ABSTRACT Climate change is more and more considered to be a major global environmental risk. To s t i m u l a t e the participation of Dutch scientists in the i n t e r n a t i o n a l r e s e a r c h effort a r e s e a r c h p r o g r a m m e was established jointly by Ministries involved in Dutch policy actions. The aim of this scientific long term, policyoriented p r o g r a m m e (NRP) was to support Dutch and i n t e r n a t i o n a l climate change policy. To conclude the first phase of the NRP an international conference was held in Maastricht (The Netherlands) from 6 through 9 december 1994. The proceedings of this conference cover a wide range of subjects including: * key note papers of internationally leading scientists on relevant aspects of the climate problem; * assessments of NRP research on the climate system, the causes of potential change in the system, the possible effects and consequences of climate change, and possible alternative policy actions (including technological and/or social); * short papers of the NRP projects and other ongoing research projects, with final conclusions per project.
vii PREFACE
The Proceedings of the International Conference on Climate Change Research: Evaluation and Policy Implications give an excellent impression of both the state of the art of climate change research in general as well as the research projects carried out during the first phase of our National Research Programme on Global Air Pollution and Climate Change (NRP). A large number of experts gathered in Maastricht, one of the most hospitable cities in our country. During discussions with participants it became clear t h a t the quality of the research presented and the organization of the conference itself were considered to be well above average. For this excellent achievement I would like to commend the Programme Committee, the Organising Committee and above all the Conference Secretariat for a job well done! The results of the conference laid down in these proceedings, the result of the international expert review brings us to our conclusion: we have to proceed on the road chosen. Also in the second phase of the continued programme (till 2001) we will put a bit of emphasis on carefull programming and accurate evaluation and presentation of the research projects. In particular the incorporation of the projects and their results within international joint efforts will be promoted. Again my appreciation for these proceedings that will attract the attention of a large research (and policy) community, also due to the timely production and distribution by Elsevier Science B.V.
T. Schneider Programme Director of the Dutch National Research Programme on Global Air Pollution and Climate Change
Editorial The D u t c h N a t i o n a l R e s e a r c h P r o g r a m m e on Global Air P o l l u t i o n and Climate Change The International Conference on Climate Change Research: Evaluation and Policy Implications, held from 6 t h r o u g h 9 December 1994 in M a a s t r i c h t , The N e t h e r l a n d s , concluded the first phase of the Dutch N a t i o n a l R e s e a r c h Programme on Global Air Pollution and Climate Change (NRP). The second phase of this programme started in 1995 and will last to 2001. The conference covered a wide range of subjects, including the climate system, the causes of potential change in the system, the possible effects and consequences of climate change and possible a l t e r n a t i v e responses ' w i t h i n the context of s u s t a i n a b l e development. About 350 scientists, r e s e a r c h m a n a g e r s a n d policymakers from in and outside the Netherlands participated in this succesfull conference. These proceedings contain the texts of the opening statement made by the Dutch Minister of Housing, Spatial Planning and the Environment, invited papers of internationally recognised experts which give a state of the art assessment for several areas of climate change research (part one), assessment reports of the various parts of Dutch climate change research and short papers about Dutch and foreign ongoing research projects (part two). History Climate change is regarded as a serious global problem. Over the years the insight has grown that climate change may pose a serious threat to the world (and also to the Netherlands). To know more both about the nature and seriousness of the problem as well as about the possibilities for countering its effects an intensive international research effort is needed. The NRP was established in 1990 with the aim of providing a scientific basis for the development of climate change policies, and to increase the involvement of the Dutch research community both nationally and internationally in this field. The design of the programme around specific policy-relevant goals distinguishes it from many traditional approaches. Central questions for climate change policy and research The threat of climate change has posed governments a new problem and, in terms of its size and nature, one which is very difficult to manage. Central questions for policy making are: what is going on; what is at stake; what can be done about it, how and with what consequences; what should be the timeframe and what actions should be t a k e n by whom? Such questions have to be translated into research terms in order to arrive at a research programme. A first question with respect to the global climate problem is: How does the climate system work? W h a t processes are going on? Which and w h a t kind of climate
fluctuations can be expected? When are such deviations abnormal and do they have an anthropogenic origin? How predictable is all this? A further question is how seriously the global carbon cycle is disturbed? What is the role of human activity in this? What is the influence on the global system of the greenhouse gases that are emitted? Another question is related to the impacts and consequences of climate change on nature and society. Are the risks associated with climate change larger than those related to other changes in society, especially in the developing countries? How does all this fit in with the goal of a world-wide sustainable development? And finally there is also the important question related to the way in which society deals with the climate change problem. Is it possible to mitigate emissions by the introduction of new technology? Or will it also demand the adaptation of societal structures and institutions, together with changes of lifestyle.
Organisation of the Dutch programme The nature and scale of the climate problem demands a far broader approach than hitherto has been used. The programme comprised 150 research projects, the contents of which were distributed over five themes. The research effort represents approximately 700 man-years, of which 60% is contributed by the research institutions while 40% is funded through NRP. Over 30 research institutes and universities partcipated in the programme. The NRP is presenting itself in the international arena with "Change", a research and policy newsletter on global change. Because of its policy-oriented mission, the NRP has a broad framework. The programme embodies fundamental scientific research to study the m a n n e r in which the climate system works and the physical and chemical processes t hat may produce climate change. Research on causes (emissions) contributes also to the knowledge of the climate system. The programme includes research towards the potential impacts of climate change and possible responses (technical, economic and behavioural/social options). The assessment component of the programme involves the synthesis, integration and communication of research results which provides the basis for decision making and policy actions. In practice NRP is structured according to five themes, which also constitute the structure of these proceedings: 9 the climate system: functioning, modelling and monitoring 9 greenhouse gases: underlying causes of changes in the climate system 9 impacts and consequences of climate change 9 sustainable solutions 9 integration of climate change research. Besides these proceedings, final project leaders and the final report the end of 1995. The report of the 1995) which was held during and Programme Office.
reports of all projects are available from the of the first phase as a whole will be published by international review of the programme (S&PA, just after the conference, is available from the
Acknowledgements We would like to thank Marianne Vonk who did an excellent job in organising the Maastricht conference. She also took care of the preparations for this proceedings.
xi Ottelien van Steenis and Mini Schneider's assistance was a welcome contribution both before and during the conference. We also like to recognise the work of the chairman of the conference, dr. B. Metz, the chairs of the different sessions and the rapporteurs. Last but not least we are greatful for the work of the Programme Committee and the assistance of Sue Postle H a m m o n d and Chris Bernabo of Science and Policy Associates.
The Editors Bilthoven, August 1995 NRP Programme Office P.O. Box 1 NL-3720 BA Bilthoven The Netherlands e-mail:
[email protected] OPENING ADDRESS M A R G A R E T H A DE BOER M I N I S T E R OF HOUSING, SPATIAL P L A N N I N G AND THE E N V I R O N M E N T Ladies and Gentleman, It was w i t h g r e a t p l e a s u r e t h a t I accepted the invitation to address you at the s t a r t of this i n t e r n a t i o n a l conference. You are here to evaluate the results of the first N e t h e r l a n d s ' R e s e a r c h P r o g r a m m e on Global Air Pollution a n d C l i m a t e Change, N R P I. And you will also be discussing strategies for the second N R P I have come here today as a politician, to tell you w h a t the Government is expecting from the scientific c o m m u n i t y over the next few years. M a n y of the world's climate scientists are contributing to the invaluable work of the I n t e r g o v e r n m e n t a l P a n e l on Climate Change. The IPCC has concluded t h a t the serious risk of human-induced climate change justifies immediate action. Going on t h e s t r e n g t h of the p r e c a u t i o n a r y principle, the i n t e r n a t i o n a l c o m m u n i t y , therefore, took steps to minimize these risks by establishing the U n i t e d Nations' F r a m e w o r k Convention on Climate Change. The developed countries c o m m i t t e d t h e m s e l v e s to curbing their emissions of carbon dioxide to 1990 levels by the y e a r 2000 as a first step. The N e t h e r l a n d s h a d a l r e a d y i n t r o d u c e d a climate change policy before t h e Convention was finalised. It aims at a 3% reduction for Carbon Dioxide (CO2) by the y e a r 2000 relative to the 1990 level, a 10% reduction of m e t h a n e emissions and a stabilization of Nitrous Oxide (N20) emissions. This policy has always rested on a broad consensus in P a r l i a m e n t and society. M e a s u r e s to support the 2000 t a r g e t are also beneficial from other points of view. In the N e t h e r l a n d s , we w a n t to do as m u c h as we can to link the i n t e r e s t s of economy and the environment. With the m e a s u r e s t a k e n so far, we have certainly m a d e a good start. Nevertheless, our most recent forecasts indicate t h a t we will need an incentive tax on energy if we are to meet our commitment. Of course, I am still hoping for a common energy tax w i t h i n the E u r o p e a n Union. B u t the new cabinet has committed itself to the introduction of such a tax in the N e t h e r l a n d s in 1996 if the preferred option fails. An i m p o r t a n t e l e m e n t in our climate policy has been the N a t i o n a l R e s e a r c h P r o g r a m m e , the reason for this conference. M a n y of you have participated in this p r o g r a m m e which was set up to increase our knowledge on the climate system, the causes and effects of rapid climate change, and s u s t a i n a b l e solutions. I t s unique integration of strategic and applied research has received international acclaim. It was m e a n t to encourage our scientists to p a r t i c i p a t e in i n t e r n a t i o n a l r e s e a r c h programmes.
These p r o g r a m m e s have greatly enhanced our u n d e r s t a n d i n g of the climate system. And the results so far have not changed the basic conclusions of the 1990 IPCC report. Of course, there are still many uncertainties left, but we must realize t h a t every answer may raise new questions. We have to learn to cope with this. Climate change is a difficult problem to handle for everyone: for scientists, for politicians, for the media, and for the general public. The atmosphere is a very complex system, and the links between h u m a n activities and the climate are not always clear. There are m a n y things we do not yet know precisely. Furthermore, there are long time lags between causes and effects. The picture is very confusing, especially for the ordinary man in the street. Let me say a little bit more about this. Until now, the emphasis on uncertainties has dominated the public discussion about the greenhouse effect. And as I said, this leaves m a n y people, in the N e t h e r l a n d s too, confused about the seriousness of the matter. Some react by playing down the probability and consequences of climate change or by denying the problem outright. Others argue t h a t we can afford to wait and see, and rely on adopting measures should threat of flooding become a reality. The u n c e r t a i n t i e s could contribute to a wait-and-see attitude. I t h i n k such attitudes are fundamentally flawed. We m u s t realize that by the time scientific uncertainties have been resolved it will be too late. The consequences of rapid climate change could become very costly and serious -maybe even irreversible- as time goes on. In a country like the Netherlands we may see damage to ecosystems and a decreasing supply of fresh water. Developing countries may even have to endure worse problems. We cannot possibly afford to be indifferent about these issues. In general, however, I t h i n k t h a t the public expressions of doubt t h a t I j u s t mentioned should be a signal for us to use a different type of communication. I n s t e a d of e m p h a s i s i n g u n c e r t a i n t i e s , we need to express our scientific understanding in clear terms of risk. That is the type of information policymakers, but also decision-makers in the business community and ordinary citizens are accustomed to handling. Most of our economic decisions are taken in the light of various risks. So that is the type of information politicians and the public need to receive from the scientific world! I t h i n k t h a t scientists involved in the National Research Programme should take the lead in changing tracks to a risk-based discussion of the n a t u r e and the consequences of climate change. This implies translating the results of effects studies into operational terms, attuned to the most vulnerable areas of the world. There is another fact that is becoming increasingly important for the progress of international climate policy. In international discussions there are many questions t h a t need to be answered on the basis of the latest scientific findings. Research efforts should therefore be closely attuned to the questions t h a t arise in the international policy arena. So, let me now give you an impression of what is going on in the run-up to the first meeting of the Conference of the Parties to the Climate Convention, which is to take place in March next year in Berlin.
Emission stabilization by the developed countries is only the first step towards achieving the ultimate objective stabilization. And it is a very small step if we look at the m e a s u r e s t h a t will ultimately be necessary! At their first meeting, the Parties will have to decide what the next step should be. IPCC has been requested to bring t o g e t h e r and assess the scientific information t h a t would help in determining the need for future commitments. Preliminary results show t h a t the development of global emissions over the next decade or so is crucial to w h a t we can ultimately achieve. If we take no further action we will probably not attain the u l t i m a t e objective of the Convention. Consequently, Prof. Bert Bolin, rightly advised INC delegates to address the question of commitments for the first decades of the next century as soon as possible. Now, how are we going to set out our course. We do not yet know exactly w h a t a dangerous level of greenhouse gases in the atmosphere means. What we can say right now, is that it would be foolish to cut off certain options at an early stage. We m u s t keep in mind t h a t the risks are two-way. There is not only the risk of inaction. There is also the risk of being more aggressive than necessary in dealing with the problem. The challenge is to strike the right balance by designing a kind of step-by-step risk optimization process. We need to adjust our course at regular intervals on the basis of the best available information at the time. We must try to m a n a g e the risk of climate change -even if the problem t u r n s out to be more serious t h a n we think- nor should we disregard the risk of major changes in our economies. Let me illustrate the dilemma as follows. When you are driving a car in the mist you cannot look far ahead. So what do you do? Of course, you drive slower and more carefully so t h a t you can anticipate in time. You also look attentively to any signs by the roadside to see what lies ahead. One example of applying this approach would be to assess potential investment decisions in certain areas with respect to their effect on the overall energy intensity. In areas such as the transportation infrastructure, buildings and energy supply systems we should avoid development in the direction of more energy and Carbon Dioxide (CO2) intensive patterns. Everything should be done to think ahead in this areas and launch innovative approaches in them. During the first meeting of the Conference of Parties, therefore, I will actively support initiatives aimed at reaching international agreement ion specific areas, such as energy conservation and economic instruments. For m a n y developed countries, including the Netherlands, it is probably already going to be difficult to achieve the Carbon Dioxide (CO2) target for the year 2000. Nevertheless any credible strategy m u s t soon include a scenario beyond 2000. Investors with a long-term horizon need guidance with respect to long-term developments in climate policy. I would urge participants in the NRP to provide policy-makers with the basic information they need to design the policy options for the period beyond 2000. Finally, I m u s t conclude t h a t ongoing scientific research into all the relevant aspects of the climate change problem is needed to lay a solid basis for climate
policy. The second National Research Programme should thus continue to investigate the workings of the climate system. It should also provide ammunition for developing strategies and measures for adaptation and mitigation for the period beyond the year 2000. Finally, I would strongly recommend the risk-based approach as a new dimension in effect research and in the communication of results. Both politicians and citizens urgently need information of this kind to clarify their thinking and bring the climate change problem closer to solutions. Ladies and Gentleman, I hope that you will be able to shed light on the issues I have mentioned during the days ahead. I wish you a very fruitful and inspiring conference. Thank you very much!
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
Current Progress in the Study of Global Biogeochemical Cycles Michael H. Unsworth and Gordon Wolfe College of Oceanic and Atmospheric Sciences Oregon State University, Corvallis, Oregon, 97331-6511, USA
Abstract
Aspects of global cycles of carbon, nitrogen and sulfur are reviewed. New work defining carbon source and sink strengths in oceans, northern forests and tundra, and wetlands is discussed. Effects of CO2 and N fertilization on the carbon cycle may be large but are currently ill-defined. Recent changes in rates of increase of C02, CH4 and CO in the atmosphere are probably related to volcanic eruptions. The global nitrogen cycle is grossly disturbed by human activity. Land use change and fertilizer use in the tropics may be major sources of N20. Ammonia merits further study as a regional pollutant and because of its role in tropospheric aerosol formation. Sulfate aerosols are now recognized as having significant negative forcing on climate. Whilst the direct (radiative scattering) effect of aerosols is well-understood, the indirect effect (altering cloud properties) is very uncertain, and inclusion of aerosol effects in climate models is limited by lack of data on aerosol regional distributions. In the future, industrial growth in developing countries will alter the amounts and global distribution of greenhouse gases and aerosols, and the change and distribution of aerosols will have particularly important implications for future regional climate change.
1. INTRODUCTION
Studies of biogeochemical cycles play an intrinsic part in research programs into climate change. In particular, cycles which involve radiatively active gases and particles have received a great deal of attention in recent years, partly to establish the strengths of sources and sinks and how these strengths are altered by human activity, and partly to investigate the processes by which the cycles are driven. Increasingly, it is being recognized that biogeochemical cycles of elements cannot be treated as entirely independent; improved knowledge of the chemistry which takes place in the atmosphere, soils and the oceans is revealing many ways in which cycles are intricately linked.
l0
In the space available in this paper, we have chosen to concentrate on some of the current issues associated with the biogeochemical cycles of three elements: carbon; nitrogen; and sulfur. Inevitably this assessment is biased by our personal interests, but we have attempted to include examples of some of the exciting recent developments in the atmospheric, oceanic, and terrestrial sciences. 2. THE CARBON CYCLE
Three issues have dominated research into the carbon cycle in recent years. First, balancing the C02 budget - establishing the strengths of sources of atmospheric C02 arising from human activity and natural systems, and of sinks in the ocean and on the land. Second, understanding feedbacks by which the sources and sinks of a number of carbon trace gases interact with climate change and with increasing C02 concentration. Third, most recently, investigating possible causes of sudden changes in the rates of atmospheric accumulation of a number of trace gases that have been observed in the early 1990s. 2.1
Balancing the CO 2 budget
Table 1 shows the global C02 budget for the decade of the 1980s as proposed by the IPCC (1994). Emissions are relatively well-established, and occur mainly (90%) in the northern hemisphere from fossil fuel combustion and cement production. Emissions from tropical land use change remain poorly quantified, and improved data from southern Asia, Africa, and tropical South America, preferably collected with a common methodology such as high-resolution satellite imagery, are urgently needed. Until recently it was believed that tundra ecosystems were globally a net accumulator of carbon at a rate of about 0.1 to 0.3 Gt yr 1, but recent work by Oechel et al (1993) suggests that warming in the Arctic may have changed these regions to sources with a global strength of about 0.2 Gt yr -1. The accumulation of carbon dioxide in the atmosphere is very well defined by a global network of monitoring stations. Analysis of the stable isotope 13C02 shows clearly that the seasonal amplitude in concentration which varies around the globe is dominated by the activity of the terrestrial biosphere in the northern hemisphere, rather than by seasonal changes in fossil fuel emissions or in ocean sink strength. Recent work by Farquhar et al. (1993) on the isotope composition of oxygen in atmospheric C02 leads to the possibility of distinguishing influences of different terrestrial biomes in the global C02 monitoring network, and when this approach is combined with atmospheric mixing models, it may be possible to resolve some of the present conflict about the relative role of the oceans and the terrestrial biosphere in the net uptake of C02.
1!
Table 1
Annual average budget for anthropogenic carbon for 1980-1989 in GtC/yr
Sources Fossil fuel Changes in tropical land use Total emissions Partitioning to Reservoirs Storage in the atmosphere Ocean uptake Northern Hemisphere forest regrowth Other terrestrial sinks (CO2 and N fertilization, climate effects)
GtC/yr 5.5 + 0.5 1.6_+ 1.0 7.1 _+ 1.1 3.2 2.0 0.5 1.4
+ 0.2 + 0.8 + 0.5 _+ 1.5
Data from IPCC 1994
Estimates of carbon uptake by the oceans have been made by t w o independent approaches. Methods using radiocarbon, produced naturally in the upper atmosphere or artificially during nuclear weapons testing, as a tracer give larger estimates of uptake by oceans than methods based on air-sea exchange. A recent analysis (Sesshaimer et al. 1994) suggests that estimates by the radiocarbon tracer method may need to be reduced by about 25 %. The air-sea exchange method, which is based on differences in the partial pressure of CO2 between the water and the air, and on an exchange function described in terms of wind speed and temperature, has also been reassessed (Robertson and Watson 1992) These authors pointed out that the upper 1 mm or so of the oceans is generally cooler than the bulk mixed layer by about 0.3 o C, and when this thermal skin effect is taken into account, the air-sea exchange method results in CO2 uptake estimates that must be increased by about 0.7 Gt C yr 1 . Thus recent work has brought the two approaches into much closer agreement in defining ocean uptake. There has been debate about whether increases in ocean phytoplankton productivity may have increased the ocean sink strength for CO2 over the last 100 years or more. Analysis by Falkowski and Wilson (1992) of changes in phyto- plankton biomass in the north Pacific Ocean over the last 70 years shows that changes are too small to have a significant effect on the sink strengths for atmospheric CO2. Although there are very few historical data for other main ocean basins, it seems likely that this conclusion applies on a global scale. In contrast to the open ocean, coastal zones have undoubtedly experienced large increases in nutrients associated with human population changes, but Falkowski and Wilson concluded that there is no conclusive evidence yet to suggest that these coastal zones represent a significant new sink.
12 As an example of interacting biogeochemical cycles between the atmosphere and ocean, it has been proposed that, because iron concentrations limit phytoplankton productivity in some parts of the ocean, deposition of iron from either human activities or volcanic eruptions might increase the ocean sink strengths for CO2. An experiment has recently been conducted in the equatorial Pacific Ocean to test this hypothesis, and Watson et al. (1994) reported some early results. When a small patch (8 x 8 km) was enriched with iron, there was a significant depression in the surface concentration of CO2 within 48 hours of the iron release, but the effect was only a small fraction (about 10%) of the CO2 drawdown that would have occurred if the enrichment had resulted in the complete utilization of all other available nutrients. Reasons why the fertilization procedure was much less effective in the open ocean than in the laboratory are not yet clear, but at present the results do not support the idea that iron fertilization significantly affects the oceanic CO2 sink strength. The weight of evidence at present suggests that in order to balance the sources of CO2 in Table 1 against sinks and atmospheric storage, there must be an illdefined terrestrial sink in northern temperate latitudes. This conclusion is reached by inversion of the observed atmospheric CO2 distribution combined with atmospheric tracer models, making constraining assumptions about ocean and terrestrial sinks (Tans et al. 1990, Enting and Mansbridge 1991). The likeliest terrestrial processes contributing to this sink are: changing forest management; and enhancement of productivity due to atmospheric CO2 increases and/or nitrogen fertilization from atmospheric deposition. A number of recent carbon inventories of temperate forest systems have concluded that such systems may have been a sink for about 0.5 GtC/yr in the last 20 years or so, partly because of natural regrowth and replanting after forest harvesting, and partly through fire suppression (IPCC 1994, Auclair, personal communication). Although there is much evidence from laboratory and controlled environments to show that plant productivity can be increased by 30% to 40% when CO2 is doubled, there is no conclusive evidence from the field to show long-lasting increases in northern temperate ecosystem productivity in response to increased CO2. There are sound biochemical reasons indicating that the interaction of CO2 and temperature is such that the benefits of CO2 fertilization cannot be achieved at low temperatures (Long, 1991 ). This may explain partly why, when Oechel et al. (1994) exposed natural arctic tundra to doubled CO 2 concentration, there was no long-term boost in carbon sequestration. An alternative explanation is that, after an initial burst of productivity, plants exhaust the supply of soil nutrients, and this limits future productivity. There have been insufficient long-term experiments with perennial systems such as forests and tundra to determine whether the equilibrium longterm response involves eventual changes in soil nutrient availability that would allow productivity to be enhanced. IPCC (1994) estimated that CO2 fertilization may have accounted for a sink of 0.5 to 2.0 GtC/yr during the 1980s, but such estimates must be regarded as extremely tentative.
13 In and around industrialised regions, ecosystems may receive substantial inputs of nitrogen, arising from fossil fuel burning and agriculture, and this input can act as a fertilizer. IPCC (1994) speculated that this fertilizer effect could have increased terrestrial carbon storage by 0.2-1.0 GtC/yr in the 1980s. One of the most important current programs aimed at understanding the interaction between boreal forests and the atmosphere is the international multi-agency BOREAS Program taking place in northern Canada (Sellers, et al. 1995). Analysis of the many detailed records of CO2 exchange collected during BOREAS should help in quantifying the scale of net carbon input to northern forest systems. 2.2 Feedback processes
We have already mentioned a number of studies which indicate interactions between changing climate and atmospheric CO2 concentrations and the net exchange of CO2 between the surface and the atmosphere. There has also been interest in the sensitivity of methane fluxes to climate change and CO2 concentration. Whiting and Chanton (1993) found that, for wetlands of varying productivity around the world, higher net primary production was associated with higher emissions of methane. It has therefore been suggested that, if CO2 increased the productivity of wetland vegetation, some of the benefits of carbon sequestration would be lost because of increased methane emissions. Dacey et al. (1994) recently presented results supporting this view. They studied methane emissions from a marsh that had been exposed in open-top chambers to twice ambient atmospheric CO2 for the previous 7 years, and found that methane emission over a 10-day period from the CO2-enriched sites was nearly 80% higher than in control sites. If confirmed in longer-term work, the implications of this observation are important, not only for methane fluxes from natural ecosystems but also for fluxes from wetland rice production where much effort is put into increasing productivity. Most soils that are not flooded consume methane, but the extent varies with soil water content and land use. A number of recent reports have shown that inputs of nitrogen in the form of ammonium to soils strongly inhibit soil methane consumption (King and Schnell 1994). Ammonium concentrations in many soils have increased in recent years as a result of land use changes and increases in the ammonium concentration of precipitation. Similar responses are not observed in soils treated with nitrate-based fertilizer or farmyard manure (Goulding et al., in press). Goulding et al also analyzed soils from long-term experiments at Rothamsted, England, and showed that extended (150 years) cultivation of land for arable crops reduced methane uptake rates by 85% compared to those in soil under calcareous woodland. King and Schnell argued that past increases in atmospheric methane concentration may have increased the inhibitory effect of ammonium on soil methane uptake, and this mechanism would provide a positive feedback on future atmospheric methane concentrations.
14 2.3
Recent changes in atmospheric accumulation of trace gases
One of the most puzzling and yet instructive aspects in the study of trace gas biogeochemistry occurred in the early 1990s. Until this time, CO2 concentrations around the world had increased rather consistently over the previous 30 years at about 0.5 - 1.5 vpm per year, with the rate tending to increase With time (Fig. 1).
.-..
3.0
2.5 "~ 2.0 o o
c? O
"6
~.5
1.0 0.5
~! ii
ii I !, ii I
o
;5 1960 1965 1970 1975 1980 1985 1990 1995 Year
Figure 1
Growth rate of CO2 at Mauna Loa, Hawaii. The smooth curve is filtered to suppress short-term (-~10 yr) variations. (From IPCC 1994)
After mid-1991, continuing throughout 1992 and 1993, the growth rate of atmospheric carbon dioxide slowed by an unprecedented amount (Keeling, 1993, Sarmiento, 1993). Concentrations of carbon monoxide CO and methane CH4, which had also been increasing steadily up to 1991, grew slowly from 1991 to 1993 (Fig 2). The cause of these large changes was almost certainly the eruption of Mount Pinatubo in the Philipines in June, 1991, but the mechanisms that brought about the changes are a matter for debate. As we discuss later, it seems likely that the carbon dioxide anomaly is associated with the volcanic aerosols that were injected into the stratosphere, reducing solar radiation at the surface and producing cooling on a global scale. Cooling, and
15 perhaps associated changes in rainfall and evaporation, could alter the balance between photosynthesis and respiration on the land, and the sink strengths of the ocean for carbon dioxide.
175O ..-.. > JO
o_ 1725
g "=
1700
co
1675
c 0 o r-
1650
*
1625 I
'83
'84
I
'85
I
'86
!
'87
I
'88
'89
I
'90
I
'91
I
'92
'93
Year
Figure 2
Globally averaged CH 4 concentration showing low growth rates during 1992 and 1993. (From IPCC 1994)
The changes since 1991 in atmospheric methane and carbon monoxide are probably even more complex, because both gases have a major atmospheric sink by reaction with the hydroxyl radical OH. A recent analysis by Bekki et al. (1994) suggested that an unprecedentedly large depletion of stratospheric ozone over this period may have contributed to the sharp decrease in growth rates of both gases. The decrease in stratospheric ozone would allow more ultraviolet radiation to reach the troposphere and this would have resulted in increased concentrations of OH. Bekki et al. concluded that they could account for almost half of the 1992 decrease in growth rates of both gases by this mechanism, but there may also have been changes in source and sink strengths at the surface to account for the remainder. It seems likely that the ozone depletion in the stratosphere was also caused by the Pinatubo eruption, because the fine aerosol of sulfuric acid droplets resulting from the injection of 15 to 20 million tonnes of sulfur dioxide into the stratosphere interacted with other stratospheric chemicals to destroy ozone. The consequences of this natural event provide an excellent test of our understanding of the carbon cycle, and serve as a reminder of the complex interactions that are contained within the carbon cycle.
16 3. THE NITROGEN CYCLE In comparison to the global carbon cycle, the global nitrogen cycle has been much more grossly disturbed by human activities. Table 2 summarizes annual terrestrial fluxes prior to substantial human alteration, and lists current changes arising from human activities.
Table 2 Global fluxes of nitrogen (in Tg/yr) in unperturbed terrestrial ecosystems and as a consequence of human activity
(i)
(ii)
(iii)
Unperturbed systems Biological N fixation Fixation by lightning Denitrification Perturbed systems Biomass burning Tropical land clearance New fixation Manufacture of fertilizer N N fixation by legume crops N fixation by fossil fuel combustion
TaN/vr 100 10 9O 40 22 80 30 25
Data from Vitousek and Matson, 1995
Since the late 1970s, the production of nitrogen fertilizer has probably increased by about one-third (Vitousek and Matson, 1995). Prior to that time, most of this fertilizer was used in developed countries of the temperate zones, but since then, the rate of increase in nitrogen fertilization has been extremely rapid in the tropics, so that by the late 1980s, more than half of the global nitrogen fertilizer use was in the developing world (including China). This trend is likely to continue, with the implication that the distribution of trace gas fluxes to the atmosphere, discussed in the following sections, will change substantially. Changes in the source strengths of t w o trace gases as a result of the disturbed nitrogen cycle have attracted particular attention in recent years. Nitrous oxide emissions are influenced by fertilizer use and land use changes; and ammonia arising from both fertilizer use and animal production has been shown to create important regional problems, especially in Europe.
17 3.1
Nitrous Oxide
The global budget of nitrous oxide has been revised recently by the IPCC (1992), based on new information on soil fluxes from tropical ecosystems and temperate forests, further work on cultivated soils, and new estimates of emissions from biomass burning. Large tropical sources are required to explain the N20 latitudinal gradient revealed by atmospheric monitoring over the last ten years. Keller and Matson (1994) recently reviewed sources of N20 in the tropics and evaluated the effects of land use changes. Wet undisturbed tropical forests appear to account for the largest natural terrestrial source, and this is associated with the large rates of nitrogen transformation in the soil and cycling through vegetation in these systems. Even so, it seems necessary to include additional tropical sources to balance the global budget, and Keller and Matson proposed that tropical land use change and intensification of tropical agriculture may be significant contributors towards this missing source. In particular, the creation of young pastures from previously undisturbed systems seems likely to considerably increase N20 fluxes to the atmosphere, and the increasing use of nitrogen fertilizer in tropical systems appears from the few measurements available to cause larger fluxes of N20 than would be found from similar practice in temperate crop systems. It seems likely that crop and soil management practices can be manipulated to control nitrous oxide flux and there is clearly a need in temperate and tropical areas for further work to explore this possibility. Since nitrogen fertilization increases emission of nitrous oxide to the atmosphere and may decrease absorption of methane by soils (see earlier), the potential of improved soil management for slowing the buildup of radiatively active trace gases in the atmosphere is very important. 3.2
Sources and Sinks of Ammonia
Concern about ammonia fluxes from intensive agriculture in Europe and eastern North America arises principally because dry and wet deposition of reduced nitrogen compounds can make a substantial contribution to the acidification and nitrogen eutrophication of semi-natural systems (Fowler et al. 1989). The importance of ammonia fluxes in global radiative forcing has not been adequately explored yet, but reactions between ammonia and sulfur dioxide to create ammonium sulfate aerosols, discussed in the next section, are an important contributor to the global anthropogenic aerosol burden. One of the most important advances in recent years has been the development of micrometeorological methods for studying ammonia fluxes between vegetation and the atmosphere. These techniques have allowed investigations over grazed pastures, fertilized agricultural crops, and natural ecosystems. Sutton et al. (1995) recently reviewed this work. The measurements have clearly demonstrated that there is an NH 3 'compensation point' associated with plant tissue, so that emission from plants occurs when atmospheric concentrations are below the compensation point, and deposition occurs at
18
higher atmospheric concentrations. Figure 3 summarizes measurements over agricultural crops and semi-natural vegetation (Sutton et al. 1995), showing that at low NH 3 concentrations, agricultural crops are a source of NH 3 to the atmosphere and semi-natural vegetation is a sink. The figure also indicates that the compensation point is different for the two systems, as would be expected as a result of nitrogen fertilization of crops. It seems likely that natural vegetation on which there has been substantial NH3 deposition could also have a higher compensation point, and so the sink strengths of natural vegetation for NH 3 may decrease in polluted regions with time.
30i
8~
20
o '7,
E u.. z v
00
OOC~
9 9
9
agricultural croplands
0
semi-natural vegetation
:
-10 0
0
00~
-20 -30
o
Oo
O 0
-
o~
o
00 0
0
O0
-40 -50 -60 O0 -70 0.0
i
i
i
I
i
I
0.5
1.0
1.5
2.0
2.5
3.0
3.5
NH3 (~g m "~)
Figure 3
Micrometeorological measurements of the variation of NH 3 fluxes (positive = away from the surface) with NH 3 concentration over agricultural crops and semi-natural ecosystems (From Sutton et al., in the press)
The global ammonia budget has been relatively neglected in studies of the nitrogen cycle. Schlesinger and Hartley (1992) concluded that the major uncertainties and emissions were associated with fluxes from undisturbed soils. They also particularly commented that deposition of ammonium in rain in the eastern United States declined over the 20-year period from 1963 to 1982, which they regarded as surprising, given the increasing use of urea fertilizer. They concluded that atmospheric interactions with sulfur dioxide to form sulfate
19 aerosol may account for this observation, but it also seems likely from recent work that ammonia deposition on natural vegetation close to agricultural fields has substantially increased, and the consequences of this nitrogen input deserve more attention. 4. THE SULFUR CYCLE
The global sulfur cycle has been severely perturbed by human industrial activities for many years. Because world industrial activity and fossil-fuel consumption have been concentrated in the northern hemisphere, there are virtually t w o global sulfur cycles: a relatively "natural" cycle in the southern hemisphere, and a vastly perturbed cycle in the northern hemisphere. Some consequences of the perturbed cycle, such as acid rain, have been recognized for decades. But we have only in the past few years begun to appreciate the potential importance for global climate of human impacts on the sulfur cycle. In the following sections we will briefly review recent findings about the climatic roles of both the natural and perturbed sulfur cycles. 4.1
Sulfur, aerosols, and climate
Scattering of solar radiation by atmospheric aerosols effects the atmospheric radiation balance because it reduces the amount of energy from the sun that reaches the surface of the earth. Scattered radiation may be either lost to space or absorbed by aerosols, but both processes result in a net loss of energy to earth's surface and therefore net cooling. Thus, aerosol scattering has a negative forcing influence on climate - the opposite of greenhouse gases such as CO 2. Although the mass of sulfur in the atmosphere is only about 10 .5 that of carbon, it dominates scattering in the atmosphere. This is because sulfur oxidizes and hydrates to form sulfate aerosol particles that are in the size range (0.1 - 1pro) that scatters visible radiation very effectively. Such scattering is roughly 105 times more efficient per atom as a climate forcing mechanism than the radiation absorption of greenhouse gases such as CO2 (Shaw, 1983), but the radiative influence of aerosols on climate is reduced by their much shorter atmospheric residence time, less than a week, compared to an effective residence time of decades for CO2. Sulfate aerosols contribute to radiation scattering in two distinct ways: by scattering directly and by modifying cloud optical properties and thereby influencing radiation scattering by clouds. The latter "indirect" effect has been appreciated as an important feature in global climate for some years, but is an extremely complex process and is still poorly understood.
