inert compounds, for which the concentration along streamlines is constant over long distances (kilometres). The resulting outer limits, including streamlines from the union of the Model 1 and Model 2 results, are relevant for delimiting the possible source zone location of inert compounds (Figure 5.4.7(b); thick black line). The inner limits, including streamlines from the intersection of the Model 1 and Model 2 results (Figure 5.4.7(b); thick grey line), are relevant for delimiting the source absence zone for inert compounds at the same level of confidence as for the previous case, as further explained in Jarsjo¨ et al.14 However, most organic compounds are subject to, for example, (bio-) degradation and sorption, which may lead to the development of relatively

Figure 5.4.7

Principle of (a) using particle backtracking from different, equally plausible flow models, to delimit the source zone location (or a source absence zone) at the same significance level, for (b) inert and (c) degrading (reactive) compounds. (After Ref. 14.)


255

short (down to 10 m or so in length) and stable plumes. Since degradable contaminants cannot be detected in the flowing groundwater after a certain downgradient distance from the source (the plume length), predictions about upgradient source absence (for the result that no contaminant was found in the pumping well) are invalid for distances larger than this plume length. Further (for the result that contaminant was found in the pumping well), the source cannot be located further upgradient than this critical length. We here define the plume length as the longest distance from the source, among the bundle of streamtubes leaving the source, to a downgradient point in which the concentration c is lower than some fixed value clow. Recognising that site- and contaminant-specific prediction of attenuation functions is essentially beyond state-of-the-art knowledge, we further make the assumption that it at least is possible to estimate (e.g. based on empirical, historical data) the total plume lengths LMAX and LMIN that will not be exceeded, and will be exceeded, respectively (at some chosen confidence level a), when allowing the plume to develop fully in a given setting (i.e. considering type of contaminant, the aquifer type, ambient physical and chemical conditions, etc.; see Section 5.4.3.3 for an example). For example, Ru¨gner and Teutsch20 provide empirical plume length statistics for different compounds (e.g. benzene, chlorinated hydrocarbons and polyaromatic hydrocarbons), which is useful in estimations of LMAX and LMIN. Alternatively, first-order decay rate models or sophisticated multispecies– multiprocess reactive transport models can, in principle, be used to estimate the required lengths. Using a similar reasoning as for inert compounds, the possible source zone for degrading compounds can now be spatially delimited by considering LMAX and the outer streamline limits (Figure 5.4.7(c); thick black line). Furthermore, the source absence zone for degrading compounds can be delimited by considering LMIN and the inner streamline limits (Figure 5.4.7(c); thick grey line).

5.4.3.2

Decision Tree Approach

Regarding IPT evaluations of mass flows and average concentrations, Figure 5.4.8 shows a general scheme as to how the associated main sources of uncertainty, identified in the previous sections, in practice can be quantified and addressed, using various models and methods, considering a specific site. Also, Figure 5.4.8 indicates how the inversion results can be used for comparison with, for example, regulatory limits, providing a basis for clean-up decisions or related measures. In particular, the uncertainty analyses imply that a set of different, equally plausible, mass flow and average concentration values are obtained, corresponding to different groundwater and contamination models (boxes I to VI in Figure 5.4.8). Relevant statistics describing the set of values are then compiled (box VII in Figure 5.4.8) and compared with regulatory limits (box VIII in Figure 5.4.8), which implies that decisions can be taken at different (chosen) levels of confidence. Furthermore, considering delineations of source zones and source absence zones, Figure 5.4.9 shows that, depending on site-specific conditions and

256

Chapter 5.4 I. For each groundwater model i

II. Run hydraulic model→ Flow velocities vi ; streamlines si III. For each contamination model IV. Perform inversion → mass flow, mi,j ; concentrations ci,j V. Next contamination model j VI. Next groundwater model i VII. Compile statistics of mi,j & ci,j (e.g., mean value and standard deviation of box IV results) VIII. Compare m and c statistics with regulatory limit(s), L. (Example outcome: m < L with 95% confidence.) DECISION

Figure 5.4.8

Flow chart for assessing uncertainty in IPT estimates of mass flows and average concentrations.

general knowledge about the contaminant under investigation, one of the following end results can be expected (grey boxes in Figure 5.4.9). (I) Source zone location can be delimited both in transverse and longitudinal directions. (II) Source zone location can be delimited only in the direction transverse to flow. (III) Source absence zone can be delimited both in transverse and longitudinal directions. (IV) Source zones and source absence zones cannot be delineated for the considered contaminant. Type IV results occur if the considered contaminant is not detected and at the same time the state-of-the-art knowledge is too limited for predicting plume lengths under the ambient conditions. Alternatively, the same result is obtained if the contaminant is not detected and at the same time the present analytical

257

Characterisation of Groundwater Contamination Previous studies of plume lengths under similar conditions?

For each considered compound

Enough data to perform statistical analysis?

yes

yes

no no Contaminant detected?

no yes

no RESULT – type IV: Cannot delimit the source zone, or draw conclusions regarding the absence in groundwater of upgradient contaminant.

Contaminant detected?

yes

Fewer than a handful studies?

yes

RESULT – type II : Source zone can be delimited, but only in the direction transverse to the flow.

Measured concentration above regulatory concentration limit?

LMAX = documented longest plume length LMIN = documented shortest plume length

LMAX = 75 to 99 percentile length LMIN = 1to 25 percentile length

yes

no

RESULT – type I: Source not further upgradient than LMAX. Source zone delimited by LMAX. in combination with particle tracking results.

no

Analytical detection limit considerably below the regulatory concentration limit?

yes

Measured concentration considerably below the regulatory concentration limit?

yes

(Historical) Data shows that contaminant is old enough for the development of a stable or shrinking plume?

yes

no no

RESULT – type IV: Cannot delimit the source zone, or draw conclusions regarding the absence in groundwater of upgradient contaminant.

Figure 5.4.9

no/ no data

RESULT – type III: No source within upgradient distance LMIN. Clean zone delimited by LMIN in combination with particle tracking results.

Flow chart for assessing uncertainty in source zone, or a source absence zone, delineations.

procedures are insufficient for the purpose, or if the plume is young/not at steady state. Note, however, that a no-detect result otherwise (i.e. if analytical procedures, state-of-the art process knowledge and historical records are appropriate) implies that the source can be excluded with high certainty from a specific area (a type III result). A detection of contaminant at a CP implies that the source causing the contamination (of the specific CP) can be delimited

258

Chapter 5.4

in the direction transverse to flow (at different levels of certainty), through particle tracking (a type II result). Furthermore, if the state-of-the-art knowledge of plume lengths under ambient conditions is appropriate, a detection of contaminant implies that the location of the source can be delimited both along and transverse to the direction of flow, at a chosen level of certainty (a type I result).

5.4.3.3

Example of Application at an Industrial Site

Figure 5.4.10 shows a comparison of regulatory concentration limits (State of Baden-Wu¨rttemberg, Germany) and IPT estimations of average benzene concentrations across 19 considered CPs at the application site in southwest Germany (see above). If the contaminant concentration in the groundwater at the CP is above the limit (dark grey CPs in Figure 5.4.10), or even up to one order of magnitude (OM) below the limit (lighter grey CPs in Figure 5.4.10), we consider it contaminated and use the methodology described in previous sections to delimit the location of the contamination source. If, on the other hand, the concentration at the CP is more than one OM below the limit, or if the contaminant was not detected analytically at all, we consider the groundwater clean (corresponding CPs are white in Figure 5.4.10) and use the methodology to delimit zones absent of source.

Figure 5.4.10

Predicted zones delimiting benzene source locations, and zones absent of benzene source, at a field application site in southwest Germany. (After Ref. 14.)


259

In Figure 5.4.10, the borders of the predicted zones delimiting contaminant sources are shown with black lines (thick if the concentration at the CP is above the regulatory limit). Furthermore, the zones predicted to be absent of contaminant source are dashed. Since observations of benzene plumes (which are generally biodegradable) are relatively frequent and well-documented, and statistical analyses of benzene plume lengths are performed and summarised in Ru¨gner and Teutsch,20 we can here present results of type I and III, using the notation of the flow chart in Figure 5.4.9. Figure 5.4.10 shows, as an example, results for a confidence level of 75%. Considering the Ru¨gner and Teutsch20 database and referring to the notation introduced in Section 5.4.3.1.2, the parameter LMAX is taken as an upper threshold value, chosen such that 75% of the reported plume lengths will fall below the threshold. Furthermore, LMIN is taken as a lower threshold value, chosen such that 75% of the reported plume lengths will fall above the threshold. For benzene, the analysis resulted in LMAX ¼ 420 m and LMIN ¼ 60 m. As a consequence, at our field site the benzene source could be delimited to within 420 m in the upstream direction from the measurement well (see wells 5, 9, 10, 20, 32, 62, 64, 2202 and 2058 of Figure 5.4.10), whereas source absence of benzene (for the wells where no benzene was found) could be delimited to within 60 m at the same level of confidence (see wells 2, 3, 7, 15, 16, 17, 66 and 359 of Figure 5.4.10). Performing the analysis at a higher confidence level would imply larger extents (in the flow direction) of the source presence zones and smaller extents of the source absence zones. This methodology for zone delineations at given levels of confidence can be considered a basis for planning of future land use and/or in necessary additional investigation and remediation activities.

5.4.4

Quantification of Natural Attenuation Rates Using Integral Measurements

5.4.4.1

Principles

After screening an area at large scale as described above, if compound-specific total mass flow rates (MF) have been quantified within selected contaminated smaller scale sub-areas at different distances from a contaminant source zone using the integral groundwater investigation method, and if the average travel time Dt in groundwater between the two operated control planes CP(I) and CP(II) is known, it is possible to quantify compound-specific effective firstorder natural attenuation (NA) rates using the following equation (assuming a retardation factor equal to one for the effective values): MFCPðIIÞ ¼ MFCPðIÞ elDt

ð5Þ

MFCPðIIÞ 1 l ¼ ln MFCPðIÞ Dt

ð6Þ

leading to

260

Chapter 5.4

with l [T1], MFCP(I) and MFCP(II) representing the effective natural attenuation rate constant and the measured compound-specific mass flow rates at control planes CP(I) and CP(II), respectively.4 Total mass flow rates and average concentrations can be simultaneously estimated for a number of target compounds at each abstraction well. This may include not only the original contaminants, but also potential degradation products or hydrogeochemical indicators for natural attenuation processes, e.g. pH, EH, sulfate, nitrate, dissolved iron. It should be noted that the differences in mass fluxes between any two control planes can be due to degradation, sorption and volatilisation of the target compound. In general, a calculation of NA rates with eqn. (6) does not allow one to differentiate between these processes. NA rates estimated in this way consequently incorporate all mass flow reducing factors such as sorption and degradation. Recharge and dispersion, however, do not affect the results because of the spatial integration inherent to the mass flow rate estimation. To be able to estimate the relative contribution of the different mass flow reducing processes to the measured effective contaminant mass flux reduction between two control planes, and hence to obtain evidence of degradation, the field-scale measurements must be accompanied by reactive transport modelling.

5.4.4.2

Example of Application at a Former Gasworks Site

The studied former urban gasworks site is situated in a river valley in southwest Germany4 (Figure 5.4.11). The contaminated aquifer is composed of shallow Quaternary gravels with locally embedded sand, silt and loamy clay. Based on pumping tests, the arithmetic average of the hydraulic conductivity at the site was estimated as 2.5 103 m s1.22 Hydraulic heads were monitored over a 3-year period with no indication of significant seasonal changes or temporally variable groundwater flow directions.23 The steady state of the local groundwater flow field is due to the artificial regulation of the water level of the river that runs parallel to the eastern border of the field site (Figure 5.4.11). Transport parameters have been determined in the field employing a natural gradient multi-tracer test.24 At this site, two control planes were installed4 (Figure 5.4.11). The contaminant source is formed by NAPL phase covering an area of approximately 20 000 m2. The NAPL originates from a number of point sources. In the source zone, total BTEX concentrations in groundwater range up to 12 mg l1, whereas PAH concentrations are up to 3.2 mg l1. The resulting PAH plume has an approximate width of 120 m. Of the 16 EPA-PAHs, only acenaphthene shows high concentrations of 190 mg l1 at distances of about 280 m downstream of the source zone. The overall length of this plume is unknown yet, as no monitoring wells are available further downstream. The overall length of the BTEX plume is assumed to be less than 280 m, as only p-xylene showed concentrations exceeding 0.2 mg l1 at control plane 2 (Figure 5.4.11). Figure 5.4.12 illustrates an example of estimated mass flow rates of BTEX and other hydrocarbons based on the analytical inversion of the concentration

261

Characterisation of Groundwater Contamination Estimated PAH-plume extension

?

2

e1

l

ro

nt

Co

Control plane Employed well Existing well NAPL phase Flow direction from tracer tests

an pl

2069 B72

B73

NT01

ne1

Va

l tro

n Co

y lle bo

P1

B41 P2

un

ree St

r ve

ry

Ri

B42

da

N

pla

t

01

50

00m

NAPL free phase

Figure 5.4.11

Site overview with the location of the employed monitoring wells and the installed control planes for the integral groundwater investigation.4

10.0000

Flux [g/d]

1.0000 0.1000 0.0100 0.0010

Figure 5.4.12

ne

ne

de In

B

da In

TM

M B

2, 31,

TM B

4T

2, 1,

PB 1,

3,

5-

PB

yl

yl

X

oIs

o-

nz

X p-

l

Be

To

E-

Be

nz

0.0001

Mass flow rates of BTEX and other aromatic hydrocarbons at control plane 1 (solid bars) and control plane 2.21

262

Chapter 5.4

Effective NA rate constant [1/d]

0.15

0.10

0.05

Figure 5.4.13

e en

an

B

e In d

In d

B

3TM

1,

2,

B

4TM

1,

2,

TM

PB 1,

3,

5-

PB

yl

Is o-

yl

oX

nz

pX

l

Be

To

E-

Be

nz

0.00

Effective natural attenuation rate constants for BTEX and other aromatic hydrocarbons.4

time series measured at the abstraction wells situated at control plane 1 and control plane 2 during the integral site investigation. Based on the measured compound-specific mass flow rates at the two control planes, effective first-order natural attenuation rate constants could be estimated using eqn (6). The results are shown in Figure 5.4.13. The estimated NA rate constants for the BTEX compounds agree well with biodegradation rate constants described in the literature (e.g. Ref. 25). It should be mentioned that it is only the irreversible (bio-)degradation process that really removes mass from the investigated aquifer system. In order to estimate the individual contributions of reactive transport processes such as sorption, (bio-)degradation, etc., to the reduction of the total mass flow rate between the two (or more) control cross-sections, a process-based reactive transport model has to be applied, such as PHT3D26 or MT3D-IPD.27 The numerical models can account for sorption, (bio-)degradation, etc., as well as for the heterogeneity of both the hydrodynamic and reactive transport (e.g. sorption) parameters. The individual contributions of contaminant mass flow reducing reactive transport processes such as sorption and (bio-)degradation to the total mass flow reduction can be quantified by comparing the measured contaminant mass flow reduction with the results from the numerical model within a combined forward-inverse modelling framework.

5.4.5

Multilevel Integral Investigation of Contamination

5.4.5.1

The Multilevel Integral Investigation Method

At large scale, the IPT method is usually applied in a two-dimensional depthaveraged approach, i.e. assuming the concentrations and the IPT capture zone extent to be constant over the aquifer thickness. However, in many cases


263

contamination may be limited to distinct aquifer levels, and, due to aquifer layering, the capture zone width may vary with depth. Therefore, after screening an area at large scale as described above, for investigations at selected contaminated smaller scale sub-areas the IPT method was improved to allow multilevel measurements and to account for aquifer layering.28,29 Employing the improved investigation method, site assessment and the total contaminant mass flow rate to be removed by a remediation measure can now also be obtained with a vertical resolution. The concept of the improved integral investigation method for multilevel measurements is shown in Figure 5.4.14. To obtain the required multilevel concentration time series, a flow separation technique was developed, allowing multilevel groundwater sampling also within fully screened pumping wells.28 The algorithm used for the numerical inversion of the measured multilevel concentration time series at the pumping wells is implemented in the program code CSTREAM.5 In case of multilevel concentration time series, the twodimensional, depth-integrated formulation5 (see also above) is applied to distinct aquifer layers, which are defined by the vertical extent of the multilevel sampling sections. This layered approach is applicable to situations with no significant vertical mass transport.

5.4.5.2

Example of Application

The multilevel integral investigation method was first applied at a former chlorobenzene manufacturing site in Germany. At this site chlorobenzene concentrations in groundwater of up to 55 mg l1 could be measured. The porous aquifer has a thickness of about 15 m with a mean hydraulic conductivity of 5 1041 103 m s1. A total of three pumping wells, each with four multilevel sampling sections and a maximum pumping rate of 7.5 l s1, were positioned along a control cross-section of about 140 m length. The pumping time per well was up to about 8 days. During pumping, groundwater concentration time series of BTEX, chlorobenzene and chlorotoluene compounds were measured at each of the pumping wells and at each multilevel sampling section with variable sampling intervals, allowing a spatial resolution of about 1–2 m along the control plane. Further details are given in Ptak et al.[28] Following the measurements, the multilevel concentration time series were evaluated using the CSTREAM code, and a local-scale three-dimensional flow and transport model of the site. As an example of the results, Figure 5.4.15 shows the relative distribution of mean concentrations in the vertical direction across the control plane for benzene and 1,2-dichlorobenzene. The multilevel capture zones, the concentration values and the mass flow rates obtained from the multilevel inversion are significantly variable in space, indicating an irregular distribution of contaminant mass within the plume. Distinct zones with relatively high concentrations can be identified. It should be noted that the maximum concentrations may be found in aquifer zones between the pumping wells, and that the high concentration and the high mass flux zones are not necessarily coincident.

264

Chapter 5.4 Pumping tests with measurement of multilevel concentration time series Well i+1

Well i

Position z j Position z j+1

Mass flux

Contaminant source

Control plane

Well i+1

Well i

Position z j+1

C

C Position z j+1

Position z j

t0

t1

Position z j

t1

t2

Multilevel concentration time series (compound specific)

Multilevel inversion algorithm based on a 3D numerical flow and transport model

Level-oriented contaminant mass fluxes and concentrations at control plane

Figure 5.4.14

Multilevel integral investigation method.29

Employing the multilevel integral investigation method, the spatial distribution of contaminant concentrations and mass flow rates across control planes can be estimated without the need for a dense sampling network. This information can then be used for example to obtain an optimal design of remediation measures by focusing on the highly contaminated aquifer zones. Backtracking starting from the identified high mass flux sections at the control plane allows a more detailed local-scale delimiting of the contaminant source

265

Characterisation of Groundwater Contamination 1,2-DCB (BLACK=2 mg/L)

Benzene (BLACK=14 mg/L) 41

40

38

75

75

70

70

65

65

60

60

55

55

50 5720100

5720150

Figure 5.4.15

5720200

5720250

5720300

50 5720100

41

40

38

5720150

5720200

5720250

5720300

Relative distribution of mean concentrations in the vertical direction across the control plane for benzene and 1,2-dichlorobenzene (1,2-DCB) (greyscale related to the maximum concentration shown in black).29

zone, compared to the two-dimensional approach described above. When applied upstream and downstream of a remediation measure, the efficiency of the remediation can be quantified by comparing the total upstream and downstream mass flow rates. Multilevel integral measurements at two or more control planes positioned in the downstream direction allow the quantification of the natural attenuation potential at a high spatial resolution. In this way the new multilevel integral investigation approach may significantly contribute to an improvement of contaminated land assessment and revitalisation.

5.4.6

Conclusions

Employing the control plane-based integral investigation method, the compound-specific average contaminant concentration, the spatial distribution of concentration values and mass flow rates along a control plane, as well as the total contaminant mass flow rates downstream of an area under investigation can be estimated quickly and with a high level of certainty. The information obtained from this analysis can be considered a basis for planning of future land use. The results from the integral investigation can be used for risk assessment purposes, for the quantification of the natural attenuation potential and for the design of remediation measures. In addition, a consistent quantification of uncertainties in the results from the application of the integral groundwater investigation method is possible, considering uncertainty in the boundary conditions and uncertainty in the hydraulic property values of the aquifer. Finally, the delimiting of the source zone extent and its uncertainty allows one to define priorities for further investigation measures at a smaller scale and to

266

Chapter 5.4

develop cost-optimised clean-up strategies at sites with complex contamination patterns. In this way, employing the large- to small-scale screening procedure it is possible to obtain optimal results for the efforts spent. Therefore, the new approaches may become a basis for the development and implementation of EU directives on contaminated land assessment and revitalisation in urban industrial areas.

Acknowledgements Financial support for the presented work was provided by the European Union, by the Landesanstalt fu¨r Umweltschutz Baden-Wu¨rttemberg, by the Projekt Wasser-Abfall-Boden Baden-Wu¨rttemberg, by the BMBF, by the Deutsche Forschungsgemeinschaft and by the UFZ Leipzig-Halle GmbH. The authors gratefully acknowledge the contributions of Luca Alberti, Marti Bayer-Raich, Sebastian Bauer, Diego Bianchi, Sara Ceccon, Philippe Elsass, Thomas Ertel, Jadwiga Gzyl, Thomas Holder, Hermann J. Kirchholtes, Christian Kolesar, Dietmar Mu¨ller, Caterina Padovani, Georgia Spausta, Gilles Rinck, Gerhard Scha¨fer, Maria Giovanna Tanda, Georg Teutsch and Andrea Zanini within the INCORE project.

References 1. G. Teutsch, T. Ptak, R. Schwarz and T. Holder, Grundwasser, 2000, 4, 170–175. 2. T. Ptak, R. Schwarz, T. Holder and G. Teutsch, Grundwasser, 2000, 4, 176–183. 3. T. Ptak and G. Teutsch, Development and application of an integral investigation method for the characterization of groundwater contamination, in Contaminated Soil 2000, Thomas Telford, London, 2000, pp. 198–205. 4. A. Bockelmann, T. Ptak and G. Teutsch, J. Contam. Hydrol., 2001, 53, 429–453. 5. M. Bayer-Raich, J. Jarsjo¨, T. Holder and T. Ptak, Numerical estimations of contaminant mass flow rate based on concentration measurements in pumping wells, ModelCare 2002: A Few Steps Closer to Reality, IAHS Publication no. 277, 2003, pp. 10–16 (ISBN 1-901502-07-4). 6. M. Bayer-Raich, Integral pumping tests for characterization of groundwater contamination, PhD thesis, Center for Applied Geoscience, University of Tu¨bingen, 2004. 7. R. Schwarz, Grundwasser-Gefa¨hrdungsabscha¨tzung durch Emissions- und Immissionsmessungen an Deponien und Altlasten, PhD thesis, Center for Applied Geoscience, University of Tu¨bingen, 2002. 8. S. Bauer, T. Holder, M. Bayer-Raich, T. Ptak, Ch. Kolesar and D. Mu¨ller, J. Contam. Hydrol., 2004, 75, 183–214. 9. S. Bauer, M. Bayer-Raich, T. Holder, J. Jarsjo¨, T. Ptak and G. Teutsch, The integral groundwater investigation method: inversion of


10.

11.

12. 13.

14. 15.

16. 17. 18. 19. 20.

21.

22. 23. 24.

267

concentration-time series and field application at the INCORE Strasbourg site, in IMAGE-TRAIN, Innovative Management of Groundwater Resources in Europe, ed. G. Prokop, Training and RTD Coordination Project, Proceedings of the 1st IMAGE-TRAIN Cluster Meeting, Karlsruhe, Germany, Federal Environment Agency, Vienna, 2002, pp. 75–79. M. G. McDonald and A. W. Harbaugh, MODFLOW. A modular threedimensional finite difference groundwater flow model, investigation of United States Geological Survey, Washington, DC, 1988. D. W. Pollock, User’s Guide for MODPATH/MODPATH-PLOT, Version 3: a particle tracking post-processing package for MODFLOW, US Geological Survey finite difference ground-water flow model, US Geological Survey, 1994. M. Bayer-Raich, J. Jarsjo¨, R. Liedl, T. Ptak and G. Teutsch, Water Resour. Res., 2004, 40, W08303. T. Holder, T. Ptak, R. Schwarz and G. Teutsch, Groundwater risk assessment at contaminant sites. A new approach for source zone characterization: the Neckar valley study. Groundwater ouality: remediation and protection, IAHS Publication no. 250, Tu¨bingen, 1998. J. Jarsjo¨, M. Bayer-Raich and T. Ptak, J. Contam. Hydrol., 2005, 79(3–4), 107–134. A. Peter, Assessing natural attenuation at field scale by stochastic reactive transport modelling, PhD thesis, Center for Applied Geoscience, University of Tu¨bingen, 2002. M. Bayer-Raich, J. Jarsjo¨, R. Liedl, T. Ptak and G. Teutsch, Water Resour. Res., 2006, 42, W08411. A. Zeru and G. Scha¨fer, J. Contam. Hydrol., 2005, 81, 106–124. M. Bayer-Raich, J. Jarsjo¨ and G. Teutsch, J. Contam. Hydrol., 2007, 90, 240–251. J. Jarsjo¨ and M. Bayer-Raich, Water Resour. Res., in press. H. Ru¨gner and G. Teutsch, Literature study, Natural attenuation of organic pollutants in groundwater, Final Report for EU-FP5 project INCORE, 2001. A. Bockelmann, T. Ptak, R. Liedl and G. Teutsch, Mass flux, transport and natural attenuation of organic contaminants at a former urban gasworks site, in Prospects and Limits of Natural Attenuation at Tar Oil Contaminated Sites, Dechema eV Texte, Frankfurt am Main, 2001, pp. 325–336. M. Herfort, T. Ptak, O. Hu¨mmer, G. Teutsch and A. Dahmke, Grundwasser, 1998, 3(4), 159–166. M. Herfort, Reactive transport of organic compounds within a heterogeneous porous aquifer, PhD thesis, Universita¨t Tu¨bingen, 2000. D. Bo¨sel, M. Herfort, T. Ptak and G. Teutsch, Design, performance, evaluation and modelling of a natural gradient multitracer transport experiment in a contaminated heterogeneous porous aquifer, in Tracers and Modelling in Hydrogeology, ed. A. Dassargues , IAHS, Liege, 2000, pp. 45–51.

268

Chapter 5.4

25. T. H. Wiedemeier, H. S. Rifai, C. J. Newell and J. T. Wilson, Natural Attenuation of Fuels and Chlorinated Solvents in the Subsurface, Wiley, New York, 1999. 26. H. Prommer, D. A. Barry and C. Zheng, Ground Water, 2003, 42(2), 247–257. 27. R. Liedl and T. Ptak, J. Contam. Hydrol., 2003, 66(3–4), 239–259. 28. T. Ptak, M. Bayer-Raich and S. Bauer, Grundwasser, 2004, 4(9), 235–247. 29. T. Ptak, M. Bayer-Raich and S. Bauer, Multilevel integral investigation of contamination in large polluted aquifers, in MODFLOW and MORE 2006: Managing Ground Water Systems, Conference Proceedings, International Ground Water Modeling Center (IGWMC), Colorado School of Mines, 2006, pp. 574–578 (www.mines.edu/igwmc/).

CHAPTER 5.5

Improved Risk Assessment of Contaminant Spreading in Fractured Underground Reservoirs CHRISTOS D. TSAKIROGLOU FORTH/ICE-HT, Stadiou Street, Platani, PO Box 1414, GR-26504 Patras, Greece

5.5.1

Introduction

Soil and groundwater contamination by hazardous substances is tending to become one of the most significant problems for the environmental and economic policies of the European Union (EU). Fractures are widespread in various types of soils and rocks (e.g. clay till, sandstone, chalk, granite, limestone) and influence drastically the pathways of liquid pollutant migration in the subsurface with the potential contamination of underground aquifers. In the past, numerical simulators were developed to forecast the various contaminant transport (e.g. gravity flow, sorption, dissolution/dispersion) and reaction (e.g. biodegradation) processes taking place in the fractured zones of the subsurface. Such simulators may be used as tools in the risk assessment as well as in the design of remedial actions on fractured sites contaminated by non-aqueous phase liquids (NAPLs). Nevertheless, in spite of the progress that has been accomplished in the development of intelligent numerical solvers of the equations modelling the contaminant transport/biodegradation in fractured media, there is still a lack of fundamental knowledge concerning: the quantitative description of fractured media at different scales (e.g. single fractures, fracture network); the correlation of the NAPL transport pathways with fracture morphology and NAPL rheology (mobility); 269

270

Chapter 5.5

the quantification of the up-scaled transport coefficients of highly heterogeneous fractured media in terms of reliable phenomenological models; and the introduction of the foregoing information into macroscopic numerical simulators to produce reliable data for the risk analysis of contaminated fractured sites.

5.5.1.1

Literature Review

The design and implementation of remediation strategies for fractured soils and aquifers contaminated by NAPLs is one of the most intractable problems. In most cases, excavation down to the fractured rock, soil or sediment and the removal of the contaminated material are very expensive tasks. Potential remediation alternatives are dewatering of the contaminated zone at high pumping rates and removal of the volatile NAPLs through soil vapour extraction,1 surfactant-enhanced NAPL dissolution and mobilisation,2 in situ bioremediation,3 steam injection,4 etc. The design and installation of the most suitable remediation scheme on a contaminated fractured site requires information about the spatial and temporal distribution of pollutants throughout the subsurface. Macroscopic simulators of NAPL transport in the subsurface offer a cost-effective method for the long-term mapping of the distribution of pollutants in fractured sites.5,6 The characterisation of a contaminated fractured site at the scale of single fractures and fracture networks is a prerequisite for the determination of the transport properties of such media.7–9 Contaminant transport in fractured permeable formations is typified by the interaction of the hydraulic properties of fractures and matrix. In most cases, the permeability of the fracture network is much higher than that of the host rock (matrix), but most of the capacity for storing a pollutant is provided by the matrix porosity.10 The simulation of the contaminant transport in such formations is based on dual continuum models which separate the heterogeneous formation into two homogeneous media: one representing the fracture system and one representing the matrix.11 Depending on whether the permeability of the matrix is neglected or taken into account, the model is characterised as dual porosity/single permeability (DPSP) or dual porosity/dual permeability (DPDP).12–15 The abovementioned approach minimises the computational requirements for field-scale simulations, with the accuracy of the numerical predictions depending strongly on the use of representative up-scaled transport properties8,16 and reliability of the mathematical models.17 Much attention has been focused on the modelling of NAPL migration through fractured and low-permeability media with significant matrix porosity, such as clay till. Simulating a scenario with a sandy aquifer situated beneath a fractured clay aquitard showed that the lower aquifer was vulnerable to contamination from DNAPLs leaking on the ground surface; the DNAPL migration pathways towards the aquifer are mainly governed by the fracture aperture, matrix porosity and form of relative permeability curves. Whether or

Improved Risk Assessment of Contaminant Spreading

271

not the DNAPL can enter and persist within the underlying aquifer depends on a variety of factors such as the DNAPL release rate and its composition at the source, the thickness of the fractured clay layer and the diffusion/sorption properties of clay.18 The presence of sand lenses intersecting the fractured clay has been found to increase the time required for the non-wetting phase (DNAPL) to migrate through the vertical extent of a clay sequence from a few days to several years.19 Parametric analysis of the influence of the main fracture characteristics (mean aperture, roughness, and correlation length) on the hydraulic properties of single fractures has shown that the mean fracture aperture is the most important parameter affecting strongly the permeability and entry pressure.20 The drainage/imbibition hysteresis effect increases with increasing capillary number.20

5.5.2

Objectives and Approach of the TRACE-Fracture Project

The overall objective of the EU-funded TRACE-Fracture project (1/2/2000-31/ 1/2003, contract no. EVK1-CT-1999-00013)21–24 was (1) to develop a novel method for the characterisation of fractured media at the scales of single fractures and fracture networks, (2) to develop reliable and predictive phenomenological models that provide the single-phase flow, two-phase flow and solute dispersion effective coefficients of fractured porous media as functions of fracture morphology, fluid rheology and hydrodynamics, (3) to integrate the new phenomenological models into a novel numerical simulator of the macroscopic contaminant transport in fractured underground reservoirs, (4) to integrate the new numerical tool into a generalised methodology of risk assessment and rational design of remedial strategies for contaminated fractured aquifers and (5) to implement the results in two geologically different fractured sites contaminated by NAPLs. The general concept of the approach used in the TRACE-Fracture project is illustrated in Figure 5.5.1. Accurate geostatistical properties and reliable effective transport coefficients of fractured soils and rocks are determined at multiple scales ranging from single fractures to fracture networks, by combining properly data from field-work and photo-geological analysis with laboratory-scale experiments and computational methods. This information is integrated into an updated numerical simulator of the organic pollutant transport in fractured media (SIMUSCOPP). The simulator is used as a tool for the cost-effective calculation of the spatial and temporal distribution of the saturation of the bulk NAPL in unsaturated/saturated zones, and concentration of NAPL compounds in groundwater. Long-term numerical predictions of the chemical status of groundwater under various scenarios of pollution that simulate the site contamination history are coupled with additional information (e.g. exposure pathways, potential receptors, toxicity of substances) for the risk assessment of fractured sites contaminated by organic pollutants. Two very different highly heterogeneous fractured sites which have been contaminated by

272

Chapter 5.5 Fi Fieldwork on site & llab-scale experiments multig eological multi- scale geological informat ion

History matching of ma experiments expe

Risk Ri assessment of of f fractured sites

Por Pore-network si simulators

Dual Dual-porosity macroscopic simulator

Figure 5.5.1

Concept of TRACE-Fracture project.

the waste oils of industrial facilities were investigated to assess the risks threatening human health and the ecosystem: one site overlying clay till sediment and situated in an urban area of Denmark (Ringe site) and one site overlying granite rock and situated in northern Spain (Spanish site). The results of the project might be helpful in formulating protocols for the risk assessment of fractured contaminated sites, designing cleanup strategies and setting the boundaries of the protection zones for aquifers underlying fractured areas. In the following, our attention is focused on the Ringe site, whereas the application of the methodology to the Spanish site has been published elsewhere.15

5.5.3

Description of Ringe Site

The investigated Ringe site is situated in an abandoned asphalt and creosote factory at Ringe, on Funen island, Denmark. The subsurface consists of clay till overlaying a primary sandy aquifer. The clay till is approximately 8–12 m thick and is dissected by multiple fractures, which are responsible for the transport of pollutants through a normally tight layer. The site was contaminated by leaking storage tanks of a creosote and asphalt factory (1929–1962) as well as by waste oils of several companies and automobile workshops (1962–1988). In 1988, it was discovered that the subsurface of the site was strongly contaminated by creosote, various compounds of which were traced in the aquifer, 22 m below ground, and a long-term site remediation programme was adopted by the Danish Environmental Protection Agency. The site was used for field studies in a number of research projects (1994–2003): a great number of wells were established and four open pits were excavated. A geological model was developed (Figure 5.5.2), and numerous investigations were performed to collect data concerning the hydraulic and solute transport


Figure 5.5.2

273

Macropore distribution on a representative region of the Ringe site.

characteristics of clayey fractured sediments.25,26 It is worth mentioning that clay till is a sediment that is widespread in the subsurface of regions covered by ice during earlier glacial times, such as Canada, the USA, northern Europe, as well as all Alpine and Alpine marginal areas.27

5.5.3.1

Conceptual Fracture Network Model

The upper 5 m of till may be separated into three zones differing with respect to the characteristic distribution of pores and fractures (Figure 5.5.2): (i) an upper zone (0.5–2.5 m below ground surface) dominated by bio-pores (burrows), desiccation fractures and a highly porous matrix; (ii) a central zone (2.5–4 m

274

Chapter 5.5

below ground surface) dominated by well-connected desiccation and glaciotectonic shear fractures; and (iii) a lower zone (44 m below ground surface) dominated by glaciotectonic shear fractures. Three distinct fracture systems as well as a number of randomly oriented fractures were recognised in the upper layer (0–5 m below ground surface) (Figure 5.5.2). Systematic measurement of the aperture of desiccation and glaciotectonic fractures was done by analysing 2D BSEM images of resin-impregnated samples (Figure 5.5.3). It was revealed that the fracture aperture resembles a 2D network of elliptical channels and such a model was employed in experimental and theoretical approaches to estimate the single- and multiphase effective transport coefficients at the scale of a single fracture. Based on the foregoing information and statistics of fracture intensity/ spacing, a conceptual model of the fracture network/porous matrix system was established (Figure 5.5.4). The weathered microporous matrix is dominant

Sample: SF-2D

1000 µm

Figure 5.5.3

BSEM image of resin-impregnated single fracture. The 2D aperture area was digitized with ScanPro 5.0 and appears dark. Matrix Matr zones

Clay Till High porosity CaCO3-poor

Clay Till Medium porosity CaCO3-rich

Macro-pore zones Borrows and rootholes

Desiccation fractures and tectonic fractures

Tectonic fractures

Sand

Figure 5.5.4

Fracture network/porous matrix model of Ringe site.


275

in the upper zone (o2.5 m below ground surface). In this zone, the limestone was dissolved thus leaving a highly porous matrix with a porosity ranging from 0.3 to 0.45 (Figure 5.5.4). The unweathered matrix is dominant in the lower layers, and consists of firmly consolidated clay till with a weight fraction of limestone close to 0.25–0.3 and porosity ranging from 0.25 to 0.30 (Figure 5.5.4). In depths between 3.5 m and 15 m sand-lenses of higher permeability appear.