20 4.1.1
The natural sulfur cycle and climate
We have surprisingly little quantitative understanding of the role of the natural sulfur cycle in climate. This is partly because the subject has only recently attracted much attention, and partly because the natural cycle has been so perturbed by human activity. There are few data on the distribution of sulfate aerosols even today, much less from prior eras, and trying to find the climate signal of natural sulfate aerosols in a vastly perturbed world is extremely difficult. Nonetheless, there are several likely mechanisms by which the natural sulfur cycle plays a role in climate. The marine dimethyl sulfide (DMS) effect - the brightening of marine stratus clouds hypothesized by Charlson et al. (1987) - may have been the dominant process by which the natural sulfur cycle affected climate before the industrial revolution. An intriguing aspect of this hypothesis is the suggestion of a potential feedback loop which, if negative, would act to stabilize climate. However, studies of ice-core concentrations of methane sulfonic acid and non-seasalt sulfate (atmospheric oxidation products of DMS) (Legrand et al., 1988) suggest instead that there was a positive feedback during previous ice ages. To date, evidence supporting the marine DMS hypothesis has been difficult to gather (Ayers and Gras 1991, Ayers et al. 1991, Bates et al. 1987) because of the difficulties associated with measuring simultaneously all the necessary variables (Bates et al., 1990) and the interference from anthropogenic sulfate aerosols (Falkowski et al. 1992, Schwartz 1988). Field studies planned for 1995 - 96 in the southern Pacific ocean are aimed at determining whether the 'cloud brightening' mechanism actually exists, and its potential strength. Volcanic eruptions (e.g. Mt. Pinatubo in June, 1991 in the Philippines) have important impacts on the natural sulfur cycle by injecting huge pulses of sulfur into the stratosphere which oxidize to form sulfate aerosols. Unlike tropospheric aerosols, these stratospheric aerosols remain in the atmosphere for several years due to the lack of aqueous removal mechanisms in the extremely dry stratosphere. They therefore achieve circumglobal distributions and may cause global cooling for a significant period. Such aerosols also deplete stratospheric ozone, as discussed earlier.
4.1.2 Anthropogenic changes to the S cycle Sulfur is an integral element in all biological materials, and all biogenic oil and coal contain approximately 1% - 10% S by mass. Therefore, production of sulfur gases is an inevitable by-product of fossil fuel combustion.
21 Emissions of sulfur to the atmosphere from human activities are now 2 - 3 times natural emissions annually. In the past few decades, the major areas of sulfur flux to the atmosphere have been the eastern United States and western and central Europe, and globally, about 94% of the emissions are in the northern hemisphere (Schwartz 1988). Because aerosols are removed from the atmosphere before they can be transported across the equator, this leads to vastly different distributions of aerosols in the northern and southern hemispheres (Langner et al. 1992). The sulfate haze that blankets much of the northern hemisphere is now recognized to be a direct result of fossil fuel burning rather than from natural sources. Further evidence for the disparate hemispheric cycles comes from 200% - 300% increased sulfate deposition to Arctic ice but not to Antarctic ice in the past hundred years (Mayewski et al. 1990). Because no global or even regional aerosol sampling network exists, and satellite observations are lacking, our best estimates of aerosol distributions come from computer models which begin with known sources and simulate atmospheric transport, chemistry, and physics to predict aerosol distributions (Langner et al. 1992). Although the mechanism of cooling by direct aerosol scattering is relatively simple, its impact was until recently underestimated, largely because it was not realized how much of the sulfate aerosol haze in the northern hemisphere is actually from industrial emissions. Although one recent estimate suggests that perhaps less than 10% of sulfur emissions result in the formation of new aerosol particles (Langner et al. 1992), the rate of new sulfate particle formation may have doubled since pre-industrial times. Recent re-evaluations of the direct climate forcing by radiation scattering from anthropogenic aerosols have suggested cooling of similar magnitude to the C02 warming (Charlson et al. 1992, IPCC 1994), leading to speculation that the "greenhouse signal" predicted in the late 1980's has been partially masked by concomitant sulfate aerosol production. Because the cooling due to aerosol scattering is localized, it is thought to be heavily concentrated around eastern North American and central Europe. Aerosol modification of cloud albedo (indirect climate forcing by aerosol) is a much more difficult problem. The distribution and radiative properties of clouds are probably the major uncertainties in climate prediction models, and there are no models today which treat clouds in a wholly realistic manner. Worse, the increased reflectivity of clouds from sulfur aerosol condensation nuclei is extremely non-linear and poorly understood. Nonetheless, several recent attempts to estimate the impact of human sulfur emissions on cloud properties (Jones etal. 1994, Wigley 1989) suggest cooling which may be similar to that produced from aerosol scattering - that is, roughly comparable to C02 warming.
22 Climate models are only just beginning to include sulfur emissions and direct and indirect effects of aerosols (Jones et al. 1994, Kiehl and Briegleb 1993, Taylor and Penner, 1994, Wigley 1989). Initial results suggest that predicted climate responses when aerosols are included may be quite different than for radiative gases alone. Feedbacks within the climate system may lead to cooling not just in regions of sulfur emission but also in far-removed areas, such as in the sub-Arctic (Taylor and Penner, 1994). One consequence of the northern-hemisphere enhancement of sulfur aerosols is that warming associated with increases in greenhouse gases may occur more quickly in the southern hemisphere, where it is not partially offset by aerosol cooling. 5. CONCLUDING REMARKS
Changes in global economic and social systems are likely to have profound effects on emissions of radiatively active gases and particles in future decades, with implications for global climate change. The breakup of the Soviet Union, explosive growth in the third world, and the economic emergence of Asia, in particular the "industrial revolution" in China, will lead to geographically changing patterns of fuel consumption over the next decade that are unlike anything in the past 25 years. Carbon and sulfur emissions from rapidly industrializing nations are likely to soar. At the same time, emissions in the developed world may decrease, as more stringent controls take effect. In general, we may expect an increase in emissions from low latitudes on both sides of the equator, and a possible stabilization in the higher-latitude emissions from North America and Western Europe. Emission changes in Eastern Europe and the countries of the former Soviet Union are major uncertainties. Geographic patterns of fossil fuel emissions of greenhouse gases are not particularly important, since these species are well mixed around the globe and have lifetimes of years to decades in the atmosphere. Although climate response to greenhouse gas forcing will certainly vary between regions, the distribution of responses is likely to be relatively insensitive to where the gases are emitted. The situation is very different for short-lived sulfur aerosols. Their climatic effects are intrinsically regional, since they do not exist in the atmosphere long enough to be dispersed globally. Consequently, changes in the distribution of sulfur emissions will result in different local climatic impacts. However, it is likely that regional climate response will not be limited only to areas of strong forcing, because of feedbacks in the climate system (Taylor and Penner 1994). We have barely begun to explore the complex interactions between the climatic forcing of industrial carbon, sulfur, and other emissions, as well as our other diverse impacts on global biogeochemical cycles. If world economic changes are more rapid than scientific advances necessary to understand the climatic effects of these coupled emissions, we may be chasing a "moving target" of
23 climate forcing. Climatic change policy decisions based on today's economic and social scenarios may be wrong for tomorrow's world unless we understand the effects on climate of our modifications of the major biogeochemical cycles. 6. REFERENCES
Ayers, G. P. and Gras, J. L. (1991). Seasonal relationship between cloud condensation nuclei and aerosol methanesulphonate in marine air. Nature 353: 834-835. Ayers, G. P., Ivey, J. P. and Gillett, R . W . (1991). Coherence between seasonal cycles of dimethyl sulphide, methanesulphonate and sulphate in marine air. Nature 349: 404-406. Bates, T. S., Charlson, R. J. and Gammon, R. H. (1987). Evidence for the climatic role of marine biogenic sulfur. Nature 329: 319-321. Bates, T. S., Clarke, A. D., Kapustin, V. N., Johnson, J. E. and Charlson, R. J. (1990). Oceanic dimethylsulfide and marine aerosol: difficulties associated with asssessing their covariance. Global Biogeoch. Cycles 3: 299-304. Charlson, R. J., Lovelock, J. E., Andreae, M. O. and Warren, S. G. (1987). Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate. Nature 326: 655-661. Charlson, R. J., Schwartz, S. E., Hales, J. M., Cess, R. D., Jr., J. A. C., Hansen, J. E. and Hofmann, D. J. (1992). Climate forcing byanthropogenic aerosols. Science 255: 423-430. Dacey, J.W.H., Drake, B.G. and Klug, M.J. (1994) Stimulation of methane emission by carbon dioxide enrichment of marsh vegetation. Nature 370: 47-49. Falkowski, P.G. and Wilson, C. (1992) Phytoplankton productivity in the North Pacific ocean since 1900 and implications for absorption of anthropogenic CO2. Nature 358: 741-743. Falkowski, P. G., Kim, Y., Kolber, Z., Wilson, C., Wirick, C. and Cess, R. (1992). Natural versus anthropogenic factors affecting low-level cloud albedo over the north Atlantic. Science 256:1311-1313. Farquhar, G.D., Lloyd, J., Taylor, J.A., Flanagan, L.B., Syvertsen, J.P., Hubick, K.T., Wong, S.C. and Ehleringer, J.R. (1993) Vegetation effects on the isotope composition of oxygen in atmospheric CO2. Nature 363: 439-443.
24 Goulding, K.W.T., Hutsch, B.W., Webster, C.P., Willison, T.W. and Powlson, D.S. (1995) The effect of agriculture on methane oxidation in soil. Philosophical Transactions, Royal Society of London: in the press. Hesshaimer, V., Heimann, M. and Levin, I. (1994) Radiocarbon evidence for a smaller oceanic carbon dioxide sink than previously believed. Nature 370: 201-203. IPCC (1992) Climate Change 1992. Houghton, J.T., Callander, B.A. and Varney, S.K. (Eds.) Cambridge, England: Cambridge University Press. IPCC (1994) Radiative Forcing of Climate Change. Geneva, Switzerland: Intergovernmental Panel on Climate Change. Jones, A., Roberts, D. L. and Slingo, A. (1994). A climate model study of indirect radiative forcing by anthropogenic sulphate aerosols. Nature 370: 450453. Keller, M. and Matson, P.A. (1994) Biosphere-atmosphere exchange of trace gases in the tropics: evaluating the effects of land use changes. In: Prinn, R.G. (Ed.) Global Atmospheric-Biospheric Chemistry. pp. 103-117. New York: Plenum Press Kiehl, J. T. and Briegleb, B. P. (1993). The relative roles of sulfate aerosols and greenhouse gases in climate forcing. Science 260:311-314. King, G.M. and Schnell, S. (1994) Effect of increasing atmospheric methane concentration on ammonium inhibition of soil methane consumption. Nature 370: 282-284. Langner, J., Rodhe, H., Crutzen, P. J. and Zimmermann, P. (1992). Anthropogenic influence on the distribution of tropospheric sulphate aerosol. Nature 3 5 9 : 7 1 2 - 7 1 5 . Legrand, M. R., Delmas, R. J. and Charlson, R. J. (1988). Climate forcing implications from Vostok ice-core sulphate data. Nature 324: 418-420. Mayewski, P. A., Lyons, W. B., Spencer, M. J., Twickler, M. S., Buck, C. F. and Whitlow, S. (1990). An ice-core record of atmospheric response to anthropogenic sulfate and nitrate. Nature 346: 554-556. Oechel, W.C., Cowles, S., Grulke, N., Hastings, S.J., Lawrence, B., Prudhomme, T., Riechers, G., Strain, B., Tissue, D. and Vourlitis, G. (1994) Transient nature of CO2 fertilization in Arctic tundra. Nature 371: 500-503.
25 Oechel, W.C., Hastings, S.J., Vourlitis, G., Jenkins, M., Riechers, G. and Grulke, N. (1993) Recent change of Arctic tundra ecosystems from a net carbon dioxide sink to a source. Nature 361: 520-523. Robertson, J.E. and Watson, A.J. (1995) Thermal skin effect of the surface ocean and its implications for CO2 uptake. Nature 358: 738-740. Sarmiento, J.L. (1993) Atmospheric CO2 stalled. Nature 365: 697-698. Schlesinger, W.H. and Hartley, A.E. (1992) A global budget for atmospheric NH s. Biogeochemistry 15:191-211. Schwartz, S. E. (1988). Are global cloud albedo and climate controlled by marine phytoplankton? Nature 336: 441-445. Sellers, P., Hall, F., Margolis, H., Kelly, B., Baldocchi, D., den Hartog, J., Cihlar, J., Ryan, M., Goodison, B., Crill, P., Ranson, J. and Lettenmaier, D. (1995) The Boreal Ecosystem-Atmosphere Study (BOREAS): an overview and early results from the 1994 field year. Bulletin, American Meteorological Society: in the press. Shaw, G. E. (1983). Bio-controlled thermostasis involving the sulfur cycle. Climatic Change 5: 297-303. Sutton, M.A., Schjorring, J.K. and Wyers, G.P. (1995) Plant-atmosphere exchange of ammonia. Philosophical transactions, Royal Society of London: in the press. Taylor, K. E. and Penner, J. E. (1994). Response of the climate system to atmospheric aerosols and greenhouse gases. Nature 369: 734-737. Vitousek, P.M. and Matson, P.A. (1995) Agriculture, the Global Nitrogen Cycle, and Trace Gas Flux. In: Proceedings, l Oth International Symposium on Environmental Biogeochemistry, pp. 193-207. Watson, A.J., Law, C.S., Van Scoy, K.A., Millero, F.J., Yao, W., Friederich, G.E., Liddicoat, M.I., Wanninkhof, R.H., Barber, R.T. and Coale, K.H. (1994) Minimal effect of iron fertilization on sea-surface carbon dioxide concentrations. Nature 371 : 143-145. Wigley, T. M. L. (1989). Possible climate change due to SO2-derived cloud condensation nuclei. Nature 339: 365-367.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
27
The potential effects of climate change in a riverine hydrological system in northwestern Canada S. J. Cohen
Environmental Adaptation Research Group, Atmospheric Environment Service, Environment Canada, 4905 Dufferin Street, Downsview, Ontario M3H 5T4, Canada
Abstract Assessments of climate change risks, and system vulnerabilities, may benefit from a focus on a watershed. A riverine hydrological system is an integrator of natural and human systems, and so constitutes an appropriate setting for determining the regional effects of climate change scenarios. The Mackenzie Basin Impact Study (MBIS) is presented as a case study that illustrates the challenges and opportunities presented by integrated regional assessment of climate change scenarios in a large watershed in northwestern Canada.
1. Climate Change Impact Assessment: The Integration Challenge The Framework Convention on Climate Change (FCCC) is now a part of international law, committing more than 80 nations to action. Its ultimate objective is to stabilize global concentrations of carbon dioxide and other 'greenhouse gases' at a level that does not represent 'dangerous anthropogenic interference' to the atmosphere. At issue, however, is the definition of the term 'dangerous.' This is an important challenge for climate impact assessment at the regional scale. Global scale atmospheric anomalies, such as E1 Nifio-Southern Oscillation (ENSO), are known to produce region-specific impacts (e.g. Glantz et al., 1987). The same thing is likely to happen with an enhanced greenhouse effect. What might these impacts be and what might (or should?) be the nature of the adaptive responses? Regional impact assessment is a complex multidisciplinary research challenge. To make matters even more difficult, we are considering an assessment not of an observed climatic event (such as the 1993 Mississippi River flood) but of a theoretical warming of the earth's climate by increased concentrations of greenhouse gases. There are many uncertainties associated with the data and methods used to construct scenarios of a future warmer world, and some have argued for the use of analogues (Glantz, 1988; Kearney, 1994) as an alternative to scenarios based on climate model simulations, population projections, and other forecasting tools. There is little doubt, however, that if climate warming occurs, the earth and its people will feel its effects through a variety of "pathways" and "filters," and the impact assessment needs to account for these.
28
1.1 What is integration.9 In order to capture the complex linkages between climate and regions, a research framework is needed which effectively combines information about individual sectors so that the result is more than the sum of the parts. The Intergovernmental Panel on Climate Change (IPCC) defines integrated assessment as "the most comprehensive treatment of the interactions of climate and society" (Carter et al., 1992). It addresses the "net" effect of climate-related stress, so that the indirect linkages between atmosphere, land and water resources, resource management and other policy matters, can be considered in a way that can be understood by decision makers. This would prevent the implementation of strategies or policies which assist one group or sector at the expense or detriment of others. Working at the regional scale is important because at larger scales, impacts may offset each other, and the final result may hide critical details (e.g. Rosenzweig and Parry, 1993). There are two main approaches to integrated assessment: a) models, including integrated system models such as IMAGE (Alcamo, 1994), and a new model being developed by Battelle/Pacific Northwest Laboratories in the United States (Frederick and Rosenberg, 1994), and b) assessments of policy instruments (e.g. development plans, conservation plans (e.g. Inuvik, 1993)), or regulatory bodies (e.g. river authorities (e.g. Arnell et al., 1994)). There are a wide range of options available within these two sets. Rather than relying on only one approach, a regional study could make use of several integrating techniques if they provide unique and complementary assessments. This could be called the "family of integrators" approach. With the growing interest in global-scale assessment models, and their potential application in policy gaming exercises, there remains a need for detailed information on smaller scales, which could provide the foundation for global models to produce regional simulations that are plausible to stakeholders. Integrated regional assessments could provide this information. At the same time, however, there is a requirement to convince policy makers that these decision support tools are useful for assessing response options. Policy makers do not represent a global constituency, so there is a need to address issues at their regional/national scales of interest. It is suggested that an integrated assessment should not rely exclusively on integrated system models, since most of these do not necessarily involve the stakeholder, nor make direct use of the stakeholder's perception of the climate change issue. This perception is not based solely on whether a climate change has been noticed, but on whether any observed or simulated changes in the landscape or economic production can be linked to observations or simulations (scenarios) of climate change. Stakeholders can be an important source of "ground truth," and that is the frame of reference they would use when considering responses to future scenarios of climate change (e.g. Aharonian, 1994; Bielawski, 1994). Henderson-Sellers (1993) warns that integrated impact assessments might still be circumvented in the rush towards responding to the climate change threat, and that uncertainties make such assessments premature. If full integration is impossible to achieve due to insufficient information, what about partial integration, in which there are some aspects that remain outside of the assessment, and assumptions have to be made about their level of influence. For example, the study of the Corn Belt in the United States (Crosson and Rosenberg, 1993) considered an area bounded by four states (Missouri-Iowa-Nebraska-Kansas or MINK). Water resources issues were still addressed even though upstream sub-basins were located outside the MINK region. Despite some obvious problens, can partial
29 integration provide useful input to the debate on policy responses to climate change? How could this more limited framework be designed so that climate-society issues could still be addressed, while recognizing the limitations that prevent consideration of all factors? 1.2 Purpose In order to attract the breadth of expertise and interests needed for an integrated assessment with stakeholder collaboration, some common ground must be laid out. Many impact assessments have focused on individual sectors (e.g. agriculture, wildlife, water resources), and while these can provide important technical information on direct 'first-order' impacts (IPCC 1990; Tegart and Sheldon, 1993), a wide range of external factors are often assumed to remain unchanged (Carter et al., 1992). Regional and national assessments have been produced elsewhere (e.g. Henderson and Coils, 1993; Hulme et al., 1992; Liverman, 1992; New Zealand Climate Change Programme, 1990; Ninh et al., 1991; Nishioka et al., 1993; Smith and Tirpak, 1990), but these have generally consisted of parallel sectoral studies. Crosson and Rosenberg (1993) and Parry et al. (1992) have attempted integrated assessments based on regions defined by sectoral dominance (e.g. agriculture) and/or political borders. The purpose here is to suggest that riverine hydrologic systems may provide an appropriate setting for producing an integrated regional assessment of climate change scenarios. What follows is a description of a watershed-based case study from northwest Canada, which is still in progress. The regional/watershed focus has been used to attract scientific expertise and diverse stakeholders with local knowledge. The common ground for all of them is an interest in the future of this place, with water serving as an important link.
2. Watersheds as Integrators of Natural and Human Systems The choice of study area can influence many aspects of an impact assessment, including the identification of issues and the collection of data. There are various administrative and ecological settings that might be considered (Carter et al. 1992), but the focus here is exclusively on watersheds as integrators. Land cover and land use affects hydrology and water quality, so water users (e.g. hydroelectric utilities, fisheries, navigation, domestic users, wildlife, agriculture, recreation) are necessarily linked with forces that modify the landscape (e.g. agriculture, forestry, hydroelectric utilities, industrialization, fire, pests). Governments have used watersheds as the basis for the creation of customized management structures (e.g. basin commissions, water boards). These attempt to reconcile the goals of competing interests while providing direction for regulation, water allocations and other matters. For the purpose of climate impact assessment, this is an important source of information on regional issues and their stakeholders. Although economic data are rarely collected on a watershed basis, it should be possible to at least partially address this requirement through the use of census data or other local/regional sources of information. Watershed-based assessments have been tried for the North American Great Lakes (Smith and Tirpak, 1990; Mortsch et al., 1993) and a series of international basin studies including the Nile, Indus and Zambezi (Strzepek and Smith, forthcoming). Arnell et al. (1994) provide a case that focusses on a water management authority in the United Kingdom. It is expected that climate warming would lead to an acceleration of the water cycle, with increased throughputs along the pathways linking atmosphere, ocean, landscape, freshwater
30 bodies, and society (Falkenmark, 1991). Climate warming may impose its most significant effects on water sensitive sectors, through changes in a) the frequency and severity of extreme events (floods, drought, etc.), b) timing of seasonal and annual events (e.g. spring runoff peak, autumn low flow, ice formation and break up, etc.), c) thresholds and ranges (e.g. maximum summer water temperatures), and d) land cover (e.g. erosion, fire, etc). Riverine hydrological systems will exhibit basin-specific adjustments to global climatic changes. Most warming scenarios tend to show increases in precipitation, but this does not necessarily mean wetter land surfaces or more soil moisture. Gleick (1993) concludes that if the future climate will not look like the past, there will be a great increase in the overall uncertainty associated with water management and supply. Understanding impacts is a necessary prerequisite for determining the kind of measures that could promote both limitation of greenhouse gas emissions and adaptation to environmental stresses. Assessing adaptation options requires a greater understanding of how individuals, companies and governments operate when faced with environmental stresses (Smit, 1993).
3. Case Study: Mackenzie Basin, Canada The Mackenzie Basin Impact Study (MBIS) is part of the Government of Canada's Green Plan, and has passed the halfway point in its six-year mandate to assess the potential regional implications of global climatic change (Cohen, 1993, 1994). The program includes studies on water resources, permafrost, vegetation, wildlife, economic activities, resource-based and subsistence-based communities, and applications of remote sensing and geographic information systems (GIS). Attention is also given to the challenges of producing an integrated assessment, and to incorporating traditional ecological knowledge into the MBIS.
3.1 Setting This region was chosen because it is a major high latitude watershed, 1.7 million km 2 in area, with many climate-sensitive landscapes and transition zones: tree lines (Arctic, montane, aspen parkland), discontinuous permafrost, wetlands and deltas, the edge of multiyear sea ice, and the northern limits of commercial forestry and agriculture. Freshwater and terrestrial migratory wildlife might be sensitive to climate-induced changes in the landscape. The study area is defined by the watershed boundary of the Mackenzie River and its ti'ibutaries, plus the southern Beaufort Sea and coastal zone north and east of the Mackenzie Delta. The large size precludes detailed study of all areas, though there are some activities that are conducted on the Basin scale. For the MBIS to achieve its objective, however, the range of impact-related policy questions are limited to the following: a) interjurisdictional water management, b) sustainability of native (aboriginal) lifestyles, c) economic development opportunities, d) maintenance of infrastructure, and e) sustainability of ecosystems. Additional focus has been provided by defining critical regions within the study area (Figure 1). Each of these represent potential flash points due to the intersection of potential biophysical changes with human activities. For example, the Upper Peace River region includes a major hydroelectric facility (Bennett Dam), agriculture, expanding forestry operations and communities with a history of flooding. The operation of the dam has led to concerns about the viability of the freshwater wetlands and delta, and consequently, the
32 wildlife and subsistence-based communities in the Peace-Athabasca Delta region, located downstream. How would a scenario of climate warming affect dam operations, water levels at the Delta, fisheries, migrating waterfowl, agriculture and forestry operations? Would resource-based and native communities experience the same impacts, or would climate change be felt in different ways depending on lifestyle (wage economy, subsistence/non-wage economy)?
3.2 Objective If climate warming occurs, governments and their constituents will need advice on how to adapt to the new climate. Since decision making occurs in an environment where different stakeholders compete for resources, any response options will have to account for tradeoffs between these various interests. Land and water use patterns today represent the result of historic and current compromises between these various interests, combined with knowledge gained from research and personal experience. At the scale of most current GCM-based impact assessments (e.g. grid sizes larger than 2 ~ latitude x 2 ~ longitude), land in a grid cell is not necessarily assigned to a single optimal use today, so it is unlikely that this would be different in the future. The assessment, therefore, should not restrict itself to changes in physical capability to support a particular activity (e.g. crop production). The objective of MBIS is to provide an integrated regional assessment of scenarios of climate warming for regional stakeholders and the scientific community. As a high latitude watershed, the Mackenzie Basin has been seen as an area that might benefit in certain ways by a warmer climate. These include a) longer growing season for agriculture, b) greater productivity for forestry, c) longer ice-free season for navigation, d) reduced energy demand for space heating, e) longer summer tourist season, and f) reduced cold weather stress on infrastructure. Taken individually, economic impacts could be quantified, and these might show substantial benefits for the region. Other factors need to be considered, however, and some of these may constrain the potential benefits. This list includes: a) current use of land for subsistence hunting and trapping, b) current system of land transportation, much of which is based on a stable ice and snow cover for winter roads, c) current ranges and habitats of wildlife, which underpin conservation plans and native land claims (currently being negotiated between aboriginal people and governments in Canada), and d) scientific uncertainty which hampers anticipatory responses to projected beneficial conditions. Potential negative impacts of climate warming must also be considered, because they may offset possible benefits. Examples are: a) increased erosion due to permafrost thaw, b) increased frequency and severity of forest fires, c) extension of mid-latitude pests and diseases into high latitudes, and d) reduction of habitat suitable for cold climate species of vegetation and wildlife. 3.3 Study Framework MBIS is attempting to produce an integrated regional assessment of global warming scenarios, as a way of identifying the indirect linkages between climate and regional policy concerns, such as land and water management. Several exercises are being tried, including 1) resource accounting with input-output modelling, 2) land assessment (including goal programming and multiobjective program modelling), 3) review of water resources policy instruments and their sensitivity to hydrologic changes, and 4) study of settlement patterns and their sensitivity to landscape changes. Each of these utilize the outputs of various
33 individual studies in order to address some of the human dimensions of climatic change (Cohen, 1993, 1994). All of these approaches are being tried because there is no consensus on which method is best for producing an integrated study. System models (1 and 2 above) provide a closed integrated model or set of linked models that describe particular components of the system. Analyses based on planning/management instruments (3 and 4 above) consist of a mixture of models and expert judgement. These instruments (e.g. plans, policies, regulations, indices) represent the integration of scientific information and stakeholders' preferences, and their performance under climate change scenarios would provide an important measure of impact. There is a difference between the level of control exerted by the researcher in these approaches. While the idea of "megamodels" (Frederick and Rosenberg, 1994) is growing in popularity in Europe and North America, the family of integrators concept presented here serves to provide an opportunity for other forms of input to contribute to the assessment, particularly those which are difficult to quantify. Policy analysis has both quantitative and qualitative aspects, and may be preferred by stakeholders who are leery of 'black box' models. There are several opportunities to facilitate linkage between individual study components. Within MBIS, integrated system models, economic models, and other similar tools, are being used to address complex issues related to land use and economic growth (e.g. Lonergan, 1994; Yin and Cohen, 1994; Huang et al., 1994). These mathematical or statistical techniques require a wide range of inputs, including census data, outputs of other models, and/or indices obtained from remote sensing, thereby serving as integrators of information obtained from other disciplines.
3.4 Preliminary Results One theme that has clearly emerged in the MBIS is that climate is a complex agent of change. Although scientific and political discussions have tended to focus on atmospheric change, the land and its people will likely experience climate warming through changes in streamflow, water levels, ice and snow cover, permafrost, plant growth, wildlife patterns, fire, pests and diseases. Some changes may occur gradually while others may come in the form of large steps or new extremes. The linkage between changes in air temperature and regional socio-economic concerns is largely through these landscape 'filters.' Biophysical changes are what people will notice before they pay attention to climate statistics. Has the winter road season changed? Is anything new with the caribou migration? Are current fire management strategies still working satisfactorily? What is the status of permafrost along the Mackenzie Valley and the Beaufort coastal zone? Some preliminary indications of landscape and socioeconomic impacts for the scenarios being assessed by MBIS are shown in Tables 1 and 2, respectively. Many MBIS activities are not yet at the stage where scenario results can be reported, but some information is available.
34 Table 1 MBIS Preliminary Summary of Landscape Im )acts of Climate Warming Scenarios
PARAMETER
DETAILED IMPACTS
Permafrost thaw occurs, but rate of change varies with site
*thaw would occur primarily in discontinuous zone *seasonal active layer would increase *rate of thaw in wetland areas would lag behind other sites *slopes and Beaufort Sea coastal zone may experience accelerated erosion
Water Supply changes slightly, with earlier spring peak
*annual Basin runoff changes -7 % to -3 % in GCM-based scenarios, +7 % in composite analogue scenarios *increased precipitation offset by increased evapotranspiration in many subbasins *spring snowmelt peak begins up to 1 month earlier *longer snowmelt season, lower peak in some subbasins (including Williston, upstream of Bennett Dam)
Peace River Ice Cover reduced in duration and extent
9ice cover reduced by up to 4 weeks 9upstream progression of ice reduced by up to 200 km 9runoff reduction (or reduction of discharge from Bennett Dam) would offset effects of temperature increase on ice cover
Soil Capability for Agriculture increases
9increase in availability of marginal and suitable land for spring seeded small grains and forages due to longer growing season and frost free period 9decrease in soil moisture supply
Pine Weevil Hazard increases
9increase in temperature-based pine weevil hazard index 9low elevation sites particularly vulnerable 9non-temperature factors not yet included
Fire Weather Index increases
9median index for four GCM-based scenarios corresponds to change o f - 15 % to + 81% in burned area
Summarized from Cohen (1993, 1994).
35 Runoff for the Basin was obtained using a square grid model (Soulis et al., 1994), and for the Williston subbasin with the UBC Watershed Model (Chin and Assaf, 1994). Although increased runoff was anticipated (e.g. see Miller and Russell, 1992), this does not appear to be the case for the GCM-based scenarios (Canadian Climate Centre or CCC, Geophysical Fluid Dynamics Lab or GFDL (R30 version)) for the Basin as a whole. Only the composite analogue scenario shows an increase. Newton (1994) has therefore concluded that scenario spring flood risks for vulnerable communities may not be that different from current climatic conditions. What is not clear as yet is the implication of hydrologic and landscape changes on water management agreements currently being negotiated by various governments (Felton, 1994). Peace River ice cover, for example, will be affected by both temperature changes and changes in outflow from the Bennett Dam at Williston subbasin (Andres, 1994). This may not be the final word on runoff impacts, since the Global Energy and Water Cycle Experiment (GEWEX) is pursuing a research programme in the Mackenzie (Lawford, 1994). It would appear that the other main threats to the Mackenzie landscape are a) accelerated erosion caused by permafrost thaw, especially in sloping terrain and the Beaufort Sea coastal zone (Aylsworth and Egginton, 1994; Solomon, 1994), b) increased fire hazard (Kadonaga, 1994), and c) invasion of new pests and diseases from warmer regions (Sieben et al., 1994). These landscape impacts could lead to changes in plant succession (Wein et al., 1994), thereby affecting wildlife habitat and subsistence activities of native communities. Additional information on ecosystem impacts should become available for the MBIS Final Report in 1997. First-order and second-order impacts eventually lead to others which are considerably more difficult to address. Will land claims or water resources agreements be affected? Could there
Table 2 MBIS Preliminary Summary of Socio-Economic Impacts of Climate Warming Scenarios SECTOR/LOCATION
DETAILED IMPACTS
Tourism/Nahanni National Park would experience mixed impacts
9little impact from projected minor changes in streamflow 9extended season for water-based recreation would provide economic benefits to communities near the Park 9increased Fire Weather Index (fire frequency and severity) could affect runoff, landscape character, visitor safety
Community Vision of Impacts depends on vision of lifestyle
9response to flood hazard varies by community, according to the interplay of individual, community and government responses 9significance of landscape impacts depends on whether community maintains subsistence lifestyle, or switches to wage economy
Summarized from Cohen (1994).
36 be new conflicts over land use, especially if agriculture expands northward to take advantage of improved soil capability to support crop production (Brklacich and Curran, 1994)? What might be the effects on parks and other protected areas (Pollard and Benton, 1994)? Could climate change affect the economics of oil and gas production in the Beaufort Sea (Anderson et al., 1994)? Expressing socio-economic impacts in monetary terms is going to be difficult, but it should be possible to do so for agriculture, forestry, energy, and some aspects of tourism. In the case of Nahanni Park located in the Liard subbasin (see Figure 1), water-based recreation is expected to benefit from the longer summer, but this could be offset by the threat of increased fire (Staple and Wall, 1994). There is no assessment, yet, on the potential costs of increased fire or fire protection. Community impacts could be quantified, but the effects of climate warming scenarios may vary depending on whether a traditional aboriginal lifestyle of hunting and trapping is maintained, or a shift to greater reliance on the wage economy occurs. Aharonian's (1994) case study of Aklavik, in the Mackenzie Delta region (see Figure 1), shows that residents can provide detailed visions of both "futures." In their view, community vulnerability to climate warming scenarios will change if their lifestyles changes. This may parallel circumstances that could be experienced in some developing countries during the next several decades. The integration component is currently focussed on data collection. One activity is on the development of a resource accounting framework, including a Mackenzie Basin input-output model. This will be used to determine impacts of changes in energy and forestry on the region's employment and economic productivity (Lonergan, 1994). A second modelling exercise is the integrated land assessment framework or ILAF. Its purpose is to compare changes in land capability with stakeholders' goals in order to identify possible land use conflicts in a climate warming scenario (Yin and Cohen, 1993, 1994). Potential expansion of commercial agriculture and forestry could create a conflict with existing subsistence activities, so there is a need to determine whether this is possible within the scenarios. Additional activities in multiobjective programming (Huang et al., 1994), and a study of the non-wage economy in a native community, will complement ongoing MBIS socioeconomic studies in agriculture, forestry, energy, tourism and community development (Cohen, 1994). Impacts and responses will not be felt by individual sectors in an isolated manner. A unit of land (at a scale comparable to GCM output) is not likely to end up becoming exclusively devoted to one kind of land cover or use. This set of research activities will hopefully enable MBIS to address some important cross-cutting issues at a scale comparable to regional stakeholders' interests.
4. Conclusions
A riverine hydrologic system is presented as an appropriate setting for integrated regional assessment of climatic warming scenarios. The Mackenzie Basin Impact Study (MBIS) illustrates the application of the "family of integrators" approach, consisting of several integrated system models and analyses of policy instruments. We have considered the difficulties in producing a fully integrated assessment of climate warming scenarios, and acknowledge that in the case of the MBIS, several aspects are not covered (e.g. marine wildlife in the Beaufort Sea, native communities in Alberta, future
37 economic linkages with the rest of Canada and other countries). MBIS includes population and economic growth scenarios (Lonergan and Difrancesco, 1993), but technological and institutional change scenarios have not been constructed. Although it is unlikely that MBIS can achieve full integration, we hope that partial integration can provide relevant information on sectoral and cross-cutting regional impacts. MBIS is an exercise in interdisciplinary research with stakeholder collaboration. Maintaining linkages between researchers and stakeholders has been a challenge. It may be difficult at this stage to appreciate the long term value of the MBIS experience, but it is clear that collaboration with stakeholders is vital for there to be any hope of producing an assessment that could be useful and relevant to the region of interest. In fact, partially or fully integrated assessments may be impossible without stakeholder involvement during all phases of research. For example, stakeholders participating in MBIS planning meetings contributed to the selection of economic growth scenarios, and the identification of communities and individuals willing to be interviewed as part of surveys conducted by MBIS investigators. During the remainder of the MBIS program, investigators will be completing biophysical and socio-economic impact studies, transferring information to the "integrators" (i.e. systems modellers, policy analysts, etc.), and completing integration exercises. There will be a workshop on water management, and a larger gathering in 1996 similar to the event that facilitated the production of MBIS Interim Report #2 (Cohen, 1994). MBIS investigators are expected to exchange information with each other before and after their components are completed. There are also plans for more discussions on the MBIS within the region, before and after publication of the final report in 1997.
5. ACKNOWLEDGEMENTS My thanks to Krystyna Czaja for producing Figure 1. Any opinions expressed in this chapter are my own, and not necessarily those of Environment Canada.
6. REFERENCES Aharonian, D. (1994). Land use and climate change: an assessment of climate-society interactions in Aklavik, NWT. In Cohen, S.J. (ed.), 410-420. Alcamo, J. (ed.) 1994. IMAGE 2.0: Integrated modelling of global climate change. Water, Air and Soil Pollution, 76, 1/2 (special issue), 1-318. Anderson, W.P., R. Difrancesco and M. Kliman. 1994. Potential impacts of climate change on petroleum production in the Northwest Territories. In Cohen, S.J. (ed.), 433-441. Andres, D. 1994. Peace River ice regime: an interim report. In Cohen, S.J. (ed.), 237-245. Arnell, N.W., A. Jenkins and D.G. George. 1994. The Implications of Climate Change for the National Rivers Authority. Institute of Hydrology R&D Report 12, National Rivers Authority, Bristol, United Kingdom. Aylsworth, J.M. and P.A. Egginton. 1994. Sensitivity of slopes to climate change. In Cohen, S.J. (ed.), 278-283. Bielawski, E. 1994. Lessons from Lutsel k'e. In Cohen, S.J. (ed.), 74-76.
38 Brklacich, M. and P. Curran. 1994. Climate change and agricultural potential in the Mackenzie Basin. In Cohen, S.J. (ed.), 459-464. Carter, T.R., M.L. Parry, S. Nishioka and H. Harasawa. 1992. Preliminary Guidelines for Assessing Impacts of Climate Change. Environmental Change Unit, Oxford, and Center for Global Environmental Research, Tsukuba. Chin, W.Q. and H. Assaf. 1994. Impact of global warming on runoff in Williston Basin. In Cohen, S.J. (ed.), 210-236. Cohen, S.J. (ed.). 1993. Mackenzie Basin Impact Study Interim Report #1. Environment Canada, Downsview, Ontario. Cohen, S.J. (ed.). 1994. Mackenzie Basin Impact Study Interim Report #2. Environment Canada, Downsview, Ontario. Crosson, P.R., and N.J. Rosenberg. 1993. An overview of the MINK study. Climatic Change, 24, 159-173. Egginton, P. 1993. Permafrost south of the Beaufort coastal zone. In S.J. Cohen (ed.), 5258. Falkenmark, M. 1991. The Ven Te Chow memorial lecture: Environment and development: urgent need for a water perspective. Water International, 16, 229-240. Felton, G. 1994. A review of interjurisdictional water management in Canada. In S.J. Cohen (ed.), 67-73. Frederick, K.D. and N.J. Rosenberg (eds.). 1994. Assessing the impacts of climate change on natural resource systems. Climatic Change, 28, nos. 1-2 (special issue), 1-219. Glantz, M.H. (ed.) 1988. Societal Responses to Regional Climatic Change: Forecasting by Analogy. Westview Press, Boulder. Glantz, M.H., R. Katz and M. Krenz (eds). 1987. The Societal Impacts Associated with the 1982-83 Worldwide Climate Anomalies. National Center for Atmospheric Research, Boulder. Gleick, P.H. 1993. Water in the 21st century. In Gleick, P.H. (ed.), Water in Crisis." A Guide to the World's Fresh Water Resources. Oxford University Press, New York, 104-113. Henderson-Sellers, A. 1993. An antipodean climate of uncertainty. In Henderson-Sellers and Colls (eds.), 203-224. Henderson-Sellers, A. and K. Coils (eds.). 1993. Climatic impacts in Australia. Climatic Change, 25, nos. 3-4 (special issue), 201-438. Huang, G.H., Y.Y. Yin, S.J. Cohen and B. Bass. 1994. Interval parameter modelling to generate alternatives: a software for environmental decision-making under uncertainty. In Brebbia, C.A. (ed.), Computer Techniques in Environmental Studies. Kluwer Academic Publishers, Dordrecht. Hulme, M., T. Wigley, T, Jiang, Z.-c Zhao, F. Wang, Y. Ding, R. Leemans, and A. Markham. 1992. Climate Change due to the Greenhouse Effect and its Implications for China. World Wide Fund for Nature, Gland, Switzerland. Intergovernmental Panel on Climate Change (IPCC). 1990. Climate Change: The IPCC Impacts Assessment. W.J.McG. Tegart, G.W. Sheldon and D.C. Griffiths (eds.). Australian Government Publishing Service, Canberra. Inuvik, Community of. 1993. Inuvik Inuvialuit Community Conservation Plan. Available from Wildlife Management Advisory Council, Inuvik, Northwest Territories. Kadonaga, L. (1994). Fire in the environment. In Cohen, S.J. (ed.), 329-336.
39 Kearney, A.R. 1 9 9 4 . Understanding global change: a cognitive perspective on communicating through stories. Climatic Change, 27, 419-441. Lawford, R.G. 1994. Knowns and unknowns in the hydroclimatology of the Mackenzie River Basin. In Cohen, S.J. (ed.), 173-196. Liverman, D. 1992. The regional impact of global warming in Mexico: Uncertainty, vulnerability and response. In Schmandt, J., and J. Clarkson (Eds.), The Regions and Global Warming: Impacts and Response Strategies. Oxford University Press, New York, 44-68. Lonergan, S. 1994. Natural resource/environmental accounting in the Mackenzie Basin. In Cohen, S.J. (ed.), 39-42. Lonergan, S., and R.J. Difrancesco. 1993. Baseline population and economic growth simulation. In Cohen, S.J. (ed.), 131-139. Miller, J.R. and G.L. Russell. 1992. The impact of global warming on river runoff. Journal of Geophysical Research, 97, D3, 2757-2764. Mortsch, L., G. Koshida and D. Tavares (eds.). 1993. Adapting to the impacts of climate change and variability. Proceedings of the Great Lakes - St. Lawrence Basin Project Workshop, 9-11 February, 1993, Quebec City. Environment Canada, Downsview, Ontario. Newton, J. 1994. Community response to episodes of flooding in the Mackenzie Basin. In Cohen, S.J. (ed.), 421-430. New Zealand Climate Change Programme. 1990. Climatic Change: Impacts on New Zealand. Ministry for the Environment, Wellington. Ninh, N.H., M.H. Glantz and H.M. Hien. 1991. Case Studies of Climate-Related Impact Assessment in Vietnam. UNEP Project Document No. FP/4102-88-4102. United Nations Environment Programme, Nairobi. Nishioka, S., H. Harasawa, H. Hashimoto, T. Ookita, K. Masuda and T. Morita (Eds.). 1993. The Potential Effects of Climate Change in Japan. Center for Global Environmental Research, Tsukuba, CGER-I009-'93. Parry, M.L., M. Blantran de Rozari, A.L. Chong and S. Panich. 1992. The Potential SocioEconomic Effects of Climate Change in Southeast Asia. United Nations Environment Programme, Nairobi. Pollard, D.F.W. and R.A. Benton. 1994. The status of protected areas in the Mackenzie Basin. In Cohen, S.J. (ed.), 23-27. Rosenzweig, C. and M.L. Parry. 1994. Potential impacts of climate change on world food supply. Nature, 367, 133-138. Sieben, B.G., D.L. Spittlehouse, R.A. Benton and J.A.McLean. 1994. A first approximation of the effect of climate warming on the white pine weevil hazard in the Mackenzie River Drainage Basin. In Cohen, S.J. (ed.), 316-328. Smit, B. (Ed.). 1993. Adaptation to Climatic Variability and Change. Report of the Task Force on Climate Adaptation, Canadian Climate Program. Department of Geography, University of Guelph, Occasional Paper No. 19. Smith, J.B. and D.A. Tirpak (Eds.). 1990. The Potential Effects of Global Climate Change on the United States. Report to Congress, United States Environmental Protection Agency, Washington. Solomon, S. 1994. Storminess and coastal erosion at Tuktoyaktuk. In Cohen, S.J. (ed.), 286-292.
40 Soulis, E.D., S.I. Solomon, M. Lee and N. Kouwen. 1994. Changes to the distribution of monthly and annual runoff in the Mackenzie Basin using a modified square grid approach. In Cohen, S.J. (ed.), 197-209. Staple, T. and G. Wall. 1994. Implications of climate change for water-based recreation activities in Nahanni National Park Reserve. In Cohen, S.J. (ed.), 453-455. Strzepek, K.M., and J.B. Smith (eds.), forthcoming. As Climate Changes: The Potential International Impacts of Climate Change, Cambridge University Press, New York. Tegart, W.J.McG., and G.W. Sheldon. (eds). 1993. Climate Change 1992: The Supplementary Report to the IPCC Impacts Assessment. Australian Government Publishing Service, Canberra. Wein, R., R. Gal, J.C. Hogenbirk, E.H. Hogg, S.M. Landh~iusser, P. Lange, S.K. Olsen, A.G. Schwarz and R.A. Wright. 1994. Analogues of climate change - fire vegetation responses in the Mackenzie Basin. In Cohen, S.J. (ed.), 337-343. Yin, Y., and S.J. Cohen. 1993. Integrated land assessment framework. In Cohen, S.J. (ed.), 151-163. Yin, Y., and S.J. Cohen. 1994. Identifying regional policy concerns associated with global climate change. GlobalEnvironmental Change, 4, 246-260.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
41
Assessing the impacts of climate: The issue of winners and losers in a global climate change context Michael H. Glantz Environmental and Societal Impacts Group National Center for Atmospheric Research* PO Box 3000 Boulder, Colorado USA 80307-3000
1. INTRODUCTION Most reviews of the greenhouse issue begin with the works in the mid-1890s of Swedish scientist Arrhenius. The physical processes have been well known for more than a century. Interest in the possible impacts on climate of CO2 emissions as a result of human activities has waxed and waned since that time, with temporary peaks of interest appearing in the mid1930s (Callendar, 1938), the mid-1950s (Revelle and Suess, 1957), and again in the mid1970s (e.g, Kellogg, 1977). Today we are inundated by assessments of the prospects of a global warming and its possible impacts on society and the environment produced by national, international, and nongovernmental organizations. Discussions of such a prospect have steadily increased during the past twenty years, reaching very high political levels in the late 1980s and early 1990s. The century-long interest in the science and impacts of the human-induced enhancement of the greenhouse effect has been interrupted partly by other more pressing and urgent historical events such as two world wars, a worldwide depression, decolonization, the Cold War and then its demise, and a temporary global cooling; and partly by the fact that the impacts of a CO2-induced global warming were originally believed to be beneficial to society. For example, Callendar (1938) suggested that a greenhouse warming would help to thwart the emergence of an apparently imminent Ice Age. Scientific and anecdotal evidence was cited to suggest that the earth was coming to the end of an interglacial period, and that soon processes leading to an Ice Age would begin again. From about 1940 to the late 1960s, the global atmosphere underwent a yet-to-beunderstood cooling. Scientists provided scientific as well as anecdotal evidence (convincing both to the lay public and segments of the scientific community) to support the view that the earth was possibly on the threshold of an Ice Age: the growing season in England had been shortened by two weeks, fish species formerly caught off the northern coast of Iceland began
*The National Center for Atmospheric Research is sponsored by the National Science Foundation.
42
appearing only off its southern coast, sea ice in the North Atlantic had increased in its southward extent in the early 1970s and was appearing in shipping lanes that were normally ice-free; and hay production in Iceland declined by 25% as a result of less hospitable weather. In the United States, the fact that the armadillo, which had migrated as far north as Kansas in wanner decades, was starting to retreat toward the south was also used as evidence to support the Ice Age hypothesis. Geologic records were invoked as well, to show that an Ice Age was near. During this brief period of concern with global cooling, one issue widely considered was how it might affect the relative economic and political positions of different countries. Even the US Central Intelligence Agency undertook a set of studies to show how the cooling might affect the agricultural production and energy demand in the USSR (CIA, 1974, 1976). The Ecologist examined the potential impacts of a few degrees of cooling on agriculture in the Canadian prairies (Goldsmith, 1977). Some books and articles on the topic went so far as to identify specific countries that would become climate-related world powers in the event of a cooling. For example, Ponte (1976) suggested that "adapting to a cooler climate in the northern latitudes, and to a drier climate nearer the equator, will require vast resources and almost unlimited energy .... A few countries, such as equatorial Brazil, Zaire, and Indonesia, could emerge as climate-created superpowers." He also suggested that "We can say with high probability today that the global monsoon rainfall will be below average for the remainder of the century." Another book on the possibility of a global cooling (Impact Team, 1977) suggested that with a cooling "there would be broad belts of excess and deficit rainfall in the middle latitudes; more frequent failure of the monsoons that dominate the Indian subcontinent, south China, and western Africa; shorter growing seasons for Canada, northern Russia, and north China. Europe could expect to be cooler and wetter. Of the main grain-growing regions, only the United States and Argentina would escape adverse effects." There was no reluctance whatsoever to discuss who might win and who might lose, or to identify specific countries or specific economic sectors within a country as winners and as losers in the event of a global cooling. There is a striking difference between the scientific and political responses in the 1970s to a potential cooling and those of today to a warming. Today there is a strong reluctance, if not opposition, within scientific as well as policymaking circles to recognize (or address or discuss) the existence and identity of specific winners and losers, especially winners. When he was a US Senator, US Vice President Albert Gore (1992), for example, argued that there would be no winners in the event of a global warming, a view that is apparently also held by the US Environmental Protection Agency (EPA). Soviet scientist Mikhail Budyko (1988), in contrast, asserted that everyone would benefit from a global warming based on scenarios, plausible from his perspective. Perhaps the comments that US Senator Tsongas (1982) made about diametrically opposing views on the energy crisis of the 1970s and 1980s apply to the views of Gore and Budyko on winners and losers: Both of these approaches are equally absurd, equally rhetorical, and equally successful. When talking to the convinced, they are very powerful. And that is basically how most people address the issue: we are awash in rhetoric, not to mention hypocrisy, when what we need is a careful sorting and weighing of the facts and values involved in making ~ or not making m a decision. Many people seem to believe that discussing winners and losers (or, as some prefer to call them, advantaged and disadvantaged) will be divisive and could ultimately undermine
43 efforts to put together a global coalition truly intent on combating anthropogenically induced global warming. Reaction to a 1989 speech by Barber Conable, then-President of the World Bank, illustrates that discussion of winners and losers has, at least up to the recent past, been politically taboo. Environmental groups, which have been marching lock-step on this particular issue, opposed his public comments. In addition, some US congressmen even went so far as to suggest the need for a closer scrutiny of the World Bank's activities and budget. For example, the Washington Post (12 September 1989) reported, "In a letter to Conable, Wisconsin Senator Robert Kasten, Jr. wrote, 'the Bank's failure to be on the front lines of efforts to fight global warming threatens the Bank's long-term financial support from Congress.'" A similar argument was raised with respect to preventive versus adaptive response strategies. When the US EPA released two reports in 1983 suggesting that global warming was inevitable (Seidel and Keye s, 1983) and, as a result, people should plan for a rising sea level (Hoffman et al., 1983), the Friends of the Earth publication Not Man Apart denounced the Agency for "throwing in the towel," while at the same time, President Reagan's science adviser denounced the EPA reports as "alarmist." There was a feeling that "premature" discussions about adaptive strategies with respect to global wanning would break down the development of a united effort to support the pursuit and enactment of preventive strategies. Proponents of preventive strategies wanted attention to focus mainly on prevention as the best way to cope with global warming. There is, however, one projected impact of global warming for which one is allowed to identify specific winners and losers M sea level rise. This is probably because it is the one impact of a global warming for which there may be no obvious winners at the national level. No one has been reluctant to identify specific losers associated with sea level rise (papers have identified winners at the subnational level, such as coastal engineering firms and people who would have beachfront property as a result of a neighbor's misfortune). In this regard, one could argue that the sea level rise problem is similar to the stratospheric ozone depletion problem m no readily apparent national winners can be identified. Such would probably not be the case for changes in rainfall distribution, water resources availability, agricultural production, fisheries productivity, and energy production and consumption. The purpose of this presentation is to foster discussion of issues associated with the process of identifying winners and losers. What factors, for example, must be taken into account in labeling a region, an activity, an economic sector, or a country a winner or a loser? How do perceptions compare with reality? Can wins and losses be objectively and reliably identified? What are the costs and benefits of not addressing this issue as opposed to addressing it openly? My intention is not to label specific countries as winners or losers. To do that, one could simply use any of the GCM-generated scenarios, the scenarios generated by paleoecological reconstructions, or assessments of recent environmental changes and label specific countries and regions within countries accordingly. My purpose is to draw attention to the importance of addressing the winner-loser issue. As a note of caution, any attempt to identify potential winners and losers could only be viewed as a preliminary first step, because of the possibility of climate change surprises. For example, when I sought to include an assessment of the impacts of freezes on citrus production in the state of Florida as part of a larger set of analogues to possible global warming regional impacts, EPA advised me to drop that case study, asserting (not suggesting) that with a global warming "there would be no more freezes in Florida"! We did the study
44 anyway (Miller, 1988). As it happened, the 1980s, cited by scientists as the warmest decade in North America on record, witnessed the largest number of freezes in central Florida in its 130 years of record. Thus, regional counter-intuitive climate surprises must be expected. Nevertheless, identification of winners and losers is happening behind the scenes and should be brought out into the open. I realize that there is a risk associated with identifying winners and losers. If winners and losers are identified with some degree of reliability, the potential for unified action against the global warming may be reduced. Winners will not necessarily want to relinquish any portion of their benefits to losers in order to mitigate the impacts of their losses. On the other hand, there is also a risk in not making such a distinction between winners and losers. While scientists and policymakers formally discuss only losses associated with a global wanning, others may perceive that there will be positive benefits as well. The result is that the proponents for action on global warming could be likened to the fable about the emperor's new clothes, professing there are no winners, while everyone agrees with them in public but privately believes the opposite. This could sharply reduce the credibility of the proponents for taking action, lessening the chances for any response, preventive, mitigative, or adaptive.
2. SCENARIOS OF WINNERS AND LOSERS In the following section, the notion of winners and losers is discussed in terms of climatic conditions. These conditions include today's global climate regime, an altered climate regime, and varying rates of change. Winners and Losers with Today's Global Climate Regime It seems obvious that, say, fifty years hence there will be some societies that benefit from whatever climate exists at the time. After all, with today's climate, we can identify climaterelated winners and losers. As an example, the following map (Figure 1) shows droughtprone regions in sub-Saharan Africa, some of which could be considered climate-related losers. Such maps, depicting drought-prone and flood-prone areas, exist for other regions around the globe. Gains and losses at all levels of social organization, from local to international, may result directly from climate changes or from human responses to those changes. While there are several spokespersons for the extreme views (i.e., that all will win or all will lose), in all cases of changes (both relative and absolute), some will benefit, while others will be adversely affected. In addition, some nations, sectors, and groups may be in a better position to respond or adapt to climate change, turning this to their future advantage. The currently identifiable relative advantages and disadvantages of different nations, sectors, and groups result from a combination of climatic factors (such as climate variability and the frequency and intensity of extreme meteorological events) and a wide range of unique (by country, region, sector, or group) economic, social, and political factors that must be taken into serious consideration in any analysis (for more discussion of this issue, see ESIG, 1990). The differences, attributable to climate factors (e.g., recurrent droughts or floods), are likely to persist, although the relative positions of those affected might change. Furthermore, if such differences become extreme, they can lead to population movements by the disadvantaged (i.e., generating environmental refugees) and to conflicts either within national borders or across them).
45
~ M O S T CRITICALLYAFFECTEDBYTHEDROUGHT As of June 1985
ALGERIA
I
I ...... LIBERIA GUINEA-BISSAU
GAMBIA
RWANDA. BURUNDI 7
ZAIRE
~!.:~. . . . . . . . ~iili~i~i U
Critically affected
~
Most critically affected
Localized drought is prevalent in several other countries.
~ w York Tim~, 8120185
Figure 1 One could easily argue that there has been little sustained (or effective) effort to date by climate-related winners to assist those who might be considered climate-related losers. Examples that reinforce low expectations about adequate humanitarian assistance from the industrialized countries are not difficult to find. We have seen, for example, that in the past several decades, foreign assistance has been frequently tied to political considerations (e.g., aid to Cambodia and South Vietnam in the 1960s and 1970s, or to Ethiopia in the 1980s). In the early 1970s when there were widespread droughts throughout the world (except in the United States), then-US Secretary of Agriculture Earl Butz spoke about how food exports from the United States would be a new tool in the nation's foreign policy negotiating kit. Despite public statements to the contrary, few leaders in countries chronically affected by the adverse impacts of today's climate believe that they can rely on long-term, politically neutral assistance from those favored by today's global climate. The Colorado River Compact of 1922 provides an example of a recent "climate change" in which winners and losers have been identified. The Colorado River Basin in the southwestern United States was divided into two parts, the Upper and Lower Basins. The flow in the system was estimated at about 15 million acre-feet (mar), based on the record for the previous 20-year period, 1900-20. The representatives of the various states in the basin agreed to divide in absolute terms 15 maf average annual flow equally between the two
46 basins: 7.5 maf for each basin (75 maf over a ten-year period). However, because the Upper Basin states thought that there would be more water in the system than 15 maf, they agreed to provide the lower basin states with 7.5 maf, thinking that they would benefit from any surplus that might exist (see Brown, 1988). Shortly after the agreement was signed, however, the Colorado River entered a period of low streamflow, setting record lows in the 1930s (referred to as the US Dust Bowl decade). Today, the average annual streamflow (Figure 2) is estimated at about 13.5 maf. The loss of streamflow has to be absorbed by the Upper Basin. Thus, in this situation, one can
LEE'S
/
m ~
13.5 Million Acre-ft./Year Long-term Average Flow
/'vvx. /'1~,
,,,UJ~oO:'~'----4 1- ~ k0. , ~
V
Compact Drafted
20
O~z 15. w~
FERRY
5 0 , 1564
,
-"%/ ,
I 1600
,
,
,
~J ,
I 1650
i
i
_..
~b/ ~ , . / i
l
I 1700
r,,
"%.I i
~ t
^ k./
~ I 1750
i-, ~
~ J i
i
,,,k,
"~#k/''"
~'j
I
I
I
l
1800
A k/N-,/
I
I 1850
I
...~
k.~-I
I
v I
I 1900
,
~ ~ ,
I 1950
Figure 2 (from Stockton and Jacoby, 1976) identify winners and losers as a result from what might be considered a climate change that has, to date, lasted about six decades. Carrying this analysis a step further, one might ask what those who benefited from the Compact have done to compensate those who have not? What lessons for climate change responses by society might be drawn from this situation? Should future water compacts be based on proportional divisions of a variable resource, instead of absolute amounts? What does this case study suggest about when to reach agreement on a variable resource m before or after winners and losers are identified? Finally, an important question that merits attention, but has yet to be addressed among discussions about possible strategic responses to global wanning, is the following: who loses and who wins if no action is taken and the global climate regime remains as it is today? If it could be ascertained that no global wanning were to occur, what actions would today's climate-related winners take to alleviate the climate-related problems of today's climaterelated losers?
Winners and Losers with an Altered Global Climate Regime While we do not yet know the global, let alone regional, specifics of the havoc (or windfall) that a climate change will bring, we can assume that there will be winners and losers with a global climate warming. Some researchers and policymakers who are primarily concerned about regional impacts believe that, compared to the present climate of their region, it is possible that their climate could improve rather than worsen with a global warming. Saudi Arabia is one such example; Ethiopia might be another. Given their current climate, they might consider the risk of change worthwhile. Bandyopadhyaya (1983), an Indian social scientist, as well as Budyko
47 (1988) of the Russian Federation have made this argument at length in favor of a climate warming. Often, when people talk about the possibility of increased rainfall in a given region, a counter-argument is raised that ambient temperatures (and, therefore, evaporation rates) will also increase. This would tend to negate any benefits that might come from additional rainfall. Yet, history shows that societies have devised ways to capture rainfall and reduce evaporation, thereby improving the percentage of rainfall that they can effectively use (Glantz, 1991). Can we trmd examples of environmental conditions that different societies might have to cope with in the advent of a global warming? Are there existing climate change analogues for most places in the world? For example, in the United States, it has been suggested that the state of Iowa would become hotter and drier. Might Nebraska or Kansas provide a glimpse at Iowa's possible future environmental setting and, therefore, a glimpse of Iowa's future? Attempts to identify climate analogues are not new. The following maps of the former USSR (CIA, 1974) and of China (Nuttonson, 1947), Figure 3 and 4, respectively, depict agroclimate analogues from North America. Similar analogue maps could be created pertaining to climate warming, once we have an improved picture of the regional impacts of a global warming.
Analogies "Forecasting by analogy" provides social scientists with another approach to identifying possible societal scenarios associated with climate-related environmental change. The objective of this approach is not to forecast future states of either the atmosphere or society. Its purpose is to identify present-day societal strengths and weaknesses in human responses to environmental changes in order to forecast society's ability to cope with stresses that might accompany an unknown climate future. It can provide researchers with a low-tech approach to scenario development that encourages researchers to rely on existing, thus reliable, information about the regional impacts of extreme meteorological events. It can also provide a first approximation of societal preparedness for coping with an as-yet-uncertain climate future. Each methodological approach to develop a global warming scenario generates highly speculative glimpses of the future. To date, no one has successfully identified a method to forecast with any degree of reliability future states of the atmosphere. It would, therefore, be misleading to rely on any one of these scenarios as a basis for making specific policy recommendations in a specific region or locale. Such scenarios should not be taken as predictions or forecasts. They can, however, be used to create awareness among policymakers of the need to assess the regional consequences of climate change (e.g., Glantz, 1988). Winners and Losers and Rates of Change Many environmental changes with which decisionmakers are concerned today derive from human activities: climate change, tropical deforestation, desertification, mangrove destruction, and varying lake and inland sea levels. Climatologists, environmentalists, and policymakers have sought to obtain numbers that characterize the rates of these changes at global, national, and regional levels. These rates of current environmental change are determined directly or indirectly from space and ground observations, combined with
49
,~
,i
~
Alberta .
9
Climatic Analogues Wyoming ....~.
'~: i~:. j.i .....
Minnesota ~: ~ ~ ..:o
North Monta~ 4::.
The NorthAmedcan ~ n s |ndcsted have temperatures aml minfail somewhatsimilarto areas in China, C o ~ n s such as these can only be suggestive.
Figure 4
Puerto ~ Rico
statistical measurements. Projections of future rates are obtained from modeling activities, extrapolations of present-day trends, and subjective "guesstimates." These rates are extremely important to the development of scenarios about environmental conditions (including climate) in future decades. They also have an important impact on the particular policies pursued to mitigate or adapt to those rates and processes of environmental change, as well as on attempts to fine-tune the methods of detection. Perceptions about rates of change of global warming can affect one's views of the costs or benefits of such a climate change. Rates of environmental changes are often as important as the magnitude of those changes over the long term. When rates take on a crisis element (e.g., high stakes involved, perceived threat, short time to act, thereby challenging a society's ability to adjust), decisionmakers appear to take a more serious view of environmental changes that might affect them. High rates are more likely to cause alarm, while slow rates foster a "business as usual" attitude. One would even argue that it is often perceptions about the rates of change that prompt human action and not necessarily the order of magnitude of that change, even if that magnitude of change is unprecedented and beyond recent human experience. Environmental change appears to be used frequently as a bargaining chip in domestic as well as international negotiations. More specifically, in this context rates of change will affect response by decisionmakers. Negotiating tactics and strategies will clearly be affected by the rates of environmental changes that are proposed or used in intra-national and international negotiations. Varying rates are used by different protagonists in debates about
50 policy responses that might be required to deal with such changes. Thus, the importance of taking the subjective elements out of rates and processes can lead to more informed, more integrated, and more objective decisionmaking and perhaps to a more realistic view of what regional and local gains or losses might accompany global warming. It is also important to note that for any group, relative advantages and disadvantages are likely to change over time and that what might appear to be an advantage from climate change in the near term may, in the long run, turn into a disadvantage and vice versa.
3. RELATED QUESTIONS Before any attempt to identify specific winners and losers from a global wanning, there are several "prior" questions that must be addressed. In this section, some of these questions are posed and only briefly discussed to stimulate more critical examination. The following is meant to be suggestive of the kinds of concerns that must be raised when assessing the societal impacts of a global wanning. What Do W e Mean By a Win or a Loss?
It is not sufficient, meaningful, or realistic to equate more rainfall than normal with a win, and less rainfall than normal with a loss. In reality, the actual amount of rainfall in a given location does not by itself tell much about agricultural production. There are numerous articles about definitions of drought (e.g., Wilhite and Glantz, 1985). Researchers have identified differences between meteorological, agricultural, and hydrologic droughts. For example, if the expected annual amount falls (no meteorological drought) but is distributed throughout the growing season at the wrong time with respect to crop growth and development, a sharp decline in agricultural production (an agricultural drought) could occur. Likewise, defining a win or a loss according to changes in evaporation rates also may not be very useful. If evaporation rates increase, and all else remains the same, then there will be a depletion of water resources. However, as noted earlier, people in many add and semiadd areas have devised ways to minimize the impacts of high evaporation rates by the way they collect, store, and use their available, often scanty, water resources. Thus, the dependence on a single physical parameter to identify the costs or benefits to a society of a climate change has severe limitations. How Does One Measure a Win or a Loss? One might suspect that Canada will be a winner because, as temperatures increase and the growing season lengthens, agricultural productivity will improve. But what will be the impacts on Canadian fisheries, the timing of seasonal snowmelt, or the Canadian ski industry? Another local-scale example of the difficulty associated with measuring wins and losses is provided by historical attempts to augment precipitation in a semiarid part of central Colorado (USA). Cloud seeders were hired to suppress hail, augment rainfall during the growing season, and reduce rainfall during harvest, in order to improve the productivity of hops for beer production. Another group of farmers growing other crops (e.g., lettuce) and ranchers with different moisture requirements in the same valley opposed these cloud-seeding activities. The conflict between the two factions became violent, and the operation was eventually halted. Thus, even within small areas there can be different responses to changes in rainfall, making an objective determination of a win or a loss exceedingly difficult.
51
Finally, if one group loses, but loses less than others, should they be considered as an absolute loser or relative winner?
Can Wins and Losses Be Aggregated? While wins and losses can be "added" together to produce a net figure, one must question the value of that figure. The wins (or losses) are not shared commodities. Those who lose may not benefit in any way from those who win. For example, when the Peruvian fishery collapsed, those fishermen who had focused their activities (fishing gear, fishmeal processing factors, etc.) on exploiting anchoveta were not prepared to take advantage of exploiting the sharp increase in shrimp populations that appeared along the Peruvian and Ecuadorian coasts. A country can expect to have both winners and losers within its borders in the event of a climate change. While the winners may be in a position to take care of themselves, someone will have to help the losers. Wins and losses cannot be meaningfully aggregated. A win is a win, and a loss is a loss. What Is the Relationship between Perceptions of Wins and Losses and Actual Wins and Losses? Given the uncertainties surrounding the regional impacts of a global warming, actual winners and losers within and between countries cannot be identified with any degree of confidence. Perhaps we will learn that in reality everyone will lose (or win) with a global wanning of the atmosphere. However, as long as some regions or countries perceive themselves to be winners (or losers), they will act according to this perception. Thus, the issue of winners and losers must be addressed openly, objectively, and scientifically, if we wish to minimize the chance that actions taken in response to a global wanning will be based on misperceptions (Jamieson, 1994). How Should One Deal with the Issue of Intergenerational Equity? Identifying winners and losers spatially, as well as temporally, must become a concern of those dealing with the global warming issue. Arguments about intergenerational equity have been invoked to generate support for taking action now against global warming. We are asked to take actions today to protect future generations from the environmental insults wrought by the present generation. But how can we generate support for inter-generational equity when we cannot even achieve intra-generational equity among the various groups and generations now living? It appears that we have come to believe that any change in the status quo is, by definition, a bad change. But the real answer to this question will depend on who is asked to respond. A Saudi Arabian might believe that any change in the current climate regime will most likely be better for future generations of Saudi Arabians than the existing one. The opposite belief might be held by a farmer in the US Great Plains. The truth of the matter is that most people fear change (e.g., Hoffer, 1952).
4. CONCLUSION Every discipline has dealt with the concept of winners and losers in one way or another n biology, political science, history, sociology, economics, geography, law, ecology, conflict resolution, risk assessment, game theory, and so on. Climate-related impact assessment as
52 a result of global warming is only the latest topic that requires consideration of winners and losers. There have been conflicting views on whether to identify specific countries as winners or losers in the event of a global warming of the atmosphere. There has also been a reluctance to discuss the possibility that there may be any winners at all. It is time to get beyond that reluctance and to ask questions that need to be addressed so that the notion of winners and losers can be assessed on a more objective and realistic level. There is a calculated risk in such a discussion. Once specific winners have been reliably identified, there may be a reluctance on their part to lend support for global action to combat a greenhouse warming. We must take this risk. Many issues must be resolved before we will be in a position to identify with any degree of confidence who those specific winners might be. In the meantime, other issues, such as equity, definition, measurement, and perception versus reality, must be addressed if we ever hope to identify with some degree of confidence how specific countries, economic sectors, and regions within countries can develop response strategies to climate change in the 21st century. With such information in hand, governments and nongovernmental organizations would be in a position to devise tactics and strategies for coping with global-warming-induced national and regional changes.
The Regionalization of Environmental Problems Given the resurgence of worldwide concern about "global" environmental issues and their regional causes or consequences, regional organizations could provide an effective arena for discussing, resolving, or averting regional conflicts related to environmental change. Perhaps the 1990s provide a "window of opportunity" for a review of regional organizations and their potential contribution toward resource-related cooperation and conflict resolution at the regional level. Whereas regional international organizations (functional as well as geographic) have often been relegated to marginal roles in the international political arena with regard to resources issues, this could change in the future. Such a change can be expected because, as climate regimes shift in response to global warming, so too will the location of some highly valued natural resources, sometimes across national boundaries, such as water resources and fish populations. As resources and people dependent on them "migrate" on land and in the marine environment, the risks of regional conflicts, as well as the opportunities for regional cooperation, are likely to increase. The time may be fight to talk about the "regionalization of environmental problems."
5. REFERENCES Bandyopadhyaya, J., 1983: Climate and World Order: An Inquiry into the National Causes of Underdevelopment. New Delhi, India: South Asian. Brown, B.G., 1988: Climate variability and the Colorado River Compact: Implications for responding to climate change. In M.H. Glantz (ed), Societal Responses to Regional Climate Change: Forecasting by Analogy. Boulder, Colorado: Westview Press, 279-305. Budyko, M.I., 1988: Anthropogenic climate changes. Paper presented at the World Congress on Climate and Development, 7-10 November 1988, Hamburg, Germany.
53 Callendar, G.S., 1938: The artificial production of carbon dioxide and its influence on temperature. Quarterly Journal of the Royal Meteorological Society, 64, 223-241. CIA (Central Intelligence Agency), 1974: Potential Implications of Trends in World Population, Food Production, and Climate. Report OPR-401, August. Washington, DC: CIA. CIA (Central Intelligence Agency), 1976: USSR: The Impact of Recent Climate Change on Grain Production. Report ER 76-10577 U. Washington, DC: CIA. ESIG (Environmental and Societal Impacts Group), 1990: On Assessing Winners and Losers in the Context of Global Warming. Report of Workshop held 18-21 June 1990 in St. Julians, Malta. Boulder, Colorado: ESIG, National Center for Atmospheric Research. Goldsmith, E., 1977: The future of an affluent society: The case of Canada. The Ecologist, 7, 160-194. Glantz, M.H., 1991: The use of analogies in forecasting ecological and societal responses to global warming. Environment, 33, No. 5, 10-33. Glantz, M.H. (ed.), 1988: Societal Responses to Regional Climatic Change: Forecasting by Analogy. Boulder, Colorado: Westview Press. Gore, A., 1992: Earth in the Balance: Ecology and the Human Spirit. New York: Houghton Mifflin Co. Hoffer, E., 1952: The Ordeal of Change. New York: Harper and Row Publishers. Hoffman, J.S., D. Keyes, and J.G. Titus, 1983: Projecting Future Sea Level Rise: Methodology, Estimates to the Year 2100, and Research Needs. EPA 230-09-007. Washington, DC: US EPA, Office of Policy & Resource Management. Impact Team, 1977: The Weather Conspiracy: The Coming of the New Ice Age. New York: Ballantine Books. Jamieson, D., 1994: Global environmental justice. In R. Attfield and A. Belsey (eds), Philosophy and the Natural Environment. Cambridge: Cambridge University Press. Kellogg, W.W., 1977: Effects of Human Activities on Global Climate: A Summary with Considerations of the Implications of a Possibly Warmer Earth. WMO Tech. Note 156 (WMO No. 486). Geneva, Switzerland: WMO. Miller, K.M., 1988: Public and private sector responses to Florida citrus freezes. In M.H. Glantz (ed), Societal Responses to Regional Climatic Change: Forecasting by Analogy. Boulder, Colorado: Westview Press, 375-406.