5.5.4

Hierarchical Methods for the Determination of Transport Properties

Experimental techniques and numerical methods were developed for the determination of the single-phase transport properties of fractured media. With the aid of the critical path analysis (CPA) of percolation theory and effective medium approximation (EMA), accurate phenomenological models were developed to relate explicitly the absolute permeability and electrical formation factor of single fractures with microscopic properties of their aperture.28 The results of the geological characterisation (Figure 5.5.4) were employed for the computer-aided construction of networks of glaciotectonic and desiccation fractures intersecting the clay till29 (Figures 5.5.5(a)–(d)). The up-scaled transport properties of fractured clay till (Figures 5.5.5(a)–(d)) were calculated by using hierarchical pore network simulations (Table 5.5.1) and were found to be comparable to results of field- and laboratory-scale hydraulic tests (Table 5.5.2).25,26 Phenomenological models were developed to relate the flow velocity of inelastic shear-thinning NAPLs (e.g. asphalt, emulsions of creosote with water, crude oil) in single fractures and fractures embedded into porous matrices with NAPL rheology (e.g. parameters that specify the viscosity as a function of shear rate) and fracture aperture properties.30,31

5.5.4.1

Multiphase Transport Coefficients of Single Fractures and Fracture Networks

Experimental procedures and numerical methods were developed to determine the two-phase flow properties (capillary pressure and relative permeability curves) of fractured media. Network-type hierarchical simulators of the immiscible displacement of an aqueous phase by a NAPL in single fractures, single fracture embedded into porous matrix and fracture networks were developed to compute the up-scaled relative permeability and capillary pressure curves29 (Figures 5.5.5(e) and (f)). Visualisation experiments of drainage and imbibition performed on glass micromodels revealed that the transient immiscible displacement growth patterns in fractures are affected strongly by NAPL rheology, flow rates and fracture surface wettability32–34 (Figure 5.5.6). With the aid of numerical algorithms of history matching, the capillary pressure (Figures 5.5.7(a) and (d)) and relative permeability (Figures 5.5.7(b) and (e)) curves of porous and fractured media were estimated simultaneously from datasets of unsteady-state displacement experiments and were found strongly

276

Chapter 5.5

–2

–2

–2.5 –3

–2.5 –3

–3.5 –4

–3.5

–4.5 –5 10

–4.5

–4

–5 10

8 6 4

(a)

8

6 2 0 0

2

8

10

6 4

4

2

2

0 0

(b)

10

8

6

4

–2 –2.5 –3 –3.5 –4 –4.5 –5

(c)

0

(d)

regions 1-2-3 region 1 region 2 region 3 region 4

0.01

1E-3 0.0

(e)

Figure 5.5.5

10

9

8

7

6

5

4

3

2

1

0

1.0

Relative Permeability, krw, k rnw

Capillary pressure, Pc (bar)

0.1

5

0.2

0.4

0.6

Water Saturation, Sw

0.8

0.8

0.6

0.4

(f)

regions 1-2-3 region 1 region 2 region 3 region 4

0.2

0.0 0.0

1.0

k rw

k rnw

0.2

0.4

0.6

0.8

1.0

Water Saturation, Sw

Generation of the two fracture systems. (a) Fracture system 1: desiccation fractures. (b) Fracture system 2: tectonic fractures. (c) Upper view and (d) side view of the fracture network; region 1 consists of tectonic (light grey) and desiccation (mid grey) fractures (number ratio ¼ 1); region 2 consists of tectonic (light grey) and desiccation (dark grey) fractures (number ratio ¼ 1/3); region 3 consists of tectonic (light grey) and desiccation (mid grey) fractures (number ratio ¼ 1/6); region 4 consists only of tectonic fractures. (e) Capillary pressure and (f) relative permeability curve for the entire fracture network, and sub-networks corresponding to different depth intervals (Table 5.5.1).

277


Table 5.5.1

Calculated fracture network permeability at various depths (Figure 5.5.5).

Permeability (mD)

Three regions

Region 1

Region 2

Region 3

Region 4

Kvertical//gravity Khorizontal//tectonic Khorizontal//desiccation

1073

1372 836 230

775 594 76

538 514 29

372 372 0

Table 5.5.2 Depth (m) 0–2 2–4 4–6 6–13

Figure 5.5.6

Experimental values of the permeability over the various zones of Ringe site. Test Infiltration tests Column tests Infiltration slug tests Falling head tests Infiltration tests

Hydraulic conductivity (m s1) 5

4

1.5 10 1.5 10 3 106 106–1.2 105 3 107 1.2 107

Permeability 1.5–15 Da 300 mD 100 mD–1.2 Da 30 mD 12 mD

Final fluid distribution after the displacement of NAPL by water in a dual-pore network (artificial glass-etched model of a single fracture surrounded by the clay till porous matrix; the displacement is from left to right): (a) the NAPL is paraffin oil (Newtonian fluid); (b) the NAPL is a suspension of ozokerite (natural wax) of concentration 2% in paraffin oil (shear-thinning fluid).

sensitive to the ratio of viscous to capillary forces (capillary number) for Newtonian (Figure 5.5.7(c)) and shear-thinning (Figure 5.5.7(f)) NAPLs.33–36 A fully automated technique, based on the high sensitivity of the colour intensity of an aqueous solution to pH, was devised35,37 for performing solute dispersion visualisation experiments on transparent porous media models, and measuring precisely the transient solute concentration profiles with image analysis (Figure 5.5.8(a)). Chaotic, buoyancy-driven solute dispersion regimes were identified38 at the scale of pore network/single fracture (Figures 5.5.8(b)–(d)). An inverse modelling method was developed for the simultaneous estimation of the longitudinal and transverse dispersion coefficients of fractures from transient solute concentration profiles measured with miscible displacement and single-

278

Chapter 5.5 Relative permeability, krw krnw

Capillary pressure, Pc (Pa)

3000 2000

1000 900 800 700 600 500

(a)

400 300 0.0

0.2

0.4

0.6

0.8

1.0

Non-wetting phase saturation, Snw Relative permeability, krw krnw

Capillary pressure, Pc (Pa)

1000 900 800 700 600 500 400 300 0.0

(d) 0.2

0.4

0.6

0.8

1.0

Non-wetting phase saturation, Snw

Figure 5.5.7

0.9 0.8 0.7 0.6

krnw

krw

0.5 0.4 0.3 0.2

(b)

0.1 0.0 0.0

0.2

0.4

0.6

0.8

(c) 1.0


3000 2000

1.0

1.0 0.9 0.8 0.7

k rw

0.6

k rnw

0.5 0.4 0.3 0.2

(e)

0.1 0.0 0.0

0.2

0.4

0.6

0.8

(f)

1.0


Estimated two-phase (a) capillary pressure and (b) relative permeability curves from (c) transient data of the displacement of an aqueous (dark grey) phase by a Newtonian (light grey) NAPL (paraffin oil), at various values of the capillary number. Estimated two-phase (d) capillary pressure and (e) relative permeability curves from (f) transient data of the displacement of an aqueous (dark grey) phase from a shear-thinning (light grey) NAPL (ozokerite 1.5% in paraffin oil), at various values of the capillary number.

source solute transport experiments.37 The longitudinal dispersion coefficient increases and tends to be dominated by macrodispersion as the variability of the fracture aperture is enhanced (Figures 5.5.9(a) and (b)) and is correlated with fracture morphology and Peclet (ratio of convective to diffusive flux) number.35 Conclusively there is a variety of parameters that should be taken into account in the deternination of the effective transport coefficients of fractured media: (1) the geometry and topology of the aperature of single fractures; (2) the morphology of fracture networks; (3) the rheology of NAPL; (4) the ratio of viscous to capillary forces (capillary number); and (5) the ratio of convective to diffusive flux (Peclet number).

5.5.5

Numerical Modelling of NAPL Fate in Unsaturated and Saturated Zones

A macroscopic numerical tool was developed for quantifying the NAPL spreading in fractured clay till sediments and underlying aquifers. The macroscopic


Figure 5.5.8

279

Visualization single source-solute transport experiments performed on a single fracture. A low solute concentration aqueous solution (groundwater/dark colour) flows steadily from the left to the right, whereas a high solute concentration aqueous solution (pollutant/light colour) is injected at a low flow rate through a hole (single source). (a) Steady-state solute dispersion regimes across the single fracture at various values of Peclet (Pe) number (Pe ¼ u0lp/Dm, u0 ¼ pore velocity, lp ¼pore length, Dm¼solute/solvent diffusion coefficient) without the action of gravity. (b-d) Buoyancy-driven (chaotic) successive steady-state solute dispersion regimes across a single fracture at various Pe values (the gravity acts vertically to the main flow direction and lobe-shaped instabilities are created by the downward flow of the heavier liquid and the upward flow of the lighter liquid).

simulator SIMUSCOPP (owned by Institut Francais du Petrole) was updated to simulate the contaminant transport in fractured porous media.39 The effective transport coefficients of the clay till fracture systems were up-scaled to the block size of the numerical grid.29 The geological model was transformed to an equivalent macroscopic dual porosity–dual permeability numerical model of the fractured clay till site.29 The spatial and temporal distribution of NAPL saturation in the unsaturated zone of clay till was determined with the aid of SIMUSCOPP for the following pollution scenario: creosote was leaking from tanks for a period of 30 years on a surface of 25 m2, at a constant flow rate of 3 105 m3 per day. The migration pathways of pollutants over the period of 60 years were simulated29 (Figure 5.5.10(a)). The matrix is highly saturated by

280

Chapter 5.5 x

x y

y Solute injection

1

Solute injection

2

1000

(a)

1000

3

DL/Dm=0.04 Pe1.48 100

DL/Dm=0.087 Pe0.9

DL/ Dm

DL/ Dm

2

(b)

DL/Dm=7.4e-4 Pe1.78

100

10

1

DL/Dm=0.21 Pe1.06 10 Dual pore network

1

Single pore network

1

Estimated values Taylor dispersion in large pore Taylor dispersion in small pore

Estimated values Taylor dispersion in a pore

0.1 1

10

100

1000

Pe=up0Lp /Dm

Figure 5.5.9

10000

0.1 1

10

100

1000

Pe=up0Lp /Dm

Longitudinal dispersion coefficient estimated from miscible displacement experiments performed on (a) simple (single fracture) and (b) dual (single fracture surrounded by matrix) pore networks, as function of Peclet number.

water and is bypassed by the NAPL, most of which flows downwards through the fracture network pathways (Figure 5.5.10(a)). Moreover, in the fractured zone, the NAPL is retained in the sand lenses rather than in the clay till, which is fully saturated by water (Figure 5.5.10(a)). In the simple medium simulation, contrary to the dual medium simulation results, NAPL does not reach the aquifer (Figure 5.5.10(b)). Without the presence of the fracture network, an important back flow of NAPL and water occurs at the surface (soil/air interface). The water-saturated and low-permeability matrix acts as a barrier that prevents the downward NAPL flow (Figure 5.5.10(b)). The dual-porosity simulations performed on the cross-section (Figure 5.5.10(a)) provided the NAPL flux towards the aquifer. This flux was used as a boundary condition in the single-porosity SIMUSCOPP simulator to predict the transport of the dissolved compounds in the aquifer. The groundwater pollution at a distance 500 m downstream from the source was examined. The aquifer was modelled as a 2D cross-section 770 m long (500 m downstream and 250 m upstream from the source), and 10 m deep. The spatial and temporal distribution of naphthalene and phenol concentration in the homogeneous sandy aquifer (saturated zone) underlying the fractured clay till zone was determined with the aid of SIMUSCOPP (Figure 5.5.11). The predicted concentrations of naphthalene and phenol


Figure 5.5.10

281

Simulation of the spatiotemporal evolution of NAPL saturation within the unsaturated zone of wet clay till (a rainfall rate equal to 5.5 m3 per year or equivalently 73 mm annual recharge was assumed for the entire surface of 25 m2): (a) simulation by accounting for the presence of fractures; (b) simulation by ignoring the existence of fractures.

in groundwater exceed significantly the maximum ones measured on the field, because some parameters introduced in the simulator, such as the weight fractions of the compounds in creosote and the rate of NAPL flux on the groundwater table, might be overestimated.

5.5.6

Risk Assessment of Contaminated Sites

The RAGS (Risk Assessment Guidance for Superfund) approach of the US Environmental Protection Agency (EPA),40 focused on human health risk assessment, was selected for implementation in the TRACE-Fracture project. The risk analysis included the following steps: (1) a hydrological model of Ringe site was established; (2) the database containing field measurements of the spatial and temporal distribution of the concentration of oil pollutants (PAH, phenols, BTEX) in the subsurface of Ringe site was revised and updated (Table 5.5.3; Figure 5.5.12); (3) the current practices were combined with the numerical predictions of the updated SIMUSCOPP model (Figure 5.5.11); and (4) risk analysis of the threats posed to potential receptors of the contaminated groundwater of clay till site was done. By using field measurements (Table 5.5.3) and simulation results (Figure 5.5.11), along with RBCA software, there were found risks associated with the high levels (exceeding the Dutch intervention limits) of several species (BTEX, naphthalene, phenols) concentration in soil and groundwater, at long distances from the source of

282

Chapter 5.5

Figure 5.5.11

Simulation of naphthalene and phenol dissolution/dispersion in the primary sandy aquifer underlying the fractured clay till site.

pollution. The latest available data (Table 5.5.3) indicate that the contaminants have travelled long distances (hundreds of metres) away from the site (Figure 5.5.12). However, the groundwater is used by the local community as a source of drinking water and hence effective methods are to be selected to clean up a volume of about 5000 m3 of contaminated soil.

5.5.6.1

Site Remediation

The following technologies were screened as alternatives to decontaminate the Ringe site: (1) bioremediation, (2) bioventilation (3) enhanced bioremediation, (4) chemical oxidation, (5) soil flushing, (6) thermal treatment, (7) electrical resistance heating, (8) radio frequency/electromagnetic heating and (9) hot air/ steam injection. Moreover, the following technologies were examined as alternatives for groundwater remediation at the Ringe site: (1) enhanced bioremediation, (2) slurry wells, (3) pump-and-treat, (4) bioreactors, (5) sprinkler irrigation, (6) granulated active carbon/liquid-phase carbon adsorption and (7) separation.

0 0 0 0 0 0 0 0 0

0 0.06 0.006 0 0 0.01 0 0 0

0 0 0 0 0 0 0 0 0

0.1

1003 1/1/ 98

0.8 0 0 0

0 0 0 0 0 2 0.06

1/7/ 98

0.9 0 0 0

0.08

1001 1/1/ 98

0 0 0 0 0 0 0 0 0

0 0 0 0 0 3 0

1/7/ 98

0 0 0 0 0 0 0 0 0

0 0.29 0.19 0 0 0 0 0 0 0 0 0 0

1999

0 0 0 0 0 0 0 0 0

0 0 0 0

0

1004 1/1/ 98

0 0 0 0 0 0.008 0 0 0

0 0 0 0 0 5 0

1/7/ 98

0 0 0 0 0 0 0 0 0

0 0 0 0

0

1009 1/1/ 98

0 0.07 0.01 0.08 0 10 0 0 0

0 0 0.8 0 0 7 0.08

1/7/ 98

0 0.07 0.01 0.08 0 10 0 0 0

0 0 0.8 0 0 7 0.08

1/7/ 98

0.5 0.7 0.6 0.2 0.4 0.3 0.3 0.04 0.03

0.5 0 0 8

8

1010 1/1/ 98

Chemical species concentration (mg l1) in groundwater of Ringe site.

Benzene Toluene Xylenes 1,2,3-Trimethylbenzene 1,2,4-Trimethylbenzene 1,3,5-Trimethylbenzene Naphthalene 1-Methylnaphthalene Phenol o-Cresol m-/p-Cresol 2,6-Dimethylphenol 2,4- and 2,5Dimethylphenol 2,3-Dimethylphenol 3,4-Dimethylphenol 3,5-Dimethylphenol Carbazol Quinolin Thiophen Benzothiophen Benzofuran Dibenzofuran Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a]pyrene Dibenzo[a,h]anthracene

Date

Well no.

Table 5.5.3

0.7 0.8 0.75 0.3 0.5 0.5 0.4 0.04 0.05

1100 3 60 3 6 7 3

1/7/ 98

0.8 1 0.1 0 0.07 0 0 0 0

0.15 0 0 1

12

1021 1/1/ 98

0.008 0.01 0 0 0 0.007 0 0 0

200 0 10 2 4 4.5 0.15

1/7/ 98

0.28 0.53 0.33 0.46 0 0 8.1 0.78 0.34 0 4233 0 0 0 0 0 0 0 0 0 0 0

0 0.4 17 0.83 6.5 0.76 2.5 0 0 0 0 17 0.55

1999

0 4152 0 0 0 0 0 0 0 0 0 0 0

21

2000

1100

0 1.4 0.29 0 0 4.3 11 0.56 0

244 0.74 18 0.87 4.7 0 12 0.12 7.9 0 0 150 0.38

1999

9811

7.1 12 4.3 0 0 73 173 3.7 0

1800 12 365 16 88 0 254 0.75 34 0.85 1 450 39

1999

9911

430 150 62 3.4 0 52 128 168 0.58

1800 43 296 10 58 13 1000 43 44 1.7 6.7 430 1200

1999

9912

Improved Risk Assessment of Contaminant Spreading 283

284

Figure 5.5.12

Chapter 5.5

Locations of supply and monitoring wells around Ringe site.

The abovementioned methods of remediation were the output of the screening process included in the RAGS methodology. Once different remediation alternatives were screened, they should be evaluated in detail with respect to the specific characteristics of the Ringe site. The RAGS methodology utilises nine criteria with equal weight fractions to evaluate and compare different remediation alternatives (Table 5.5.4). It seems that the most suitable method of soil decontamination might be a thermal treatment combined with an extraction system (Table 5.5.4). Abstraction of groundwater in conjunction with adequate treatment processes (pump-and-treat) is likely the best option to decontaminate the groundwater at the Ringe site (Table 5.5.4). On the basis of the foregoing analysis, the suggested strategy for remediation of the Ringe site is a combination of different systems and technologies, including: steam injection or electrical heating of soil; dual-phase vacuum extraction of near surface liquids and vapours; groundwater abstraction to create a hydraulic barrier to prevent additional contaminant migration; and

Bioremediation

+ + + + 2

Overall protection Long-term effectiveness Reduction of toxicity Short-term effectiveness Implementation Cost Score

Soil

Soil flushing + + 2

Chemical oxidation + + +/ 0.5

Evaluation of potential remediation technologies.