54 Nuttonson, M.Y., 1947: Ecological crop geography of China and its agro-climatic analogues in North America. International Agro-Climatological Series, Study No. 7. American Institute of Crop Ecology. Ponte, L., 1976: The Cooling. Englewood Cliffs, NJ: Prentice-Hall. Revelle, R., and H.E. Suess, 1957: Carbon dioxide exchange between atmosphere and ocean and the question of an increase of atmospheric carbon dioxide during the past decades. Tellus, 9, 18-27. Seidel, S., and D. Keyes, 1983: Can We Delay a Greenhouse Warming? Washington, DC: US Environmental Protection Agency. Stockton, C.W., and G.C. Jacoby, 1976: Long-term surface water supply and streamflow trends in the Upper Colorado River Basin. Lake Powell Research Bulletin, 18. Tsongas, P.E., 1982: Foreword. In K.A. Price (ed), Regional Conflict and National Policy. Washington, DC: Resources for the Future, xi-xiv. Wilhite, D.A., and M.H. Glantz, 1985: Understanding the drought phenomenon: The role of definitions. Water International, 10, 111-120.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
55
Sustainable Development and Climate Change R.K. Turner Centre for Social and Economic Research on the Global Environment (CSERGE), University of East Anglia, Norwich and University College London, United Kingdom
Abstract Sustainability is defined in terms of four overlapping positions, ranging from very weak to very strong sustainability. The core idea is of a non-declining capital stock (including natural capital) over generational time. Weak sustainability positions emphasise capital substitution possibilities and the power of technical process to mitigate resource depletion and pollution problems. Climate change and its associated risks and strong uncertainty are characterised by features which f a v o u r ' a strong sustainability approach incorporating the precautionary principle. Strong sustainability positions recognise constraints on substitution processes and incorporate ethical concerns such as intergenerational equity as a moral duty. Cost-benefit analysis is moderated via safe minimum standards which set GHGs concentrations and emissions abatement targets.
Introduction: Sustainable Development Concept Economists define sustainable development in terms of non-decreasing levels of utility, or income per capita, or real consumption per capita over time. In broad terms it involves providing a bequest from the current generation to the next of an amount and quality of wealth which is at least equal to that inherited by the current generation. This requires a non-declining capital stock over time and is consistent with the intergenerational equity criterion. The most publicised definition of sustainable development credited to the World Commission on Environment and Development also included an intragenerational equity criterion (WCED, 1987). Sustainability therefore requires a development process that allows for an increase in the wellbeing of the current generation (with particular emphasis on the welfare of the poorest members of society), while simultaneously avoiding uncompensated and 'significant' costs (including environmental damage costs) on future generations. Such a cost liability would reduce the 'opportunities' for future generations to achieve a comparable level of well-being (Pearce, Barbier and Markandya, 1990). The sustainability approach therefore is based on a long-term perspective, it incorporates an equity as well as an efficiency criterion, and it may also emphasise the need to maintain a 'healthy' global ecological system (Costanza et al., 1992). A spectrum of overlapping sustainability positions (from very 'weak' to very 'strong') can be distinguished, see Figure 1 (Turner, 1993). Weak sustainability requires the maintenance of the total capital stock- composed of
57 K~ (manufactured or reproducible capital); Kh (human capital, or the stock of knowledge and skills); K (natural capital: exhaustible and renewable resources, together with environmental structures, functions and services) - through time with the implicit assumption of infinite substitution possibilities between all forms of capital. The Hartwick Rule (Hartwick, 1978) is also used to buttress the weak sustainability position by regulating the intergenerational capital bequests. The rule lays down that the rent obtained from the exploitation of the natural capital stock by the current generation, should be reinvested in the form of reproducible capital which forms the future generations' inheritance. This inheritance transfer should be at a sufficient level to guarantee non-declining real consumption (well-being) through time. The implicit capital substitutability assumption underpins the further argument that extensive scope exists over time for the decoupling of economic activity and environmental impact. The decoupling process is mediated by technical progress and innovation. While total decoupling is not possible, and with the important exception of cumulative pollution, society's use of resources can be made more efficient over time (i.e. the amount of resources used per unit of GNP goes down faster than GNP goes up and the aggregate environmental impact falls). From the weak sustainability perspective a key sustainability requirement will be increased effective research and development, i.e. new knowledge properly embodied in people, technology and institutions. From the strong sustainability perspective some elements of the natural capital stock cannot be substituted for (except on a very limited basis) by manmade capital and therefore there is a concern to avoid irreversible losses of environmental assets. Some of the functions and services of ecosystems in combination with the abiotic environment are essential to human survival, they are life-support services (e.g. biogeochemical cycles) and cannot be replaced. Other multi-functional ecological assets are at least essential to human wellbeing if not exactly essential for human survival (e.g. landscape, space and relative peace and quiet). We might therefore designate those ecological assets which are essential in either sense as being 'critical natural capital'. Supporters of the "deep ecology " [VSS] position argue for a particular type of nonsubstitutability based on an ethical rejection of the trade-off between man-made and natural capital. The strong sustainability rule therefore requires that we at least protect critical natural capital and ensure that it is part of the capital bequest. The combination of the risk of irreversible environmental losses and a high degree of uncertainty surrounding past rates and future trends in resource degradation and loss, as well as the full structural and functional value of ecosystems (Gren, Folke, Turner and Bateman, 1994), leads strong sustainability advocates to adopt the precautionary principle. Conservation of natural capital and the application of a safe-minimum standards (Bishop, 1993) approach are therefore important components of a strong sustainability strategy. This message is that environmental degradation and loss of natural resources represent one of the main ways in which today's generation is creating uncompensated future costs. Hence restoration and conservation of natural resources and the environment is crucial to achieving sustainable development.
58 A number of sustainability rules (which fall some way short of a blueprint) for the sustainable utilisation of the natural capital stock can be outlined:
II)
III)
IV) V)
Market and policy intervention failures related to resource pricing and property rights should be corrected. The regenerative capacity of renewable natural capital should be maintained, i.e. harvesting rates should not exceed regeneration rates; and cumulative pollution Which could threaten waste assimilation capacities and life-support systems should be wherever feasible avoided. Technological changes should be steered via an indicative planning system such that switches from non-renewable natural capital to renewable natural capital are fostered; and efficiency-increasing technical progress should dominate throughput-increasing technology. Resources should, wherever possible, be exploited, but at a rate equal to the creation of substitutes (including recycling). The overall scale of economic activity must be limited so that it remains within the carrying 'capacity of the remaining natural capital. Given the uncertainties present, a precautionary approach should be adopted with a built-in safety margin.
Figure 2 summarises some of the measures and enabling policy instruments that would be involved in any application of a very weak sustainability (VWS) through to a very strong sustainability (VSS) strategy (Turner, 1993). From our review of sustainability, the emphasis on equity and social issues in sustainability as well as on the physical constraints is important. For development to be sustainable it must incorporate (under the strong sustainability view) non-depletion of natural capital; both intergenerational and intragenerational equity principles; and in the latter context must be capable of providing sustainable livelihoods to those whose livelihoods are primarily natural resource dependent. Agenda 21 sets out principles for sustainable development without advocating any explicit definition of sustainability and with a tendency for focusing on global issues which may not be of greatest concern to those poorest sections of the world. The implicit definition of sustainability within Agenda 21 however would seem to be closely related to the concept of strong sustainability discussed above, though the lack of operational details and the prevailing obstacles to change mean that implementation of such an agenda represents a very formidable task. In the context of climate change the sustainability concept would favour the adoption of a general response strategy that was based on the following ethical arguments: there is an obligation to avoid harm to future generations, either in an absolute sense, or so long as the avoidance measures themselves do not impose unacceptable cost on society;
59
Figure 2 Sustainability Mode (overlapping categories)
VWS
WS
SS
VSS
Sustainability Practice Management Strategy (as applied to projects policy or course of action)
Conventional Cost-Benefit Approach: Correction of market and intervention failures via efficiency pricing; potential Pareto criterion (hypothetical compensation); consumer sovereignty; infinite substitution Modified Cost-Benefit Approach: extended application of monetary valuation methods; actual compensation, shadow projects etc; systems approach, "weak' version of safe-minimum standard F i x e d S t a n d a r d s Approach: Precautionary Principle, recognition of the full value of natural capital; constant natural capital rule; 'strong' version of safe minimum standard A b a n d o n m e n t of Cost-Benefit Analysis: or severely constrained cost-effectiveness analysis; bioethics (i.e. an acceptance of the rights and interests of non-human species which then constraints human activity on moral grounds, e.g. the loss of tropical forests is in some circumstances morally wrong)
Policy Instruments (most favoured)
Pollution Raw Control and Materials Waste Policy Management e.g. pollution taxes, elimination imposition of property rights
Conservation and Amenity Management of subsidies,
e.g. pollution taxes, permits, deposit-refunds; ambient targets
e.g. Ambient standards; conservation zoning; process technology-based effluent standards; permits; severance taxes (i.e. taxes on resource extraction); assurance bonds (a sort of market-based insurance fund to mitigate environmental damage impacts) standards and regulations; birth licences
Source" Turner (1993)
the avoidance of harm to future generations is important because the future has no power to influence decisions taken now which may cause them harm; and current generations have moral obligations to future generations (either via overlapping generations or via the acceptance of interests/rights for future people). Climate change may strain an economy's capacity to achieve sustainable development by imposing unpredictable and significant damage, damage mitigation and adaptation costs. Resource investment and development planning may then be badly disrupted, pushing the sustainability goal further into the future. Developing economies will be faced with disproportionality severe dislocation costs because of their 'vulnerable' socio-economic systems and supporting ecological systems. Climate change is only one component of global environmental change (i.e. a complex flux of factors - population growth, increasing urbanisation, increasing industrialisation and intensification of agriculture, increasing rate of economic growth and international economic interdependency, the globalisation of information transfer and communications
60 and an increasing rate of attitudinal and lifestyle changes - the impacts of which can manifest themselves at a number of different spatial and temporal scales). Many developing countries, and to a lesser extent some regions (e.g. coastal zones) of developed countries, are already under heavy environmental pressure and potential climate change impacts on, in particular, agricultural sectors and coastal zone resources will further exacerbate their developmental problems. Climate risk is therefore very much an equity issue because the cost of riskbearing is not evenly distributed across societies, and more significantly across countries. Developing countries wiI1 face an especially high risk-bearing cost burden. We now turn to a closer examination of climate change risk and its implications for sustainable development.
Climate Change Risk Climate risk and other Global Environmental Change risks are shrouded by strong uncertainty and this is especially so at the regional level. The global scope of the potential changes means that there is collective risk which affects very large numbers of people. These risks are not statistically independent and the effectiveness of risk "pooling" is reduced. They are also endogenous risks in the sense that the global systems changes are being driven by human economic activity, because of its sheer 'scale' (Chichilnisky and Heal, 1993). The economy and the environment are jointly determined systems and the overall scale of economic activity is now very significant. Climate change impacts are potentially therefore part of a wider set of impacts and consequences. There is also a degree of permanent unpredictability present because the dynamics of the jointly determined system (coevolutionary process) are characterised by discontinuous change around critical threshold values both for biotic and abiotic resources, and for ecosystems functions. The stability of the jointly determined economy-environment systems depends less on the stability of individual resources, than on the resilience of the system, i.e. the ability of the system to maintain its self-organisation in the face of stress and shock. Unfortunately even if the critical threshold values could be discovered, neither the transition time to a new system state, nor the form of the new system state could be predicted. It is now a matter of some debate in the context of climate states whether the most 'natural' behaviour to be expected is a gradual warming trend process, or an abrupt phase change, as one climate region gives way to a new one (either globally or regionally). The characteristics of climate and related risks and the pressure of strong uncertainty provide a compelling rationale for the deployment of strong sustainability/precautionary instruments in ecological economic systems. The existence of possible threshold effects involving irreversible loss of potential productivity, and the failure of markets to signal the nearness of such thresholds, both imply the need for instruments that maintain economic activity and its pollution and waste generation consequences within appropriate bounds. The economic perspective, in principle, suggests the following analytical sequence for GHGs abatement and mitigation of other global environmental change effects - a general acceptance and application of extended cost-benefit analysis; recognition and quantification wherever feasible of environmental
5! risks; deployment of the precautionary approach via safe minimum standards (subject to their social opportunity costs) in the presence of strong uncertainty; deployment of a portfolio of enabling policy instruments to meet the chosen GHGs concentrations abatement targets and other sustainability goals.
Climate Change Decision-Making Strategies Greenhouse gases (GHGs) induced climate change poses a multifaceted challenge which has to be addressed via a collective decision-making framework operating at both the national and international levels. The decision-making contexts are characterised by 'strong' uncertainty and irreversibility and therefore favour the adoption of more, rather than less, risk averse strategies. A priori, a strong sustainability approach would seem therefore to be appropriate since it recommends the avoidance of those options which may generate the worst outcomes ('unacceptable' cost burdens) and encompasses the precautionary principle. What is and what is not an acceptable cost, from the strong sustainability, perspective is only partly measured by reference to individuals' preferences (conventional 'economics approach). Individuals may not be well informed about climate risk and expert opinion is constrained by strong uncertainty. Further, human preferences may not fully capture intrinsic values in nature - see Figure 3. In the market place, a product's value is encapsulated in its market price which in turn is determined in part by consumers' willingness-to-pay. But environmental resources often have no price tag and information is lacking concerning their 'true' value and significance. Many of these environmental assets are also public goods and this is another characteristic that makes it difficult for markets to evolve in such assets. To make the comparisons of environmental and other costs and benefits, within cost-benefit analysis, economists have therefore to impute a value for non-market environmental assets. A range of valuation methods and techniques have been deployed in order to estimate the value of various components of the environment. Environmental economists have developed a terminology of valuation which distinguishes between individual (private) use value (direct and indirect use of the environment), option value, quasi-option value, bequest and existence (non-use) value. Debate continues over the precise boundaries between these different components of total economic value. The social value of environmental resources is then simply the aggregation of private values. However, ecological economic research findings indicate that the social value of environmental resources committed to some use may not be equivalent to the aggregate private value of the same resources in any given system, because of the following factors: o
The full complexity and coverage of the underpinning 'life support' functions of healthy evolving ecosystems is currently not precisely known in scientific terms. A number of indirect use values within systems therefore remain to be rediscovered and valued.
62
Figure 1.
3
A General
Value
Typology
Anthropocentric I n s t r u m e n t a l Value
This is equivalent to "Total economic v a l u e " = use + non-use value. = DUV + [direct use value]
IUV -+ [indirect use value]
OV [option value]
+
QOV + [quasi-option value
BV [bequest value]
+
EV [existence value]
The non-use category is bounded by the e x i s t e n c e v a l u e concept which has been the subject of much debate. Existence value may therefore encompass some or all of the following motivations: interpersonal altruism, resource conservation to ensure availability for others; vicarious use value linked to self-interested altruism and the "warm glow" effect of purchased moral satisfaction; intergenerational altruism ( b e q u e s t motivation and value), resource conservation to ensure availability for future generations; stewardship nature;
motivation, h u m a n responsibilities for resource conservation on behalf of all
" Q - a l t r u i s m " , motivation based on the belief t h a t non-human resources have rights and/or interests and as far as possible should be left undisturbed. If existence is defined to include stewardship and "Q-altruism" then it will overlap into the next value category outlined below 2.
Anthropocentric Intrinsic Value _
This value category is linked to "Weak anthropocentrism" in a subjectivist sense of the term value. It could be culturally dependent. The value attribution is to entities which have a "sake" or "goods of their own", and instrumentally use other parts of nature for their own intrinsic ends..." It remains an anthropocentrically related concept because it is still a h u m a n valuer that is ascribing intrinsic value to non-human nature ("Q-altruism"). 3.
Non-Anthropocentric I n s t r u m e n t a l Value
In this value category entities are assumed to have sakes or goods of their own independent of h u m a n interests. It also encompasses the good of collective entities, e.g. ecosystems, in a way t h a t is not irreducible to t h a t of its members. But this category may not demand moral considerability as far as h u m a n s are concerned. 4.
Non-Anthropocentric Intrinsic Values _
This value category is viewed in an objective value sense, i.e. "inherent worth" in nature, the value t h a t an object possesses independently of the valuation of valuers. It is a meta-ethical claim, and usually involves the search for constitute rules or t r u m p cards with which to constrain anthropocentric i n s t r u m e n t a l values and policy. Source" Adapted from Hargrove (1992)
53
.
o
.
Because the range of use and non-use value that can be instrumentally derived from an ecosystems is contingent on the prior existence of such a healthy and evolving system, there is in a philosophical sense a 'prior value' that could be ascribed to the system itself. Such a value may not, however, be measurable and may not be commensurate with the economic (secondary) values of the system. The continued functioning of a health ecosystem is more than the sum of its individual components. There is a sense in which the operating system yields or possessed 'glue' value, i.e. related to the structure and functioning properties of the system which holds everything together. A healthy ecosystem also contains a redundancy reserve, a pool of latent keystone species and processes which are required for system maintenance in the face of stress and shock.
The adoption of a systems perspective, the recognition of primary ecosystem value (in addition to secondary value related to components of the system) and the nature of much environmental risk, i.e. high cost, low probability risks, emphasise the need for policy instruments that safeguard the range of options to future generations. Such precautionary instruments ensure that irrespective of the actual outcome of current activity, the next generation is left with an equivalent resource endowment (allowing for some trading between different forms of capital-physical capital, human capital and natural capital) and opportunities for economic development. These are commonly identified as sustainability constraints, e.g. safe minimum standards. Uncertainty about system boundaries and the effects of scale and thresholds underline the value of a precautionary approach, and many sustainability instruments have the property that they are precautionary. Sustainability requires each generation to maintain the self-organising systems that provide the context for all human activity and therefore possess 'primary' value. This does not imply that all assets should be preserved. Rather it implies conservation of opportunity. Thus one criterion, for example, for decision-making under 'strong' uncertainty is the 'maximim' criterion (i.e. minimise the worst outcome strategy). This is also complementary to the Rawlsian equity criterion (i.e. maximise the conditions of the least well off). On the other hand, since the uncertainties are so great and potential mitigation costs are so high, a better strategy might be to 'wait and see' and not to adopt any extensive policy interventions. As scientific, economic and technological data cross some of the uncertainties may diminish and we will be better able to discern and quantify the GHGs abatement cost and damage cost functions. Policymaking could then be aided by the application of the costbenefit method and techniques. Such an approach would be more in line with the 'weak' sustainability perspective. For some commentators the 'wait and see/business-as-usual' stance is attractive because they argue that if the forecast of a gradual trend rate of temperature rise (+0.3k per decade) is accurate (Houghton et al., 1990), then the global temperature signal will be discernible sometime between 2010 and 2020.
54 Policymakers therefore ought to defer any significant GHGs emission abatement measures to the next generation, and take out 'partial cover' by encouraging insurance schemes applicable to individuals and nations (e.g. Alliance of Small Island States International Insurance Pool proposal). Critics have countered that the mere existence of strong uncertainty cannot justify a no policy response option. The potential climate-induced damage costs could be very high, and in any case are only one possible element in the aggregate global environmental change impacts, many of which have already put heavy pressure on ecosystem resilience and adaptation capacities: The futures' 'opportunities' set is therefore being threatened, especially as GHG impacts may be irreversible. Whether or not future generations do possess moral rights or interests, most people would support the view that the present's 'coefficient of concern' for the future is not zero. Finally, a range of policy options are either 'no regret' negative net cost options, or are moderate cost options because once implemented they carry with them secondary benefits in addition to avoided climatic change damages. 9 The decision-making context and process are complex because uncertainty and irreversibility characteristics are compounded by the existence of a range of conflicting decision criteria, e.g. economic efficiency, intragenerational/ intergenerational equity, sustainability and precaution. The process probably has to be both hierarchical and sequential. Taking the weak sustainability position, we might assume that given moderate rates of technical progress (1 to 2% per annum) and actual global warming adaptation costs of up to 3% of GNP, future generations will be substantially better off than the current generation. So if the future benefits outweigh the present abatement costs, the future should pay those costs. But recall the weak sustainability capital substitution axiom, which in this context assumes that atmospheric capital is substitutable by manmade capital. If no such extensive substitution is possible then delaying GHG abatement measures in favour of providing a capital bequest (and investing is knowledge) for future generations cannot be justified. Alternatively, taking the strong sustainability position, possible significant future damage costs and irreversible impacts suggest that the future may be made worse off than the present. The passing on of a 'net liabilities' bequest to the future is morally questionable. Therefore a global level GHGs concentrations/emissions abatement target and interregional allocations need to be exogeneously set, guided by the precautionary principle. Once the commitments have been made then the search should be for cost-effective enabling measures, which should be adopted sequentially - 'no regret' energy conservation and efficiency improvement measures first, followed by fuel switching and other measures requiring longer lead times and significant capital investments. W e a k Versus Strong S u s t a i n a b i l i t y P o l i c y O p t i o n P o r t f o l i o s In principle, the full set of climate change response measures include the following: (I)
science-based research to reduce climate change and impact uncertainties;
65
(II)
technological research focused on more cost-effective GHGs mitigation measures (energy conservation and efficiency measures etc);
(III)
reversal of policies which in the past have encouraged the inefficient use of resources and waste sinks (i.e. correction of market and intervention 'failures');
(IV)
joint implementation, technology transfer and other forms of international cooperation to limit climate change;
(V)
measures to reduce GHGs emissions and/or to increase sequestration of GHGs;
(VI)
measures to enhance adaptation capacities of both socio-economic and natural systems facing the consequences of climate change;
(vii) insurance schemes to hedge against climate risk and the costs of adaptation; (VIII) policy interventions directed at the main drivers of global environmental change (such as population growth) and its related pressures. The different sustainability perspectives encourage the adoption of different policy option portfolio configurations. A weak sustainability strategy would seek to promote a portfolio based on measures (I), (II), (III) and (VII). It would be a more reactive rather than a proactive strategy and would lay great stress on the ability of research and development to reduce uncertainty and promote efficient resource usage. Cost-benefit thinking and analysis would be used as an i m p o r t a n t aid to decision-making. A strong sustainability strategy would seek to implement a more proactive and a more comprehensive policy portfolio including all the measures (I) to (VIII) in the list above. It would lay stress on the need for a precautionary approach and would recognise an obligation to future generations, not to pass on net liabilities. It would further seek to incorporate climate change impacts within a more general recognition of the GEC 'scale' problem. In this context integrated resource m a n a g e m e n t strategies would be highlighted (e.g. integrated coastal zone management). Cost-benefit analysis would be constrained by the precautionary principle via a safe-minimum standards (SMS) approach. The latter would be applied in an absolute sense (regardless of costs) when and if 'critical' natural capital assets were identified as being under threat; and more generally in a relative sense depending on the social acceptability of the SMS's own cost implications. References
R. Bishop, Economic efficiency, sustainability and biodiversity. Ambio 22 (1993) 69-73.
55
10
G. Chichilnisky and G. Heal, Global environmental risks. Journal of Economic Perspectives 7 (1993) 65-86. R. Costanza et al., Ecosystem Health: New Goals for Environmental Management, Island Press, Washington, 1992. I-M. Gren, C. Folke, R.K. Turner and I. Bateman, Primary and secondary values of wetland ecosystems. Environmental and Resource Economics 4 (1994) 55-74. C. Hargrove, Weak anthropocentric intrinsic value. The Monist 75 (1992) 183-207. J. Hartwick, Substitution among exhaustible resources and intergenerational equity. Review of Economic Studies 45 (1978) 347-354. J.T. Houghton et al. (eds), Climate Change: The IPCC Scientific Assessment, Cambridge University Press, Cambridge, 1990. D.W. Pearce, E. Barbier and A. Markandya, Sustainable Economic Develpment, Edward Elgar, Aldershot, 1991. R.K. Turner (ed), Sustainable Environmental Economics and Management, Belhaven Press, London, 1993. World Commission on Environment and Development, Our Common Future, Oxford University Press, Oxford, 1987.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
67
Global Climate change" social and institutional options M.R. Redclii~
Global E n v i r o n m e n t a l C h a n g e P r o g r a m m a (ESRC), W y e College, Wye, K e n t T N 2 5 5AH, U n i t e d K i n g d o m
Introduction In exploring the available social and institutional options over global climate change we need to address some fundamental questions about individual behaviour, social responsibility and 'globalisation'. Before doing this, however, it might be useful to establish some points of entry.
(1) Responses to the condition of the environment We need to know more about how environmental problems are perceived. For example, in the case of climate change, policies to combat global warming, to be effective, require some understanding of the links between individual behaviour and climate (both atmospheric concentrations and emission levels). Much better public information and media attention are essential before people can assess their responsibility for what is happening and what they can do about it. There are a number of social mechanisms which enable us to distance ourselves from the full implications of our behaviour. These need to be looked at - how our 'underlying social commitments' help establish this distance - before behaviour can be changed.
(2) Responses to existing policies We also need to know more about public responses to existing policies, many of which are not viewed as 'environmental'. Societies are not homogenous. Some ecological benefits carry distributive costs. How do people become enrolled in more sustainable practices like recycling and companies in green accounting? Pressure points exist where public opinion and values are more amenable to change. We should not forget that societies are reflexive systems; unlike inanimate objects, what we do reflects what we understand. More work needs to be done on sustainability indicators, to enable us to place normative goals into an operational context. Before going further, however, we need to examine global environmental change itself.
68 Global Environmental Change
Although it is usually conceded that values play a large part in the way we approach the environment, particularly the environment on our doorstep, the same concession is rarely made for the global environment. Global environmental change is often identified with physical processes "out there", such as ozone depletion, biodiversity losses and, particularly, global warming. The global environmental agenda has, to some extent, been established by the natural sciences, working within a positivist tradition (Newby 1993). The reports of the Intergovernmental Panel on Climate Change (IPCC) are a case in point. The authority of the IPCC's deliberations stems, to some extent, from its "scientific" objectivity, which influenced people like the former British Prime Minister, Mrs Thatcher, in lending it their support (Boehmer-Christiansen 1993). This paper examines whether global environmental change is as free from value considerations as many people believe, or hope. It goes on to explore three clusters of issues which suggest that an alternative approach needs to be taken. Global environmental change can be understood in terms of three sets of issues, each of which forces us to examine our part in its construction: human relations with "Nature", the need to live with increased uncertainty, and the extent to which our management of the environment reflects essentially human, rather than environmental, concerns. Each of these issues influences not merely the way we understand environmental problems, but also the way in which we can act to change them. They are also represented in the three major policy initiatives to have developed out of the Earth Summit in Rio de Janeiro in 1992: the Framework Climate Convention; the Biodiversity Convention and the institutions responsible for establishing more sustainable practices at the international level (particularly the Commission for Sustainable Development and the Global Environment Facility)(Grubb 1993). In the final part of this paper their relevance to the global climate agenda is considered. Human Relations and "Nature"
The nineteenth century was a period in which the physical sciences saw spectacular progress, and most of the scientific disciplines assumed the identity they possess today. It was also a period, in Europe and North America, of enormous economic growth, and with economic progress came confidence. Looking back from the vantage point of the end of the twentieth century this belief in progress, and the confidence that went with it, are the hallmarks of modernism (Redclift 1993). Relatively rapid industrialisation, and the growth of towns, were "global" phenomena because they served to incorporate other economic systems, and other cultures. Globalisation in the latter part of the twentieth century has served to underline these links, changing the international economic division of labour, using technology and communications to provide global images, as well as markets, and seeking to preserve the exotic and unfamiliar ("the other") whether through tourism or environmental campaigning, as items of consumption (Featherstone 1990, King 1991).
69
During the late nineteenth century, and early twentieth century, the opposition between nature and culture, made room for the social sciences as autonomous disciplines, they grew up in the interstices between the ethical concerns of the humanities and the positivism of the "hard" sciences. The insistence that human cultures were distinctive brought into question both the 'external' environmental determinism of some of the new sciences, and the "naturalism" of others, which saw human behaviour as the outcome of "internal" biological forces, equally beyond our control (Benton and Redclift 1994). Both of the imperatives provided by nature, the external environment and the human biological condition, were found wanting. It is not an accident that many of the issues which proved (and still prove) difficult for the social sciences to confront, such as eugenics, racism and the measurement of intelligence, lie at the crossroads of biology and social conditioning. In this sense the nature/culture dichotomy was both the springboard for the social sciences' advance, and the irresoluble problem they confronted (Benton and Redclift 1994). Within the social sciences the discipline which benefited most from its identification with human purposes in the nineteenth and twentieth centuries was economics. Neo-Classical economics grew out of the increasing confidence of industrial capitalism with its own success, and its refinements were linked to the problems faced by twentieth century industrial economies (welfare economics, Keynesianism, development economics). Many of the issues which will confront us as we approach the twenty first century - the relationship between the production of goods and services and the satisfaction of our needs, as well as the social and environmental consequences of their production - elude mainstream economists. Many of the underlying assumptions which influenced economic reasoning, such as the effects of scarcity, now appear much less important than issues like the environmental consequences of economic behaviour, which played little part (Yearley 1991). In the view of many Neo-Classical economists the significance of scarcity could be grasped through concepts like the economic costs of resource acquisition. Pollution and the proliferation of so-called "externalities" can be seen as manifestations of profligacy, rather than scarcity, and our inability to manage its consequences. As our dependence on economic techniques increases, the need for more inclusive systems of thought appears more urgent. We are forced back, inevitably, to consider our relations with nature, from which resources derive. Our increasing knowledge of biological systems has not enabled us to utilise them sustainably, and this is due in some part to the divorce which was effected in the nineteenth century between our understanding of the laws of nature and those of "man". We are faced by an interesting paradox. On the one hand the degradation of nature has called into question some of the values which contributed to the Promethean successes of the past. The rights of non-human species, and the primary obligations which we have to nature, are now regarded as politically important, and not merely by Deep Ecologists. At the same time, many of those who espouse environmental concerns refuse to acknowledge that it is the way that human societies are organised, and structured, which determines environmental problems. What are the values generated from the management of the environment today? They clearly reflect the interface between society and nature, and the difficulty we experience in dealing with this interface. Environmental management itself suggests a mastery of nature, and an ability to control the environmental consequences of our behaviour. The growing
70 importance of scientific knowledge, and "rationality" as the coda for this knowledge, together with our institutionalised behaviour and social commitments, has served to increase the appeal of technical solutions to human-induced problems. To provide solutions to environmental problems, however, we need look no further than the human societies which produce them; something which we seldom do (Beck 1992).
Living With Uncertainty: the Importance of Time and Space Another consequence of the growing confidence of science has been the expectation of certainty. With the development of scientific techniques and methods the status of scientific prediction rose, and with it the status of scientists. Predicting environmental consequences has proved to be difficult, however, partly because of the complexity of environmental systems, and partly because of the unpredictability of human actions. Science is apparently successful in offering predictions which reduce uncertainty. However, science also collapses time and space, and increases the flow of knowledge and information available. This, in turn, tends to increase uncertainty, and to fuel speculation about the basis on which decisions have been taken. We need to give close attention to the factors have buttressed the claims of science to reduce uncertainty (Brown 1989). First, many environmental problems involve high levels of human anxiety, associated with risks to human health, which appear to increase with the expansion of our knowledge. Second, since environmental science is an essential part of the solution to environmental problems, it follows that improved regulation, and greater technical expertise in addressing environmental problems, also serve to demonstrate the shortcomings in the application of science (Read 1994). Global environmental problems, in particular, such as ozone depletion and global warming, are not only complex in terms of their chemistry or biology, they are also apparently inaccessible to technical "fixes". Unlike the administration of antibiotics, or the inoculation of patients against the risk of contracting life-threatening diseases, changes in behaviour induced by environmental awareness, such as the purchase of aerosols free from CFCs, or the use of lead-free petrol, do not ensure environmental safety. We know more but we are able to do less. In addition, there is evidence from the growth of campaigning groups, around environmental issues, that the gap between "lay" perceptions of the environment and "expert" opinion, is actually widening (Yearley 1991). Faced with a barrage of increasingly complicated, and contradictory, information about environmental risks, the layperson is likely to question the authority of science, and the confidence politicians place in scientists. It soon becomes clear that the "critical thresholds" which are endorsed by political leaders and expert witnesses, are themselves political compromises, framed to manage public apprehension. The more that the official environmental discourse may seek to dampen public apprehensions, the more it becomes clear that "certainty" does not prevail. Public anxiety is only part of the picture. If the environment exists in the specialist knowledge that we possess about it, there is less for the "non-expert" to regard as their area of competence. This effects the "ownership" of environmental issues. Research from developing countries has shown that the growth of specialist knowledge is related to the
71
growth of non-specialist "ignorance", and this observation is equally appropriate in the North. Doubts about the degrees of certainty associated with formal scientific knowledge are matched by alternative, holistic models of human relations with nature, which interpret "facts" differently, and which seek new ways of understanding, rather than an enlarged databank of information. It is clear that different values are held by different groups of people. Some groups, at least, are using the opportunity presented by scientific uncertainty to re-evaluate their values (Thompson et al. 1986). The two dimensions of uncertainty which deserve particular attention are the spatial and the temporal. We are accustomed to make most decisions on the basis of present time, and any future consequences play a smaller part in our calculations than immediate consequences. However, environmental choices often bear little relationship to the decision-making and dislocation of everyday life. They require an imaginative leap into the future, to the next generation or subsequent generations. The timescale of ecological processes, particularly those operating at the global level, makes it imperative that we attach weight to the future, and that what economists call the rates of discount reflect this importance. Many environmental changes are also "systemic" in the sense that they can only really be understood through the way that systems change. Biodiversity is a case in point, since threats to individual species carry serious implications for ecosystems as a whole. The loss of one plant variety from a local ecosystem can jeopardise the survival of animal populations which are dependent upon it. Since the timescale of ecological processes bears so little relationship to everyday decision-making, it is important that we attach value to the loss of flexibility and variety in future environments. The spatial dimensions of the environment are also important in any consideration of values. The environmental consequences of human activity are often experienced at several removes, not only in time but in space. The economic development of the industrialised countries, their diets and lifestyles, have been responsible for transforming the environments of developing countries located thousands of miles away. The "ecological footprints" left by industrialisation, and consumer wants in the North, are not easily erased. This serves as a reminder that while in the North we tend to regard the protection of nature as a fundamental ingredient of environmentalism, in the developing countries environmental issues often present themselves in terms of protection from nature. Perhaps we need to consider whether the driving forces behind global environmental change, including industrial growth and consumerism, increase the environmental security of people in the South or seriously threaten it? The values generated in our society carry implications for the environment that are only dimly perceived most of the time. Consumerism implies a commitment to aspiration, to "improve" one's lifestyle. A desire to own the fruits of technoscience is apparently within everyone's grasp. At the same time we are concerned should the environmental costs of progress arrive on our doorstep. The response of local communities to environmental problems - "Not In My Back Yard" (NIMBY) - is a product of contemporary lifestyles in the industrialised countries, every bit as much as concern about protecting the whale or tropical forests. The process through which we are removed from the consequences of our actions, sometimes called "distanciation", is illustrated in a number of ways. Among them is the way
72 the enhanced greenhouse effect, through its impact on climate, is likely to increase perturbations in weather conditions, especially in the tropics, with increased occurrence of freak storms, drought and sea-level rise. The measures necessary to avert these risks are not difficult to determine, but the political will to act confronts widespread apathy and indifference.
Economic Values and Environmental Management Neo-Classical economics developed through making a number of assumptions about the environment. Natural resources such as water, soil and clean air, were often depicted as "free goods", meaning that they were available freely; they did not involve a charge. However, it is clear that environmental "goods" are qualitatively different, in significant ways, from goods for which we do pay a charge. Clean water and air, unpolluted soils, are not available "freely" in nature once human beings have had a hand in economic development. Environmental economics has been forced to consider the costs of cleaning up the environment, and of conserving natural resources to ensure their supply (Winpenny 1991). Ecological economics is also concerned with wider questions which have eluded most economists since the nineteenth century. Attempts are being made to distinguish between "wants" and "needs", and between the way our needs are satisfied, for example through more consumer goods, and the needs themselves. The conditions under which goods and services are produced is a key question. At the same time the social and ecological consequences of their production is a concern to Green economists. Many argue that we should develop methodologies for arriving at "utilisation values", that is, the value of goods and services throughout their lifetime. Such values would include the cost of waste disposal, the benefits from reuse or recycling, and the pollution or resource degradation associated with the use of raw materials in their manufacture. Within environmental economics there are broadly three camps. The first camp argues that there is nothing to prevent us from placing economic value on the environment. Using prices and market instruments we can assign the real costs of environmental degradation. What is required is further refinement of methodologies such as contingent valuation, which enable us to approximate individual preferences for environmental goods and services. In the view of these economists the "logic" of economic rationality can be used to manage the "randomness" of nature (Pearce 1993). A second camp takes a very different view. They argue that we cannot place a value on the environment, like that for human-made goods. Natural capital, in their view, is qualitatively different from human-made capital, and should be treated as qualitatively different. Following Oscar Wilde's famous aphorism, we are in danger of knowing "..the price of everything and the value of nothing". In the view of radical ecologists the logic of nature cannot be geared to the randomness of the market. As human beings we are part of nature, and cannot subject nature to our laws as we are subjected to natural laws (Ekins and Max Neef 1992). Between these apparently irreconcilable positions are others which probably attract considerably more support than is immediately evident. Some institutional economists like Jacobs argue that we can, and should, develop economic methodologies which, in effect,
73 "value" nature (Jacobs 1991). However, we should also recognise that Neo-Classical economics is itself a social construction, and its development reflects the preoccupations of industrial capitalism. We can develop methodological tools which place more, or less, emphasis on the importance of market forces. If we wish we can propose guidelines, indicators for "sustainability planning", which allow radical shifts in economic policy and thinking. Unlike some radical political ecologists, people in this third camp, propose that we intervene and regulate the environment, essentially to meet human purposes rather than follow imperatives in nature itself. They also agree that we will all be the richer if we examine the underlying social commitments which govern our lives, the maintenance of our present "lifestyles" and patterns of consumption. However, unlike Deep Ecologists, for example, mainstream environmental economists believe changes in human behaviour can be induced through policy instruments and interventions. It is also important to distinguish between analytical positions like those found within environmental economics or the sociology of science, and the value commitments of a society. To some extent analytical positions can play the part of, or even displace, other systems of values. We have only to reflect on the central role which Neo-Liberal economics has attributed to the "choices" of individuals in the market-place, or what Huber has called "ecological modernisation", through which business has sought to incorporate environmental costs in its range of products and services (Mol and Spaargaren 1990). These are examples of the close relationship between the values of the wider society and those that govern environmental questions. It would be surprising if core values such as "individualism", "private property", "choice" and "independence", the political values which govern everyday actions and desires, were unrelated to the way in which we interact with our environment. However, it is much more difficult to specify the nature of this interaction, the variables at work, and the lines of causation. These positions themselves reflect a modernist discourse that still sees the human subject as universal and all knowing (Redclift 1993). They do not address the fallibility of human beings, most notably in our inability to reflect upon the increased knowledge we possess about the wider universe. If science is continually widening the frontiers of what we know, it is also revealing the extent of what we do not know. We are, in fact, seeking to interpret what we do not know in terms of what we know. At the very least this is a hazardous procedure. Global environmental agreements and human values
The international agreements which were signed at the Earth Summit in 1992 give expression to environmental values, many of them widely shared. At the same time these agreements, if they are to succeed in changing the way we manage our resources globally, require that we pay more than lip-service to the values we espouse. The institutional apparatus established at Rio de Janeiro, as much as the agreements themselves, provides evidence of the difficulty in providing a consensus for global environmental management (Thomas 1993).
74 It is clear that values are implicit in what we take for granted from natural systems, as well as what we propose to do to protect these systems. At the same time, the process of economic development enshrines a different set of values. The Brundtland Commission, which reported in 1987, sought to enlarge this debate, and to make our value preferences more explicit (Brundtland 1987). Unlike the reports of the I.P.C.C. it did not purport to be a value-free document, but freely admitted to political objectives, many of which were subsequently incorporated in Agenda Twenty One. The idea of "sustainable development" as a way of informing policy cannot be divorced from the attempt to integrate quite different systems of values. Much of the confusion accompanying the discussion of sustainable development, and the drawing up of international agreements, stems from the relationship between our values and our knowledge about global environmental problems. The scientific controversy accompanying global climate change, and the deliberations of the I.P.P.C., has suggested that increasing ou/" knowledge about future climate change, and its impacts, will enable us to adopt more appropriate values, emphasising long-run sustainability over short-run economic gain. However, the evidence for this assumption is weak. Rather, it might be asserted that until we address the environmental problems associated with our current values, there is little likelihood that we will be able to make much use of the knowledge which is accumulating about the global environment. As Tickell argues, "... our ignorance of species and ecosystems is profound, not only of present ecosystems and species, but of their future uses and services. It is an understatement to refer to this level of ignorance as mere uncertainty" (Tickell 1994, 4). The major provisions of the Framework Convention on Climate change mark an important watershed in international agreements to protect the environment. The Convention established the principle that action to start addressing the problems of climate change should not wait upon the full resolution of scientific uncertainties. It also asserts that developed countries should take the lead in introducing measures to reduce the threat of global warming. Finally, it endorses the idea that developed countries should compensate the developing countries for any additional costs that they might incur in taking measures under the Convention. Superficially, at least, the goal of sustainable development is one publicly espoused by most governments. Most of the governments in the North have signed, and in some cases ratified, agreements which endorse a set of principles and values that place global sustainability above vested interests and short-term economic advantage. However, at a more profound level there is little agreement about the "values" which need to inform sustainable development. The "natural services" provided by the environment are acknowledged, but the assumption that they will continue to be provided, is still made. Real environmental costs and benefits are scarcely acknowledged in the day-to-day economic management that determines their use. Similarly, rather than using the precautionary principle to help provide for more flexible responses to uncertainty, most policy is still formulated against unsustainable assumptions, about population, military expenditure and economic growth. Global inequalities, particularly between North and South, are part of the "taken-for-granted" assumptions behind international agreements in "non-environmental" areas such as the liberalisation of trade. Inequalities within developing countries, we are regularly told, are part of the price that such countries pay for the absence of "development". However, evidence that economic growth has particularly
75 adverse effects on the Newly Industrialising Countries' environments, should lead us to question whether successes in market economies really are a prerequisite for better environmental management in these countries. This paper has argued that the options available to us over global climate change need to be placed in their context; our societies. Environmental consciousness is indelibly linked to social and political unease; it does not spring from the physical 'environment' alone. It follows that measures to combat possible climate change need to be located within socially meaningful categories, and we need to develop a better understanding of the reasons people assume social responsibilities towards the environment in the first place.
REFERENCES Beck, U. (1992). Risk Society, Sage, London. Benton, T., and Redclift, M.R. (1994). 'Introduction' in Social Theory and The Global Environment, Routledge, London. Boehmer-Christiansen, S. (1993). 'Scientific consensus and climate change: the codification of a global research agenda'. Energy and Environment 4(4). Brown, J. (1989)(edited). Environmental Threats, Belhaven Press, London. Brundtland (1987). World Commission on Environment and Development, Our Common Future, Oxford University Press. Ekins, P., and Max-Neef, M. (1992)(edited). Real-Life Economics, Routledge, London. Featherstone, M. (1990)(edited). Global Culture, Sage, London. Grubb, M. (1993). The Earth Summit Agreements: a Guide and Assessment, Earthscan/Royal Institute of International Affairs, London. Jacobs, M. (1991). The Green Economy, Pluto Press, London. King, A. (1991)(edited). Culture, Globalization and the World System, Macmillan, London. Mol, A.P.J., and Spaargaren, G. (1990). 'Sociology, Environment and Modernity', Paper presented to International Sociological Association Conference, Madrid. Newby, H. (1993). 'Global Environmental Change and the Social Sciences: Retrospect and Prospect' Economic and Social Research Council, Swindon. Pearce, David (1993). Blueprint Three - measuring sustainable development, Earthscan, London. Read, P. (1994). Responding to Global Warming, Zed Books, London. Redclift, M.R. (1993), "Sustainable Development: Needs, Values, Rights', Environmental Values 2(1), Spring. Thomas, C. (1993)(edited). 'Rio: Unravelling the Consequences', Environmental Politics (Special Issue) 2(4), Winter. Thompson, M., Warburton, M., and Hatley, T. (1986). Uncertainty on a Himalayan Scale, Milton Ash Edition, London. Tickell, C. (1994). 'Socio-political perspectives on Biodiversity', Green College, Oxford. Winpenny, J.T. (1991). Values for the Environment, HMSO, London. Yearley, S. (1991). The Green Case, Harper Collins, London.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
77
National and International Economic Instruments for Climate Change Policy J. B. Opschoor Department of Spatial Economics, Vrije Universiteit, De Boelelaan 1105, 1081HV Amsterdam, Netherlands
Abstract Policy instruments addressing sources and sinks of climate change can be applied at two levels: the national and the international level. The international focus on costeffective policies points at the need to evaluate the merits of "economic" or incentive based instruments in particular. An overview is given of experiences with such instruments at the national level and prospects for applying similar instruments internationally are explored. Although a tradable system of (net) emissions quota may have high potential benefits, its acceptance meets with relatively large problems. Carbon taxes or charges may be a better alternative. An important recommendation is that countries and regions could move ahead in locally optimal but partial ways, to be extended and harmonised after some more trial-and-error. Experimentation with joint implementation is a promising avenue for the long run.
1.
INTRODUCTION
Global environmental issues arise when specific forms of environmental degradation turn into a problem in which a large number of countries have a stake" they may be the results of activities in specific individual countries but often the welfare of other countries is adversely affected, potentially or factually. In the absence of a global environmental regulator international coordination through voluntary agreements is required. Such co-ordination may extend to the principles to be collectively applied, targets with respect to environmental quality to be achieved, individual countries' shares in these targets, and instruments to be applied. On the latter issue, a distinction can be made between instruments to be applied at the international level, and instruments to be applied within national jurisdictions. Below, we shall look at possible instruments at both levels and their relative merits and demerits. We will look at some more promising instruments at the international level in particular: a carbon/energy charge and tradable emission quota.
2. INTERNATIONAL INSTRUMENTS
2.1
ENVIRONMENTAL
POLICIES:
PRINCIPLES
AND
International Environmental Policies: General Principles
Transboundary environmental issues are the results of actions in individual countries that adversely affect the welfare of other countries. In the case of significant transboundary externalities, one possible approach is that some countries undertake
78 unilateral action (see Hoel 1991 for some caveats in this respect), but normally international coordination through agreements is more effective. This requires decisions by countries to become part of such agreements, which may imply a weighing of costs and benefits by the countries concerned. Individual countries might opt for not entering such agreements and to become free riders: then they would not share in the costs of measures whilst enjoying the benefits of the collective efforts of the other countries. In the case of unidirectional externalities such free riding behaviour cannot be countered by others polluting more in order to punish the free rider (OECD 1991a); hence other forms of establishing reciprocity would have to come in force. Even in cases of reciprocal externalities, countries might decide to not participate in a collective effort, if their valuation of environmental quality is relatively low. In such cases, incentives may be needed to make such countries engage in the agreement, such as side payments (compensations in one way or another). Also, efficiency arguments might be raised in favour of some countries financing environmental measures in other countries. Generally, therefore, some sort of international cooperation is needed if global issues such as climate change are to be addressed effectively and this may include a mechanism for burden sharing. The main perimeters of international environmental policy in the decades ahead, have been set by the principles as laid down in the Rio Declaration on Environment and Development (UN 1992)and specific international agreements such as the Framework Convention on Climate Change. The principles of the Rio Declaration most relevant to the development of national and international policy instruments, are: - t h e souvereign right of states to exploit their own resources and the responsibility to ensure that activities within their jurisdiction do not cause damage to the environment of other states or areas beyond the limits of national jurisdiction (Principle 2); -the duty to cooperate (with common, but differentiated responsibilities) to conserve, protect and restore the health and integrity of the Earth's ecosystems (Principle 7); - the application of "the precautionary approach": where serious or irreversible damage may occur, cost-effective measures to prevent environmental degradation must be taken (Principle 15); - the promotion of the internalisation of environmental costs and the use of economic instruments, taking into account the approach that the polluter should, in principle, bear the cost of pollution (Principle 16). The Framework Convention on Climate Change aims at stabilising concentrations of greenhouse gases in the atmosphere at levels that would prevent dangerous anthropogenic interference with the climate system are prevented, in a time frame that allows ecosystems to adapt and economic development to continue sustainably (Art. 2). Developed countries have to take the lead (cf. Principle 7) and proceed (cf. Principe 1 5 ) w i t h cost-effective measures and policies so that they generate global advantages at least cost (Article 3); efforts to abate climate change may be implemented jointly by parties to the treaty (ibid.).
2.2
E n v i r o n m e n t a l Policy I n s t r u m e n t s
So far, agreements
and conventions
have often
sought consensus
about
79 national efforts in terms of targets for emission reductions (e.g. the SO2-"clubs", CFCs). Given such agreed targets it would be a matter for national policies to decide how these targets would translate into national policies and instruments. However, from an economic perspective such agreed and fixed national targets might be a costly way of achieving overall environmental quality (Hoeller et al 1992), especially when (marginal) abatement costs differ between countries or regions. We thus have two levels at which climate change policy instruments should be considered: the national and the international. These shall be discussed in Sections 3 and 4; here we proceed with a general introduction to environmental policy instruments. Environmental policy can make use of two basic strategies (Fig. 1, routes a and b). Firstly, public projects and programmes can be set up that aim at preventing, compensating and eliminating environmental degradation or at providing substitutes for traditional behavioural patterns, such as: collective treatment facilities, environmental sanitation and (re)construction programmes, afforestation, etc.; the costs of such programmes would have to be borne by the relevant agents (governments, firms, individuals) according to some mechanism for burden sharing (e.g., from public funds, through levies and charges, etc.). Secondly, the decision making process may be influenced at the micro level through policy intervention. The second strategy is discussed in more detail below. Rational decision makers will base their decisions about their activities on a comparison of the various options open to them. They will compare the costs and benefits of these options, defined as all (dis)advantages relevant to the decision maker as aggregated by his/her individual weighing system. In such a situation, decisions can basically be influenced in three different ways (Fig. 1, left hand side): 1) alteration of the set of options open to agents; 2) alteration of the cost and/or benefits relevant to agents; 3) alteration of the priorities and significance agents attach to environmental change (i.e. changing the structure of agents' costs and benefits). Route 1) is referred to as direct regulation, defined as: institutional measures aimed at directly influencing the environmental performance of polluters by regulating processes or products used, by abandoning or limiting the discharge of certain pollutants, and/or by restricting activities to certain times, areas etc. The polluter is left no choice: he has to comply, or face penalties in judicial and administrative procedures. Route 2)entails economic incentives or market stimuli. The motivation relied upon here is that if environmentally more appropriate behaviour is made more rewarding in the eyes of the agent involved, then attitudes and behaviour will 'automatically' shift in favour of these socially more desirable alternatives. Options can be made more or less (financially or economically) attractive by applying charges or levies, granting subsidies, implementing tax differentiation etc. Such instruments will be referred to below as economic instruments. In this way environmental concerns can in a certain sense be 'internalised' by altering the agent's context rather than the agent's value structure or preferences. Route 3) includes approaches such as: education, information extension, training, but also: social pressure, negotiation and other forms of 'moral suasion' leading to a change of perceptions and priorities within the agent's decision framework. It aims at full 'internalisation' within the preference structure of the agent.
80 In this paper we will concentrate on economic instruments, i.e. route 2. Basically, one may distinguish the following categories of economic instruments: 1) charges, 2) subsidies, 3) deposit-refund systems, 4) market creation, and 5) financial enforcement incentives (OECD 1989). Charges may, to some extent, be considered as a "price" to be paid for pollution. Polluters have to pay for their implicit claim on environmental "services", which thereby enters at least in some part into private cost-benefit calculations. There are various types of charges, including charges on emissions (e.g. CO2), and charges on products (e.g. fossil fuels, CFCs, cars). In deposit-refund systems a surcharge is laid on the price of potentially polluting products. When pollution is avoided by returning these products or its residuals to a collection system, a refund of the surcharge follows. Markets can be created where actors might buy "rights" for actual or potential pollution or where they can sell "pollution rights" or their process residuals (for reuse or recycling). In emissions trading, dischargers operate under some multi-source emission limit and trade is allowed in permits adding up to that limit. Given certain quota under agreed emission reductions, also countries could trade. Criteria for selecting specific instruments relate to (Opschoor and Turner 1994; OECD 1994): (environmental) effectiveness, economic efficiency, and (social and political) acceptability. Amongst the acceptability criteria distributional considerations are especially important, particularly in the context of global environmental issues. Economists have argued that economic instruments are to be preferred to especially direct regulation, as they tend to invoke least cost technical and economic responses; moreover, it is believed by many of them to also generate a more effective incentive for technocal innovation. Using simple models, it can be shown that a pro rata agreed emissions reduction approach is less efficient in economic terms, than applying either charges or trade in permits to achieve the same reduction (e.g. Barrett in OECD 1991b) - but this is theory: it ignores much of the lack of data and knowlegde on effectiveness and efficiency, and overlooks the acceptability problem and all that it stands for (Opschoor and Turner 1994). Environmental policy instruments typically come in "cocktails" or "mixes" of the pure elements described above. Thus, a carbon trading system is a combination of a regulatory measure (i.e. the total volume accepted) and an economc one (trading quota within that maximum). Economic instruments do have a part to play in tackling global environmental issues. The wider the geographical extent within which they are applied, the larger the efficiency and flexibility-benefits of applying these instruments are likely to be. However, it is also fair to observe that economic instruments meet with a number of problems. In fact, one can discern a set of different dilemmas: Firstly, the 'administrator's'-dilemma: direct regulation may be inefficient but yet efficient economic alternatives may be inacceptable to the policy makers for cultural or political reasons. Charges and trading schemes will then not be used where this could be practical and soacially advantageous. Secondly, the 'second best' -dilemma : a first best approach making use of economic instruments may, in a distorted world with government and market failures, create more inefficiency and be less effective than introducing second best instruments such as direct and uniform regulations. Thirdly, the 'revenue'-dilemma: returning the revenues of charges and levies to
81 the sector that they came from (e.g. for subsidising CO2-reduction by fuel shifts) enhances acceptability and effectiveness but is socially inefficient. This often results in discussions between public finance and fiscal experts on the one hand, and environmental managers on the other. Finally, there is the 'leverage point'-dilemma: policy instruments may affect economic agents and processes in different stages of the product life cycle. Subtle and tailor-made emissions charges may be optimally effective without loss of environmental potentials for economic use, but at the same time they may be administratively costly. However, administratively easier but more clumsy inputs charges (e.g. on energy or Chlorine) may be environmentally and economically inefficient.
2.3
Climate Change and Climate Change Policy Instruments Climate change, CC-policies and Cooperation
The overall objective of the Climate Convention has been given in Par. 2.1. It is assumed here that under a "Business as Ususal"-scenario of global economic development, GHG-concentrations would rise to levels beyond those aimed at within the Climate Convention. This then would imply levels of damage to ecosystems and to development that are to be avoided at least partially. One of the principles of the Convention is to deal with climate change in a cost-effective way; that is, to achieve objectives at least social costs. The costs of climate change include avoidance or abatement costs (cost of emissions reduction and sink enhancement) and accommodation costs (costs of coping with residual GHC-concentration rise). Both abatement and accommodation costs include direct costs (i.e. the cost of measures and of policies aimed at reducing climate change or coping with it) and indirect costs (the net socio-economic effects elsewhere in society as a spinoff of these measures and policies). One may wonder whether the level of admissible GHG-concentration could not also be determined in a least-cost approach, i.e. by minimising the sum of abatement and accommodation costs associated with alternative levels of abatement (this is one of the hot issues in the IPCC Working Group III at the time of this Conference). This would imply that all relevant elements of accommodation costs can be measured with a reasonable degree of accuracy - including damage costs due to exposure to concentrations- over a very long time horizon. However, according to many researchers (including the present author) the uncertainties inherent in such calculations justify a more political approach, where the overall net emission levels are to be below some negotiated maximum path, satisfying conditions such as those of Article 2 of the Convention. Within such an overall maximum, various different patterns of national net emissions (i.e. emissions corrected for sink enhancement) could be accepted. This allows for negotiation over, or trade in emissions quota, etc., to achieve cost-effectiveness.
Climate Policy-related Economic Instruments In terms of environmental impact, there are several ways in which environmental policies could operate technically in order to become or remain compatible with some environmental target or standard. One is, to move or relocate sources (i.e.
82 activites) to areas where they will contribute less to the global issue at stake. In cases such as the climate issue, relocation does not provide a real solution. Environmental degradation can also be avoided by reducing source strengths through reduction of activity levels, diffusion of existing cleaner technology and/or innovation. A final strategy is that of circumventing the environmental impact by enhancing the environment's capacity to absorb or otherwise deal with the pollution. Sink enhancement e.g. through afforestation is one relatively inexpensive method of dealing with notably carbon dioxide emissions -the major greenhouse forcing substance. In order to trigger economic agents responsible for GHG-sources and sinks to move towards either one of these technical options, they will have to be commanded or convinced, and especially in the global context, economic approaches to this may warrant attention. Prospective modelling work suggests that the impacts of instruments (charges and permits) on revenues and income transfers between countries can be very important (Hoeller et al 1992). In the case of (transboundary and) global environmental problems a special issue arises out of the combination of the criteria of effectiveness and efficiency. It may well be more effective to allocate a certain amount of money to financing environmental activities elsewhere ("joint implementation"). Several types of economic instruments for addressing global environmental issues have been proposed: i) Emissions charges or products charges (e.g. taxes on energy use) or combinations such as an international carbon/energy charge; ii) Global permitting systems for emissions or for using a global environmental resource, allowing for trade betweeen countries in such permits or in emissions offsets (cf. the Montreal Protocol); iii) Sanctions against free riding or non-compliance in relation to environmental treaties, including trade sanctions; iv) Motivation to participate in agreements by compensation payments and by socalled "joint implementation' programmes; v) Deposit-refund systems. Below, we shall first look at national policy instruments (section 3) and then move on to options for climate policy instruments at the international level (section 4).
3.
REVIEW OF NATIONAL POLICY INSTRUMENTS Currently, most countries operate a range of environmental policy instruments including economic instruments. In this section climate change related economic instruments as currently in use or under consideration in OECD countries, will be briefly reviewed (after OECD 1994): charges and trading schemes. 3.1
Charges Charges on processes and products that generate pollutants contributing to global environmental problems could make these inherently less attractive. Several existing charges on products relate to global environmental problems (e.g. CFCs, fuelrelated CO2). Product charges have several advantages. First, the administrative framework for collecting the charge may be relatively easy to conceive, or may -in some cases, already function in some parts of the world: in most countries and for many fuels, systems already exist for taxing them. Second, changing the level of a charge or
83 tax is a relatively simple intervention, compared with altering other instruments (such as tradeable permits). In the area of charges on emissions most air pollution charges relate to non-GHG emissions, especially of acidifying substances (US, Canada, France, Japan, Scandinavia, Portugal). One succesful example is the Swedish NO x-charge on heat and power producers: the charge has speeded up compliance to sharper emission standards to be imposed in 1995. The accelerating mechanism was the rechanneling of the charge's revenues to the producers according to their final energy production; thus, heavy emitters subsidised clean energy producers and this provided an incentive for rapid innovation. More than half of the OECD countries have differentiation in car sales tax rates or annual vehicle tax, according to the levels of emissions (based on e.g. car weight, catalytic converter, emission standards compliance). Several countries have explicit and implicit carbon taxes with an intended incentive impact. Carbon taxes are now applied in the Scandinavian countries (including Denmark), Italy and the Netherlands, but only in the Norwegian and Swedish case are they significant enough to have an incentive impact. All OECD countries have energy taxes, though at different levels and using different operational systems. Some countries have or are considering effective energy/carbon charges, and others look at raising energy tax levels. Charges on Ozone depleting chemicals exist in Australia, Austria, Denmark and USA, possibly with incentive impacts in the latter two countries.
Domestic Policy Impacts National charges will have nationally relevant policy impacts, notably in the fields of income distribution, sectoral activity levels, public finance, macroeconomic policies (Piacentino 1994). Energy taxes are applied to a commodity with fairly low short term price elasticities. Hence, in order to achieve a given quantity impact on its consumption, the price rises must be high. In order to achieve a levelling off of CO2 emissions in 2020 at the 1990 level, calculated emission taxes vary between $30 to $150 per ton of carbon. This implies high tax revenues and raises the question as to the distributional impacts of such a taxation policy. There is a case to be made for using these revenues as part of the overall public finance, which, in the case of an assumed fiscal neutrality would imply that other taxes could be reduced accordingly. In addition to the fiscal aspect, the income effects of the charge are to be considered and perhaps compensated. Emissions reduction efforts may induce significant feedbacks to the economic process. Research has been done to explore the economic costs of e.g. reducing carbon emissions in various countries. Achieving large reductions of energy-related CO2-emissions may depress growth rates of world GDP by .2 percentage points, but this can nevertheless imply reductions in the long run levels of global GDP of some 3-8% (between 2025 and 2050); some national models (e.g. for Norway, Netherlands, Sweden) predict higher growth rate depressions with higher GDP-impacts earlier on (2000-2010) at much lower levels of reduction (Hoeller et al 1992). Moreover, charges could lead to shifts in the industrial structure. In one country's exploration of the economic impacts of unilateral and joint action via carbon/energy charges (without exemptions), national (and even some collective)
84 sectoral implications could be dramatic (CPB 1992); in the very energy-intensive sectors there might be relocation of industries to other countries or regions. Calculated reductions in energy use might result from reduced or replaced activity, rather than from fuel shifts, energy conservation or new (leaner) technology (ibid.). Unless agreements could be reached with all important trading partners, countries will not easily impose extra costs on their industries through high energy or carbon taxes. To protect their international competitive positions, countries usually consider tax schemes only if associated with substantive exemptions for domestic energy-intensive industries operating on international markets. In the context of regionally applied charges, one alternative to such exemptions could be: mitigating measures on transboundary transactions correcting for cost differentials due to nonparticipation in the charge scheme. The proposals for an EC CO2/energy tax appears to favour an approach based on such exemptions. Providing exemptions and mitigation may easily lead to different consequences in terms of internal and external support for policies aimed at global environmental issues and little is known of these and other indirect effects of such corrective measures.
3.2
Other Policy Instruments
Apart from charges (or their less efficient and effective reverse: subsidies), countries may use instruments such as trading schemes and of deposit-refund systems. On national trading schemes, there is very little experience outside the US and it has been reviewed elsewhere (OECD 1994). Tradable permit systems exist in the US, Canada, Australia and Germany; the Canada and US schemes include trading in CFC-quota, air pollution emissions, and some car and fuel-related emissions. There are no direct climate change related trading schemes. An extensive literature exists on the efficiency of trading and its prospects for air pollution abatement and prevention (see, e.g. NAPA 1994; Klaassen and Pearce 1994). In theory emissions trading is as efficient as charges, and there may be less uncertainty as to its environmental effectiveness than would be the case with charges. We shall come back to this below, in Section 4 on international instruments. A second approach could be that of deposit-refund systems in relation to greenhouse gases. Deposit-refund systems would imply putting a charge or tax on bringing a unit of e.g. CO2 into the atmosphere whilst reimbursing for removal, disposal or sequestering of a unit of CO2. CO2 removal and fixation might enable economies to seek least cost options for GHG-reductions by comparing the costs of combustion reductions with those of fixation, etc. Compensiations for CO2-fixation could thus be combined with a CO2 charge into a deposit-refund analogue. This could be done both nationally (e.g. when fixation in new forests is financially facilitated) and internationally (see e.g. Huppes et al. 1993).
4.
CLIMATE POLICY INSTRUMENTS: INTERNATIONAL OPTIONS
As in the case of national policy instrumems, international instrumems may aim at an incentive effect or they could be intended to raise financial resources to
85
undertake other activities including emission abatement. In fact, the revenue raising character of some of these instruments might appear as an advantage given the large amounts of financial resources needed to address adequately the global problems and the distributional aspects that would be encountered in attempts to obtain global commitments. One argument against this is that if prices are correctly reflecting environmental costs in consequence of a "proper" tax level, then earmarking would lead to a distortion of the optimal allocation.
4.1
Emission charges
Especially in the case of charges there are grounds for engaging in an international dialogue on account of possible repercussions on international trade and investment. In relation to global warming, taxation may be based on energy content or on Carbon content (or some combination of these two). All OECD countries have energy taxes, though at different levels. Some countries have or are considering effective energy/carbon charges, and others look at raising energy tax levels. Carbon taxes are now applied in the Scandinavian countries (including Denmark) and the Netherlands. The European Community is considering a carbon/energy tax on energy raising primary energy prices initially with $3, going up to $10 per barrel of oil equivalent some 7 years after the scheme becomes operational. This might result in a reduction of CO2 emissions in 2000 in the order of 6-7% (DRI as quoted by Carraro and Siniscalco 1993). The charges/tax option is an appropriate one given the wish to improve market signals and to raise public awareness. Due to the elasticities involved however, they may not be the most effective way to modify behaviour in the short run. Macro economic impacts or the fear for such impacts, might make individual countries reluctant to move ahead of others in introducing such charges, or in making them high enough to have an incentive impact (see par. 3.2). Especially in the case of charges there are very solid grounds for engaging in an international dialogue on them, explicitly taking into account impacts on international trade and investment, on relocation etc., as resulting from a substantial tax.
4.2
Tradeable Permits Systems (TPS)
The main idea of tradeable permits is achieve an environmental target at least cost to society, by setting an emissions reduction target, distributing or auctioning permits up to the total set by the reduction target, and by allowing trade in these permits. Compared with charges, permit trading has the advantage of a more certain result in terms of emissions reductions: the amount of permits issued sees to that, if it is enforced. These permits can be subjected to market forces: there can be more of them for sale if further technological innovation results in cleaner technologies; demand -if mobilised- will competetively force prices down to their appropriate level. There must be an information system (with information about potential buyers and sellers) and an auction procedure. Also, there must be some agreed initial endowment and this is one of the main difficulties with a TPS. In addition, there are operational conditions to be met e.g. on the definition of the market, the 'size' of the market in terms of numbers of buyers and sellers, of real possibilities for trade in terms of actual or potential cost differentials, etc. (see e.g., UNCTAD 1992). Several of these issues will be briefly reviewed.
86 The commodity in which trading is to occur, has to be defined clearly. For obvious economic reasons the flexibility-and hence the potential cost-effectiveness- of the scheme would be enhanced if apart from emissions also sink-enhancement strategies could be incorporated, and if the whole range of GHgasses (i.e., not only CO2) could be considered. Even in the case of greenhouse forcing alone, there are several options (e.g. fossil CO2 emissions, ibid. plus C-sequestration by plantation, net CO2-emission, equivalent CO2 including other gases etc.). The choice of definition is likely to reflect the performance to be expected on a number of criteria, such as compatibility with sustainable development, efficiency, etc. There seems to be a growing consensus that limiting trade to emissions of energy-related carbon dioxide is the most feasible initial option (Swart, in OECD1992a), with that of including other CO2sources and sinks as a very promising alternative especially as it may do more justice to the claims of developing countries (Agarwal and Narain 1991). With limited numbers of market actors there is the risk of parties being capable of manipulating permit prices, or to influence prices in related commodity markets (Tietenberg, OECD 1992a) in order to affect the distribution of rents. According to Roland (in OECD 1992a), this risk is relatively low (see also Bohm 1991). Yet, there is another distributional issue related to market power: with unequal purchasing power and in a buyers' market, the possession of permits might concentrate in the portfolios of a few rich nations. One solution to the problem of market power confusing the performance of trading schemes would be to limit the period during which rights to emission remain valid: under such circumstances hoarding and accumulating emission rights would be much less economically rewarding (Boorsma et al 1988, OECD 1994). The question of how to distribute initial rights is a crucial one in obtaining international support for any large scale trading system. This initial endowment or distribution could be based on e.g. current emissions level, past responsibility, equality of effort, GDP, population, etc. The initial allocation will have to be a compromise, based e,g. on both a per capita allocation and current emission levels.
4.3
Deposit Refund Systems
Countries, apart from being sources of substances or other interventions giving rise to global environmental problems, may also engage in activities that enlarge the environment's capacity to absorb (or otherwise handle) human activities. In the case of climate change, sink enhancement e.g. through agricultural and reafforestation policies is a case in point. Economic instruments may support such activities. In the case of charges: if only net emissions are charged, or emissions corrected for the annual impact in terms of sink enhancement, then the latter would be economically attractive; and so would, of course, innovation. A similar argument holds for allowing trade in credits built up by sink enhancement or technological innovation. However, the issue of the measurement of the contribution to resolving global environmental problems through sink-enhancement is such an intricate one, that it might preclude advancement in international environmental agreements, unless put aside until better monitoring is possible. In principle, however, if countries engage in activities that enhance their environments' capacities to absorb or buffer global pollutants, then they could be compensated for that within a trading system if such activities would yield additional permits or credits to them that could subsequently be offered on the
87 permits market. Future extensons of the Climate convention might include provisions for compensating countries for carbon removal and fixation, presumably certified, or by allowing these to be credited to in the form of equivalent additions to their allowed emissions.
4.4
Joint Implementation
A step towards a full permit trading situation, may be that of "joint implementation". Under a regime of joint implementation, countries might find it in their interest to participate in an agreement and to take on emissions reductions responsibilities, whilst the financial consequences would be shared with or adopted by other participants. From Western Europe, acid deposition offsets are being sought and financed in Central and Eastern Europe as an alternative to carry out costly abatement programmes in e.g. the Netherlands (1993). The Climate Convention allows certain parties (notably OECD countries and economies in transition) to implement jointly with other parties (including developing countries) to the Convention, of activities to reduce GHG-emissions and enhance sinks ("joint implementation", Article 4.2.a). For countries that have accepted emissions targets, joint implementation would effectively allow more efficient emissions reduction within the total target. If joint implementation is undertaken with countries that are not (yet) committed to any target, then the effectiveness becomes more dubious, whilst no doubt average (and most likely total) emission abatement costs would be lower. From the side of the developing countries a risk of joint implementation would be, that it might be a disincentive to developed countries to change their patterns of production and consumption. Special attention should be given to the conditions under which such schemes are appropriate (see e.g. Kuik et al 1994). Criteria to be applied include: (i) additionality of net emissions reduction, (ii)certified environmental effectiveness, (iii) complementarity to reduction of own emissions. In a way joint implementaton schemes can be regarded as justified by the fact that often the countries providing these compensations are those that now and in the past have been largely responsible for the present state of the environment; they could be regarded as paying their 'environmental debt'. In fact, joint implementation enables one category of polluters to engage in environmental policies when its income constraint is an over-severe impediment. Such schemes, while avoiding some of the problems posed by straightforward trading schemes, may lead to other difficulties: i) In as far as co-operation of developing countries is required, these might show reluctance in going along with programmes that apparently imply financial transfers with "new conditionalities" (the proviso that they be spent on specific activities targeted to deal with specific reductions in emissions) involved. ii) Difficulties in establishing the environmental effectiveness are likely, as this involves difficult assessments of time paths of emissions with and without the programmes, against the background of often relatively impredictable developments at the level of the underlying economic activity levels. Certification arrangements may at least partially address this issue. iii) Some of these schemes entail the exchange of property rights on natural resources in other countries; it could be difficult to disentangle environmental effectiveness from considerations in terms of expected capital gains. A growing literature (e.g. Kuik et al 1994) is geared towards designing effective,
88
efficient and equitable arrangements for joint implementation. A tradable carbon quota system could easily arise out of a joint implementation situation, when the countries involved all develop targets for (net) carbon emissions.
5.