Criteria

Table 5.5.4

+ + + + + 4

Thermal treatment and bioslurping + + + 0

Bioremediation

+ + 2

Slurry walls

Underground water

+ + + + + 5

Pumping and treatment

Improved Risk Assessment of Contaminant Spreading 285

286

Chapter 5.5

water treatment plant to treat the abstracted groundwater before reinjecting it to the upstream.

5.5.6.2.

In Situ Stimulation/Remediation of Contaminated Sites

Ringe site is representative of numerous contaminated fractured and lowpermeability sediments/soils covering a major part of the surface, especially in northern and central Europe. Such soil types possess a special problem in relation to the spreading of contaminants into groundwater. The fractures form hydraulic avenues through the otherwise low-permeability clayey sediments/ soils. Traditional remediation technologies used in high-permeability soils (extraction, ventilation, etc.) are primarily based on vertical wells that are installed on the subsurface. Regarding the fractured low-permeability sediments, the problem is that the transport takes place predominantly in the vertical fractures whereas NAPL is accumulated in the impermeable matrix. Therefore any remediation method based on vertical wells is expected to be very inefficient. In order to perform effective in situ remediation of fractured sediments a large number of fractures have to be connected to a well or to a highly permeable sediment layer. The bulk hydraulic conductivity in the fractured sediments may be stimulated either by increasing the fracture aperture and/or the connectivity between fractures and/or the density of the fractures. During hydraulic fracturing, new fractures are introduced into the system and the aperture of the existing fractures is increased due to the uplift of the soil above the fracture. Fracturing is a method whereby a gas (pneumatic fracturing) or water/slurry (hydraulic fracturing) is injected into the subsurface at pressures exceeding the in situ pressure at flow rates exceeding the flow rates corresponding to the natural in situ permeability. The induced fracture itself is commonly a sheet-like feature with maximum dimensions of roughly 20 m and a thickness of 1 to 20 mm depending on the type of injected fluid. Hydraulic fractures are commonly filled with granular material, which keep the fractures open. Pneumatic fractures are not filled with granular material and are kept open due to irregularities along the fracture walls. Investigations over the past 15 years in North America have shown that fractures can be created in contaminated, fine-grained sediments, where they increase the flow rates to and from wells by one or two orders of magnitude.41 The technique appears to offer the possibility of significantly reducing the costs of remediation of contaminated sites underlain by clay till by increasing the rate at which remediating agents can be introduced into the subsurface and the rate at which contaminated fluids can be extracted. Induced fractures can be established either from vertical wells (most common in groundwater) or from angled/horizontal wells. Hydraulic fracturing is widely used in the petroleum industry where the fractures are created at great depth in rock to improve the productivity of oil


287

wells. It has been shown that hydraulic fractures may be created at shallow depths in sediments to increase their hydraulic conductivity and improve the remediation of contaminated sites.42,43 Most of the environmental applications have been developed by researchers in Cincinnati, Ohio, with the applications conducted in silty and clayey glacial drift similar to the deposits found throughout Scandinavia, the Baltic countries and large parts of Germany, the Netherlands, the UK, Poland and other areas that were transgressed by glaciers during former ice ages. The technique is applicable to the remediation of a wide range of contaminant types, including petroleum hydrocarbons, chlorinated solvents, pesticides and other compounds.44 The properties of hydraulic fractures vary considerably, but many demonstrations have shown that the rate of remediation can be increased by one to two orders of magnitude.42 The technique appears to offer the possibility of significantly reducing the remediation costs of contaminated sites underlain by especially silty clay till.

5.5.7

Socioeconomic Relevance and Policy Implications

Liquid pollutants penetrate into the subsurface from industrial waste disposal ponds or municipal waste landfills, leaking underground storage tanks, agricultural chemicals, oil spills, etc. Unavoidably, pollutant infiltration through soils leads to the contamination of underground aquifers and the introduction of hazardous chemicals in plant, animal and human tissues. The European Inventory of Existing Chemical Substances lists over 100 000 compounds.45–47 The threat posed by many of these chemical remains uncertain because of the lack of knowledge about their concentrations and the ways in which they move through and accumulate in the environment and then impact on humans and other life forms. For most European countries, underground aquifers remain the main sources of water supply to urban and rural areas. The pollution of water by organic and inorganic wastes of domestic, industrial and agricultural activities in combination with climatic changes, affecting the aquatic cycle, has led to a dramatic reduction of the reserves of ‘‘drinking’’ groundwater. All water, polluted by households, industry or agriculture, returns back one way or another and may cause environmental damage. Over 300 000 potentially contaminated sites have been identified in Western Europe, and the estimated total number in Europe adds up to 1 500 000.45–47 In Eastern Europe, soil contamination by fuels (e.g. gasoline, diesel, kerosene, crude oil) around abandoned military bases, airports and oil transportation pipes poses the most serious risk. Current practices used by environmental companies for the risk assessment of contaminated sites follow some protocols suggested by environmental agencies. Commonly, the heterogeneous/fractured nature of soils is overlooked in these protocols. The new methodologies developed in the course of the TRACE-Fracture project might be integrated as options into such protocols. The suggested methodology of site characterisation reduces substantially the number of excavations/monitoring wells and the number of field-scale/

288

Chapter 5.5

laboratory-scale experiments and tests that are commonly required for characterising highly heterogeneous formations. The methods and approaches developed for fractured clay till are generic and can be extended to (1) any other fractured site with similar geological characteristics, (2) any other numerical simulator that includes the module of dual porosity/dual permeability medium and (3) any type of organic pollutant. The market is full of commercial numerical codes for the simulation of contaminant transport in the unsaturated and saturated zones of the subsurface. However, the main shortcoming of all these simulators is the lack of efficient up-scaling procedures that transform the actual heterogeneous geological models to equivalent numerical models. For this reason, the reliability of the predictions of most macroscopic simulators is questionable. The generalised procedures, developed in the context of the TRACE-Fracture project for the accurate transformation of conceptual geological models into numerical grids, could be useful not only for updating the capabilities of the SIMUSCOPP, but also for improving the predictability of any numerical code used in risk assessment of such contaminated lands (e.g. QUMPFS, TOUGH2, COMPFLOW). The economic impact of the project within the EU is hard to quantify with any reasonable precision. In the USA, the various sources place the clean-up costs of contaminated land, water and structures between $500 billion and $1 trillion over the next 50 years. In Europe, the corresponding figures are deemed to climb to the same order of magnitude or even higher, given the increased drilling and excavation costs compared to those in the USA. The accurate characterisation of the fractured polluted sites coupled with reliable estimates of pollutant migration can contribute substantially to the design and application of restoration strategies that can deliver safe, reconstituted fractured soils at reduced cost and effort compared to current practices. The number of necessary laboratory-scale experiments and field tests will decrease, companies will save money and time and the cost of the decision-making procedures will be reduced substantially. Roughly, with the use of the new methodologies of site characterisation, the cost may be reduced from h1 000 000 to h300 000.

Acknowledgements The TRACE-Fracture project (1/2/2000-31/1/2003) was supported by the European Union under the Energy Environment and Sustainable Development (EESD) sub-programme of the 5th FP (contract number EVK1-CT199900013). I would like to express my thanks to many people who contributed to the TRACE-Fracture project by mentioning just a few of them: Dr M. Theodoropoulou (TEI of Patras and University of Patras, Greece); Dr V. Karoutsos (University of Patras, Greece); Dr K. E. Klint (GEUS, Denmark); Dr P. Gravesen (GEUS, Denmark); Dr C. Laroche (IFP, France); Dr P. LeThiez (IFP, France); Mr L. Molineli (CH2M-Hill, Spain); and Mr F. Sanchez (CH2M-Hill, Spain).


289

References 1. A. S. Goldford, G. A. Vogel and D. E. Lundquist, Waste Manag., 1994, 14, 153. 2. K. D. Pennell, M. Jin, L. M. Abriola and G. A. Pope, J. Contam. Hydrol., 1994, 16, 35. 3. E. L. Madsen, J. L. Sinclair and W. C. Ghiorse, Science, 193, 252, 830. 4. B. R. Keyes and G. D. Silcox, Environ. Sci. Technol., 1994, 28, 840. 5. Y.-S. Wu, K. Zhang, C. Ding, K. Pruess, E. Elmorth and G. S. Bodvarsson, Adv. Water Resour., 2002, 25, 243. 6. K. Zhang and A. D. Woodbury, Adv. Water Resour., 2002, 25, 705. 7. X.-H. Wen and J. J. Gomez-Hernandez, J. Hydrol., 1996, 183, ix. 8. C. T. Miller, G. Christakos, P. T. Imhoff, J. F. McBride and J. A. Pedit, Adv. Water Resour., 1998, 21, 77. 9. G. S. Bodvarsson, Y.-S. Wu and K. Zhang, J. Contam. Hydrol., 2003, 62–63, 23. 10. J. Birkholzer, H. Rubin, H. Daniels and G. Rouve, J. Hydrol., 1993, 144, 1. 11. J. E. Warren and P. J. Root, SPE J., 1963, Sept., 245. 12. F. W. Schwartz and L. Smith, Water Resour. Res., 1988, 24, 1360. 13. R. H. Dean and L. L. Lo, SPE Res. Eng., 1998, May, 638. 14. H. H. Gerke and M. T. van Genuchten, Water Resour. Res., 1993, 29, 305. 15. K. E. Klint, P. Gravesen, A. Rosenbom, C. Laroche, L. Trenty, P. LeThiez, F. Sanchez, L. Molinelli and C. D. Tsakiroglou, Water, Air and Soil Pollution: FOCUS, 2004, 4, 201. 16. B. Bourbiaux, M. C. Cacas, S. Sarda and J. C. Sabathier, Revue de l’IFP, 1998, 53, 785. 17. S. M. Hassanizadeh and W. G. Gray, Adv. Water Resour., 1993, 16, 53. 18. K. J. Slough, E. A. Sudicky and P. A. Forsyth, J. Contam. Hydrol., 1999, 40, 107. 19. D. A. Reynolds and B. H. Kueper, J. Contam. Hydrol., 2001, 51, 41. 20. K. Vandersteen, J. Carmeliet and J. Feyen, Transp. Porous Media, 2003, 50, 197. 21. C. D. Tsakiroglou, TRACE-Fracture, 1st Annual Progress Report, March 2001. 22. C. D. Tsakiroglou, TRACE-Fracture, 2nd Annual Progress Report, March 2002. 23. C. D. Tsakiroglou, TRACE-Fracture, 3rd Annual Progress Report, March 2003. 24. C. D. Tsakiroglou, TRACE-Fracture, Final Report, March 2003. 25. R. C. Sidle, B. Nilsson, M. Hansen and J. Fredericia, Water Resour. Res., 1998, 34, 2515. 26. B. Nilsson, R. C. Sidle, K. E. Klint, C. E. Boggild and K. Broholm, J. Hydrol., 2001, 243, 162. 27. M. Houmark-Nielsen, Bull. Geol. Soc. Denmark, 1987, 36, 1. 28. C. D. Tsakiroglou, Ind. Eng. Chem. Res., 2002, 41, 3462.

290

Chapter 5.5

29. C. Laroche, C. Henrique, S. Bekri, L. Trenty, P. LeThiez, Proceedings of 8th International Conference on Contaminated Soil CONSOIL 2003, Gent, Belgium, 12–16 May 2003. 30. M. Theodoropoulou, V. Karoutsos and C. Tsakiroglou, Environ. Forensics, 2001, 2, 321. 31. C. D. Tsakiroglou, J. Non-Newtonian Fluid Mech., 2002, 105, 79. 32. C. D. Tsakiroglou, M. Theodoropoulou, V. Karoutsos, D. Papanicolaou and V. Sygouni, J. Colloid Interf. Sci., 2003, 267, 217. 33. C. D. Tsakiroglou, M. Theodoropoulou and V. Karoutsos, AIChE J., 2003, 49, 2472. 34. C. D. Tsakiroglou, J. Non-Newtonian Fluid Mech., 2004, 117, 1. 35. C. D. Tsakiroglou, M. A. Theodoropoulou, V. Karoutsos and D. Papanicolaou, Water Res. Resour., 2005, 41, W02014. 36. M. A. Theodoropoulou, V. Sygouni, V. Karoutsos and C. D. Tsakiroglou, Int. J. Multiphase Flow, 2005, 31, 1155. 37. M. Theodoropoulou, V. Karoutsos, C. Kaspiris and C. D. Tsakiroglou, J. Hydrol., 2003, 274, 176. 38. C. D. Tsakiroglou, M. Theodoropoulou and V. Karoutsos, Oil Gas Sci. Technol. La Revue de l’IFP, 2005, 60, 141. 39. L. Trenty, C. Henrique, C. Laroche, P. LeThiez, Proceedings of 8th International Conference on Contaminated Soil CONSOIL 2003, Gent, Belgium, 12–16 May 2003. 40. US EPA, Risk Assessment Guidance for Superfund, Vol. I, Human Health Evaluation Manual (Part A), Interim Final, USEPA/540/1-89/002, 1989. 41. US EPA, Risk Reduction Laboratory and the University of Cincinnati, Hydraulic Fracturing Technology: Applications Analysis and Technology Report, USEPA/540/R-93/505, 1993. 42. L. C. Murdoch, W. Slack, B. Siegrist, S. Vesperm and T. Meiggs, Am. Soc. Civ. Eng., 1997, 10A. 43. L. C. Murdoch, M. C. Kemper, M. Narayanaswamy, A. J. Wolf, Demonstration of Hydraulic Fracturing to Facilitate Remediation, HWRIC RR-068, Illinois DNR Hazardous Waste Research & Information Center, Urbana-Champaign, 1997. 44. US EPA, Alternative methods for fluid delivery and recovery, EPA/625/ R-94/003, 1994. 45. EEA, Europe’s Environment: The Second Assessment, European Environment Agency, Luxembourg, 1998. 46. EEA, Environment in the European Union at the Turn of the Century: Summary, European Environment Agency, Copenhagen, 1999. 47. EEA, Environment in the European Union at the Turn of the Century: Appendix to the Summary. Facts and Findings per Environmental Issue, European Environment Agency, Copenhagen, 1999.

CHAPTER 5.6

Groundwater Risk Assessment at Contaminated Sites (GRACOS): Test Methods and Modelling Approaches PETER GRATHWOHLa AND HANS VAN DER SLOOTb a

Centre for Applied Geoscience, Universita¨t Tu¨bingen, Sigwartstrasse 10, DE-72076 Tu¨bingen, Germany; b Energy Research Centre (ECN), P.O. Box 1, NL-1755 Petten ZG, The Netherlands

5.6.1

Introduction

All over Europe there are numerous contaminated sites where top layers of soil or other materials are contaminated. Remediation of all these sites would be economically impossible and therefore site prioritisation is needed based on an evaluation of the risk to subsoil and groundwater contamination. The definition of ‘‘groundwater risk assessment’’ as used here is the risk of a compound migrating from a contaminated source in the unsaturated zone to the groundwater. The goal is usually to predict the contaminant concentration at different points of compliance, which include (1) the boundary between disposed materials and the natural subsurface environment, (2) the transition zone between groundwater and vadose zone below disposed or contaminated materials and (3) the groundwater downstream of a contaminated zone (often the property boundary). The European Union (EU) project GRACOS and similar projects have addressed these issues and the relevant processes as depicted in Figure 5.6.1. For the release of inorganic (major, minor and trace elements) and organic constituents from contaminated materials into water, it is important to identify a few key issues: the nature of the constituents of concern, which may have very different release behaviours based on their chemistry and the local conditions; 291

292

Chapter 5.6 Groundwater Risk Assessment in Case of Contaminated Soil / Contaminated Soil Air Source zone: The contaminant concentration in a soil leaching test or in soil air is above the legal limit !!! → Is the concentration in the groundwater above the legal limit ???

Contaminant Concentration

Depth (unsaturated zone)

Groundwater recharge (A)

Soil air sampling CO , CH

Column test

Possible scenarios: (A )

Residual oil phase petroleum hydrocarbons

(B)

B) No recharge (immobile Contaminants)

(C)

D)

(D)

datio

n (A+

A) Contaminant transport by seepage water C) no recharge: vapor diffusion only

egra

O -Diffusion

(E)

Groundwater recharge: Persistent contaminants can reach the groundwater No recharge: Nonvolatile compounds are immobile Volatile compounds, no recharge: Peristent compounds can reach the groundwater by vapor diffusion Biodegradable compounds may reach the groundwater in very low concentrations (below legal limit ?) Further attenuation in the shallow plume due to volatilization and biodegradation ???

Biod

D) Biodegradation ??? CO

With

E) Volatilization

from

biodegradation

O -Supply Capillary fringe

Shallow plume E) Biodegradation ???

Aqueous concentration

Contaminant transport across the capillary fringe by diffusion and transverse dispersion

Contaminant volatilization across the capillary fringe downgradient from the source zone

Contaminant source zone

Groundwater

Oxygen transport across the capillary fringe aerobic biodegradation of hydrocarbons

Downgradient attenuation zone Distance from source zone

Figure 5.6.1

The GRACOS scenario.

the major release mechanisms in soil and soil-like materials, which will usually be dominated by percolation or (vapour) diffusion; the hydrology of the site under consideration, i.e. how much infiltration will occur and what are the relevant/preferential flow paths; the nature of the contamination source in terms of release controlling parameters such as pH, EC, redox conditions and dissolved organic carbon (DOC); the changes in release conditions with time due to processes such as depletion of source material and shifts in the main release controlling parameters; the biodegradation/natural attenuation of released constituents in the subsoil and in the groundwater; and the targets to be evaluated for the consideration of proper mitigating measures, including obtaining information on the allowable release rates post-remediation and the applicable monitoring strategies (whether active remediation took place or not).

Groundwater Risk Assessment at Contaminated Sites (GRACOS)

293

Almost all of these questions can be traced back to a quantification of the release rates and/or aqueous concentrations of inorganic and organic contaminants at a specified point of compliance. Obviously, judgment based on potential leaching is more appropriate for assessing long-term impacts of a contaminant release than evaluations based on total composition (i.e. contaminant concentration in solids). Consequently, in material testing and soil contamination analysis, a paradigm shift from concentrations in solids towards concentrations in the aqueous phase has occurred during the last decade. For these tests, the main question is how to achieve a reliable groundwater risk assessment without spending too much money, so that efforts are deemed economically viable. The economics behind these studies are very relevant because of the large material streams involved. These streams include releases from contaminated soils, demolition wastes, dredged sediments and waste materials from mining and industry. Since test applications depend on compound properties, two types of contaminant classes have been distinguished in GRACOS: volatile contaminants which travel in the gas phase in the unsaturated zone (e.g. volatile solvents and gasoline constituents); and non-volatile compounds which more or less exclusively migrate with the seepage water (e.g. most inorganic contaminants, heavy metals and high molecular weight organic compounds). For the assessment of volatile compounds, vapour-phase monitoring and Henry’s law constant allow for the derivations of in situ aqueous concentrations. For non-volatile and ionic compounds, aqueous leaching tests are needed to determine the aqueous concentrations. The subsequent sections will specify and present key conclusions based on selected datasets form the GRACOS project. More detailed information is available from www.uni-tuebingen. gracos.de and from the numerous GRACOS publications, of which the most important ones are listed as Refs. 1–4.