A TENTATIVE ASSESSMENT AND CONCLUSIONS
Climate change policies may lead to substantial social costs; attempts to identify least-cost approaches can therefore have high social benefits. Economic approaches to global issues are important to build into the emerging conventions and institutions aspects such as flexibility and efficiency. Several types of instrument may be considered. Recently most attention has been given to systems of charges and tradeable permit systems. These systems have different characteristics in terms of their environmental and economic performance (see e.g. IPCC 1994). But their most important common characteristic is that they will help in minimising overal emissions reduction costs by shifting effective emissions reductions to countries with lowest marginal costs; that is, if we are comparing trading in GHG-permits with GHGcharging (e.g. a CO2-tax) and not with an energy tax. Energy taxes would be relatively inefficient ways of achieving climate-relate objectives compared with GHG-emissions charges (Carraro and Siniscalco 1993; Zhang 994). Advantages of TPS over a charges system include (Bohm 1991): (i) the relative certainty of meeting emissions standards; (ii)fewer complications with non-convertible currencies than when handling charges' revenues; (iii) TPS does suffer from harmonisation problems with existing national taxes, as carbon/energy charges would. In addition, a tradeable quota regime would put developing countries in a position where transfers would be based on agreed upon rights, whereas a tax-cum-transfers system might keep developing countries in a situation of structural dependency on industrial countries. Finally, a tradable permit system would probably induce a forward market with associated intertemporal efficiency gains. Relative advantages of a tax system over TPS are (ibid.): (i) the revenue raising nature of the instrument; (ii)the familiarity of governments and other actors with the excise tax principle; (iii) low transaction costs; (iv) tax systems would not give rise to a compromising dominant position of large industrial countries. Furthermore, appropriate international institutional frameworks to operate a charges system are relatively easy to conceive. Both systems, that of tradeable permits and of charges, suffer from difficulties in obtaining wide support. Obviously, introducing an international system of charges will not be easy, given the differences in the levels at which energy and -implicitly carbon are charged in different countries. Trading schemes appear to, at best, pose a future option only: practical difficulties, especially related to establishing an acceptable initial endowment, provide impediments to their rapid introduction. A rapid and full-scale introduction of any one of these theoretical alternatives is therefore very unlikely. Rather, one should expect experimentation on smaller scales and with partial approaches, from which a broader and harmonised approach might develop later. Various ways can be envisaged: 1) Second-best compensatory introduction.
89 In the case of tradeable permits, side payments, an "equitable" initial endowment and temporary exemption of poorer countries could help in (gradually) introducing a worldwide system. In the case of taxes the revenue raised by it could be (partially) recycled on the basis of income effects of the tax, initial income differentials, efforts in developing or installing cleaner technology or augmenting pollution sinks, etc., so as to enhance international support. 2) A less than across-the-board-approach which focuses on certain main elements of the climate change problem, operates in groups of contributing countries or sources, with selected compounds, gradually building up from there towards a more complete system. For a number of reasons (OECD 1992b), it appears that a system of national taxes on energy or carbon could be achieved easiest at the regional level. This might hold particularly in the European region (some details on an EC carbon/energy charge were given above). Subsequent introduction of similar charges elsewhere could lead to harmonisation towards e.g. an OECD-wide charge. The larger the geographical scope of the charge, the less need there is to exempt industrial sources on the basis of distortions in long term comparative cost differentials and competitive positions, or to use other complicating additional measures. A system of trading in emissions permits or offsets would presumably only start with a group of developed economies as well, on the basis of internationally agreed emission targets. Such countries could be allowed to buy reductions elsewhere (offsets). It is important to design procedures that could facilitate the gradual development of a true market for entitlements out of bilateral transfers of such entitlements in the initial phases (Roland in OECD 1992a). 3) Mixed systems might arise with charges in some regions and emissions permit trading elsewhere. Other instruments of a mixed or hybrid nature may exist: (i) national or international funds drawn from the revenue of national charges for e.g. energy use and/or carbon emission could be set up to finance environmental expenditure in developing countries; (ii) tradable credits could be built up by abating emissions in other countries, based e.g. on unit rates to be decided in international agreements; etc. The partial and hybrid systems discussed here, may have institutional advantages in that they often build on well known procedures such as negotiations about reduction efforts. Conclusion
Looking at the economic efficiency (and disregarding a number of institutional and political impediments), fully fledged schemes of emissions charges or tradeable emission permits appear most (and equally) attractive. If their effectiveness were to be the criterion, tradeable permits hold the promise of more certainty in comparison with charges. However, such schemes are very unlikely to come about in the near future. In terms of environmental effectiveness a step-by-step approach starting with some system applied by a small number of countries, gradually increasing its geographical scale and incorporating more elements of charging and/or trading, with a gradual level of price rise in the case of a charge, appears to be the most promising and practicable avenue for an international climate change policy. Considering the acceptability of the various instruments, there are difficulties
90 Fig. 1. Environmental Policy Approaches
I control activities
facilities/ options i
I
i
I b. 1: regulation I" . . . . . . . . . . . . . .
[ I
!
l--
!
I
I
I
i 1
i
1
perceived options
decision
--"
behaviour
ul
I
I
1
L
environmental impacts
t * win.-
mmu
I
F I I I !
1I
i
I
',b.2: incentives
1
I
options costs and benefits
I
1
t
I_ . . . . .
i P 1
other impacts
I
I
I
',b.3: internaliI
9
zatlon ...,..
=.,=,.== . . . . .
!
other impacts of options
1 I
I I
[ I
1
. . . . . . .
I I b.3: information
approach b
approach a [
policy choices | I I
institutional context
~,fter: Opschoor 1991
j
.J
91 with the equity aspects (both initial endowments and issues related to accumulation) and the institutionalisation rendering tradeable permit schemes politically ill-acceptable compared with charges or taxes on energy. On-going discussions about a carbon/energy charge in Europe, perhaps to later on be introduced OECD-wide as a second step to a global scheme, are very important in this respect. "Joint implementation" of emissions reduction where some countries participate financially and technically in other countries' abatement or sink enhancement efforts, could be a way of developing towards an international trading system. However, given the current status of joint implementation it will not be a major instrument in achieving targets for the year 2000.
6.
REFERENCES
Agarwal A. and S. Narain (1991). Global Warming in an Unequal World: a Case of Environmental Colonialism. Centre for Science and Environment, New Delhi. Bohm, P. (1991). Policy Problems of International Environment. Paper presented at the International Environment in the '90s",Neuchatel, Aug. 1991.
Cooperation Conference:
on the Global "Economy and
Boorsma P., P.C. Gilhuis, B.M.S. van Praag and J.B. Opschoor (1988). An Anti Acidification Fund. (in Dutch). Ministry of Public Housing, Planning and Environmental Mnagement, Publication Series Air, Nr. 77, 55 pp. Carraro C. and D. Siniscalco (1993). The European Carbon Tax: An Economic Assessment. Kluwer Ac. Press Dordrecht. CPB (Central Planning Bureau) (1992).Long Term Economic Consequences Energy Charges (in Dutch). CPB Working Documents No. 43, The Hague._
of
Hoel M. (1991). "Global environmental problems: the effects of unilateral actions taken by one country". Journ. Envir. Econ.& Man., Vol. 20 Nr. 1, Jan. 1991: 55-71. Hoeller P.,A. Dean and M. Hayafuyi (1992). "New Issues, New Results: the OECD's Second Survey of the Macroeconomic Costs of Reducing CO2 Emissions". Working Papers, No. 123. OECD Economics Department, OECD, Paris. Huppes G., E. van der Voet, W. van der Naald, P. Maxson and G. Vonkeman (1993). New Market-OrientedInstruments for Environmental Policies. Graham & Trotman Ltd, London and Kluwer Ac. Publ. Norwell MA. IPCC (intergovernmental Panel n Climate Change) (1994). Climate Chane: Policy Instruments and Their Implications. Proceedings of the Tsukuba Workshop of IPCC Working Group III. Ceenter for Global Environmental Research, National Institute for Environmental Studies, Environmental Agency Japan.
92 Klaassen G. and D. Pearce (1994). "Economic Incentives and Air Pollution Control". Environment and Resource Economics Special Isue on Economic Incentives and Air Pollution Control, forthcoming. Kuik, O. P. Peters and N. Schrijver (1994). Joint Implementation to curb Climate Change: Legal and Economic Aspects. Kluwer Ac. Press, Dordrecht/Boston/London. NAPA (National Academy of Public Administration (1994). The Environment Goes to Market: the Implementaton of Economic Incentives for Pollution Control . NAPA Washington, July 1994 OECD (1989). Economic Instruments for Environmental Protection. Paris, 1989. OECD (1991a). How to Apply Economic Instruments. OECD, Paris OECD (1991b). Responding to Climate Change: Selected Economic Issues. OECD. Paris 1991. OECD (1992a). Climate Change: designing a tradeable permit system. Paris 1992. OECD (1992b). Climate Change: designing a practical tax system. Paris 1992. OECD 1994. Manageing the Environment: the Role of Economic Instruments. Paris, 1994 Opschoor J.B. (1991). "Economic Instruments for Controlling PMPs". Hans Opschoor and David Pearce (eds) (1991). Persistent Pollutants: Economics and Policy. Kluwer Ac. Press Dordrecht. Opschoor J.B. and R.K. Turner (eds) (1994). Economic Incentives and Environmental Policues: Principles and Practice Kluwer Ac. Publ. Dordrecht?london/Boston. Piacentino D. (1994). "Carbon Taxation and Global Aspects". In Opschoor and Turner 1994.
Warming:
Domestic
Policy
UN (1992). Declaration on Environment and Development. UNCED, A/CONF.151 /PC/ WG.III/L.33/Rev. 1. UNCTAD (1992). Combating Global Warming: Study on a Global Tradeable Carbon Emission Entitlements. UN New York
System of
Zhang Z.X. (1994). "Setting Targets and the Choice of Policy Instruments for Limiting CO2 Emissions". Wageningen Economic Papers 1994-2, Wageningen Agricultural University
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
93
M A G I C C a n d S C E N G E N " I n t e g r a t e d m o d e l s for e s t i m a t i n g r e g i o n a l c l i m a t e c h a n g e in r e s p o n s e to a n t h r o p o g e n i c e m i s s i o n s T.M.L. Wigley University Corporation for Atmospheric Research, P.O. Box 3000, Boulder, CO 80307-3000, USA EXTENDED ABSTRACT
MAGICC and SCENGEN are a suite of models that determine the regional details of future climatic change for specified emissions scenarios, together with estimates of their uncertainties. These models follow through and compare the consequences of a "policy" emissions scenario and a "reference" scenario. MAGICC (Model for the Assessment of G__reenhouse-gas I_nduced Climate Change) converts emissions to concentrations, to radiative forcing, to globalmean temperature and sea-level change. SCENGEN uses this temperature change output together with information from General Circulation Models (GCMs) to develop regional scenarios for climate change. Input emissions data (from an editable "library") are required for CO2 (fossil and land-use emissions separately), CH4, CO, NOx, NMHCs, halocarbons and fossil SO2. CO2 concentration changes are calculated using the carbon cycle model of Wigley (1993), which uses CO2 fertilization to give a contemporary carbon budget consistent with observations. CH4 concentrations are determined using the variable-lifetime model of Osborn and Wigley (1994). For both CO2 and CH4, user, best-guess, low and high projections are given to allow an assessment of uncertainty. N20 and halocarbon concentrations are computed using simple constant-lifetime mass-balance models. For the halocarbons, input is required only for four key species, C F C l l , CFC12, HCFC22 and HFC134a. A scaling method calibrated against a range of more comprehensive analyses is used to account for other halocarbons. The effect of halocarbon-induced stratospheric ozone depletion is included using a modification of the chlorineloading method of Wigley and Raper (1992). Fossil-based SO2 emissions are used to determine both the direct and indirect radiative forcing effects of sulfate aerosols, following the method of Wigley and Raper (1992). Tropospheric ozone and carbonaceous aerosol effects are also accounted for, albeit in a relatively simple way. For the gas-cycle and radiative forcing models, all parameters are consistent with the latest (1994 and 1995) recommendations of Working Group 1 of the Intergovernmental Panel on Climate Change (IPCC). Radiative forcing values are transformed to global-mean t e m p e r a t u r e changes and oceanic thermal expansion using t h e upwelling diffusion energybalance climate model of Wigley and Raper (1992). The temperature change values are used to drive ice-melt models for Greenland, Antarctica, and small glaciers and ice caps in order to obtain total sea level rise. The models currently
94 used are those of Wigley and Raper (1993), but these are in the process of being updated as a part of the 1995 IPCC exercise. Uncertainty ranges for globalmean temperature and sea level change are also calculated. The temperature and sea level results from MAGICC are being used by IPCC for their 1995 assessment of climate c h a n g e ~ i n this sense, the models used represent the current state of the art. MAGICC t e m p e r a t u r e output is used to drive the SCENGEN climate scenario generator. Regional patterns of climate change, AC(t) (the underlining here indicates a two-dimensional pattern), are calculated using AC(t)=AT(t)AC* where AT(t) is the global-mean temperature change and AC* is the normalized pattern of climate change. AC(t) may be temperature, precipitation, humidity, cloudiness, etc. on monthly, seasonal and annual time scales. AC* (based on more than 10 GCMs) may be either for single models or averages of a number of models. AC* values are obtained by dividing the results from individual GCMs by the corresponding global-mean temperature change values. This scenario generation method allows time-dependent patterns of climate change to be developed from either equilibrium GCM or transient coupled A/OGCM results (or both), under the assumption of a time-invariant "signal" pattern ( which can be justified by analysis of coupled A/OGCM results). It also allows results from models with widely different climate sensitivities to be combined. For the globe, SCENGEN gives output at the 5 ~ latitude by 5 ~ longitude level. For Europe and the USA, output is available at 1~ by 1~ or better. To obtain the higher resolution, the 5 ~ by 5 ~ data are smoothly i n t e r p o l a t e d t o 1~ by 1 ~ and added to high-resolution, high-quality baseline climatologies. Uncertainties are quantified at two levels, either by driving SCENGEN with low, mid or high global temperature changes from MAGICC, and/or by using the 90% confidence bands for AC* obtained from an analysis of inter-model differences. The whole system is embedded in a user-friendly shell, and is designed to run rapidly on a high-level (e.g. 80486-based) microcomputer. A full simulation can be completed in a few minutes.
Acknowledgments MAGICC and SCENGEN were developed using funds from the U.S. Department of Energy, the European Community (DGXI), the U.K. Department of the Environment, and the Electric Power Research Institute. Most of the work was carried out in the Climatic Research Unit, University of East Anglia, Norwich, UK, by the author, Sarah Raper, Mike Hulme, Mike Salmon, Jiang Tao and Tim Osborn.
REFERENCES I T.M.L. Wigley, Tellus 45B (1993) 409-425. 2 T.J. Osborn and T.M.L. Wigley, Climate Dynamics 9 (1994) 181-193. 3 T.M.L. Wigley and S.C.B. Raper, Nature 357 (1992) 293-300. 4 T.M.L. Wigley and S.C.B. Raper, (In) Climate and Sea Level Change: Observations, Projections and Implications (eds. R.A. Warrick, E.M. Barrow and T.M.L. Wigley), Cambridge University Press, Cambridge, U.K., (1993) 111-133.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
95
The Process of Developing Policy Based on Global Environmental Risk Assessment D J Fisk Department Environment UK
Introduction I have been invited to give a short presentation on developing policy based on a global environmental risk assessment. I very much look forward to an exchange of views later in the morning about how policy and science interact, and what we have learnt from the process so far. For my part I am going to take the title literally and look at the global warming issue as if it were a formal problem in risk assessment. I want to use that framework to make one or two suggestions as to how the interaction of policy and research might evolve as the climate convention progresses.
Background The Intergovernmental Panel on Climate Change which began its work in 1988 and gave its first report in 1990, was a landmark in the development of technically based global environmental policy. Of course the assessment process has moved on since then. The Framework Convention for Climate Change has come into force, and national plans by most Annex 1 countries have been sent to the interim secretariat. In Berlin this March the Conference of Parties will meet for the first time. Amongst its tasks will be to set up the Subsidiary Body on Science and Technology Assessment. Characteristic of the stepby-step approach of a modern international environmental instrument, the convention has set the Conference of Parties a review deadline of 1998 to set post 2000 goals. With all this new process coming on board, I very much commend the conference's wish to look at how interactions between policy and research can be improved. IPPC90 as a Risk Assessment The first IPPC process was in many ways a classic risk assessment the report identified the hazard - IR absorption and the gases which exhibited this property
2)
the risk of these gases accumulating in the atmosphere through scenarios and chemistry
96
3)
assessed the consequences - the impacts on sea level, agriculture, health etc
4)
looked at the options for risk management
This IPPC process, which in its formal risk assessment form is familiar to most of us handling micro-environment problems, has not been without its critics. Certainly, since 1990, the peer review process has been improved, but at the penalty of a much more time consuming process. Gone for ever are the pre-IPPC days when a small group of experts could sit down in Bellagio and write their conclusions up in an afternoon! There is the perennial problem of immediacy that plagues any risk assessment that is based on an active area of research. The early steps in the risk assessment can become outdated by the time that the final steps have been completed. IPPC90 you will recall had to use the earlier National Academy Science review for its consequenceanalysis because the IPPC90 risk assessment was still in progress. There was no opportunity to test the effect of the risk management responses in the climate and impact models. Despite these criticisms the IPPC has proved a powerful tool to assemble and assimilate the state of research. Indeed it picks up an enviable number of citations in its own fight in learned journals. Other global environmental areas, such as the state of the global oceans, may not be any poorer in underpinning science, but are disadvantaged because of this lack of synthesis machinery. While constructing a synthesis of the science is vital, it leaves the issue of how the science is to fed into the actions which might follow from it. The IPPC has more recently been wrestling with this aspect, and I will also focus on this issue.
Development of National Hans In charting the development of international action, it would be wrong to ignore the development represented by the preparation of national plans under the convention. It is a common experience that risk management takes on a different character once we move from analysis to trying to actively manage the hazards. Those disadvantaged by the action exercise their fight to question the risk assessment and consequences analysis. Often Governments or at least democratic Governments - find that they have inadequate instruments to deliver action by others unless they are equally convinced. The national plans submitted to the convention underline this point. Governments can set the framework for action, but they need the co-operation of others to deliver real changes in emissions trends. In that sense Dr Brenabo's paper on communications between scientists policy makers and society at large is especially important.
Hazard Identification and Risk Analysis Most people I suppose ask 'Is climate change a problem?' and if the answer is yes 'what could I do that would have any effect?' These are not bad questions for the policy making process at any level. Answering these questions through risk assessment begins by establishing the hazard. No one has seriously challenged that the infra-red absorption
97
property of greenhouse gases is a hazard. Identification that there is a hazard in a risk assessment is usually sufficient to establish the case for 'best practice' in handling the hazard, or in climate change parlance 'no regrets' measures. I shall have something more to say about 'no regrets' when I come to risk management options. Hazard identification is only the start of the analysis. The next step. Risk analysis has proved more difficult. S c e n a r i o s are S c e n a r i o s
In a traditional risk analysis, situations are envisaged which might realise the hazard. The probability of each situation is assessed and the overall probability of the hazard being realised computed. Superficially IPPC have worked in a similar m a n n e r . Four scenarios were exhibited in IPPC90. The number expanded in IPPC92 to six. To these might be added the scenarios developed by World Energy Council. IPPC have often been pressed to identify the most likely scenario, or attribute probabilities to the set of scenarios. This would certainly permit a conventional risk analysis. However close scrutiny of the time axis of these scenarios which extends to 2100 shows that IPPC would be right to stand its ground - a 'scenario is a scenario not a probability weighted forecast'. Let me argue this point by looking at the oft quoted IS92a scenario. Suppose I were to treat this as a forecast. Then I can make a number of other deductions about the long term future. First, taking into account the implied cost of nuclear power in IS92a it would be clear that despite the passage of a 100 years and an ever widening technical and scientific base we had found no cure for cancer that trivialised incidence of the disease. The world, although incredibly richer would still not be at peace and would still be concerned at nuclear proliferation. Treating IS92a as a forecast we do not appear to have found a room temperature super-conductor which would of course have revolutionised energy storage and transport. No doubt with that knowledge we could save a guilder or two elsewhere in the Dutch national research budget! If the choice of technical revolutions look as if I am biasing IS92a downwards perhaps I might add it also implies that we do not seem to have cracked the biochemistry of ageing either in 100 y e a r s . If we had it would be difficult to guess what the population driver figures might look like. Anyone of us could associate a subjective probability to these events. However the likelihood of consensus amongst 5 billion people as to what those subjective probabilities would be seems rather remote. This is in contrast to forecasts in the shorter term - or at least the shorter term to the climate scientist. These forecasts limit themselves to a time span in which even if these technical shocks were to be realised their probability of influencing the forecast is vanishingly small. For the sake of a name we might call the end of such a time span a Schumpeter horizon to acknowledge that beyond it Joseph Schumpeter's creative destruction implies that we can no longer rely in any sense on extrapolation. The recent IEA forecast for global carbon dioxide emissions, for example, falls within the Schumpeter horizon. Such a horizon is also the natural time span in which to set step by step legally binding commitments in conventions, at least for those who want to take their commitment seriously. The scenario process is inescapably normative. This is less of a problem than might be supposed at the stage I have reached in the risk analysis. If for example you turn to the
98 Brundtland Commission Report you will find the development of a normative scenario for the economic development of the world's nations. It may not happen but the scenario embodies widely held aspirations for the future. Thus while IPPC could happily construct an infinite number of scenarios, it is only those that express our aspirations that we believe we want to see actively pursued that need be included in the initial risk analysis. The scenarios ought for example to show the property of sustainable development. I deduce from this argument that the policy making process in the convention needs in due course to address which scenarios reflect the aspirations of its parties. It ought to be these scenarios which make up the feedstock of climate models and impact estimates. In the first stage of a risk assessment the key scenarios are those which reflect aspirations without being fettered by considerations of climate impact. It is a matter of taste whether the term 'business as usual' quite captures that flavour. Consequence Analysis It has become rather popular to open discussions on climate change with a recital of the uncertainties in climate modelling. From a risk assessment point of view this narrow focus is not altogether healthy. Admittedly the IPPC90 key index of modelling uncertainty - the climate sensitivity - ranges over a factor of 3 from lower to upper bound. But this is no larger than the range of climate forcings arising from the IS92a scenarios themselves. It is therefore not just a range of climate science possibilities that need to be explored. The point to note is that all but one of the scenarios have rising climate forcing, and that all estimates of climate sensitivity are positive non-zero. Thus under these scenarios the modelling uncertainty simply changes the time at which a certain climate change condition takes place. Uncertainties in climate modelling influence the risk management not the initial risk analysis. The question that is seldom answered by professional sceptics is just what scale of climate change matters, and whether that degree of climate change is within the range of scenarios, taking into account uncertainty in climate sensitivity. These are key questions that the consequence analysis must address. The degree of precision that we need from climate modellers depends critically on the degree of precision required by the impact assessment. Types of Impact Assessment In collective environmental decision making, the 'least helpful' outcome for a consequence analysis is that changes are found to be gradual. It may be gard to find consensus on a trade-off. In contrast sharp changes, sometimes called comer solutions from optimisation theory, are very important findings for gaining a consensus. By their nature they bring together a coincidence of different interests. For example there may be a rate of change at which temperate forests decline, or a sea temperature at which the Antarctic ice sheet begins to shelve, or the thermohaline circulation stops. I hope that it not too self evident if I suggest that these classes of impacts deserve special priority in impact research as a basis for collective decision taking.
99 It may of course be that such sudden changes do not exist and that climate impacts are gradual in their effect. There have been some attempts to tackle gradualist change by normalisation to some valuation criteria as a basis for contracting trade-offs. The new IPPC assessment will be reviewing some of these approaches. Personally I have some doubts that we have fully worked out how to use this methodology in the context of a long term issue like climate change. In particular it is not clear how much prior context has to be agreed before the figures have a hope of gaining a consensus. However the approach teases out one difficulty in a reductionist approach to impacts. We simply do not know what it would feel like to be living during a time that climate change was so apparent that we lacked confidence in how the climate might change around us. It is common experience in environmental policy that society's response to a consequence, changes once the consequence is realised. In climate change this state of mind presumably sets in when we are confident that we can detect the enhanced greenhouse gas signal in the global climate record. I would argue that will be an important marker in the development of the convention. I conclude that impact studies have a special importance in a risk assessment because they define the precision demanded of climate models, and their structure determines the likelihood of a consensus to respond to the risk. Risk M a n a g e m e n t - Yet to be Begun
What I have discussed so far is simply establishing the climate change consequences of pursuing our aspirational scenarios. For any scenario that breaches the conditions of the climate change convention - adaptable rate of change to a safe stable concentration - the risk management component of a risk assessment comes into play and requires us to revisit our scenario. It would certainly be true to say that we have hardly begun to articulate in the convention how the next steps in risk management analysis could be undertaken. The conventional history of risk management in an environmental instrument starts with a few suspect hot spots. These lead to some generalised early action. By the end date of this agreed action the underlying science is clearer and usually substitute technology has been developed. The final stage of the instrument is then played out. The parties in the climate convention are clearly struggling with the first stage, which we label hazard management. Countries with the technical and social means to devise ways of abating emission have drawn up national plans. The unusual aspect of the convention is its timescale. We may still have not detected man-made climate change by the early part of the next decade. Although there are some good ideas in the national plans the dream substitute technologies have yet to come into play. It is of course difficult to judge how the Conference of Parties will take this issue forward. Most countries have found that the store of 'no regrets' measures is difficult to unlock, not least because those who have interests in the older policies often take a 'I regret nothing' stance. The Conference is therefore likely to be interested in looking
100 more closely at means of better co-ordinating national measures, possibly through a Protocol to the Convention. Germany has already submitted some ideas on these lines. If risk management were to become the underlying principle through which the Conference of Parties developed its work, then progress would undoubtedly be step-bystep. Measures would be assembled to take effect over my Schumpeterian horizon against specific commitments. As each target end-date was reached, the Parties would review the effect of their measures, assess the improved knowledge of the climate science, and inspect the concentrations levels of greenhouse gases that had actually been reached. Let us suppose in conclusion that the convention was to take this route through its subsequent meetings. What kind of dialogue with science might be needed? I would suggest on the basis of the points made: For Improving Hazard Analysis (1)
Continuation of IPPC reviews of global atmospheric chemistry, as the convention attempts to treat greenhouse gases in a comprehensive fashion.
For Improving Risk Analysis (2)
Clear rationals from IPPC and SUBSTA for scienarios.
For Improving Consequence Analysis (3)
Continued Scouting for 'comer solutions' in impacts.
(4)
Differential impacts analysis (i.e. comparing impacts between different risk management scenarios).
For improving Risk Analysis
(5)
Analysis of the comparative performance of differing measures in national plans.
I have not meant to exclude work on either large scale climate modelling or extensive impacts research. I thought however that this important work might be referenced in a different context. I pointed to the time when there was general agreement that man-made global warming had been detected in the climate record. It would be an important turning point in the development of the convention, but also in the nature of file public debate. It is difficult not to have noticed how extreme local climate events in the recent past have spurred public interest and debate in this issue. In the past our meteorological advisers have been able to re-assure us that these events were not distinguishable from natural variation. In the future it may be more difficult to make that assertion. The research into large scale
101 modelling enterprises may then become important not just for projecting future change, but interpreting the change that will be seen around us.
The views expressed in this paper are those of the author, and do not necessarily represent those of the UK Department of the Environment
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
103
Communication Among Scientists, Decision Makers and Society: Developing Policy-Relevant Global Climate Change Research. J. Christopher Bernabo Science & Policy Associates, Inc. Suite 400 West Tower, 1333 H Street N.W., Washington, DC 20005, USA Abstract
Defining the research most relevant to policy is not simply a technical task that can be answered by scientists. Decision makers need and value information differently than curiosity-driven scientists. In order to link science more effectively to policy, the two communities must gain a greater mutual understanding. Decision makers must define their needs so that scientists can determine how, and by when, research can address these needs. This vital dialogue between communities typically has been more ad hoc than systematic. The complexity and urgency of the global climate change issue necessitate ongoing communication between scientists and decision makers on the information needed for policy development and what research can provide. The results of relevant science policy dialogues are discussed herein.
1. INTRODUCTION Effective communication between researchers and decision makers is a crucial ingredient for successfully addressing society's pressing environmental concerns. The increase in policy makers' demands for research that is relevant to solving societal issues highlights the communication gap between the technical and policy communities. The gap, largely caused by lack of mutual understanding, results in flawed and inadequate communication that hinders decision making and confuses the public. This paper examines the cause of this communication gap and describes the significance of recent efforts to develop more fruitful science-policy dialogues on the issue of global climate change. First, the post-Cold War shift in government priorities for research funding is described; then the underlying relationship between science and policy is explored to identify key sources of ongoing miscommunication. The paper then explains the importance of defining policy-relevantscience questions that research can address. Finally, three projects are described involving the elicitation of decision makers' information needs in The United States, The Netherlands, and internationally.
2. POLICY RELEVANT RESEARCH Fifty years after World War II, the major political, social, and economic changes sweeping the globe are causing an historic shift in the emphasis of research funded by governments. In many nations, such as the United States, national security was a major societal justification for massive public funding of the natural sciences and engineering. The end of the Cold War military competition has caused a wide spread reevaluation of science funding priorities [ 1,2].
104 Furthermore, the public's faith in science as an unquestioned source of ever increasing material living standards has been shaken by the emergence of many technologically-induced environmental problems [3]. With economic constraints to growth and global competition rapidly increasing, there are greater demands to direct government-funded science and engineering toward solving pressing societal problems. The emerging post-Cold War rationale for government funding of research has five priority factors: 9 Emphasizing science that provides societal benefits; 9 Linking research programs to the needs of decision makers; 9 Providing economic development and competitive advantages; 9 Developing partnerships with diverse stakeholders; and 9 Leveraging international research activities and programs. None of these factors are new, but the increased emphasis upon them in guiding research investments is a major development for science in the post-Cold War period. This greater attention to investment return and the societal relevance of research will require enhanced efforts to improve the communication between scientists, decision makers, and the public.
3. RELATIONSHIP OF SCIENCE TO POLICY "Science has the first word about everything and the last word about nothing," Victor Hugo observed. The truth of this is inherent in the relative roles that both "objective" scientific information and "subjective" human values inevitably play in decision making. Environmental policies are developed by interpreting and applying technical information in light of the needs and human values of society (Figure 1). Viable policies must not only be technically sound but also socially, politically, and economically acceptable.
SOCIETAL FACTORS POLICY DEVELOPMENT TECHNICAL INFORMATION
ACTION
7
~., Figure 1. The relationship of science and human values in policy development, showing primary interactions and feedbacks.
105 Science alone cannot provide answers to policy makers' ultimate questions because science necessarily is silent on the human values that underlie the decisions societies make. The scientific method itself is designed to screen out the value preferences and biases of the subjective human beings who conduct research. Technical information is useful in identifying issues, developing options, providing understanding and evaluating consequences for policy actions. But in the end, human values must be applied to determine what is "good" policy for a given society on a specific issue. Take the example of nuclear energy: is promoting it a good or bad policy'? On the surface this appears to be a scientifically answerable question yet nations with access to the same technical information have made different choices about the best policy for their societies. Indeed, there are Nobel Laureates that staunchly argue opposite sides of the case because the question ultimately involves human values. Science can only approximate the risks and benefits, but a subjective value judgement must be applied to decide what ratio between the two is acceptable to a given individual or society [4]. Many of the difficulties scientists and policy makers face in communicating and working together arise from differences in their professional cultures (Table 1). Both the scientific and decision making communities experience frustration over the paradoxical relationship between information development and policy development. The public and policy makers often perceive that science is more effective at identifying uncertain problems than it is at providing certain solutions. On the other hand, the technical community becomes frustrated by the perceived inability of policy makers to grasp the facts and take what they personally judge is the "logical" action. Table 1
Contrasting Professional Cultures of Scientists and Policy Makers Science
Policy
Objective Facts Proof Rational Measurements Incremental Progress
Subjective Values Beliefs Emotional Perceptions Deadlines and Crises
Applying technical information to decision making is a fundamentally different type of activity than discovering new knowledge. Alvin Weinberg coined the term "trans-science" to describe the process of using technical information in making decisions that inherently transcend the bounds of science [5]. He points out that facts alone are not sufficient even for weighing the benefits and costs in policy issues, because subjective values must be applied in choosing what facts to use and how. Harvey Brooks concludes that, "the facts that are selected and the way they are presented to the public may have a greater political impact than the facts themselves" [6].
106 4. SCIENTIFIC UNCERTAINTY AND POLICY DECISIONS Environmental policy debates typically involve discussion of uncertainties and how much certainty is "enough" to justify a proposed action. The question of how much information is adequate for a given policy always involves a value judgement and cannot be answered by scientific research alone. There is no objective point in science that defines enough certainty for policy. Research can only quantify the uncertainty in the science, and even that with great difficulty, but policy involves many other types of inherent uncertainties. Policy decisions must consider uncertainties about matters such as the significance of facts, the perceptions of the issue (opinion polls), the economic and social viability of the proposed solutions, and the actual versus intended consequences of the action. The degree of scientific consensus is just one part of the information needed for decision making. Brooks cautions that we "should be careful not to expect that scientific consensus should be a necessary condition for policy consensus, an expectation to which scientists tend to be too prone" [7]. For instance, we might have no fiscal policies if action required consensus on economic predictions. The policy makers' roles include making subjective judgements about which information should be acted on and how much certainty is enough for decision making. The degree of certainty that is adequate for policy can be viewed as an equation balancing scientific uncertainty and political uncertainty. Two general principles apply to environmental issues: The greater the societal consensus on an issue, the less scientific certainty required for action.
II.
The higher the societal costs of a policy, the greater the scientific certainty required for action.
The inverse of these principles also is true. They imply that enough certainty in the science is always defined relative to the political certainty in the issue. Therefore, enough scientific certainty in the policy process is a dynamic factor, not a static end point from research. Two examples illustrate these principles. The United States and Canada fully shared scientific information on acid deposition; they had joint monitoring programs and the same degree of technical certainty on the issue. Nonetheless, lower scientific certainty was required to justify policy action by Canada because there was much higher political certainty than in the United States. Over 90% of Canadians believed that acid rain was a serious problem, while there was no such consensus in the United States. Canadians saw a threat to their major industries--timber, fisheries, and tourismnfrom the potential damages. In the United States, pollution control costs were instead perceived to be a threat to industry and jobs. In essence, all details of politics aside, the reason for the national differences in the thresholds of scientific certainty required for action was simple and predictable. In the United States, chlorofluorocarbons (CFCs) were banned as spray can propellants back in 1978. At that time no ozone hole had appeared and scientific certainty about the issue was lower than about acid rain in 1980 or global climate change in 1994. The threshold
107 of scientific certainty was low for the initial CFC ban because there was political consensus that the risks of skin cancer were not judged to be worth the benefits of protecting a few jobs. Further banning CFCs from all other uses awaited higher scientific certainty because of the greater societal costs involved.
5. POLICY-RELEVANT SCIENCE QUESTIONS For scientists to assist effectively in the development of policy, their research needs to be focused on the questions of greatest value to decision makers. Examining past experiences in applying science to address environmental issues helps illustrate the importance of defining the policy-relevant science questions to guide research. Policy relevancy is determined by the specific needs of the information user (policy maker) not the interests of the information producer (researcher). Unfortunately, the questions investigated by curiosity-driven science are often different than those required to provide the most policy-relevant information. This occurs because decision makers only require the information that can materially assist their specific deliberations, whereas scientists seek greater fundamental understanding of their subjects. Other mismatches exist because of the different values attached to information in the research and decision-making realms, and because policy makers need information that cuts across fields of research. There are three general ways to define policy-relevant research questions: 9 Educated guesses: This has been the traditional means whereby scientists who study an issue presume to formulate what questions they deem relevant to decision makers. Although quick, this investigator-driven approach fails to examine the real needs of the policy users. Curiosity-driven questions tend to dominate these agendas without the benefit of decision makers' input. 9 Multi-stakeholder dialogues: This approach involves systematically eliciting the information needs of decision makers in the various stakeholder groups for the issue. Interviews and meetings are utilized to determine what the information users' need. Then scientists are involved in examining and responding to these needs in a facilitated process that ensures results reflecting the best input from both co~rununities. This process can be accomplished over several months and builds direct dialogue between the participants, helping bridge the science-policy communication gap. A limitation of this method is that it does not allow distinguishing what information participants say they need from what they may use in practice. 9 Social science research: This is the most intensive approach and goes beyond eliciting the expressed needs of decision makers to study their actual behavior in applying information. It involves carefully designed research and field studies observing the behavior of subjects involved in decision making. This approach provides valuable insights into the use of technical information in policy development. This scholarly approach requires extended periods, usually years, during which the policy relevant questions may shift. Moreover, it does not necessarily build ongoing dialogue between the science and policy communities. Whereas the educated guess approach has typically been used, a combination of the multistakeholder dialogues and social science research is most effective. The dialogues facilitate timely development of broad policy-relevant science questions and build mutual understanding as a basis for consensus between the participants. This approach directly
108 enhances the effectiveness of linking science and policy. The longer-range and more intensive social studies of decision makers' and scientists' behaviors help provide deeper understanding for designing more effective communication. Interactions between these two types of approaches is valuable in assisting each to reach its goal. The remaining sections of this paper describe three projects that represent multi-stakeholder dialogues aimed at defining policy-relevant research questions for global climate change. The general significance of the results of a pioneering study done in the United States in 1992 are reported. The second study was done in 1994 for The Netherlands, and it improved on the methods in the initial project. The third study is being conducted in 1995 by a joint team of the investigators from the U.S. and Dutch projects and applies the previously developed approaches to an international context.