5.6.2

Leaching Tests (Heavy Metals, Low-Volatility Organic Compounds)

5.6.2.1

Total Composition vs. Aqueous Concentrations

Total composition (i.e. concentrations based on dry solids) is inadequate for most environmental assessment purposes because for many of the constituents a significant fraction of their total content is essentially non-leachable by water. This potential contaminant release into water can be subdivided into (1) potentially leachable, which is a maximum amount leached under predefined worst-case conditions, and (2) actual leached amount, which is the amount leached under the conditions imposed by the material itself. In Figure 5.6.2, the leaching behaviour of cadmium from an agricultural soil heavily contaminated from a sewage sludge application is shown for illustration. It

294

Chapter 5.6 1000 ANC mode Duplicate INGESTION INHALATION

Leached at L/S=10 (mg/ kg)

100

SCE

ACIDIC ENVIRONMENTS PLANT UPTAKE

NaNO3 Hac CaCl2 CEN

NATURAL SOIL

10

1

SCE2 Total

SOIL LIMING

Cd

0.1

CEMENT STABILIZATION OF CONTAMINATED SOIL

0.01 1

3

5

7

9

11

13

pH

Figure 5.6.2

Relevant pH domains for assessing different questions in relation to different types of impact (L/S, liquid/solid ratio).

demonstrates that different test methods can be placed with respect to one another by plotting such test data as a function of pH. This even applies to the test data obtained from sequential chemical extraction schemes, provided the data are calculated as cumulative leached amounts in subsequent leaching steps. It is recommended that site evaluations be based on leaching rather than on total composition analysis, which is currently being employed by regulations. The main advantage of judgment based on leaching vs. total composition is that the assessment is taking place on the true aspects causing environmental impact. This is because the non-leachable constituents will not cause confusion or controversy, since they would not show up in a leaching test.3–6

5.6.2.2

Percolation vs. Batch or Shaking Tests: Comparison to Field

Although batch shaking tests are very popular in environmental analysis of contaminated materials, it has become obvious that for the assessment of long-term contaminant behaviour, dynamic tests closer to natural conditions

295


are needed. Column leaching tests are important laboratory techniques commonly used for the determination of desorption or dissolution rates of (mobile) contaminants from various materials (soils and sediments, mining wastes, recycling and construction materials, demolition waste, etc.). Critical parameters in leaching tests are the contact times and the mass transfer rates into the aqueous phase. Extended contact times result ultimately in equilibrium conditions between the solid phase and the water. Equilibrium concentration levels may depend on dissolved organic matter and colloids and for heavy metals additionally on pH and redox conditions. At equilibrium, results from different tests are interrelated as shown in Figure 5.6.3, provided that different test conditions have no influence on other parameters that determine the aqueous concentrations of the target compound. For example, pH, DOC and turbidity (the amount of suspended particles) typically depend on the liquid/solid (L/S) ratio used in the test (at high L/S ratios in shaking tests DOC may be diluted whereas turbidity may increase). The equilibrium concentration in the aqueous phase (Cw) depends on the L/S ratio and the sorption capacity of the sample for a specific compound: Cw ¼

Cs;ini Kd þ LS

ð1Þ

where Kd is the distribution coefficient (l kg1) defined as the ratio of the concentration in the solids to the concentration in water (Kd ¼ Cs/Cw). Cs,ini 10 LS = 0.25

concentration Cw

1 LS = 2 0.1 LS = 10 0.01

1⋅10

3

0.1

1

10

100

1⋅103

distribution coefficient Kd

Figure 5.6.3

Decrease in water concentration with increasing distribution coefficient Kd at different liquid/solids ratios (LS) starting at an initial concentration in the solids (Cs,ini) of 1 mg kg1, mg kg1, etc. The least dilution, meaning almost equal concentrations independent of LS is observed for Kd 4 100 (inverse linear relation between concentration and Kd). The highest concentrations in water are observed for small LS ratios and low Kd values (LS ¼ 0.25 represents approximately the conditions in a column experiment or a natural porous medium; LS then corresponds to the ratio of porosity and bulk density ¼ n/rbulk).

296

Chapter 5.6

denotes the initial concentration in the solids. If Kd is much larger than L/S then the aqueous concentration is independent of L/S. For small values of Kd, increasing L/S ratios cause simply a dilution. When pH changes occur, Kd can no longer be considered constant and other approaches are needed. For inorganic compounds, adsorption phenomena are often strongly nonlinear, and hence their solid/solution distribution cannot be described by a linear, i.e. concentration independent, Kd relationship. In addition, constituents cannot be considered to be transported through soil fully independent of each other. To account for multicomponent effects typical for inorganic constituents, such as competition for adsorption sites, a mechanistic modelling approach is preferable over empirical (Kd) models. Several models are available that take into account relevant processes that influence the solid/solution partitioning and subsequently the transport rates. These processes include, among others, mineral precipitation, incorporation into solid solutions, sorption onto iron, aluminium and manganese (hydr)oxide surfaces and interaction with particulate (POM) and dissolved organic matter (DOC). Depending on the choice of input parameters, such mechanistic geochemical models allow site-specific as well as generic model predictions. In addition, the calculation of element partitioning between dissolved, complexed and different particulate phases is relevant with respect to potential bioavailability of elements, as it is generally believed that complexed forms are less likely to be taken up by organisms than elements in free ionic form. DOC bound constituents are likely to be transported over larger distances than free forms due to facilitated transport. This may also be true for very poorly water soluble organic micro-pollutants. In Figure 5.6.4, the element Zinc Contaminated soil NL 0.001

Model

0.0001 0.00001 0.000001

Partitioning liquid and solid phase, [Cu+2]

0.001 0.0001

Tenorite

0.00001

Clay

0.000001

FeOxide POM-bound

0.0000001

DOC-bound

0.00000001

Free

0.0000001 1

2

3

4

5

6

7

8

9

10 11 12 13 14

0.000000001 1

Cu+2 fractionation in solution

100%

2

3

4

5

6

7

8

9 10 11 12 13 14

Cu+2 fractionation in the solid phase 100%

Fraction of total concentration (%)

Fraction of total concentration (%)

0.01

Concentration (mol/l)


[Cu+2] as function of pH 0.01

80% 60% 40% DOC-bound

20%

80%

Tenorite Clay

60%

Fe Oxide 40%

POM-bound

20%

Free 0%

0% 1

2

3

4

5

6

7

8

pH

Figure 5.6.4

9 10 11 12 13 14

1

2

3

4

5

6

7

8

9

10 11 12 13 14

pH

Prediction of copper release from zinc-contaminated soil illustrating the adequate match between test data (filled circles; TS14429) and model results (solid curve). Partitioning between dissolved and particulate phases in concentration (top right). Fractionation of copper in dissolved (free and DOC associated) and solid phases (particulate organic matter (POM), iron oxide, clay and specific minerals).

297

Groundwater Risk Assessment at Contaminated Sites (GRACOS) 100

0.1 0.01 0.001 0.0001 0.00001

Ni

0.000001 0.0000001

10

Concentration (mg / l)

Cumulative release (mg / kg)

1

Leached (mg / kg)

1

100

10

1 0.1 0.01 0.001 0.0001 0.00001

3

5

7

9

11

13

0.00001

10

0.00001

100 10 1 0.1

DOC

0.001 5

7

9

11

13

1000 100 10 1 0.1

1000

100

10

0.01

0.00001

0.001

pH

Figure 5.6.5

10

10000

0.001 3

0.1

10000

Concentration (mg/l)

Cumulative release (mg/kg)

1000

0.001

L/S (l/ kg)

100000

10000

Leached (mg/kg)

0.1

L/S (l/kg)

100000

1

0.001

0.0001 0.001

pH

0.01

0.01

0.000001

0.0000001 1

0.1

0.1

L/S (l/kg)

10

1 0.00001

0.001

0.1

10

L/S (l/ kg)

Relationships between pH dependence and percolation tests (laboratory) for a mixture of wastes (integral mix largely consisting of contaminated soil, sediments and soil cleaning residues; filled diamonds) disposed in a 12 000 m3 pilot cell with leachate data from 1.5 m3 lysimeters (triangles) and the full-scale pilot (filled circles).

distribution between dissolved and particulate phases obtained with a mechanistic geochemical modelling approach is shown for a zinc-contaminated soil. A detailed description of this modelling approach is referred to in Dijkstra et al.7 Figure 5.6.5 shows that results from column percolation tests and field lysimeter measurements can be related to field-scale data, if the sampling and testing are done in an appropriate manner. Under field conditions, the relationship between time, infiltration and L/S can be calculated simply based on the dry bulk density (rbulk [kg m3]), the height of the percolated material of interest (h [m]) and the net infiltration rate (N [mm a3]):

t ½a ¼

LSrbulk h N

ð2Þ

The consistency between the data from different scales of testing (laboratory to pilot to field scale), as shown in Figure 5.6.5, demonstrates that the underlying processes are sufficiently understood and can be quantified. In addition, this observation indicates that appropriate sampling procedures and mixing of samples to constitute a representative leachate sample lead to more meaningful

298

Chapter 5.6

results than the often reported scattered data from individual sub-samples obtained in the field or laboratory. Using the new chemical speciation/transport tool LeachXS-Orchestra,8,9 the percolation test data for the integral waste mix consisting largely of contaminated soil, sediments and soil cleaning residues were modelled (Figure 5.6.6). The mineral assemblage and the iron and aluminium oxide sorption parameters were determined from modelling the pH dependence leaching test results of the same mix. The DOC release is assumed to be a continuous decay function. In the case of the lysimeter and the field data graph showing preferential flow, the 25% of total mass involved in leaching accounts for the difference from the percolation test. The simultaneous modelling of some 25 major, minor and trace elements is ambitious and the modelling results are quite promising. [Na+] as function of L/S 1.0E+00


pH

pH as function of L/S 8.5 8.3 8.1 7.9 7.7 7.5 7.3 7.1 6.9 6.7 6.5 0.0001

0.001

0.01

0.1

1

10

1.0E-01

1.0E-02

1.0E-03

1.0E-04 0.0001

100

0.001

0.01

L/S (l/kg) 1.0E-01



1.0E-02

0.001

0.01

0.1

1

10

1.0E-03

0.001

0.01



L/S

Figure 5.6.6

10

100

[Zn2+] as function of L/S

1.0E-07

0.1

1

1.0E-03

1.0E-06

0.01

0.1

L/S

[Cu2+] as function of L/S

0.001

100

1.0E-02

1.0E-04 0.0001

100

1.0E-05

1.0E-08 0.0001

10

[SO4 ] as function of L/S

L/S 1.0E-04

1

-2

[Ca2+] as function of L/S 1.0E-01

1.0E-03 0.0001

0.1

L/S

1

10

100

1.0E-04 1.0E-05 1.0E-06 1.0E-07 1.0E-08 1.0E-09 0.0001

0.001

0.01

0.1

1

10

100

L/S

Full mechanistic modelling (open diamonds) of percolation tests (laboratory) for a mixture of wastes (integral mix largely consisting of contaminated soil, sediments and soil cleaning residues; filled circles) disposed in a 12 000 m3 pilot cell in comparison with leachate data from a 1.5 m3 lysimeter (triangles) and the full-scale pilot (squares).


299

Apart from the prediction of the release as a function of L/S (or time), information on the partitioning within the column as a function of time at a specified depth or as a function of depth at a specified time can be obtained without any extra effort. This provides further insight into the processes occurring within the column, which may not yet be apparent in the outflow. In Figure 5.6.7, results are shown for Ca, Cu and Mo. In the case of Ca, several minerals are of relevance; for Cu, organic matter dominates the release; for Mo, PbMoO4 and Fe are controlling phases. In both cases of Ca and Mo, dissolved concentrations locally increase within the column as a result of local conditions. This illustrates the kinetics between all mutually interacting inorganic constituents.

5.6.2.3

Boundary Conditions Leading to Changing Release Rates

Concentrations of many compounds, especially heavy metals, often depend strongly on pH and DOC, as shown in Figure 5.6.4 for an example of an inorganic constituent. Besides the complexation of heavy metals, DOC also causes solubilisation of organic compounds which decreases the sorption coefficients and increases leaching. The solubilisation factor (corresponding to the decrease in sorption or Kd) is given by S 0 ¼ 1 þ fDOC KDOC

ð3Þ

where fDOC and KDOC denote the fraction of dissolved organic carbon in the aqueous phase and the organic carbon normalised distribution coefficient (l kg1). Figure 5.6.8 shows the influence of increasing DOC values on distribution patterns of polycyclic aromatic compounds (PAHs) leaching from demolition waste. Low-solubility compounds are affected most but they do not contribute much to the sum of the 16 EPA PAHs, if the DOC concentration stays below 30 mg l1. It should be noted that in both cases—heavy metals and organic compounds—DOC can cause significantly enhanced leaching and in many field cases is probably more important than particle facilitated transport. In contrast to suspended particles, DOC cannot be filtered (in the laboratory or in the field) and therefore is an important parameter if present in sufficient concentrations (430 mg l1).

5.6.2.4

Release Kinetics

So far, the considerations have been concerned with contaminant release under equilibrium conditions, which may not always be the case. There are generally two types of non-equilibrium: chemical non-equilibrium and physical nonequilibrium. Chemical non-equilibrium in a leaching test may occur when the kinetics of mineral dissolution/precipitation and/or sorption processes are slow relative to the contact time with the water. Physical non-equilibrium indicates the mass transfer of contaminants from a sorbed or solid state to mobile water by film diffusion and intraparticle diffusion.

Figure 5.6.7

0.001

0.01

6.67 DOC-bound

depth (m)

10.00

13.33

3.33

POM-bound PbMoO4[c]

6.67

depth (m)

10.00

13.33

Concentration profile for Ca+2 at 3 cm

POM-bound Calcite time (days)

DOC-bound time (days)

Free FeOxide

POM-bound PbMoO4[c]

time (days)

0.000000001 0.00 2.00 4.06 6.15 8.23 10.2912.3814.4616.54

0.00000001

0.0000001

0.000001

0.00001

0.0001

Concentration profile for MoO4-2 at 3.3 cm 0.001

Free POM-bound

Concentration profile for Cu+2 at 3.3 cm 0.01 0.001 0.0001 1E-05 1E-06 1E-07 1E-08 1E-09 1E-10 1E-11 0.00 2.00 4.06 6.15 8.23 10.29 12.38 14.46 16.54

Free Anhydrite

0.0001 0.00 2.00 4.06 6.15 8.23 10.29 12.38 14.46 16.54

0.001

0.01

0.1

1

10

Concentration profile for Ca+2 at 10 cm

POM-bound Calcite

time (days)

Concentration profile for MoO4-2 at 10cm

Free FeOxide

POM-bound PbMoO4[c]

time (days)

0.0000001 0.00 2.00 4.06 6.15 8.23 10.2912.3814.4616.54

0.000001

0.00001

0.0001

0.001

Concentration profile for Cu+2 at 10cm 0.01 0.001 0.0001 1E-05 1E-06 1E-07 1E-08 1E-09 1E-10 1E-11 0.00 2.00 4.06 6.15 8.23 10.29 12.38 14.46 16.54 Free DOC-bound POM-bound time (days)

Free Anhydrite

0.01 0.00 2.00 4.06 6.15 8.23 10.29 12.38 14.46 16.54

0.1

1

10

Predicted partitioning between dissolved and particulate phases in the laboratory column of integral waste consisting of contaminated soil, sediments and soil cleaning residues as a function of depth at a specified time (15.5 days) and as a function of time at a specified depth (3 and 10 cm from the inflow into the column) for Ca, Cu and Mo as representatives for the behaviour of a major element, a metal and an oxyanion.

Free FeOxide

0.000000001 0.00

0.00000001

0.0000001

0.000001

0.00001

0.0001

Concentration profile for Mo after 15.5 days 0.001

1E-11 0.00 3.33 Free POM-bound

1E-10

1E-09

1E-08

1E-07

1E-06

1E-05

6.67 10.00 13.33 alpha-TCP depth (m)

Concentration profile for Cu after 15.5 days

3.33 POM-bound Calcite

Concentration profile for Ca after 15.5 days

0.0001 0.00 Free Anhydrite

0.001

0.01

0.1

1

0.0001





Concentration (mol/l) Concentration (mol/l)

Concentration (mol/l) Concentration (mol/l) Concentration (mol/l)

10

300 Chapter 5.6


301

100 0

3

10

20

50

Solids

concentration µg l-1

10

1

0. 1

Figure 5.6.8

C hr Bb F Bk F Ba In P de no D ah A Bg hi P Su m

Py Ba A

n Ph e An t Ft h

Fl

A

ny Ac e

0. 01

Distribution pattern of 16 EPA PAHs in aqueous leachate concentrations of soils. With increasing DOC (0–50 mg l1) the aqueous phase concentrations shift towards the distribution pattern of the solids (righthand bar for each PAH). Note that low molecular weight PAHs up to anthracene (Ant) are not much affected and that more than 20 mg l1 is needed to cause a significant increase in the sum of the 16 EPA PAHs.

Small grains in both cases cause higher release rates because of the higher surface area present per unit volume of porous media. Intraparticle diffusion is relevant for porous materials such as clay aggregates, rock fragments, concrete in demolition waste and construction products, slag, etc., i.e. the most abundant soil and waste materials of interest. Intraparticle diffusion limits mass transfer increasingly with time because of decreasing concentration gradients inside the particle during leaching. Initially, concentration gradients are steep leading to high release rates and equilibrium is established after short flow distances in the contaminated material. Therefore, most column experiments (and percolation in the field) start with equilibrium conditions which sooner or later shift into nonequilibrium. The lag time towards non-equilibrium depends on grain sizes (longer for fine-grained materials or for a high fraction of fine-grained material in a heterogeneous, coarse sample) and the sorption capacity (expressed as Kd). High sorption capacity causes extended periods of equilibrium contaminant leaching and therefore higher solubility (low sorption) compounds leach faster than low solubility compounds in a chromatographic manner (shown for PAHs in Figure 5.6.9). Diffusion-limited release causes concentrations and release rates to decrease with the square root of time: in this domain the release rates are independent of the flow rate. If intraparticle porosities and tortuosities are known, the leaching of a series of compounds can be predicted based on the relative water solubilities or relative Kd values of the compounds, as shown in Figure 5.6.9.