6. U.S. DECISION MAKERS' CLIMATE INFORMATI()N NEEDS In 1990, a number of U.S. research organizations became concerned that the governmentsponsored U.S. Global Change Research Program (USGCRP) may not provide an adequate basis for the inevitable information demands of future policy development. They decided that a first step in moving toward a policy-relevant research agenda was to determine generally what information decision makers needed, and they launched the "Joint Climate Project to Address Decision Makers' Uncertainties" [8]. This unique private-federal partnership was sponsored by the Electric Power Research Institute (EPRI), U.S. Environmental Protection Agency (EPA), the U.S. Forest Service (USFS), and the U.S. Departments of Energy (DOE), Agriculture (USDA), and Interior (DOI). The project was designed and conducted by Science & Policy Associates, Inc. The Joint Climate Project established a multi-stakeholder dialogue to help identify some major questions U.S. decision makers had about global climate change and then had scientists determine what research and time frames would be required to address those questions. 6.1. Focusing on the Needs of Decision Makers
The Joint Climate Project identified policy-relevant research using two interactive phases: U.S. decision makers first defined their information needs, then scientists gave feedback on these needs and determined the research required to address the policy-relevant questions. During the first phase of the project, the needs of the users of climate information were identified through interviews, workshops, and focus groups involving national-level decision makers. These individuals included dozens of U.S. government and private sector officials, ranging from working-level experts to members of Congress, Administration officials, and industry CEOs. They were invited to participate in the project on the basis of their active roles in climate change policy and their diverse perspectives, from federal regulators and resource managers, to industrial representatives and environmental groups. The interactive process lasted six months and resulted in a consensus set of policy-relevant general questions for researchers to address. Then, leading experts in climate-related fields were convened at a workshop to discuss the specific questions developed by the decision makers. The scientists were chosen for their activities in research or in the synthesis of research results. They represented a broad range
109 of expertise, including climate system modeling and monitoring, managed and unmanaged ecosystems, energy and technology, as well as economics and social sciences. The workshop participants examined the research needed to address the questions and the expectations for providing better information over the next two, five, and ten years, and beyond.
6.2. Findings of the Joint Climate Project The consensus-identifying approach of this project yielded several key findings that reflect the general concerns of decision makers and the responses of the research co~ruuunity. In discussions with these two communities, several common themes emerged for enhancing communication and increasing the value of research results.
6.3. The Concerns of Decision Makers The participating decision makers identified several general principles that define policy-relevant questions for research. The project was conducted during the year before the United Nations Conference on Environment and Development (UNCED). Talks were well underway to craft a Framework Convention on Climate Change. Therefore, many government policy makers focused on these and other ongoing international negotiations and conferences. The officials specifically asked for information to support follow-up actions to UNCED and preparations for future events. For their part, non-government decision makers expressed concern with the possible regulatory implications of proposed actions. 9 Climate Change Impacts and Human ReL~ponses are Key to Decision Making: Aside from pressing international policy issues, decision making is driven by concerns about the potential impacts of changing climate at the regional level, rather than predictions of changing global mean values of climate variables. Specifically, input is needed from the economic, social, and ecological sciences on the potential regional impacts of climate change and the consequences of possible response strategies. Any response to the threat of climate change must be measured against what is at stake. Therefore, more information is needed on the ecosystems, regions, and human populations that are most at risk from potential climate changes, even if atmospheric research is still unable to provide reliable predictions of the specific changes that will drive effects. 9 Implications of Uncertainties Need Clarification: Researchers need to clarify the sources and implications of policy-relevant scientific uncertainties and estimate time frames for reducing them. Many uncertainties, although scientifically profound, may be relatively insignificant for developing policies. There is a need to define better which uncertainties are most important for policy development and resource management, and the practical implications of these uncertainties for decision makers. 9 Certainty is Not a Prerequisite for Action: During the project, several decision makers stressed that the resolution of all scientific uncertainties is not a prerequisite for policy action. Decisions are regularly made in the face of some uncertainty. Decision makers will apply their constituents' values to determine how much certainty they judge is enough to take political action. 9 International Perspectives Drive Policy:
110
6.4. The Response of Researchers In the next phase of the project, a diverse group of U.S. experts in climate-related fields were convened to examine how research could best address the questions posed by decision makers. Specifically, the scientists examined what types of research are needed to reduce the uncertainties in the policy-relevant questions and estimated the time frames for possible results. 9 Timely Results: Some of the key questions decision makers have about climate change can be addressed within a short time frame on the basis of analysis and interpretation of currently available scientific information. Although more complete scientific understanding of climate change may be decades away, much of the information needed to begin addressing decision makers' questions can be provided within two to five years. This could include a comprehensive evaluation of indicators of global climate change, a preliminary vulnerability analysis for systems and regions most sensitive to climate change, and an assessment of the sources and levels of greenhouse gas emissions for use in identifying potential mitigation and adaptation options. 9 Parallel Approach to Climate and Human Responses Research: Scientists need not wait for accurate climate predictions before beginning their research on potential impacts and response options. It is neither necessary nor practical for research to progress sequentially from the climate system, to the impacts, and then to the potential human responses in order to provide useful results for decision makers. Much can be done to improve the understanding of impacts without waiting for accurate regional climate predictions. For example, integrated regional and multi-sectoral models~using climate, ecological, demographic, economic, and social data collected at the regional level--can provide essential information on potential climate responses, the vulnerability and adaptability of key systems, the extreme ranges of change, and the impacts of climate change on the global marketplace. 9 Greater Emphasis on Impacts and Human Responses Research: Information on climate change impacts and response strategies has the greatest potential for assisting decision makers, yet these fields are the least researched. Many of the key questions identified by decision makers involve a significant amount of new socioeconomic, behavioral, and ecological research. However, only modest increases in funding for these disciplines would be necessary to achieve useful information for policy within a few years. Social science and economic research, in particular, receive a small percentage of federal funding, but are critical for making decisions about climate change. 9 Integrated Assessments and Case Studies: Integrated assessments of the causal linkages from emissions through impacts and human responses would help structure information for effective use in decision making. Such assessments would incorporate natural and physical sciences, economics, and social factors, including technological change and adaptation. In addition, a coordinated examination of case studies of regional climate variability is needed--based on historically documented events that show how societies have responded to past climatic variations. This information would provide valuable insights on how to treat future events. 9 Expect the Unexpected: Multi-disciplinary research on potential surprises is also important, given their potentially serious implications for decision making (i.e., climate change could be much worse than anticipated, or it could be insignificant). Decision makers and scientists should frequently re-examine research on potential surprises, given that scientific progress is
111 incremental and new information may become available. Based on this information, contingency plans could be developed to prepare for unforeseen events. 9 International Perspective: Because of the global dimensions of the issue, an international perspective for research is essential. Although decision makers may be most concerned with regional and local consequences, developing world issues (such as population and economic development as well as the pace, quality, and sustainability of development) will be critical. Assessing the ability of the international community to implement mitigation and adaptation measures is important for evaluating the effectiveness of response strategies on the climate system. The project asked researchers to identify the potential types of information that research could provide to address decision makers' concerns in two, five, and ten years. The participants provided educated estimates of the potentially available information for time frames of interest to decision makers. These estimates were developed without regard to financial or other resource constraints. Furthermore, the researchers suggested what research could do, and not what currently planned efforts will do. 6.5. Lessons in Communication
Discussions during the Joint Climate Project with representatives of both communities provided ample evidence that decision makers and researchers are uncomfortable with the present situation. Both are anxious to develop and sustain a productive dialogue. Both would like to increase the effectiveness of the research community in the decision-making process. Both agree that a two-way bridge must be developed to span the communications gap between the two communities. But to truly close this gap, to construct a bridge between the two communities, will take more than wistful expressions and lofty pronouncements. There is no substitute for sustained effort and innovative institutional arrangements. The decision makers and researchers who participated in the project agreed that greater attention must be paid to the development of systemic communications processes. In particular, both sides need to recognize the following points. 9 N o t an Either~Or Decision: Decision makers' choices are not simply between pursuing research or implementing response strategies. Rather, the challenge is to define the appropriate levels of each over time. Researchers need to provide a broad array of information to address the complex and interacting decisions on global climate change. Decision makers, for their part, need to recognize the long time scales involved in research and, thus, the importance of continuity of funding and program goals. 9 Global Climate C h a n g e in a Relative Risk Context: Prediction of changes in mean global temperatures does not give an adequate picture of the societal risk that can be related to every-day experiences. The risk of global climate change needs to be compared to the risks of other economic, social, and environmental issues. Because the public tends to respond to perceived crises, assigning relative risk would help decision makers distinguish between verifiable serious threats and possibly misplaced public concern. Given that risk is a function of both the probability and the magnitude of the expected consequences, better data on possible impacts are critical to better estimates of societal risk. 9 Urgent N e e d f o r Education: A concerted effort is needed to educate decision makers on the facts and uncertainties of global climate change. Since public concern is often the
112 impetus for formulating policy, scientists need to communicate technical information to the public more effectively and more frequently. In addition, scientists need to learn more about the decision-making process and the types of information most useful for policy. Frequent, two-way communication between decision makers and researchers is essential if research is to play an effective role in the decision-making process. 9 Research Does Not Always Provide the Answer: Decision makers should understand that additional research can increase the amount of uncertainty in some areas. Researchers should inquire about how much certainty decision makers require to take a specific action. To this end, uncertainties that are not relevant to decision making should be identified early in the process. Decision makers and researchers should also seek ways to manage continuing uncertainties. For example, building resilient institutions would provide a flexible response to any future changes in climate, albeit at potentially significant costs. Contingency plans allow decision makers to prepare for possible climate outcomes through R&D on response technologies, without needing to deploy them. 9 Develop an Ongoing Assessment Process for Research: To improve communication and better inform decision makers, research efforts should include an iterative assessment process. These assessments not only help to identify the relevant questions, but also serve to structure the research results and, thus, facilitate clearer communication between the two communities. Furthermore, the assessment process provides valuable input to the planning of policy-relevant research.
6.6. Project Significance The Joint Climate Project represents a preliminary step in determining how researchers can assist U.S. decision makers over the coming years and decades, thereby helping to bridge the communication gap between these two corrununities. A more frequent and systematic twoway dialogue will be needed between decision makers and researchers in order for research to inform the decision-making process. Discussions with decision makers and researchers during the project revealed that both communities are very interested in developing and sustaining a productive dialogue. Both would like to increase the effectiveness of the research community in the decision-making process. Following the successful dialogue established by the Joint Climate Project, other similar efforts were initiated for climate change in The Netherlands and for biodiversity in the United States [9]. These types of dialogues also need to be supplemented by more in-depth social science studies to elicit greater understanding of the behavior of decision makers in applying science. A better mutual understanding of the professional cultures of researchers and decision makers is required to enhance the effectiveness of linking science to policy.
7. NETHERLANDS POLICY OPTIONS STUDY "Policy Options Addressing the Greenhouse Effect," a climate change project conducted in The Netherlands, had an approach and goals that were consistent with the Joint Climate Project. The Policy Options study was conducted for the Dutch National Research Programme on Global Air Pollution and Climate Change (NRP) by Prof. Pier Vellinga with his colleagues at the Institute for Environmental Studies (IVM) at the Free University of
113 Amsterdam and Prof. Jan Klabbers, with consultation by Dr. Chris Bernabo [10]. Within the project a dialogue has been initiated between policy makers, scientists, and other societal actors to look at how Dutch society can cope with the risks of climate change and the challenge of sustainable development. The project produced two types of results. The first included various policy options and related actions. The second, and probably more important, results were related to the process itself. There was an improvement in the communication and discussions among all the stakeholders which, over the longer term, can lead to a more solid foundation for action.
7.1. Project Objectives The Policy Options study was designed to bridge the gap between perceptions of policy makers, researchers, and public interest groups. The specific objectives were to: 9 Reinforce communication between the three communities; 9 Illustrate the perceptions of the communities; 9 Examine policy development options; and 9 Inject the options into the Dutch policy development process.
7.2. Project Approach The first step in the process to develop climate policy options was to identify the issues through interviews and workshops with policy makers. Natural and social science researchers then assessed the issues in position papers and workshops. Next, round table discussions linked the science and policy perspectives. The outcomes of these discussions provided the basis for the development of a range of policy options and related actions.
7.3. Resulting Policy Options The key policy options that emerged from the study were: no-regrets (actions which may be economical regardless of climate change considerations, although they may not be considered no-regrets by every country), least regrets (actions which adopt the precautionary principle), acceleration (encouraging reductions through subsidies or taxes), technological innovation, and institutional(ised) cultural change. The five options that have been generated effectively illustrate the complexity of the climate change issue with respect to causes, uncertainties, international relationships, and the values and norms that are at stake. They acknowledge the divergence of the views and interests of all players, and encourage working towards convergence of actions. Full details of the policy options may be found in the final report of the Dutch study [10].
7.4. Recommendations Recommendations are based on the observation that within the natural sciences it was relatively easy to reach a shared view on climate change, but the bridge between the natural and social sciences was rather difficult to make. The project's recommendations include: 1. Enhance the Communication Between Scientists: Discussions in the science workshops revealed that economists, sociologists, philosophers still show large discrepancies in their
114 view of the problem and the paradigms used in their approach to the problem. As climate change and sustainable development require open, interdisciplinary minds, much work has to be done to improve the dialogue between the disciplines concerned. 2. Improve the Science-Policy Interface: The improvement of the science-policy interface will promote the assimilation of scientific results by policy actors and will also help in identifying the relevant research questions. This implies a broadening of the communication between the science community and policy actors from the private sector, the national and local government, and public interest groups. It is recommended that attention be paid to sector-specific constraints and opportunities of climate change and sustainable development. 3. Integrate Climate Policy into Broadened Socio-Economic and Environmental Policies: Various key societal groups do not perceive climate change as a problem that warrants stringent measures. All groups seem to agree, however, that environmental policies, including those relevant to climate change, should be integrated into broad socio-economic policy. 4. Address Two Fields of Priorities: The project revealed two fields of priorities, that necessarily need to be addressed: improving the dialogue between natural and social sciences; and improving the dialogue within the social sciences.
7.5. Project Significance The Dutch project laid a solid basis for a continuation of the fruitful communication between the researchers attached to the NRP and the policy actors from all sectors of society. During the project it became clear that there is a vast body of knowledge available outside the scientific community. Different but valid perceptions exist about the various aspects of climate change and climate change policy outside the scientific community. Through the project, these have been initiated and can now serve as an important source of information both for policy actors and researchers.
8. INTERNATIONAL CLIMATE CHANGE PROJECT The success of the U.S. and Netherlands studies encouraged the development of a project applying a similar process at the international scale. The project on "Enhancing the Effectiveness of Research to Assist International Climate Change Policy Development" (International Climate Change Project) is designed to determine the range of uncertainties and information needs of decision makers in relation to global climate change in an international context [ 11]. The project also assesses the research needed to help answer the associated questions and facilitates dialogue between scientists and policy makers at the international level. S&PA, IVM, Prof. Jan Klabbers, and Professor Bill Moomaw of the Tufts University Fletcher School of Law and Diplomacy in the United States have undertaken a project designed to address these issues at the international level. The initial phases of the project are funded jointly by the Dutch NRP and the U.S. EPA.
115 8.1. Project Goals The goals of the International Climate Change Project are to: 9 Identify and scope the range of policy options under consideration by representative countries, for which future research information is needed. 9 Determine the research required to address the information needs relevant to the range of policy options identified and to help guide the planning of policy-relevant research. 9 Enhance the dialogue between the decision making and research communities at the international level for the climate issue. The international dialogue fostered by this effort will promote a better understanding between decision makers and scientists both nationally and internationally. 9 Facilitate the planning of research that is more relevant and usable by decision makers. Exercises such as this project make a lasting contribution to improving the utilization and linking of science with policy development.
8.2. Project Approach The project compliments the current international activities relating to climate change and explores the long-term policy questions that require research-based information needs to support the range of policy options identified. The International Climate Change Project utilizes and updates the results obtained from the similar studies in the United States and Netherlands. The project will be undertaken in three phases: 9 Phase I - Project Design and Analysis: Activities in Phase I covered project planning and design to formulate the scope and tasks of the project. A project Steering Committee provided guidance and recommendations on the scope of the project. Selection criteria for choosing participating countries were developed together with the procedures and approach for undertaking interviews and workshops in the subsequent phases. These selection criteria were designed to promote the selection of countries that would provide a wide range of policy options and the information and research needs to support these options. The Steering Committee considered five types of criteria identified by the project team: environmental, economic, political, cultural/geographical and feasibility. Initially four countries, in addition to the United States and The Netherlands were chosen to be included in this pilot project: Brazil, China, India, and Poland. 9 Phase H - Identification of Policy Options: The objective of Phase II is to determine the range of policy options under consideration by policy stakeholders in the climate change issue. This will be undertaken through in-country interviews with representatives from the participating countries in February and March 1995, including an update of the information obtained during the national projects in The Netherlands and the United States. An international decision makers' workshop will be held in June 1995 to elaborate on the range, motivations, and substance of policy options. The output of Phase 1I will be a final report detailing the policy options identified and other results of the interviews and workshop. 9 Phase III- Identification of Research Needs to Address Policy Options Identified and InterCommunity Dialogue: Phase III will present these options identified in Phase II to the
research management community from the participating countries through the presentation of briefing papers and a workshop. A dialogue between the policy-making and research communities will be established through round tables or another workshop. The purpose of
116 this exercise is to identify research priorities and agendas and consider their implementation. Output from Phase III will be a report identifying key areas of research to address the policy questions identified as priorities in Phase II, synthesizing the dialogue between the policymaking and research communities, and summarizing the key areas for research and information needs identified during this dialogue.
8.3. Project Significance The selection of Brazil, China, India, and Poland as pilot countries allows for an examination of climate change policy and research options in situations that are markedly different than those found in more developed countries. Integrating the findings of these efforts with those of the U.S. and Netherlands studies will provide a preliminary picture of how the decision-making and research communities can work together to address climate change at the global level. The process used in the project can be tailored to the needs of other countries to help them establish a dialogue between the science and policy communities.
9. CONCLUSIONS Developing policy-relevant research requires the involvement of both scientists and decision makers in framing the appropriate questions. Policy users of the research results must articulate their information needs and consult scientists on the feasibility of research providing meaningful answers. Scientists can examine those initial requirements to determine the limitations and strengths of investigation and to meet them within available budgets and time frames. An iterative process between the users and producers of the information is desirable to refine and then periodicall~ update the policy-relevant research questions as both the science and policy evolve.
10. REFERENCES 1 2 3 4 5 6 7 8
Enabling the Future: Linking Science and Technology to Societal Goals, Carnegie Commission, Washington, 1992. Environmental Research and Development: Strengthening the Federal Infrastructure, Carnegie Commission, Washington, 1992. Report of the Task Force on the Health of Research, Committee on Science, Space, and Technology, U.S. House of Representatives 102nd Congress, Washington, 1992. J.C. Bernabo, Science and Policy: Notes from a Former Congressional Fellow, in Proceedings of the American Geophysical Union, EOS, 7 (1986) 82. A.M. Weinberg, Science and Trans-Science, in Minerva 10 (1972) 207. H. Brooks, Expertise and Politics: Problems and Tensions, in Proceedings of the American Philosophical Society, 119 (1975) 257. H. Brooks, The Resolution of Technically Intensive Public Policy Disputes, in Science and Human Values, 9 (1984). J.C. Bernabo and P. Eglinton (eds.), Final Report of the Joint Climate Project to Address Decision Makers' Uncertainties, EPRI Technical Document No. TR- 100772 (1992).
117 9
T.B. Carter and K.D. Smythe, Biodiversity Uncertainties and Research Needs: Interim Report, Science & Policy Associates, Washington, 1993. 10 J. Klabbers, P. Vellinga, et al., Policy Options Addressing the Greenhouse Effect, National Research Programme on Global Air Pollution and Climate Change, Bilthoven, The Netherlands, 1994. 11 S.P. Hammond, J.C. Bernabo, et al., Project Plan for Enhancing the Effectiveness of Research to Assist International Climate Change Policy Development, Science & Policy Associates, Washington, 1994.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
119
CLIMATE CHANGE, POLICY O P T I O N S A N D R E S E A R C H IMPLICATIONS P. Vellingaa, M. HisschemSllera, J.H.G. Klabbersb, M.M. Berkc, R.J. Swartc and A.P. van Uldend
a
Institute for Environmental Studies, Vrije Universiteit (IVM/VU), De Boelelaan 1115, 1081 HV Amsterdam, The Netherlands
b
Klabbers Management & Policy (KPMC), Oostervelden 59, 6681 WR Bemmel, The Netherlands
c National Institute for Public Health and Environment (RIVM), P.O. Box 1, 3720 BA Bilthoven, The Netherlands d
Royal Netherlands Meteorological Institute (KNMI), P.O. Box 201, 3730 AE De Bilt, The Netherlands
ABSTRACT Policy options for climate change and their research implications are discussed in this paper. Instead of science telling policy actors what to do, this project started by asking policy actors how they perceived the climate change issue, how it could best be handled and what science can do to address their questions and concerns. Through a series of interviews and workshops five different options have been constructed and for each a corresponding research agenda has been developed. I m p o r t a n t findings of this project are, 1) it is easier to reach consensus about actions t h a n about the nature of the problem; 2) issue linkage is crucial as the problem of climate change is complex and the benefits of emission limitation are too remote to be the single motivator for action; 3) an i m p o r t a n t condition for progress in climate change policy is a strengthening of the science-policy interface. This project is an illustration of how this can be achieved. 1.
INTRODUCTION
There are m a n y questions surrounding the climate issue. The project on 'Policy options addressing the greenhouse effect' took a look at whether and in what way our society can cope with the risks of climate change and the challenge of sustainable development. A dialogue was initiated between scientists and the broadest range of policy actors such as members of parliament, governmental
120 policy-makers, representatives of trade unions, employers' organisations, business, environmental and consumers' NGOs etc. This dialogue was organised in such a way that the risk of climate change and opportunities to deal with these risks were dealt with simultaneously. The project was set up to bridge the gap between the perceptions of policy makers, the private sector, societal organisations and scientists. Its objectives were to: * enhance the communication between the various parties; * make the various perceptions of the problem visible; * explore options for policy development; * introduce the options into the Dutch policy formation process; * identify the information needs and related research strategies per option and discuss the results in the research community. 1.2 Steps towards identification and exploration of policy options Theproject has been carried out in a number of steps as indicated in Figure 1. Step one is a serie~ of interviews, which led to three hundred questions relevant to the problem, formulated by policy actors. Through a workshop with the policy actors the questions were articulated and the number was reduced to 35. In step two these questions were assessed by natural and social scientists. Next, in step three the results were fed back to the policymakers, private sector and societal organisations (all relevant policy actors) through round-table discussions. The results generated through this process formed the basis of a workshop of all parties, including the scientists. This meeting developed options for Dutch (long-term) policy aiming at sustainability in terms of solutions to the climate question. A variety of options and related actions were formulated ranging from no-regrets to social-cultural change. As a follow-up, round-table discussions were held in which policy actors identified information needs and research strategies for each of the options to be addressed by the research community. This paper describes the process, the results of the various stages, the various policy options and the related set of research strategies.
~'dentification~~ sc'ientific~~ policy of issues assessment linking * interviews and workshop to identify major questions
* position papers and workshops with natural and social scientists
* round-table discussions linking science and policy
Figure 1. Subsequent steps in the research project
policy options * elaboration of policy options, related actions and research implications
121 2.
I N T E R V I E W S AND ELABORATION OF INTERVIEW R E S U L T S
Rather than science telling policy actors what to do, this project started by asking policy actors what they thought about the climate issue, how it could best be handled and what science could do to address their questions and concerns. The project primarily focused on the longer term policy options, with 2025 as time horizon. The people interviewed took the opportunity to blow off steam about politics and the way the issue is handled by various governmental departments and about the scientific community, which should produce "consistent signals instead of generating ambiguity and controversy". Some examples of statements that several times cropped up in the interviews are listed in Box 1. 2.1 The m e s s a g e : t w o - w a y c o m m u n i c a t i o n With the statements available, two clear messages were conveyed to the scientific community. The first was that (perceived) controversy among scientists on the nature of the problem is the main hindrance in convincing decision makers and the public at large of the urgency of emission control. Scientists tend to focus on remaining uncertainties, rather than communicating what is known and agreed upon. It was accepted that progress in science thrives on controversy. It was stressed, however, that the dissemination of results should not reinforce already existing confusion about the greenhouse effect, since this would weaken the initial commitment to action. The second message was that scientists should not restrict themselves to working in their institutes and communicating their results only through scientific publications. It was stressed that a precautionary approach can only be based on a broadly shared understanding of the nature of the problem and that this can only be gained through active communication by the scientific community with the public at large. Likewise, the societal and technological science community should not just perform disciplinary desk studies: they should carry out a number of demonstration projects in which their claims about the feasibility of certain solutions can be demonstrated to the decision makers and the general public. Finally, the interviews made it clear that knowledge of climate change is not the monopoly of the scientific community. Among the people interviewed there was a very broad and often detailed knowledge of the greenhouse effect and of the various response strategies. 2.2 One g o v e r n m e n t , different v o i c e s On the basis of the interviews, it was observed that the various governmental departments had positioned themselves in different phases of the development of climate change policies. One department (Economic Affairs) was still in the phase of conceptualising the policy issue (Is there a problem?); two were in the second phase: having accepted the risks of the problem, exploring response actions and building coalitions (the d e p a r t m e n t of Agriculture, Fisheries and N a t u r e M a n a g e m e n t and the department of Transport and Public Works. A fourth department (Environment) was in the next phase: efforts are made to implement policies; societal groups are persuaded and resources are allocated to solve the problem. As a consequence, the government did not speak with one voice. The resulting inconsistency in environmental policy development caused confusion amongst actors. Some considered the policy measures too weak, while others thought they lacked any basis and went too far. The following associated response
122 p a t t e r n s were identified: reactive (defensive response to g o v e r n m e n t policy), receptive (receptive response to government policy), constructive (acceptance of one's own responsibility) and pro-active (internalising in one's strategic planning). 2.3 I d e n t i f i c a t i o n of i s s u e s , i n t e r v i e w s a n d w o r k s h o p Questions put to the policy actors were: 1. "What are your perceptions of the greenhouse effect?" 2. "What potential impact may the issue have on your organisation, on both a short and long term basis?" 3. "How is your organisation responding to it?" 4. "Can the research community help to address your questions?"
Examples of statements from the first round of interviews * The greenhouse problem is probably the biggest environmental problem that we shall face in the twenty-first century. * Early action on emission reduction is imperative. * The m a i n challenge is how to get everybody on board for the far-reaching measures t h a t will be necessary. * Climate change is a non-issue, pushed by science and embraced by politics * If it would eventually prove to be a problem, then adaptation would be the best strategy. * Even if the Netherlands were to be in favour of reducing emissions, a unilateral Dutch policy would never succeed because of the global scale of the issue. * Strong emission reduction measures in the N e t h e r l a n d s would w e a k e n the industrial sector in international competition. Box 1. Examples of statements from the first round of interviews. The research team grouped the questions and s t a t e m e n t s t h a t were g a t h e r e d through the interviews. Next, the results were articulated through a workshop with the policy actors. 3.
SCIENCE ASSESSMENT
3.1 S c i e n t i f i c c o n t r o v e r s y d i s c u s s e d The next step in the project was a logical consequence of the previous one. Position papers were drawn up around the questions from the previous phase, and these were then discussed in two working conferences in order to provide a scientific assessment of the enhanced greenhouse effect. The first conference gathered experts from the n a t u r a l sciences and focused on biogeochemical cycles, responses of the climate system to changes in greenhouse gas concentrations and the effects of climate change. The basic question put to the n a t u r a l scientists was: does the greenhouse effect exist and, if so, w h a t are the risks? Some controversial positions were discussed and evaluated, including a
123 recent report written by professor BSttcher, an outspoken dissenter from the 'climate consensus' in the Netherlands. The second conference included experts from the social sciences and focused on technological solutions, economic aspects of Dutch greenhouse policy options and psychological, sociological as well as philosophical/ethical aspects. The m a i n question which had to be answered was: how do you perceive the climate problem and how can we cope with it from the socio-economic, technological and behavioural points of view?
3.2 S c i e n t i f i c a s s e s s m e n t of t h e g r e e n h o u s e e f f e c t P l e n a r y sessions during both gatherings resulted in confrontations between r e p r e s e n t a t i v e s of different disciplines. These confrontations resulted from 'language problems' and differences in underlying assumptions among the related disciplines. After these 'language problems' were solved and assumptions were thoroughly discussed, a consensus was reached on a total of 90 statements. A few of these are listed in Box 2. The n a t u r a l scientists (the first three s t a t e m e n t s in Box 2) discussed and evaluated the (un)certainties related to the dynamics of the climate system. The experts from the social sciences were inclined to accept the problem and tried to find ways and means to deal with it. This is reflected in some of their statements.
S e l e c t e d s t a t e m e n t s f r o m n a t u r a l a n d s o c i a l s c i e n t i s t s a b o u t the g r e e n h o u s e effect * The concentration of greenhouse gases, CO2, CH4, and N 2 0 , CFCs and tropospheric ozone have increased since 1960 as a result of h u m a n activities. * Experimenting with the global climate is not a feasible option. Risk assessment should include the possibility of irreversible changes to the climate. * The greenhouse effect is only one of man's disturbances of the t e r r e s t r i a l system. If possible, i n s t r u m e n t s to reduce the greenhouse effect should therefore also reduce other disturbances. * Estimations of the effects of a substantial reduction of greenhouse gases show t h a t it is possible t h a t macroeconomic effects will be relatively small and sectoral relatively large (employment, profitability and production). * Sustainable lifestyles should be promoted as positive changes. As a general rule, it can be concluded that lifestyle changes need to be based on the three A's: they m u s t be Achievable, Acceptable and Attractive. * Sustainable technology requires a shift in ideological, cultural and societal values within society. Ultimately, individuals will have to find new modes of behaviour within the limits of the ecospace. Box 2. Selected s t a t e m e n t s from the n a t u r a l and social scientists about the greenhouse effect
124 4.
FIVE POLICY OPTIONS
4.1 Linking science and policy In the third stage of the project, the results of the scientific assessment were fed back to the policy actors. In six round table discussions the results of the scientific assessment were linked with policy and institutional actors. Participants included representatives of several ministries, the chemical industry, the electricity sector, t r a n s p o r t and agricultural organisations, political parties, trade unions, environmental NGOs and consumer organisations. The round table discussions once again revealed the wide variety of perceptions about climate change. This part of the project showed that policy and institutional decision makers in general accept the scientific statements. The debate primarily focused on the societal aspects of a range of climate change policies. This implied a change in the perceptions of the different actors; at this stage there was more convergence than there had been at the beginning of the project (workshop 1). Differences of opinions were primarily related to the proposed policies for dealing with the problem. Six round table discussions produced five rough drafts of policy options for dealing with climate change. These were further developed in five working groups. The policy options are: * N o regrets: it is uncertain whether climate change will occur and, if so whether
substantial reductions in greenhouse gas emissions will be necessary; this implies no action regarding climate. * L e a s t regrets: climate change is a most serious problem with potentially
irreversable effects. Since the effects as yet are unclear, a risk approach should be taken. A trade-off is made between risks linked with intervention and non-intervention; this implies action now; the uncertainties are an important motivator for pre-cautionary measures. * A c c e l e r a t i o n : climate change is a serious problem, but too complex to address
head on; the climate problem can best be addressed through generally recognized short term problems in related fields; the focus is on issue linkage and on synergies and positive feedbacks presently existing in society; this implies action now but only in the context of other issues. * Technological innovation: climate change is a serious problem; but technological
development is the only way to match the demands of an ever increasing world population with the carrying capacity of the environment; this implies action focusing on technology, research, development, demonstration and diffusion. * I n s t i t u t i o n a l ~cultural change: within this option it is assumed that technological
solutions will not be sufficient to reach a sustainable society. Major societal, cultural and institutional changes are required to create a sustainable society; this implies action in all areas, not necessarily related to climate. In this project the policy options are defined not as a single type of action or instrument, but as the whole of opinions and suppositions about the climate issue, the effects and the possible solutions. The five policy options represent five different mainstream perceptions present in society. The options are typical views
125 which r e p r e s e n t a mixture of problem perception and solution perception. An i m p o r t a n t observation in this project is t h a t the perception of the 'problem' per actor is strongly linked with the perception of the 'solution'. For each of the five policy options a n u m b e r of actions were formulated. The interesting result of the project is t h a t there is much more agreement on the type of actions t h a n on the policy options. It was possible to identify eight different fields of action t h a t were mentioned under all policy options. The main difference between the options is the intensity and geographic scale of the implementation of the listed actions. The 'common' fields of action are indicated in Box 3.
Common actions * Towards an eco-tax system. * Low carbon transport systems/infrastructure. * Energy efficient housing/offices. * Redesign of industrial processes and products. * Towards renewable energy sources and renewable materials. * Joint implementation. * Towards closing the substance cycles at smallest geographical scale. * Stimulate technological and cultural innovation. Box 3. List of common actions. 5.
RESEARCH IMPLICATIONS
5.1 I n t r o d u c t i o n The next step in the project was to investigate the research implications of the various options. For each option a round-table conference was held. In these meetings the information needs and related research areas (including their focus) were identified. The r e s e a r c h implications are discussed in the following p a r a g r a p h s per policy option. This part of the project has not yet been finished. Nevertheless, a n u m b e r of observations with respect to information needs and research implications can tentatively be made. 5.2 N o - r e g r e t s Within the no-regrets option it is considered uncertain whether climate change will occur and, if so, whether substantial reductions in emissions of greenhouse gases will be necessary. Priority is given to instruments t h a t serve other (socio-economic and e n v i r o n m e n t a l ) objectives, s i m u l t a n e o u s l y r e s u l t i n g in a reduction of greenhouse gas emissions. No-regrets i n s t r u m e n t s will, irrespective of climate change, pay off anyway. Key words for this policy option are scarcity and real
politics. - Climate change is considered not to be a real problem. However, scarcity of fossil fuel resources is a problem. Moreover, policies t h a t address scarcity are likely to get much more support. Simultaneously, such policies lead to a reduction of greenhouse gas emissions.