302

Chapter 5.6 100

Concentration in water [µg l-1]

Demolition waste 10

1

0.1 Nap Ace 0.01 Phe Fth

Comparison of measured and numerical modelled datas from column test

0.001 0.1

1

10

100

1000

time [d]

Figure 5.6.9

Long-term release of PAHs (Nap, naphthalene; Ace, acenaphthene; Phe, phenanthrene; Fth, fluoroanthene) from demolition waste. Equilibrium concentration plateau followed by diffusion limited release (tailing); symbols denote data and curves are numerical simulations based on the spherical intraparticle diffusion model.

The time needed for equilibration in a leaching test increases with increasing L/S ratio (at high L/S more compound has to diffuse from the solid to the aqueous phase until the equilibration concentration is reached), therefore column tests (L/S o 0.25) more likely yield equilibrium concentrations initially than typical batch shaking tests (i.e. L/S¼10). Figure 5.6.9 also shows that, in principle, only a short-term column test (one day) is needed to determine the equilibrium concentration. This may prove of relevance for the concentration which is to be found under field conditions shortly after placement. If information about the long-term dynamics of leaching is required, than an ongoing short-term column test can be simply continued beyond one day. Experience shows that the effort for a short-term column test actually is not greater than for shaking tests, and at the same time the results are more precise and of better reproducibility.

5.6.2.5

Column Tests: Standardisation and Design

In CEN TC 292 ‘‘Characterisation of waste’’ and in ISO TC 190 ‘‘Soil’’ there have been developed standardised protocols for waste (CEN TS TS 1442910 and 1440511), soil and soil-related materials, such as sediments (ISO 21268-312 and 413). Currently, the same methods are proposed for characterisation of granular construction materials (CEN TC 351). This constitutes a horizontal (across different sector) approach to standardisation as opposed to a materialspecific test development. For a broad range of inorganic constituents, leaching

303


kinetics have been evaluated recently in pH-dependence leaching tests for MSWI bottom ash14 and percolation tests.15 Column tests can in principle be flexible in their dimensions. For practical reasons largely related to the particle size of the granular material to be tested, column dimensions of 5–10 cm diameter and a length of 15–30 cm have been proposed (resulting in 0.5 kg to several kilograms solids in the column depending on density). Generally, the percolation is run to reach a certain liquidto-solid (L/S) ratio. For quality control purposes (compliance testing) there is no need to go to high L/S. Under such conditions sufficient eluate for analysis can be obtained already after L/S ¼ 2:1 within one or two days. These dimensions allow for investigation of many materials that are received in the laboratory without too much pretreatment (grinding, etc.). This also has a positive effect on sampling requirements, which can be less severe than in the case of analysis of composition on relatively small laboratory samples. Figure 5.6.10 shows the typical set-up of a column leaching test. One limitation of column tests is given for fine-grained clayey samples which in compacted form have very low permeabilities: such materials can be percolated in a mix with coarse sand, as aggregates or as compacted bodies with flow around them, as this is how they often occur in the field. Note that the packing density does not influence the equilibrium concentration. Prerequisites for reproducible results are representative samples and a reasonably homogeneous contaminant distribution in the sample. Recently, very reproducible results were obtained for the leaching of many inorganic constituents from MSWI bottom ash, a relatively heterogeneous material.15 Usually columns have a sufficient self-filtration capacity yielding turbidities in the effluent that are too low to influence the leaching of organic compounds. It is important to note that since artefacts are very likely to occur with custom filters (such as cellulose or glass fibre filters) and since DOC cannot easily be removed, it is recommended for the release of organic compounds that only centrifugation methods be applied for when turbidity in the effluent is too high (i.e. 410 FNU). Filtration is only required for inorganic compounds if suspended particles are comprised of target metals such as iron-containing

storage tank ss tubing

glass column

(deionized pure water)

(diameter 6cm, length 16cm, volume 450ml)

PVC-tube soil sample filter (quartz sand)

peristalticpump

solvent e.g. cyclohexane for extraction of organic compounds

column effluent

Figure 5.6.10

Typical set-up of a column leaching test (see also CEN TC 29216 ‘‘Characterisation of waste’’ and ISO TC 19012 ‘‘Soil percolation tests’’).

304

Chapter 5.6

mica or clay minerals. For percolation, water free of the target compound is needed. Changes in the redox conditions or biodegradation need to be avoided by removing major electron acceptors such as oxygen, nitrate or sulfate by stripping and deionising of the water. Light-induced bioactivity can be avoided relatively simply by wrapping a column set-up (e.g. in aluminium foil). Measurements of parameters besides the contaminants, such as pH, EC, DOC and turbidity (and Eh), are necessary when further data interpretation, modelling and data comparison is foreseen.

5.6.2.6

Concluding Remarks for Leaching Tests

Most factors that influence the release of contaminants from a contaminated site can be assessed through aqueous leaching tests. Although a wide variety of leaching tests is available in the literature, only a limited set of tests suffice to address key processes and influencing factors. For a largely percolationdominated situation, as in most real-world cases, a column leaching test is the most appropriate method to evaluate contaminant release. In the case of rather impermeable clay or clayey loam samples, aggregated material can be filled into the columns, or depending on the scenario compacted material in a ‘‘flow-around-monolith’’ mode can be used (according to NEN 7347). Changing conditions of pH, EC, redox and DOC should be monitored in the leachate. By carrying out a pH dependence test (i.e. CEN/TS1442910), the importance of the chemical speciation and partitioning can be addressed, which are relevant for assessing bioavailability of contaminants and mobility in soil and groundwater. For the prediction of long-term behaviour, modelling is an important tool (bench-scale and even full-scale testing can cover only a limited time span relative to the timeframe over which answers are sought). In general, the findings of leaching of pollutants support the tiered approaches which, for example, consist of basic, i.e. comprehensive, characterisation tests for solid materials, which are not well known, and later a simple, short-term compliance test for already well-characterised materials. With compliance tests the effort and costs in assessing large material streams such as demolition wastes, incineration ash, slag, dredged sediments, etc., can be minimised.4 Basic characterisation allows for the understanding of the longterm dynamic behaviour of a specific material (including changes in redox, pH, etc.) whereas a compliance test only confirms that the sample belongs to a wellknown (i.e. characterised in a basic test) group of materials. Because of that, compliance tests can be very simple ranging from visual appearance to batch and short-term percolation tests. On the other hand, basic characterisation tests have to be designed in a way which allows for the understanding of the longterm leaching behaviour of a contaminant. Results from such tests can be used in mechanistic models which then allow for predictions of the long-term behaviour under field conditions. According to the experiences gained in GRACOS, short-term column tests are less difficult to perform and yield more robust results than the widely used shaking or batch tests. It is well known in the scientific community that the shaking tests originally developed for


305

activated sludge from waste water treatment are not suitable for materials such as demolition waste, dredged sediments, incineration ash, etc. For inorganic contaminants, column tests in combination with batch testing as a function of pH provide the most robust results, as the information covers a wide range of possible conditions relevant for judging releases from granular materials.

5.6.3

Groundwater Risk Assessment for Volatile Compounds

5.6.3.1

Vapour-phase Diffusion

Contrary to ionic or highly water soluble compounds, which are predominantly transported by seepage water, volatile compounds can reach the groundwater table via the vapour phase. Some of the most frequent groundwater pollutants such as fuel constituents (benzene, toluene, ethylbenzene, xylenes (BTEX)) and chlorinated solvents (trichloroethene (TCE), perchloroethene (PCE), etc.) are volatile and migrate in the subsurface environment by vapour-phase diffusion. Diffusive spreading follows Fick’s second law: @C @2C ¼ Da 2 @t @x

ð4Þ

where Da denotes the apparent diffusion coefficient [m2 s1] in the three-phase system soil solids/water/air, which is defined as Da ¼

Dg ðn2:5 De g =nÞ ¼ ng þ ðnw =HÞ þ ðKd r=HÞ a

ð5Þ

where n, ng and nw denote the total, gas-filled and water-filled porosities, respectively. Dg is the molecular diffusion coefficient in the gas phase and H is Henry’s law constant, defined as the concentration ratio of gas to water. The term ng2.5/n denotes an empirical relationship for calculation of the effective diffusion coefficient (De) for vapour-phase diffusion in the unsaturated zone.17 Kd is the sorption coefficient which can be estimated based on empirical relationships in terms of the organic carbon content of the soil.18 Note that partitioning into the water is often more important for retarded diffusion of volatile contaminants in the soil air than sorption onto the soil solids. With equation (4) and (5), the transient spreading of contaminants in the unsaturated zone can be assessed (the simplest case is of a concentration front, i.e. C/Co D 0.5, spreading from a source of ffi constant concentration where the pffiffiffiffiffiffiffi distance travelled is proportional to Da t). For the volatilisation of constituents from mixtures of organic compounds (e.g. benzene from gasoline) the change in composition over time causes changing vapour-phase concentrations in the source, which has to be accounted for by Raoult’s law: Ci ¼ wi

p0i mi g ¼ wi Cgsat gi RT i

ð6Þ

306

Chapter 5.6 2.0 MTBE

(b) Mixture 2

1.5 Flux (g m -2 h-1)

methylcyclohexane

toluene

1.0 TCE

1,1,1-TCA

ethylbenzene

0.5 PCE

0.0 10

0

20

30

40

50

60

70

80

90

100

Time (h)

Figure 5.6.11

Pure forward prediction (curves) of measured (symbols) diffusive fluxes of gasoline constituents though a sand column of 30 cm length.19

where Cgsat, pi0, mi, R and T are the saturation concentrations in the gas phase of the pure compound [g l1], the saturation vapour pressure [kPa] and the molecular weight [g mol1] of compound i, the gas constant [l kPa mol1 K1] and the temperature [K], respectively. Ci, wi and gi denote the equilibrium vapour concentration [g l1], the molar fraction and the activity coefficient of compound i in the mixture, which equals 1 in an ideal mixture (this is often assumed). The volatilisation rates are proportional to the vapour pressure and therefore depend on the molar fraction in the mixture. Figure 5.6.11 shows that with equation (4)–(6) the fluxes of gasoline constituents diffusing from a non-aqueous gasoline source in a sand column over 30 cm distance can be predicted in pure forward fashion (i.e. no fitting) reasonably well by a simple numerical model.19

5.6.3.2

Coupled Models for Simulation of Field Sites

Numerical models proved successful in predicting the behaviour of organic vapours in the unsaturated zone at the field scale in well-controlled lysimeter and field experiments.20–22 Such numerical models couple vapour-phase diffusion, biogeochemical reaction and flow of seepage water allowing for the simulation of the spatiotemporal contaminant concentrations in the unsaturated zone and the capillary fringe as shown in Figure 5.6.12. Furthermore, these models allow for simulating the aging of complex organic mixtures in the field by considering volatilisation of constituents from a source zone in the soil, the transport of pollutants to the atmosphere and the increased concentration gradients due to biodegradation (which accelerates the volatilisation of degradable compounds).22 Initially, the high-volatility compounds volatilise leading to an increase of the mole fraction of the residual compounds in the organic mixture which subsequently leads to an increase of the equilibrium vapour-phase

307


gaseous concentration [mg/l air] cyclo-pentane 50 20 2 1 0.2 0.02 0.002 0.0002 30

G_i, T = 3.000E+ 00 days

1

2 2

2

0

5

0.2

0.0

1

4

0.2 0.02 0.00.002 00 2 0.0002

20

0.2

2

0.02 0.2

3

0.002

1

2

0.02

1

2

0.000 2 0.002

1

0.002 0.0002

depth [m]

0

10

15 X [m]

20

25

G_i, T = 9.000E+ 01 days 0.2

2 1

2

2 1

0.02 0.0002

0.2

2

0

5

0.002

1

3

4

10

15

20

02

0.0

0.2

2

gaseous concentration [mg/l air] cyclo-pentane 50 20 2 1 0.2 0.02 02 0. 0.002 0.0002 30 0.0

1

1

2

depth [m]

0

25

X [m] G_i, T = 3.500E+ 02 days 0.002

0.000

2

1 2

0.0

0.00

2

02

depth [m]

0

3 0.002

4 0

5

Figure 5.6.12

10

15 X [m]

0.02

20

25

gaseous concentration [mg/l air] cyclo-pentane 50 20 2 1 0.00 0.2 0.02 0.002 0.002 0.0002 30

Development of an organic vapour-phase plume (cyclopentane) from a kerosene source in the unsaturated zone after 3 (initial spreading), 90 (maximum spreading) and 350 days (shrinking and detached plume in the groundwater), two-dimensional simulation, groundwater flows from left to right.21

concentrations as shown in Figure 5.6.11 and for the field case in Figure 5.6.13. Since vapour-phase pressures are dependent on temperature this has to be considered for obtaining an improved fit between observation and prediction. However, already a simple isothermic model can predict the general behaviour reasonably well as shown in Figure 5.6.14. Also important is the reproduction of the CO2 produced by biodegradation of the organic vapour-phase compounds influenced by changing water contents in the soil because this strongly influences the effective diffusion coefficients (equation (5)) as shown in Figure 5.6.15. The only fitting parameter needed in this case is the biodegradation rate constant.

5.6.3.3

Concluding Remarks for Risk Assessment for Volatile Compounds

Concentrations of volatile compounds in seepage water and the capillary fringe can be calculated from concentrations in the soil air based on Henry’s law constant. This also applies to the saturation vapour concentrations at the boundaries to non-aqueous phase liquids such as fuels which can be calculated from Raoult’s law (in most cases an activity coefficient of unity is appropriate). Both laws

308

Chapter 5.6 40 35

molefraction [%]

30 25

hexane 3D model decane 3D model iso-octane 3D model Værløse field data

20 15 10 5 0

0

100

200

300

time [days]

Figure 5.6.13

Development of mole fractions of kerosene constituents during the first year of the GRACOS field experiment; ‘‘aging’’ of the source in the three-dimensional model (curves) agrees very well with field data for hexane, isooctane and decane.21

require the ‘‘local equilibrium assumption’’ in models which can be confirmed in bench-scale laboratory tests19 as well as in lysimeters22 and a well-controlled field experiment run over a time span of one year.20,23,24 Diffusion coefficients can be estimated based on empirical relationships17,19 or tracer tests.25–27 For ‘‘real’’ field scenarios, biodegradation of organic vapours and the subsequent geochemical changes such as O2 depletion and CO2 production have to be considered. Coupled (seepage water flow–vapour-phase diffusion– biogeochemical reaction) models such as MIN3P24 could be validated with the well-controlled field experiment of a kerosene spill at the GRACOS field test site. Volatilisation to the atmosphere has been proved to be the most important natural attenuation process in the unsaturated zone for volatile organic pollutants from shallow spills when the surface is not sealed, followed by biodegradation (Figure 5.6.16). Since biodegradation causes steep concentration gradients in the vadose zone, spreading is limited but aging or depletion of compounds from the source is accelerated. Modelling results illustrate that the overall biodegradation rates depend mainly on distribution parameters such as Henry’s law constant of the fuel constituents (because degradation takes place in the aqueous phase exclusively), on the biological degradation rate constant, on the soil water content and on the temperature. Temporal changes of temperature and infiltration rates affect volatile organic compound behaviour significantly (highly dynamic system): for a first assessment, isothermal and stationary boundary conditions can be assumed. Transport to groundwater

309

Groundwater Risk Assessment at Contaminated Sites (GRACOS) x=0.0, z=-2.3m, located 1 m below center of NAPL source component gaseous concentrations in mg/l, 3D model

1.4

toluol(g) Værløse field data toluol(g) 3D model steady state toluol(g) 3D model transient

gaseous concentration [mg/l]

1.2

1

0.8

0.6

0.4

0.2

0

0

100

200

300

time [days] 20 15

5 0 air T NAPL source 2.25mdepth

Figure 5.6.14

T [˚C]

10

-5 -10

Fit of predicted toluene concentrations in the unsaturated zone below the kerosene source in the GRACOS field experiment with consideration of temperature fluctuations (dash–dot line) and under isothermal conditions (solid curve). The lower diagram shows the temperature in the air at the surface and at 1 m and 2.25 m depth.21

was found to be controlled by dispersion–diffusion processes: although only a small fraction of contaminants may reach the groundwater, legal limits in the capillary fringe are locally likely to be exceeded.4,22,28–30

5.6.4

Modelling for Groundwater Risk Assessment of Inorganic Constituents

As already explained above, for inorganic substances, mechanistic geochemical modelling approaches have more perspective over simple empirical Kd models,

310

Chapter 5.6 2D model location 1.3 m below NAPL source

CO2(g) CO2(g)transientflow

0.012 0.01

precipitation

-0.03 -0.025 -0.02

0.006

-0.015 -0.01

0.004

-0.005 0

0.002 evapot ranspiration

0

Figure 5.6.15

groundwater recharge

100

0.005

200 time [days]

0.01

300

3

total inflow total outflow

0.008

tota l outflow [m /day]

partial gas pressure [atm]

0.014

Influence of precipitation and evaporation, i.e. soil water contents, on the development of CO2 partial pressure due to biodegradation of organic vapours in the unsaturated zone: solid curve, steady state; dash–dot curve, transient infiltration conditions.16

mass in % of initial

100 biodegradation

80 degassing to atmosphere

60

1,2,4-TMB

40 20

mass still present

0 −6 10

Figure 5.6.16

−5

−4

10 10 degradation rate constant [s-1]

10

−3

Competition of biodegradation and degassing to the atmosphere as a function of the biodegradation rate constant (pseudo first order) for the residual mass of 1,2,4-trimethylbenzene (1,2,4-TMB) after one year (GRACOS two-dimensional field scenario).21,28 With increasing biodegradation, degassing to atmosphere is diminished and the volatilisation from the source is accelerated.

as their outcome has a much more predictive value.7,14,15 Unlike the apparent complexity of these models, there is a great perspective in limiting their input to obtain results of routine tests. Similarly, the choice of binding parameters for these models allow only a few degrees of freedom, as for all model components, ‘‘generic’’ parameter sets for a broad range of elements have been derived. This means that this type of modelling is characterised by a general applicability with respect to a wide variety of contaminants and a variety of soils, sediments

311


and wastes. Models that can couple chemistry with transport include, among others, PHREEQC,31 ECOSAT32 and ORCHESTRA.8 The latter has the most flexibility for incorporating new adsorption models, solid solution descriptions, surface precipitation and organic matter interaction such as Nica Donnan. PHREEQC at present (2006) does not contain a module for the binding of metals to organic matter, which is needed to predict metal mobility in the upper soil layers with high contents of dissolved and particulate organic matter. Although not always necessary for risk assessment, the cited model codes rarely have detailed hydrological capacity standards available (only onedimensional transport; two- or three-dimensional in principle is possible, but not standard). At present, there are already some models such as PHAST,33 MIN3P34 and PHT3D35 that combine detailed hydrology with detailed chemistry. A mechanistic modelling approach was used in 2006 to derive new limit values for building materials in the revised Dutch Building Materials Decree. The methodology followed to derive these limit values is similar to that described in EN1292016 and is shown in Figure 5.6.17. The source term was based on contaminant-specific decay functions derived from percolation tests (TS1440511) performed on a variety of construction and waste materials. A percolation rate was assumed of 300 mm per year (average net rainfall in the Netherlands) and different heights of application were assumed (i.e. 0.5 m to 5 m) that establishes the relationship between time and L/S ratio. Existing data from typical Dutch soil profiles were used to define the properties of the underlying soil profile (i.e. in terms of water saturation and sorbent surfaces present in the soil such as clays and organic matter). The

Establish regulatory limits (iterative process)

Percolation through waste

Transport through soil

Figure 5.6.17

Concentration. at POC

Model simulation of contaminant leaching from a waste application to soil and groundwater rain

limit value

Time

“Point of compliance” with Environmental limit value

Simplified figure showing the process of modelling groundwater contamination and establishment of regulatory limit values.

312

Chapter 5.6

calculations were performed with ORCHESTRA8 using one-dimensional flow of water along a streamline. Next, on a given ‘‘point of compliance’’, breakthrough curves were established and concentrations were compared to limit values (i.e. ecotoxicological values). Using an iterative process, limit values were then derived for building materials for each contaminant (expressed in mg kg1 leached at L/S ¼ 10) in such a way that concentrations at the point of compliance just fulfil the criteria at the point of compliance (17). The advantage of this semi-mechanistic approach over the often used Kd approach is obviously that important interactions between contaminants (and major elements such as iron, aluminium and calcium in the soil and groundwater) are captured by the model. In addition, this way of modelling is fully transparent as all thermodynamic constants used in the modelling are justified in the scientific literature. The next step in modelling is to describe both the source and the transport in the unsaturated and saturated zones by the full mechanistic modelling approach.

5.6.5

Conclusions/Recommendations

In general, for a contaminant of interest, the groundwater risk assessment should consider (1) the amount that is available for release or leaching from the material, (2) the multiphase local equilibrium distribution (i.e. aqueous/solid for non-volatile contaminants; vapour/ water/solid for volatile contaminants) and (3) any factors that may significantly alter the multiphase local equilibrium (e.g. pH, DOC, colloids and suspended particles). For assessment of volatile contaminants, identifying the extent of the contaminant source area, whether or not a separate organic contaminant phase (dispersed or continuous) is present, and the extent of the vadose zone or groundwater migration plume are important. Monitoring soil vapour concentrations and the assumption of the local vapour/water equilibrium are generally sufficient (Henry’s law). When a separate organic phase (i.e. a complex organic mixture such as fuels, lubrication oils or solvents) is present, then the constituent’s vapour pressure can be calculated based on Raoult’s law. For assessment of non-volatile contaminants, leaching is best characterised using up-flow column percolation tests in conjunction with pH controlled batch testing for pH-sensitive compounds and for when long-term release is of interest. Column testing provides initial equilibrium concentrations in the leachate and information on the long-term leaching dynamics as a function of elution liquid-to-solid ratio (L/S) or pore volumes. Results from the testing methods discussed above allow for the parameterisation and use of a contaminant release model for the source, which can then be coupled with a fate and transport model for considering dilution and attenuation from the source location to the point of compliance.36 Such models require a detailed description of hydrology at the field site, including possible preferential flow patterns. In response to the specific evaluation needs, the source term model may be either a simplistic, semi-analytical (i.e. spreadsheet)


313

model that provides an overestimate of release in order to be intentionally conservative, or it may be a more complex coupled local equilibrium model (i.e. including geochemical speciation for inorganic species) or a kinetic model that is coupled with representative mass transfer properties. Future tasks aim at the EU-wide validation of the characterisation leaching tests so that for these methods performance characteristics can be obtained. Additionally, based on many years of experience, a database/expert system has been developed which couples laboratory test data, physical aspects of release (hydrology), chemical changes in material properties with time and chemical reaction/transport modelling.9

Acknowledgement This work was supported by the EU 5th framework programme project GRACOS (Groundwater Risk Assessment at Contaminated Sites, EVK1CT1999-00029).

References 1. D. Halm and P. Grathwohl (eds), Proceedings of the 1st International Workshop on Groundwater Risk Assessment at Contaminated Sites (GRACOS), Tu¨bingen, Germany, 21–22 February 2002, Tu¨binger Geowissenschaftliche Arbeiten (TGA) C 61, 2002. 2. D. Halm and P. Grathwohl (eds), Proceedings of the 2nd International Workshop on Groundwater Risk Assessment at Contaminated Sites (GRACOS) and Integrated Soil and water Protection (SOWA), Tu¨bingen, Germany, 20–21 March 2003, Tu¨binger Geowissenschaftliche Arbeiten (TGA) C 69, 260, 2003. 3. H. A. van der Sloot, Waste Manag., 2002, 22, 693–694. 4. Guideline for Groundwater Risk Assessment at Contaminated Sites (GRACOS), www.uni-tuebingen.gracos.de. 5. D. S. Kosson, H. A. van der Sloot, F. Sanchez and A. C. Garrabrants, Environ. Eng. Sci., 2002, 19(3), 159–204. 6. H. A. van der Sloot, L. Heasman and Ph. Quevauviller (eds), Harmonization of Leaching/Extraction Tests, Studies in Environmental Science, Elsevier Science, Amsterdam, 1997, vol. 70. 7. J. J. Dijkstra, J. C. L. Meeussen and R. N. J. Comans, Environ. Sci. Technol., 2004, 38, 4390–4395. 8. J. C. L. Meeussen, Environ. Sci. Technol., 2003, 37, 1175–1182. 9. LeachXS: a database–expert system for leaching and environmental impact assessment, 2005 (http://www.leachxs.com). 10. CEN/TC292 (2005), Characterization of waste—Leaching behaviour tests—Influence of pH on leaching with initial acid/base addition, CEN/ TS 14429.

314

Chapter 5.6

11. CEN/TC292 (2004), Characterization of waste—Leaching behaviour tests— Up-flow percolation test (under specified conditions), CEN/TS 14405. 12. ISO 21268-3, Soil quality—Leaching procedures for subsequent chemical and ecotoxicological testing of soil and soil materials—Part 3: Up-flow percolation test. 13. ISO 21268-4, Soil quality—Leaching procedures for subsequent chemical and ecotoxicological testing of soil and soil materials—Part 4: Influence of pH on leaching with initial acid/base addition. 14. J. J. Dijkstra, H. A. Van der Sloot and R. N. J. Comans, Appl. Geochem., 2006, 21, 335–351. 15. J. J. Dijkstra, H. A. Van der Sloot and R. N. J. Comans, A consistent geochemical modeling approach for the leaching and reactive transport of major and trace elements in MSWI bottom ash, submitted. 16. CEN/TC292 (2005), Characterisation of waste—Methodology guideline for the determination of the leaching behaviour of waste under specified conditions, EN 12920. 17. P. Moldrup, T. Olesen, J. Gamst, P. Schjønning, T. Yamaguchi and D. E. Rolston, Soil Sci. Soc. Am. J., 2000, 64(5), 1588–1594. 18. R. Allen-King, P. Grathwohl and W. P. Ball, Adv. Water Res., 2002, 25(8–12), 985–1016. 19. G. Wang, S. B. F. Reckhorn and P. Grathwohl, Vadose Zone J., 2003, 692, 701. 20. M. Broholm, M. Christophersen, U. Maier, E. Stenby, P. Hoehener and P. Kjeldsen, Environ. Sci. Technol., 2005, 39(21), 8251–8263. 21. U. Maier and P. Grathwohl, in Reactive Transport in Soil and Groundwater, ed. G. Nu¨tzmann, P. Viotti and P. Aagard, Springer, 2005, pp. 141–155. 22. G. Pasteris, D. Werner, K. Kaufmann and P. Ho¨hener, Environ. Sci. Technol., 2002, 36, 30–39. 23. P. Gaganis, P. Kjeldsen and V. N. Burganos, J. Vadose Zone Res., 2004, 3, 1262–1275. 24. P. Gaganis, H. K. Karapanagioti and V. P. Burganos, Adv. Water Res., 2002, 25, 723–732. 25. D. Werner and P. Ho¨hener, Environ. Sci. Technol., 2002, 36, 1592–1599. 26. D. Werner and P. Ho¨hener, Environ. Sci. Technol., 2003, 37(11). 27. D. Werner, P. Grathwohl and P. Ho¨hener, Vadose Zone J., 2004, 3, 1240– 1248. 28. P. Grathwohl, I. D. Klenk, U. Maier and S. B. F. Reckhorn, IAHS Publ., 2002, 275, 141–146. 29. I. D. Klenk and P. Grathwohl, J. Contam. Hydrol., 2002, 58(1–2), 111–128. 30. I. D. Klenk, Transport of volatile organic compounds (VOCs) from soilgas to groundwater, PhD dissertation, TGA C55, Center for Applied Geosciences, Tu¨bingen, 2000. 31. D. L. Parkhurst and C. A. J. Appelo, User’s guide to PHREEQC (version 2): a computer program for speciation, batch-reaction, one-dimensional transport, and inverse geochemical calculations, Water Resource Inv. Report 99-4259, US Geological Survey, Denver, CO, 1999.


315

32. M. Keizer and W. H. Van Riemsdijk, ECOSAT: a computer program for the calculation of speciation and transport, Department of Soil Quality, Wageningen University, 1996. 33. D. L. Parkhurst, K. L. Kipp, P. Engesgaard and S. R. Charlton, PHAST: a computer program for simulating ground-water flow, solute transport, and multicomponent geochemical reactions, 2005. 34. K. U. Mayer, E. O. Frind and D. W. Blowes, Water Resour. Res., 2002, 38(9), 1174–1195. 35. H. Prommer, D. A. Barry and C. Zheng, Ground Water, 2003, 42(2), 247–257. 36. P. Grathwohl, in Boden und Altlasten, ed. Franzius, Lu¨hr and Bachmann, Erich Schmidt Verlag, 2000, vol. 9, pp. 41–60.

CHAPTER 5.7

INCORE: Integrated Concept for Groundwater Remediation THOMAS ERTELa AND HERMANN J. KIRCHHOLTESb a

Sachversta¨ndigen-Bu¨ro, Boschstr. 10, DE-73734 Esslingen, Germany; Landeshauptstadt Stuttgart, Amt fu¨r Umweltschutz, Hermann Josef Kirchholtes, 36-3.51, Gaisburgstr. 4, DE-70182, Stuttgart, Germany

b

5.7.1

Motivation and Basic Concept

European cities located in river basins are using groundwater from local shallow aquifer systems. Industrial development in the 20th century was rapid and caused urban groundwater pollution, often exceeding the legal limits. Changes in land use during this period have created complex contamination patterns, such as heterogeneous distribution of contaminants, the presence of different contaminants and large landfill areas (Figure 5.7.1). Besides the threat to the wider environment, existing soil and groundwater contamination has resulted in incalculable costs for long-term groundwater remediation. Healthy residential and working conditions can be guaranteed only on uncontaminated ground. The presence of environmental pollution can limit investment in urban development on brownfield sites. However, structural changes now in place offer an opportunity to improve soil and groundwater quality. Considerable thought must be made in order to reach sustainable improvements: this is exemplified by the integral procedures developed by INCORE. The current legal approach for the treatment of soil and groundwater pollution is focused on a particular set of problems caused by a specific polluter. All measures aim at a rapid reduction of environmental damage so that risk to the public associated with a particular property is removed. However, this approach fails in heavily polluted areas with different property owners and complex pollutant patterns (Figure 5.7.2). Large amounts of private and public money are being spent to identify and assess point sources of contamination without being able to quantify reliably their impact on groundwater quality; numerous remediation schemes are undertaken without an economic evaluation of their long-term performance. 316

INCORE: Integrated Concept for Groundwater Remediation

Figure 5.7.1

Aerial view of Stuttgart Neckar valley.

Figure 5.7.2

Complex pollution pattern schema.

317

Five European cities, Stuttgart, Linz, Strasbourg, Milan and Bydgoszcz, which share similar groundwater problems in their industrialised urban areas, committed themselves to develop jointly suitable solutions. Specific local conditions vary in these five INCORE project areas; they vary with respect

318

Chapter 5.7

to groundwater conditions, existence of public and private monitoring wells, type of pollutants, size of problem areas, rising groundwater problems, etc. Therefore they provide a representative range of the conditions to be expected across Europe. In order to achieve the INCORE project goals in a cost-effective way, different parts of the anticipated tool set were applied and evaluated at different levels of detail in the five selected areas. The proposed INCORE strategy for the investigation, remediation and revitalisation of industrial areas is based on an integrated quantification of total contaminant emissions. It considers entire industrial areas instead of particular single sites, in order to achieve a high level of confidence in the investigation results. A cyclic approach is proposed, beginning with the screening of groundwater plumes at the scale of entire industrial areas, and ending with the remediation of individual source areas or the containment of plumes. The major advantage of this approach is that the number of local-scale sites, or the size of the area to be considered, is reduced stepwise from one cycle to the next. Thus, a large potentially contaminated area would be screened but ultimately only a small area may need remedial actions. Figure 5.7.3 presents a schematic of this new approach. This new approach repeats an investigation/assessment/revitalisation cycle three times at different scales. Cycle I. The groundwater quality is screened downstream of the potential source areas. Cycle II. Only those sites where groundwater quality is not acceptable are considered further. In these cases analytical methods are used to backtrack and identify sources of contamination.

Figure 5.7.3

Cyclic approach.


319

Cycle III. The characteristics of the source zones are considered for remediation to control emissions, or implementation of monitored natural attenuation. The most appropriate technology is selected by establishing what level of contamination reduction is required, considering the proposed future use of the site. This ‘‘no further action’’ approach provides a cost effective set of tools for the optimised investigation, evaluation and management of contaminated groundwater and land in industrialised urban communities.

5.7.2

Cycle I: Plume Screening

At the beginning of the project data inventories compiling historical information and present land use data are set up, combined with the development of a conceptual site model. Based on this, first technical investigations are emission oriented and are focused on groundwater contamination. Quantification of contaminant emission is obtained from the application of a novel integral groundwater investigation method (see Chapter 5.5), which yields the total pollutant mass flux and the mean and maximum pollutant concentrations originating from contaminant source zones. The basic idea of the new integral groundwater investigation technique is that the total contaminant mass flux downstream of potentially contaminated sites is covered by the capture zones of one or more pumping wells, which are positioned along control planes perpendicular to the mean groundwater flow direction. Analyses of multiple groundwater samples obtained at the wells during pumping yield concentration time series. Results of this integral groundwater investigation method are the total contaminant mass flux (emission) and the mean and maximum concentrations within the undisturbed groundwater flow field as well as determination and the delimitation of boundaries of potential polluted source zones. Further modelling leads to estimations of probabilities of contaminant concentrations exceeding regulation limits within large areas. The size of these areas under consideration depends on the transport and degradation behaviour of different contaminants. The final result is a map (see Figure 5.7.4) of the investigation area distinguishing between areas with different levels of groundwater impact. This allows a ranking of these areas with a distinct level of confidence. This map enables the administration to set priorities for further activities. Focusing the efforts on the areas of greatest impact helps to concentrate personnel resources on those sites which cause major impacts on groundwater pollution. This leads to maximum effectiveness in administrative activities. The mapping of areas with different groundwater impact further identifies areas in which urban and economic development can take place without any hindrance by groundwater pollution. This ensures development and structural change with lower risks on investment.

320

Chapter 5.7

source area with high gw impact source area with low gw impact no gw impact gw control plane investigation area direction of gw flow

Figure 5.7.4

5.7.3

Results of plume screening.

Cycle II: Source Identification

The results of the integrated investigation provide a rough localisation of suspected source areas. However, more precise identification of the contamination source area is needed in order to apply the polluter pays principle. With the means of cost-effective laboratory and on-site analytical systems as well as isotopic fingerprinting techniques the backtracking from the control plane along the path line of the plume yields a precise localisation of the source of contamination (see Figure 5.7.5). The results of cycle II verify the sources of groundwater pollution and identify the polluter with a very high reliability. This secures the application of the ‘‘polluter pays principle.’’ These results also help to avoid law-suits which leads to an acceleration of investments and to faster administrative procedures.

5.7.3.1 5.7.3.1.1

GC-MS Fingerprinting for Petroleum Hydrocarbons General Approach

Fingerprinting of hydrocarbons was developed in the petroleum industry, so as to understand the source of crude oil and natural gases. The objectives of fingerprinting investigations (environmental forensics) pertaining to history and source of environmental pollutants are overall very similar to those of petroleum geochemistry. Fingerprinting is defined as a scientific methodology


321

source area with high gw impac source control planearea source area with low g A contributes 80 % to detected plume no gw impact 20 % to detected plume B contributes gw control plane to detected plume C no contribution investigation area direction of gw flow

C C

C

B

C

C C A A

Figure 5.7.5

Results of source screening.

developed to be used in the environmental assessment of a fuel pollutant (or fuel contaminant) to: characterise the type of the fuel contaminant; quantify the concentration of potentially environmental hazardous compounds; and identify the composition of all compounds within the fuel contaminant that can be used to reliably determine the source, and understand the rate and transport of the fuel pollutants. Hydrocarbons released into the environment are subjected to biotic and abiotic transformation reactions in the soil and groundwater media. The rate and behaviour of these hydrocarbons depends on a number of physicochemical and biological processes including (1) evaporation, (2) dissolution, (3) microbial degradation, (4) photooxidation and (5) interaction between oil and sediments. The combination of these processes known as weathering reduces the concentration of released hydrocarbons in soil and groundwater and alters their chemical composition. For a specific site investigation and oil spill identification, analytical approaches have to be considered which provide detailed compositional information, and are sensitive enough for measuring spilled oil and petroleum products in soils and groundwater. The following list provides major target analytes which are suggested for the characterization and source identification of the contaminants. Aliphatic hydrocarbons, including n-alkanes in the C8–C40 range, and selected isoprenoids like pristane (C19) and phytane (C20).

322

Table 5.7.1

Chapter 5.7

Database-base input and its variety.

Database section

Data key

n

Database section

Data key

n

Contaminant type

1 gasoline

22

Age of contamination

1o10 years

25

2 diesel/heating oil 3 bunker C oil 4 lubricating oil 5 creosote

47

2 10–20 years

18

7 16

3 20–50 years 4450 years

44 8

See list above

50

3

Chemical parameter

Number of considered samples ¼ 95.

Single-ring volatile aromatic hydrocarbons, so called BTEX compounds (benzene, toluene, ethylbenzene and m-, o- and p-xylene), and alkylated benzenes (C3–CS benzenes). Polynuclear aromatic hydrocarbon compounds (PAH; Table 5.7.1), particularly the two- to four-ring compounds (including dibenzothiophenes) can be used to identify both source and extent of degradation. These compounds include the 16 EPA priority pollutant PAHs, and their associated alkylated homologues. Biological markers (or biomarkers) are molecular fossils, meaning that these compounds are derived from formerly living organisms.1 Biomarkers are complex organic compounds composed of carbon, hydrogen and other elements which are found in crude oil and its heavy refined products. The structural characteristics of these compounds are chemically stable during degradation. Terpanes and steranes are the most common biomarkers in crude oils and their medium and heavy refined products. Besides identification of discrete fuel types, if sufficient release information is available, this technique can often discriminate whether a contaminant release was a single event, a series of events or a continuous release of a single or multiple products. By considering the results of other workers,2–5 Kaplan et al.6 have elaborated biodegradation which is presented in Figure 5.7.6, illustrating the relative level of different hydrocarbon types in fuels with a volatility range from gasoline to bunker C oil. However, this degradation chart should be used cautiously because biodegradation is a complex ‘‘quasi-stepwise’’ process that cannot be described as truly sequential alteration of compound classes.3 Weathered oils from different spills (sources) can be distinguished by using the source ratio (C3-dibenzothiophenes/C3-phenanthrenes or D3/P3) and weathering ratio (D3/C3-chrysenes or D3/C3) (Figure 5.7.7). Compounds in source ratios degrade relatively at the same ratio whereas those in weathering ratios change substantially with weathering and biodegradation.

323


Figure 5.7.6

Degradation scheme.

12

8

6

4

Increasing Oil Weathering

D3/C3 Weathering Ratio

10

North Sea, mean = 0.80 ∓ 2 SD

Alaska North Slope l crude, mean = 1.190 ∓ 2 SD

Iranian Crude, mean = 2.48 ∓ 2 SD

2 0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

D3/P3 Souce Ratio

Figure 5.7.7

5.7.3.1.2

Plot of D3/C3 vs. D3/P3.10

Contaminant Type Identification by Statistical Analysis

5.7.3.1.2.1 Objectives. The application of the fingerprinting method in practice demands much expert knowledge. The interpretation of the chromatograms

324

Chapter 5.7

and/or mass spectra needs long lasting experience and a large extent of know-how. Statistical analysis transfers the expertise-based method of fingerprinting interpretation to a tool comprehensible to each party involved in the investigation or evaluation process. This transfer was done using several statistical methods so the results of the fingerprinting tool will be both comprehensible and objective. The main goal of statistical analyses in the INCORE strategy is to develop a tool which could be used to identify the type of contaminant found in the plume and to determine source–plume relationships to potential source areas. 5.7.3.1.2.2 Database. The most detailed analytical data available from the INCORE test fields are those from samples taken at Stuttgart investigation site, which show the typical picture of a multiple used and therefore multiple contaminated industrial site. However, since the database should contain results from as many real samples as possible, the database was enlarged with other data derived from projects mainly in Germany, covering a wide range of different types of contaminants having undergone different levels of weathering (biodegradation). A set of 50 diagnostic parameters consisting of alkylbenzenes, alkylcyclohexanes, isoprenoides, PAHs and biomarkers such as steranes and terpanes were selected for further statistical consideration. For pattern identification among the parameters the original measured values were transformed in respect to the sum of the total analysed peak areas. Within the INCORE investigation and other related projects, the compiled database shows a broad range of real contaminated samples. The enlargement of the data with reference samples of ‘‘pure’’ contaminant types improves further the database. These reference samples were collected as fresh directly from service stations (diesel, gasoline) or heating oil storage tanks. An index of the database input and its variety is given in Table 5.7.1. 5.7.3.1.2.3 Discriminant Analysis. In a first step the database was filtered for the reference samples and those samples which were definitely contaminated with only one product type. This data subset was used for the model development. From the samples with known group membership (contaminant type), a set of linear discriminant functions is generated. These functions are based on a linear combination of the ‘‘predictor variables,’’ those parameters which in a stepwise procedure were identified to provide the best discrimination between the groups. In processing the stepwise method at least 17 parameters were selected which are significant for the classification of the five contaminant types. The so-called canonical correlation coefficients (CCC) reach values between 0.999 and 0.913 which testify to the very good classification of the model for the development data set. Using function 1 and function 2, the discriminant scores were calculated for each sample. The scatterplot shows that the five different contaminant types can be divided significantly. In practice most contamination belongs to multiple product types with probably different ages and different stages of degradation. Taking this information into account in a further step,

325

INCORE: Integrated Concept for Groundwater Remediation 10 discrimi. scores function 2

discrimi. scores function 2

10 5 0 gasoline middle distillates bunker C oil lubricating oil creosote

-5 -10 -15 -20 -20

gasoline middel distillates bunker C oil lubricating oil creosote

8 6 4 2 0 -2 -4 -6 -8 -10

0

20

40

discrimi. scores function 1

Figure 5.7.8

60

-5

0 5 10 15 discrimi. scores function 1

20

Calculated discriminant scores for development data set (left) and all samples (right).

all available data of the database were used to develop a new model in order to provide wider regions for the identification of real samples. As expected the classification of the enlarged sample set shows less sharp results. Regarding the CCC of function 1 which reached a value of 0.999, for the reference samples it is now reduced to 0.969. Nevertheless this result indicates a significant classification of the five contaminant types. Calculating the discriminant scores from function 1 and function 2 and plotting them together as x- and y- axis the regions of the different contaminant types can be visualised (Figure 5.7.8). It can be seen that the heavy distillates such as bunker C oil and creosote can be divided easily from the remaining samples. Also the discrimination of gasoline and lubricating oil is significant. Only the discrimination of some samples contaminated with middle distillates is not clear. For that reason a special discriminant function was estimated for separating gasoline, middle distillates and lubricating oil. The result of this approach in general shows a better discrimination between the three groups, but some samples still cannot clearly be identified. The reason for this is the sample composition and not the statistical model, because those samples show an untypical pattern of contamination, caused most likely by mixed contamination occurring at gas stations and the somehow intermediate character of so-called ‘‘winter diesel.’’ 5.7.3.1.2.4 Field Site in Stuttgart. Based on the results of the investigation performed in cycle I, this site was selected as a potential source of BTEX and PAH observed in well NT 9. For further identification of groundwater contamination and related plumes, a series of integral pumping tests (IPTs) was performed in newly installed wells given in Figure 5.7.9. The wells were placed according to potential sources known from the historical survey and previous investigations. Figure 5.7.10 gives an example of the IPT results in NT 108 showing a rapidly increasing PAH concentration, which leads to a PAH plume with its centre more than 20 m

326

Chapter 5.7

Figure 5.7.9

Results of groundwater sampling.

16

concentration [µg / l]

14 12 10 8 6 4 2 0 0

5

10

15

20

25

30

35

40

45

50

width of capture zone [m]

Figure 5.7.10

Example result of IPT.

besides NT 108. Figure 5.7.11 summarises the results of the IPT campaign giving a rough estimation of the shape of the identified plumes. According to the dynamic workplan an adaptive sampling and analysis campaign was performed to localise definitively the contaminant sources. The results are also given in Figure 5.7.11.


Figure 5.7.11

327

Results of the IPT campaign with estimated shape of the identified plumes.

A total of 10 soil samples collected from hot spots of the contaminated site, as well as 14 groundwater samples from IPTs have been studied using fingerprinting methods/parameters presented in Table 5.7.2. The correlation of the fuel contaminants identified in the monitoring wells has been performed mainly on the basis of their PAH distribution patterns. Results of this correlation lead to an interpretation of source–plume relationship as shown in Figure 5.7.11. 5.7.3.1.2.5 Hierarchical Cluster Analysis. Figure 5.7.12 shows the result of the hierarchical cluster analysis with a database subgroup consisting of all samples from Ulmer Straße. On the left a dendrogram is depicted. It shows the different steps of the cluster analysis, starting with every sample in one cluster proceeding to only one cluster left. The clusters were computed with samples which are relatively similar among the cluster group but relatively different to the other groups. From the distances, conclusions can be drawn if for example a particular test sample is nearer to one or another sample. On the right the clustering process is shown as a table. The first column indicates the cluster affiliation of the samples when eight clusters were computed. The second column indicates the cluster affiliation when only seven clusters were computed and so on.

328

Table 5.7.2

Chapter 5.7

Parameter and related laboratory methods used for fingerprinting at Ulmer Strasse. Used method H-18 GC-MS GC-MS method GC-MS method Full scan GC-MS of an extract from which aliphatic compounds are removed

Figure 5.7.12

cluster 7

cluster 6

cluster 5

cluster 4

cluster 3

cluster 2

Ulmer Straße 105/1 GW Ulmer Straße 105/2 GW Ulmer Straße 105/5 GW Ulm Str. W. NT13/6 Ulmer Straße 74 (2 m) Ulmer Straße 79 (3,8 m) Ulmer Straße 67 (5 m) Ulmer Straße 62 (4,2 m) Ulmer Straße 62 (2,2 m) Ulm Str. W. 108/1 Ulm Str. W. 108/5 Ulm Str. W. NT16 Ulmer Straße 71 (3 m) Ulmer Straße 71 (4-6 m) UlmS tr.W .N T10 Ulm Str. W. NT109/4 Ulm Str. W. NT109/8 Ulm Str. W. NT13/1 Ulm Str. W. NT9 Ulmer Straße 67 (3 m)

cluster 8

GC-MS method

Sample ID

Parameter Total petroleum hydrocarbon content (TPH) Volatile aromatic hydrocarbons (BTEX) MTBE (oxygenate) EDB and EDC (lead scavengers) Parameter pattern: Total ion chromatogram (TIC) Normal alkanes ( using m/z 85 mass chromatogram) Isoprenoids (using m/z 113 mass chromatogram) C4 alkylbenzenes (using m/z 134 mass chromatograms) Steranes and triterpanes (using m/z 217 and m/z 191 mass chromatograms, respectively) Polynuclear aromatic hydrocarbons (PAHS)

1 1 1 1 2 3 3 3 4 4 4 4 8 8 5 5 5 5 5 7

1 1 1 1 2 3 3 3 4 4 4 4 4 4 5 5 5 5 5 7

1 1 1 1 2 3 3 3 4 4 4 4 4 4 5 5 5 5 5 6

1 1 1 1 2 3 3 3 3 3 3 3 3 3 4 4 4 4 4 5

1 1 1 1 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4

1 1 1 1 1 2 2 2 2 2 2 2 2 2 3 2 2 2 2 3

1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Cluster analysis.

In contrast to the results of the discriminant analysis, the cluster analysis shows the differences between the samples in more detail (Figure 5.7.12). Not all gasoline or middle distillates samples were grouped in one class, as under real field conditions every individual sample is different in composition, age and degradation level. Nevertheless following the clustering process information can be extracted as to which sample is comparable with another and which samples do not fit with any other. For example, the samples from different depths of GW 105 and the sample from NT 13 grouped together indicate that their main contaminant is gasoline. During the clustering process the sample from 74 (2m) is grouped together with


329

these gasoline-contaminated samples which means that the contaminant pattern of this samples is most comparable to these others. This fits quite well to the real situation because 74 (2m) consists of gasoline and of minor amounts of diesel. Another group of samples which are relatively homogeneous are the samples from NT 109, NT 9, NT 13/1 and NT 10. This group is mainly contaminated with weathered gasoline and traces of moderately altered diesel. All other samples are relatively not comparable to this group which can be seen in Figure 5.7.12. The case study shows that statistical correlation can be a powerful tool to support the interpretative work of the expert and help to visualise results.

5.7.3.2 5.7.3.2.1

Isotopic Fingerprinting for Chlorinated Hydrocarbons General Approach

The isotopic composition of chlorinated hydrocarbon compounds released into the environment is in fact influenced by the different raw material sources and manufacturing processes associated with the synthesis.7 In this way, if the isotopic signature of the dissolved contaminant has not undergone significant fractionation, it still reflects the isotopic composition of the source, which can thus be identified. Nevertheless, due to extensive degradation the correlation of the pollutants with their suspected sources could be difficult, because the bulk isotopic composition can be affected. Preferential microbial degradation of the light compounds causes isotopic fractionation. This leads to an enrichment of the heavy isotopes in the remaining source compound and respectively to a depleted isotopic signature in the degradation product. Recent publications on PCE and TCE as the main pollution species in groundwater show a significant difference between initial product d13C and their degradation product counterparts. The isotopic fractionations reported are as follows: 2%

4%

12%

26%

PCE ! TCE ! DCE ! VC ! Eth: As the initial molecule is degraded, the associated carbon isotope fractionation is more and more important, reaching values up to 26% (Figure 5.7.13). The degradation process can be mathematically described by Rayleigh models. By means of these models a clear distinction between degradation and a mixture of different sources as a reason of differing isotopic signatures can be achieved.

5.7.3.2.2

Source–plume Relationship by Isotopic Fingerprinting at Nesenbach Site

5.7.3.2.2.1 Initial Situation. The site is located in the Nesenbach valley in the city centre of Stuttgart (see site map in Figure 5.7.14). From the beginning of the 20th century to 1976 a dry-cleaning facility was situated on the site. The

330

Chapter 5.7 0 -10

-30 Total

-40

PCE TCE

-50

cDCE

δ

13

C [‰ vs. PDB]

-20

-60

VC Ethene

-70 0

10

20

Time (j)

Figure 5.7.13

Experimental carbon isotope fractionation associated with biodegradation.8

cleaning agent first used was BTEX and later CHC, predominantly tetrachloroethene (PCE). In 1989 soil and groundwater contamination was detected. Initial investigations revealed concentrations up to 435 000 mg l1 CHC and 200 mg l1 BTEX in the groundwater. In 1990 the contaminated soil of the unsaturated zone was almost completely excavated. A probably remaining pool in the saturated zone still seems to feed a significant plume. The prerequisite for any further consideration on remediation technology is to prove whether this known source really is responsible for the extended plume and if there are other sources further downstream contributing to this plume. Besides classic geochemical characterisation methods, this identification of source– plume relationship was done by isotopic fingerprinting using d13C on specific CHC compounds. Beneath a covering of anthropogenic fill and quaternary sediments the subsurface consists of clay stones, weathered gypsum and dolomite layers of the triassic Middle and Lower Keuper. The different permeabilities of the clay stones and the fractured dolomite stones cause a vertical subdivision into several groundwater bearing zones: a low permeable first aquifer in the Middle Keuper and a more yielding second aquifer in the Lower Keuper. The area is tectonically disrupted by two major faults. Further consideration of the plume therefore is restricted to this second aquifer. 5.7.3.2.2.2 Plume Localisation by Local-scale IPTs. Based on the results of a sampling and analysis campaign in the wells shown in Figure 5.7.14, a rough localisation of the central area of the plume was performed by a series of localscale IPTs. Figure 5.7.15 shows the concentration time series for CHC of the IPTs in the downgradient monitoring wells 5, 7, 8, 9 and 12. The increasing CHC


Figure 5.7.14

331

Map of investigation area.

concentration during the IPTs is typical of wells located randomly to an existing plume. The rather straight lines in GWM 8 and GWM 12 indicate a central position in a plume. The lateral borders of the plume were not reached during the pumping phase.

5.7.3.2.3

Isotopic Fingerprinting

To prove the source–plume relationship and identify additional polluters which might contribute to the CHC plume in this area with many other potential polluters, several wells have been selected for isotopic fingerprinting by 13C isotopes of the different CHC compounds. Figure 5.7.16 summarises the results by giving both total CHC contents and d13C values.

332

Chapter 5.7

Figure 5.7.15

Concentration vs. time series during IPTs. δ13C [‰V-PDB]

-15

-17

-19

-21

-23

-25

-27

-29

-31

-33

-27,1

GWM 1 PCE: 37630 µg/l

-28,3

TCE: 924 µg/l cDCE: 543 µg/l

-30,1 -27,1


-24,3

TCE: 99,6 µg/l

-25,9

cDCE: 87 µg/l

-27,5 -27,5

GWM 8 PCE: 744 µg/l TCE: 56,1 µg/l

-30,1

cDCE: 67,5 µg/l

-25,4

GWM 7ku PCE: 150 µg/l

-30,6 -30,2

TCE: 26,2 µg/l cDCE: 21 µg/l

-27,4


-23,0

TCE: 11,1 µg/l cDCE:

Groundwater Science and Policy: An International Overview

Computer Science: An Overview

1, An overview of forest science

Computer Science: An Overview. 11th Edition

New Business Creation: An International Overview (International Studies in Entrepreneurship)

An overview of Eiffel

Quantum Information: An Overview

Quantum information. An overview

Information Technology Policy: An International History

Information Technology Policy: An International History

Information Technology Policy: An International History

Hotel Motel Operations: An Overview

Intelligent Tutoring Systems: an overview

Stationary Fuel Cells: an Overview

Bulk Metallic Glasses: An Overview

Stationary Fuel Cells: An Overview

The Extracellular Matrix: an Overview

An Overview of Faroese Grammar

Bulk metallic glasses: an overview

Chemiluminsence based biosensors-An Overview

The Forests Handbook, An Overview of Forest Science (Volume 2)

International economics: theory and policy

Groundwater and Wells

Groundwater and Wells

Affirmative Action in Antidiscrimination Law and Policy: An Overview and Synthesis (Suny Series in American Constitutionalism)

Groundwater Hydraulics (Developments in Water Science)

Groundwater in the Environment: An Introduction

Groundwater Geochemistry

U.S. and International Perspectives on Global Science Policy and Science Diplomacy: Report of a Workshop

Groundwater Age

Groundwater Models

Groundwater Science and Policy: An International Overview

Computer Science: An Overview

1, An overview of forest science

Computer Science: An Overview. 11th Edition

New Business Creation: An International Overview (International Studies in Entrepreneurship)

An overview of Eiffel

Quantum Information: An Overview

Quantum information. An overview

Information Technology Policy: An International History

Information Technology Policy: An International History

Information Technology Policy: An International History

Hotel Motel Operations: An Overview

Intelligent Tutoring Systems: an overview

Stationary Fuel Cells: an Overview

Bulk Metallic Glasses: An Overview

Stationary Fuel Cells: An Overview

The Extracellular Matrix: an Overview

An Overview of Faroese Grammar

Bulk metallic glasses: an overview

Chemiluminsence based biosensors-An Overview

The Forests Handbook, An Overview of Forest Science (Volume 2)

International economics: theory and policy

Groundwater and Wells

Groundwater and Wells

Affirmative Action in Antidiscrimination Law and Policy: An Overview and Synthesis (Suny Series in American Constitutionalism)

Groundwater Hydraulics (Developments in Water Science)

Groundwater in the Environment: An Introduction

Groundwater Geochemistry

U.S. and International Perspectives on Global Science Policy and Science Diplomacy: Report of a Workshop

Groundwater Age

Groundwater Models

Recommend Documents