126 In addition, greenhouse gas emission reductions in the Netherlands, including high cost for the Dutch economy, would not have a significant effect on global greenhouse gas concentrations, since the biggest countries in the world continue to grow both with respect to their populations and emissions. - Finally, there is no strong indication that the Dutch economy would significantly suffer from a changing climate. -
The following information needs and research areas are identified by the policy actors supporting the no-regrets view. 1. Scarcity of fossil fuel and mechanisms that can help to increase the efficiency of resource use are important research fields. Mechanisms and institutional arrangements that help to remove intersectoral barriers should be investigated (such as full cycle management). 2. Possibilities and impossibilities of demand side management should be investigated. What are the demands of the people and what are their priorities. Research should be carried out in the fields of price-elasticity and into the question of whether people at all are willing to adjust their consumption patterns on the basis of uncertain long-term changes in the climate system that may have both positive and negative effects. 3. Global, particularly Third World, growth of fossil fuel use and C O 2 emissions should be investigated. Special attention should be given to ways and means to increase energy efficiency in developing countries. A special point of interest is research into the leakage of greenhouse gases during exploitation and transportation of oil and gas (the total global quantity of flared natural gas is equal to the total European consumption of fuels by cars). 4. Population growth and ways to control this growth is an important area for research. 5. Regarding climate system research, priority should be given to monitoring and process analysis. 5.3 Least
regrets
Within the least regrets option climate change is perceived as a serious problem with potentially irreversable effects. As the effects are unclear a risk approach should be taken. A trade-off should be made between risks linked with the occurrence and non-occurrence of climate change in relation to the policies selected. The policies include all no-regrets instruments supplemented with anticipatory policies aimed at limitation of risks resulting from climate change. Policies anticipate a substantial reduction of greenhouse gas emissions. If reduction proves to be unnecessary, part of the effort is lost. If reduction proves necessary, further reductions will be more efficient than they would have been if only a no-regrets policy was implemented. The least regret option thus includes hedging strategies. From its inception, this option offers a long-term perspective. Keywords in this option are probability and insurance. - In this option climate change is recognized as a real problem. However it is not known how large the problem is. Therefore it is necessary to take immediate
127 action to reduce the risks. Such actions can be seen as an insurance premium. The least regrets option is seen as a rational policy based on a quantitative risk analysis. - The N e t h e r l a n d s as an energy and emissions intensive country has the responsibility to take actions according to its historic and present contribution and according to its economic and technical capabilities. The information needs and research areas as identified by the policy actors supporting the least regrets option are the following. 1. Action research in communicating the risks of climate change and the range of anticipatory actions is important. Especially the question of how all societal actors can be encouraged to implement a least regrets approach, should be addressed. 2. Regarding the climate system, the most i m p o r t a n t t a s k is to describe the uncertainties in terms of probabilities. In particular, the possibility of non-linear behaviour of the oceans and of the sources and sinks of greenhouse gases should be studied. 3. R e s e a r c h into the effects of climate change should not j u s t look at the N e t h e r l a n d s . The p r i m a r y focus should be on the global ecological and socio-economic systems such as ecosystems, food production and e x t r e m e events. Adaptation research should investigate the possibilities of decreasing the vulnerability and thus increasing the robustness of society, infrastructure and other socio-economic systems. 4. Research is also required to increase the understanding of the response capacity of society in r e l a t i o n to climate change scenarios including s u r p r i s e s . I n v e s t m e n t cycles and rates of m a r k e t penetration of new technologies need to be studied for a range of climate change scenarios. A special field of research concerns the possibilities and the potential of lifestyle changes. 5. Research to investigate and develop a range of options for drastic emission control: both in the field of rapid implementation of existing technologies as in the field of development of new technologies (renewable energy sources, sink e n h a n c e m e n t ) . Simultaneously, research should be carried out for drastic emission control through institutional changes such as new fiscal regimes and new i n t e r n a t i o n a l regimes such as joint i m p l e m e n t a t i o n and t r a n s f e r of technology. 5.4 A c c e l e r a t i o n The acceleration option focuses on synergies and positive feedbacks presently existing in society. Forces and currents t h a t are consistent with climate change policy are accelerated and barriers are removed. All policies should take into account the different time cycles of society. Key words in this option are opportunities and issue linkage. - Also in this option climate change is considered to be a real problem, but the government and intellectual elite is not capable to convince the major economic actors and society at large to substantially invest in emission control measures.
128 The only way to achieve something is "hitch hiking": riding with other issues. The direction to go is clear, but the (climate) vehicle does not have enough power on its own to get things moving so any opportunity that comes along should be grasped. - This option takes other environmental and societal problems as its point of departure. The climate issue is thus linked to other problems such as: e m p l o y m e n t , congestion, technology co-operation, u r b a n i s a t i o n , individualisation.. The following information needs and research areas are identified by the policy actors that adhere to the acceleration option. 1. Climate system research should focus on the relations between the greenhouse effect, the effects of aerosols, acidification, ozone formation, the effects of land-use changes and the various problems related to the human interference with the bio-geo-chemical cycles. Hence, climate research should be fully embedded in the global change research. Research should identify the common sources of a range of environmental problems. 2. Impacts and adaptation research should start by identifying which natural and socio-economic systems have the attention of the government and the public at large. Particularly those systems should be systematically studied for the potential of issue linkage. Research should focus on measures that help to reduce the vulnerability vis-a-vis climate change but that are originally envisaged for other reasons. 3. Similarly, the emission control type of research should start with an analyses of the various relations between climate change limitation measures and the range of technological, infrastructural, economic and other presently perceived problems that society wants/needs to address: urbanisation (housing, work, recreation and infrastructure), employment, transport/congestion, energy supply, communication, fiscal regimes, liberalisation of energy markets in Europe, agricultural problems and land surplus, world trade arrangements, development co-operation, population growth, international debt etc.. 4. Systematic research into issue linkages and development of strategies based on issue linkage, not just in the technical sense but also, perhaps even more so, in the communication domain. 5. Research into the question of how the various actions primarily driven by other issues, can be orchestrated in such a way that the total result with regard to climate is satisfactory. Research on how the climate change momentum can be maintained in a policy strategy that primarily focuses on other problems. 6. Systematic research into what people value, with the aim to identify those issues and actions that have sufficient support for implementation. 5.5 Technological
innovation
According to this strategy, technological development is the only way to match the demands of an ever increasing world population with the environment's carrying
129 capacity limited. This requires a long-term co-operation between government and private enterprise. In this option it is required that government plays a very active role in directing technological development by providing opportunities and constraints (e.g. subsidies, fiscal incentives, regulations, etc.) to stimulate the required development. The key word in this option is innovation. - The technological innovation option builds on (i) the assumption that there is sufficient technological creativity available in our society to address the climate change problem without loss of economic welfare and (ii) on the notion that technology development and implementation can be accelerated by removing existing barriers and creating positive incentives. The information needs and research areas identified by the policy actors supporting this option are the following. 1. The various relations between the climate problem and other effects of resource use should be investigated, as it is of crucial importance that new technologies address all problems and not just one single (climate) problem while increasing or creating other problems. Similarly, research into climate effects and adaptation should be relatively broad and linked with research in other environmental fields for the development of a robust technology strategy. 2. Research into the time dimension of emissions, concentrations and climate change impacts is very important for the implementation of the right technology at the right time and for minimization of the overall cost for society (e.g. a hundred year strategy for emission control and technology development). I n v e s t i g a t e the possibilities of bifurcation problems (one type of technology/infrastructure precluding the implementation of a better technology at a later stage). 3. Investigate and identify the conditions that are optimal for technology development and implementation (adoption and diffusion): e.g. market, infrastructural, institutional and cultural conditions. 4. Research and development programmes in the field of energy efficiency (supply and demand side, renewable energy, materials and redesign of industrial processes (the potential of a shift from non-renewable fossil resources to agro/biological renewable resources), and energy systems research (centralized versus decentralized systems and storage systems). The research programmes to be developed should be based on careful analyses of (potential) competitive advantages for the Netherlands. 5. Research with the aim to identify the sectors/technologies for which small incremental changes can produce large results, for example avoidance of leakages, process integrated energy efficiency schemes, more efficient cars, land use management systems etc.
5.6 Institutional/Cultural Change Within this option it is assumed that technological solutions will not be sufficient to create a sustainable society. Social, cultural and institutional changes are required
130 to reach such a goal. Furthermore, it is assumed that sustainable development can only be achieved through processes of change within society. The role of government is limited to setting conditions and providing support. Changes are promoted through support of concerted actions within society, mobilization of social organizations and the removal of institutional barriers. The focus within this option is on the achievement of the desired situation (positive motivation) r a t h e r than on the avoidance of a non-desired situation (negative motivation). Keyword in this option is quality. Quality of life is the central theme in this option, as opposed to quantity and speed. The following information needs and research areas were identified by the policy actors supporting this option. 1. With regard to climate change research, it was noted that some research will be neces-sary in order to illustrate the impact of wasteful h u m a n activities on the life support systems. However, it should be realized that the mechanisms that are advocated in climate research, such as ever larger computers trying to predict the inherently unpredictable and conventions to manage this are part and parcel of the same societal systems that are causing climate change such as ever increasing global trade and transport. Since closing the material cycles at the smallest possible geographical scale is the aim under this option, not much research would be needed to illustrate that large scale fossil fuel use and large scale landuse changes are detrimental for the environment and thus should be avoided. 2. Effects and adaptation research should focus on the relation between social, cultural, economic systems and the local climate. This relation is probably more important than presently realized, research should investigate this. 3. Research into the question of what a sustainable lifestyle looks like. 4. Research into technological innovation in support of sustainable lifestyles: technology supporting a local closure of material cycles (including carbon and nutrients). This includes research into the institutional, infrastructural and cultural systems that support a sustainable lifestyle. Research into the driving mechanisms for unsustainable production and consumption. 5. Research into the phenomenon of defensive/compensating consumption (skiing as required to compensate for stress work and intensive travelling; far away eco-tourism to compensate for lack of nearby n a t u r a l ecosystems t h a t are destroyed to make way for international airports; driving your children to school because the traffic is to dangerous for them to walk or to go on bicycle). 6. Research into the higher order effects of institutional and technological changes both to explain what has happened in the past, what is happening at present and what may happen in the future. 7. Action research and local experiments to investigate the feasibility of different lifestyles. The aim is to demonstrate that a large diversity of social/technical configurations are possible within the domain of sustainable life styles. The idea
131 of social learning and the idea of demonstration is important in the design of these experiments. P a r t of this type of research should also be how such experiments can be encouraged through generic i n s t r u m e n t s or removal of (generic) b a r r i e r s (including research into w h a t we m a y learn from earlier idealistic/utopian movements and their ideals, including research into the role of elite behaviour as a change agent, and research in the role of examples as media/agents communicating the necessity of change). 6. C O N C L U S I O N S AND R E C O M M E N D A T I O N S The evaluation of the results of this research project has not yet been finalized. Still a number of conclusions can be drawn. One conclusion came forward rather clear: the interviews and the analysis indicate that it will be very difficult, if not impossible, to reach concensus on the nature and the seriousness of climate change. It is not likely t h a t one of the five options discussed above will emerge as a concensus option as the way the risk of climate change is perceived appears to depend strongly on the societal and economic interests of the policy actors and on the individual values. For example, all participants from the energy intensive industries were convinced t h a t IPCC is trying to fool t h e m and t h a t some politicians join this game out of publicity interests. They typically favour the no-regrets option. The majority of the actors from the private sector, with interests that are relatively neutral vis-&-vis climate change policy, are in favour of either a least regrets, an acceleration or a technological innovation policy. A small part of the policy actors are in favour of the socio-cultural change option. The conclusions and recommendations that can be formulated at this phase of the project are listed below. 1. Based on the results of the project the researchers believe t h a t it is more fruitful to seek consensus about actions t h a n to seek consensus about the nature of the climate issue. 2. For all p a r t i c i p a n t s technology is an i m p o r t a n t and valued p a r a m e t e r in addressing the climate issue. However, preferences about the type of technology and the role of technology in society may differ. Still, all participants, also the ones favouring social-cultural change, favour technological research. All the promoters of technology agree that the societal needs and concerns should play the major role in technology development. 3. With regard to the climate problem, it appears as if most of the actors are fairly well informed about the n a t u r e of the problem, although the interpretation of the information differs. It seems as if society as a whole is waiting for the scientists to come with the ultimate answer about the risks. However, for m a n y of the scientists the expected response of the climate s y t s t e m and the uncertainties involved are a major reason for action. 4. A large p a r t of the policy actors implicitly favour some kind of acceleration
132 policy. Issue linkage as a basis for common action looks like a promising approach to climate change policy. However, this reveals a paradox. The general thrust of the acceleration option is that society is not willing or not ready to address the climate issue in its own right. This is expressed by the statement often made that climate change measures should piggy back on other issues. The paradox is that other issues, for example fossil fuel scarcity, usually are not perceived as sufficiently urgent to generate effective and long-term policies. If the climate issue had not been raised, all energy efficiency programmes would have been stopped in the late 1980's. It is only because of the climate issue that combined heat and power could become a success in the Netherlands. Consequently, in order to develop and implement an effective strategy, existing problems and concerns in the society other than climate change, may need to be the starting point for greenhouse gas control policies. Nevertheless, the climate issue needs to be raised continuously to keep the action going. So climate change may serve as a long-term argument for change, while short-term problems and concerns are the practical, day to day motivator for business. It should be realized that both problems are real and both arguments are needed for coherent and long-term action. This implies that the framework convention for climate change should be regarded as a meta-level policy driver, whereas more sectoral agreements should serve as implementation agents of the broader goal of limiting greenhouse gas emissions. 5. Perhaps this also holds for the research agenda: climate change research and the related research for climate change policy, play an important role as an overarching and integrating element in many fields of research. Still, their role in guiding the overall research may not be sufficiently addressed yet. The overall agenda should probably focus on global environmental change including biodiversity, land use, resource use (energy, water, etc.) and demographic issues, while climate change research should be embedded in such a global change research agenda. 6. The project has revealed that crucial societal actors may have different perceptions about climate change and the desirability of certain responses. As a consequence, they also have different information needs. In order to adequately address these needs, it is important to involve various policy actors in the development and evaluation of climate change research programmes. In this respect it is recommended to give more attention to the science-society dialogue. 7. Finally, the project was evaluated as particularly fruitful for the scientists who participated in the workshops. Many of them had never before exchanged ideas beyond existing disciplinary boundaries. This project revealed that scientists from different disciplines may come to opposite conclusions with regard to the feasibility of controlling greenhouse gas emissions. In most cases it was apparent that it is not the scientific elaborations that cause the differences, but the differences in paradigms and related assumptions that are taken for granted a priory. It is therefore recommended to stimulate multidisciplinary dialogues as a formal part of climate and global change research programmes. The effectiveness of climate change research can particularly be enhanced when also the policy actors are involved in the multidisciplinary dialogue.
133 7. D I S C U S S I O N ON " D E V E L O P M E N T OF (POTENTIAL) P O L I C Y O P T I O N S IN T H E N E T H E R L A N D S " Rapporteur: P.A. Boot Ministry of Economic Affairs, P.O. Box 20101, 2500 EC Den Haag, The Netherlands This research project did have four goals: 1) to enhance the communication between policy makers, third parties and scientists; 2) to identity and explore a range of options; 3) to input these options to the Dutch policy making process and 4) to generate a series of questions and concerns for the second phase of NRP. The activities of the project took place in a dialogue between policy and research, between policy makers themselves and researchers of different disciplines. It was organised as follows. In interviews, policy makers were asked to identify the problem. This resulted in some 300 questions for scientists. In workshop-I these were reformulated into 50 questions. Together with several position papers from scientists these questions were input into two workshops: workshop II A for the n a t u r a l scientists and workshop I I B for the social scientists and technological experts. The natural scientists felt obliged to re-arrange some of the questions to bring them more in agreement with their views. They reached consensus on 41 s t a t e m e n t s about observations on and possible effects of changes in the climate system. In the workshop of the social scientists (IIB), it proved to be more difficult to create understanding among sociologists, economists and philosophers, because of different presuppositions and disciplinary frames of reference. Once they reached consensus on the assumptions underlying the statements, they were able to formulate 49 s t a t e m e n t s conveying their joint views on causes, impacts and solutions. However, a straightforward scientific assessment could not be given and was even considered to be impossible. The statements underlined the necessity to strive for a common reference framework. The results of the workshops II A and B were fed back to policy makers and third parties during Round Table meetings. During this stage rough drafts of five policy options were formulated, which served as input for the final workshop III. This workshop did not aim at recommendations, but at the formulation of internally consistent policy options. Different perspectives on climate change response strategies were grouped into five options: 1. No regrets. In this option priority is given to i n s t r u m e n t s t h a t serve other objectives and which simultaneously result in reduction of greenhouse gas emissions. 'Climate' is no issue, and even if it eventually would prove to be a problem, then adaptation would be the best strategy. Because of the eventual scarcity of fossil fuel reserves, energy efficiency is worthwile to strive for, however. Keywords: Realism, Scarcity. 2. Least regrets. A climate problem exists, but what it looks like is uncertain. We have to m a n a g e risks. Policy has to anticipate substantial but no absolute reduction and has to provide a long-term perspective. Science m u s t support action. Keywords: Probability, Insurance.
134
3. Acceleration. Climate problems are tackled by linking it with other issues. Policy has to focus on synergies. Measures developed to contribute to other environmental and societal problems should be strenghtened in order to address climate change. Climate problems are used to solve other problems. Acceleration might be called an active form of hitch hiking or active version of no-regrets, as the climate problem is accepted as a problem. Keyword: Opportunities, Issue linkage. 4. Technological innovations. This option is based on the a s s u m p t i o n t h a t technology is the only way to match the demands of an ever increasing world population with the carrying capacity of the environment. A long-term cooperation between g o v e r n m e n t s and private e n t e r p r i s e s is required. Governments have to provide opportunities and remove constraints in order to enhance R, D & D and implementation. However, innovation itself comes from the private sector. Keyword: Innovation. 5. Institutional and cultural change. Technological solutions will not be sufficient to create a sustainable society. F u n d a m e n t a l changes are required. Societal organizations have a potential that has to be fully mobilized. Keyword: Society, Quality of life. Each of these options is most cost-effective in its own way of thinking. It might be impossible to strive for a narrowing of the range of these basic beliefs, but a promising similarity in actions may be observed t h a t are proposed by the representatives of different options. These possible actions range from low carbon t r a n s p o r t s y s t e m s and energy efficient buildings to ecotaxes and j o i n t implementation. Of course there is a difference in timing and scope of the proposed actions, but the fundamentals differ less with regard to actions that are considered useful than with regard to basic beliefs. A continuing improvement in inter- and intradisciplinary communication among scientists and policy makers is necessary to look for a more solid foundation for action. The way forward leads to utilization of shared opportunities, more than to a possibly fruitless search for consensus in basic beliefs. However, scientists and policy m a k e r s should more actively communicate those viewpoints they agree upon. The general idea in the workshop was that this important project should be known better internationally. Ideas on an international 'repetition' with representatives from different groups of countries exist. Learning-by-doing might be a fruitful way of constructing knowledge for both scientists and policy makers.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
STABILIZING GREENHOUSE CONSEQUENCES
GASES:
GLOBAL
135
AND
REGIONAL
Joseph Alcamo*, Maarten Krol, Rik Leemans E n v i r o n m e n t a l Forecasting Bureau, National Institute of Public Health and the Environment, P.O. Box 1, 3720 BA BILTHOVEN, The Netherlands *With Contributions From: Andr~ van Amstel, Johannes Bollen, Gert J a n van den Born, Alex Bouwman, Kees Klein Goldewijk, Eric Kreileman, Jelle van Minnen, Jos Olivier, Sander Toet, Bert de Vries, G~ Zuidema ABSTRACT This p a p e r assesses the environmental consequences of two t a r g e t s for C02 stabilization: 350 ppm by the year 2150 (367 ppm by 2100), and 450 ppm by 2100. As a tool for this investigation we use the IMAGE 2 integrated model of climate change. It was found t h a t these targets lead to much lower regional impacts on crop productivity, natural vegetation, and sea level rise as compared to the baseline case. Nevertheless some negative impacts do occur, and to further reduce these impacts would require more stringent stabilization targets. It was also found t h a t to achieve these stabilization targets in the atmosphere, global emissions should not s u b s t a n t i a l l y increase at any time in the future, and eventually they must be significantly reduced. 1. I N T R O D U C T I O N Article 2 of the F r a m e w o r k Convention on Climate Change proclaims the goal of achieving "stabilization of greenhouse gas concentrations in the atmosphere t h a t would prevent dangerous anthropogenic interference with the climate system." The purpose of this brief report is to review some of the consequences of two scenarios for stabilizing greenhouse gas concentrations. It is thought t h a t this information can be used in the process of selecting i n t e r n a t i o n a l policies for complying with the objectives of the Convention. Our analysis concentrates on two scenarios in p a r t i c u l a r because they have been adopted for study by Working Group I of the IPCC, as will be explained later. Our analysis draws on results of the IMAGE 2.0 model, an integrated model of climate change and the global environment1. Information about IMAGE 2.0 is given in Appendix 1. 2. W H A T W I L L H A P P E N IF NO A C T I O N IS TAKEN? In order to evaluate scenarios for stabilizing greenhouse gases, a baseline is needed for comparison. Our baseline scenario uses i n t e r m e d i a t e a s s u m p t i o n s about
136 population and economic growth.2 We note that this is not meant to be a "most likely" scenario. This scenario also assumes that no actions are taken to mitigate climate change; this allows us to estimate the possible incremental improvements that could come from stabilization versus no action. Under baseline (i.e. no action) conditions, the IMAGE 2.0 model computes that by 2100 global CO2 emissions could reach 24 Gt C/yr (within the range of IPCC emission scenarios3) and global average CO2 concentration 777 ppm. At the same time global average surface temperature could increase by 2.50C between the years 1990 and 2100. During the same period, temperatures increase by about 1.80C in the tropics and around 3 to 50C in the high latitude regions (Figure 1). Such changes to temperature and precipitation could lead to a variety of impacts. We focus on three in this p a p e r - - changes in crop productivity, disturbance of natural vegetation patterns, and sea level rise. These were selected because they are related to risks to food production, ecosystems, and economic development, which are the three impacts specifically mentioned in Article 2 of the Convention. 8
7~
" - - Base
/
-'"
/
$450
6
/
$350
.x:. ~5
~4
\
~3
/
...-"-
E2
1
,,.,
I : ;""
0" -90
. . . .. .. .. .
~'fJ1.,,''"
-60
South pole
"~i'-" "','",", .--: ..... ;
,
,
,"
-30
0
degrees latitude
30
60
90
North pole
Figure 1. Zonal Temperature Increase. The increase in surface air temperature computed by the IMAGE 2.0 model. The average temperature increase in 100C latitudinal zonal bands is given. Crop Productivity. As a result of baseline changes in climate, large portions of currently cultivated areas could experience reductions in crop yields. As one example, the potential rainfed productivity of wheat could be substantially decreased in 32% of current wheat growing areas between the years 1990 and 2100 (Figures 2a and 3a). During the same period, millet productivity could be substantially reduced in 37% of current millet growing areas (Figures 2b and 3b).
137
~ IncreasinYigeld
~~j StableYield
II
Decreasing
Yield
Figure 2 A and B. Changes in Crop Yield of Current Crop Growing Areas According to Baseline Scenario: (a) Wheat, (b) Millet. Shown are "substantial" decreases or increases in the potential rainfed productivity of w h e a t and millet over the period of the simulation, 1990 to 2100. Substantial is defined as follows: For wheat -- Substantial is taken as an increase or decrease of 0.5 t/ha/yr or more. This amounts to a roughly 10% change in the c u r r e n t level of potential rainfed productivity in c u r r e n t wheat-growing areas. For comparison, the current net yield of w h e a t is s u b s t a n t i a l l y lower -- 2.6 t/ha/yr, globally averaged. (Agrostat PC, FAO, Rome, Computerized Information Series no 1, October, 1992). Note that impacts on only current wheat growing areas are shown. New areas might be become productive for w h e a t u n d e r climate change. This is of course a very limited definition of risk to wheat growing areas, but does indicate where there is increased risk to production in current areas. For millet-- Substantial is t a k e n as an increase or decrease of 0.25 t/ha or more. This threshold is set lower t h a n wheat because millet is grown more often t h a n wheat by subsistence farmers who obtain low net yields. Indeed the current net yield of millet (0.8 t/ha/yr globally averaged, FAO, 1992, op cit.) is substantially lower t h a n t h a t of wheat. Hence, a smaller change in potential productivity for millet is of importance. It should be noted t h a t these calculations do not take into account the possible CO2 fertilization effect which could increase future yield estimates.
138 During the same period, millet productivity could be substantially reduced in 37% of current millet growing areas (Figures 2b and 3b). On the other hand, potential yield may increase in other areas, although this will not necessarily compensate for the disruption in yield elsewhere. The main areas affected would be current wheat growing areas in China, Western Europe, and parts of North America; and millet growing areas of Africa, the Middle East, India, and China (Figure 2). Area with decreasing yield of wheat
35
Base - - ' " S450 - - - -
,~ 3o~ 25c ~ 20 ~
....
iiiiiiiiiiiiii
- - - -
,,.:......
15-
..
..............................
j.J~'" I0-
2
,P~ "
5-
0
1o.
~ooo
~o~
~o~o
~o,~
~1oo
time in years
40
3s-
Area with decreasing yield of millet -----
30-
Base $450 S350
~ 25-
2o"1"t "
r lO-
t. ~ j~ ,t"o."
o
1990 2000
o................................
2025
20150
2075
2100
time in years
Figure 3. Changes in Cultivated Areas Affected by Decreasing Yields" (a) Wheat, (b) Millet. Shown are the currently cultivated areas with "substantial" decreases in potential rainfed productivity over the period of simulation, 1990-2100. As in Figure 2, "substantial" is taken as a decrease of 0.5 t/ha/yr or more of wheat, and 0.25 t/ha/yr of millet.
Natural vegetation. IMAGE 2.0 calculates global potential vegetation patterns by determining the occurrence of different plant types such as needle and broadleaved trees, shrubs and grasses. Each plant type has it typical distribution as a response to local climate and soil characteristics. Using this approach it was estimated that the baseline climate change would change the potential vegetation
139 in 42% of t h e w o r l d ' s l a n d a r e a by t h e y e a r 2100 (fig. 4), a n d in 44% of its c u r r e n t n a t u r e r e s e r v e a r e a s (Table 1). C o n s e q u e n t l y , t h e c u r r e n t n a t u r a l v e g e t a t i o n in t h e s e a r e a s will n o t be well a d a p t e d to t h e s e c h a n g e d c l i m a t e conditions. C h a n g e s of v e g e t a t i o n a t s u c h a l a r g e s c a l e could l e a d to s e v e r e d i s r u p t i o n of n a t u r a l vegetation succession, the main process through which vegetation can respond a n d a d a p t to n e w conditions. T h e s e c h a n g e s will t h e r e f o r e i m p a c t s t r o n g l y on local a n d r e g i o n a l biodiversity.
•
Soco i-Econom Facictors
I
ClimatC ehange
I
Combn ied
Figure 4. Threat to Natural Vegetation According to Baseline Scenario (1990-2100). Changes in natural land cover stemming from two main factors (i) socio-economic, (2) climate change. "Socio-economic" refers to current areas of natural vegetation that may be used for new agricultural land or fuelwood to satisfy the future food and fuel demands of the baseline scenario. These agricultural demand and land cover calculations are described in: (i) Alcamo, J., van den Born, G.J., Bouwman, A.F., de Haan, B., Klein Goldewijk, K., Klepper, O., Leemans, R., Olivier, J.A., de Vries, B., van der Woerd, H. and van den Wijngaard, R., 1994. Modeling the global society-biosphere-climate system, Part 2: computed scenarios. Water, Air and Soil Pollution, 76: 37-78, and (ii) Zuidema, G., van den Born, G.J., Alcamo, J. and Kreileman, G.J.J., 1994. Simulating changes in global land cover as affected by economic and climatic factors. Water, Air and Soil Pollution, 76: 163-198. "Climate change" refers to areas in which the potential vegetation is estimated to change because of climate change. The potential vegetation calculations employ a global vegetation model, "BIOME", described in: Prentice, I.C., Cramer, W., Harrison, S.P., Leemans, R., Monserud, R.A. and Solomon, A.M., 1992. A global biome model based on plant physiology and dominance, soil properties and climate. Journal of Biogeography, 19: 117-134. The model BIOME is embedded in IMAGE 2.0 as described in: Leemans, R. and van den Born, G.J., 1994, Determining the potential global distribution of natural vegetation, crops and agricultural productivity. Water, Air and Soil Pollution, 76: 133-161.
140 Climate is not the only factor t h a t will threaten natural vegetation patterns, and thus biodiversity. Another major factor will be the expansion of agricultural land stemming from population and economic growth (which will occur despite the more intensive use of current agricultural land). This is taken into account by IMAGE 2 in all scenarios (Figure 4). According to the baseline scenario, 23% of the world's c u r r e n t n a t u r e reserve areas m a y be t h r e a t e n e d by a g r i c u l t u r a l expansion between 1990 and 2100 (Figure 4). Also according to baseline calculations, 12% of the world's n a t u r a l reserve areas may be threatened by both climate change and agricultural expansion during this period. This includes large areas of Africa and Asia. The main point is that there is a close connection between policies to address climate change, world food production, and land use, and they will have overlapping effects on the world's natural vegetation cover and its level of biodiversity. Threat to natural vegetation in nature reserves due to climate change
-_.: s4Baseo
4540"13 r
35-
r
30-
t,_ r
(D 0
2520-
O) 1 5 t-
ml00!,-
0.. 5 0
I
1990 2000
I
I
I
2025
2050
2075
21 O0
time in years
Figure 5. Area of Nature Reserves Affected by Climate Change. Shown is the area of n a t u r e reserves where potential vegetation changes because of climate change. Calculations are performed as in Figure 4 and are then overlayed with the current location and area of nature reserves.
Sea Level Rise. Another consequence of not acting to mitigate climate change will be sea level rise due to melting of glaciers and ice caps, and thermal expansion of sea water. By year 2100 sea level is computed to be 20 to 60 cm higher t h a n in 1990, depending on location (Figure 6). Much of South Asia's coastline m a y experience a sea level rise of between 25 to 30 cm. Island states in the Caribbean could experience a sea level rise of the same magnitude, and those in the South Pacific between 20 to 25 cm (Figure 6).
141
m40.0.6o.0 30,0.40.0 ~20.0.3o.0
----~:~ ~::~ii ............. iiii~~ !7::
/~.0.
~.0
i~.o.~, o
,+~71~
~b,~:,,L
I
~:~ii::ii~i~7~iii~: i!
.':'.:~'. .;:.i. .i.iT.ii:i:i:i:i:.7..ii,lii~,:,~/~ iii!L::i~ii:i:ii!!!il;.....:::~i~.~:i.i.:.:.:~i . .i::: ........'~il~:~7!ii~
i i~i:::Li:iTii::ii,,,,,:~,:~:,,:i,,i::::}% .....
,:
~20.O -30.0 Figure 6 A and B. Regional Sea Level Rise Between 1990 and 2100 Corresponding to Baseline Scenario: (a) North America, (b) Asia and Oceania. Shown are mean increases in sea level over the period 1990 to 2100. These calculations take into account melting of ice caps, glaciers, and regional differences in the t h e r m a l expansion of sea water. They do not take into account the shifting of ocean currents nor differences in coastal wind velocity t h a t may accompany climate change.
142 2. W H A T
STABILIZATION
SCENARIOS
ARE CONSIDERED?
One way to mitigate climate change would be to stabilize the levels of CO2 and other greenhouse gases in the atmosphere. Results from the IMAGE 2.0 model show t h a t this could be an effective approach, depending on the target level and date of stabilization. In this p a p e r we examine two t a r g e t scenarios of stabilization: 9 CO2 stabilized at 350 ppm in 2150 (reaching 367 ppm in 2100). 9 CO2 stabilized at 450 ppm in 2100. These scenarios are of interest from the policy standpoint because CO2 would stabilize at about its current level (around 354 ppm), or moderately above this level. These scenarios were also p a r t of an i n t e r n a t i o n a l modeling exercise sponsored by Working Group I of the IPCC.4 For both scenarios, the atmospheric levels of CO2 are assumed to follow a smooth pathway from 1990 to their future t a r g e t date and concentration. Other greenhouse gases are also a s s u m e d to stabilize within this time frame.5 Global sea level rise
25-
///"
---- Base ----$450 9 9 9 S350
/
/
20-
t"
E 15.o. .I
._~ -6 lO-
j" .I"
j
.d"
t
I
o.
."
.t"
I .o~
.."
.,"
."
,"
eb 5-
0 1990 2000
~o~
~oso
~o~
~,oo
time in years
Figure 7. Global Average Sea Level Rise. Same calculations as in Figure 6, but averaged for the globe.
3. H O W
EFFECTIVE
ARE THE STABILIZATION
SCENARIOS?
Because of the unavoidable uncertainties of these model estimates, it is more informative to examine the relative differences between the baseline and the stabilization scenarios (Figures 1, 3, 5, 7, 8 and Table 1) r a t h e r t h a n their exact numbers. The stabilization scenarios have the following effects: 9 Regional t e m p e r a t u r e increases are subtantially smaller t h a n the baseline scenario. 9 The crop and n a t u r a l vegetation areas affected by climate change do not increase after 2050, whereas they do in the baseline scenario. 9 The total amount of area affected by climate change is significantly lower t h a n in the baseline scenario.
143 9 Sea level continues to rise beyond 2100 despite CO2 stabilization (it also does in the baseline scenario). This is because of the slow response time of the atmosphere-ocean system. 9 However, the r a t e of sea level rise is much lower than in the baseline scenario. These results show t h a t the stabilization scenarios have an overall lower negative impact t h a n the baseline. However, they also show t h a t they are not "risk-free". Impacts still occur because it takes several decades to stabilize greenhouse gases in the atmosphere, and in the meantime climate change occurs. To further reduce these impacts it would be necessary to adopt even more stringent stabilization targets.13 4. E M I S S I O N L E V E L S TO A C H I E V E STABILIZATION OF C O 2
A key question is how to achieve the stabilization of CO2 and other greenhouse gases in the atmosphere. Specifically, w h a t level of emissions would be allowed from the world's energy and industrial system? After accounting for the uptake of CO2 by vegetation and the ocean, this has been e s t i m a t e d by several global models as p a r t of an IPCC Working Group I exercise6. Results from the IMAGE 2.0 model are shown in Figure 8, and are consistent with results from other models7: 9 In order to stabilize CO2 levels by 2150 at 350 ppm, it will be necessary to i m m e d i a t e l y stabilize and t h e n s h a r p l y reduce global e n e r g y / i n d u s t r i a l emissions towards the end of the 21st century. 9 For the a l t e r n a t i v e scenario of stabilizing CO2 at 450 ppm by 2100, global energy/industrial emissions will be allowed to increase slightly above current levels, and t h e n m u s t be significantly reduced after the middle of the next century. Global CO2 emissions
25
Base
-.-s45o .
~ O15-
...s3
o
~ ~ . . . .
~ " ~ ....
/
/
I
I
I -- _|: ___ I
ffl e(~.~E ~s 110
1
~ .....
~ ~ " ~ ~ ~ " ""
~ ~ 9 ~
~ !
o
I
~176176
~176176 ~176176176
time in years
Figure 8. Allowable Global Emissions from E n e r g y / I n d u s t r y to Achieve CO2 Stabilization Goals. Shown are global CO2 emissions from energy and i n d u s t r y only (land use emissions are not included).
144
Put another way, large increases in emissions would be unacceptable at any time for either scenario. This is an i m p o r t a n t point because the allowable global emissions for stabilizing CO2 and other greenhouse gases are far lower t h a n baseline emissions (Figure 8). In the absence of policy measures, emissions are expected to sharply increase along with economic development in developing c o u n t r i e s 8 . Hence t h e r e exists a large "policy gap" between the allowable emissions for stabilizing greenhouse gases, and the emissions that will occur if no action is taken. The last issue to be raised in this report is whether emission strategies can be found to achieve the stabilization scenarios. It is possible t h a t some proposed global energy scenarios, for example from Johannson, et al.9, Shell 10, or Working Group II of the IPCCll, produce emissions low enough to achieve the stabilization scenarios. This is a key unresolved issue that needs to be resolved by the research community and reported to policymakers.12 5. S U M M I N G U P This brief p a p e r highlights some of the consequences of two scenarios for stabilizing greenhouse gases: (i) CO2 stabilized at 350 ppm in 2150 (367 ppm by 2100), (ii) CO2 stabilized at 450 ppm in 2100. Among its main findings: 9 To achieve t h e s e s t a b i l i z a t i o n ~ t a r g e t s , emissions are not allowed to substantially increase at any time, and eventually they m u s t be significantly reduced. : 9 Because of the current upward trend in global emissions, there is a large policy gap between the allowable emissions for stabilizing greenhouse gases, and the emissions that will occur if no action is taken. 9 Stabilization scenarios lead to much lower impacts on crop productivity, natural vegetation, and sea level rise as compared to the baseline case. 9 Although the stabilization scenarios show lower impacts t h a n a baseline, they are not "risk-free". Some impacts do occur, and to further reduce these impacts would require more stringent stabilization targets. 9 With regards to threats to natural vegetation and biodiversity, there is a strong need to connect policies that address climate change, world food production, and land use.
6. A C K N O W L E D G E M E N T S
The IMAGE Project is supported by the Dutch Ministry of Housing, Spatial P l a n n i n g and the E n v i r o n m e n t (VROM), and the Dutch National Research P r o g r a m m e on Global Air Pollution and Climate Change (NRP). This paper was p a r t l y funded u n d e r NRP contracts 853129, 853130, 853131, and 853132. An earlier version of this paper was prepared as a background report for the Dutch Delegation to the F i r s t Session of the Conference of P a r t i e s to the U.N. F r a m e w o r k Convention on Climate Change, Berlin, Germany 28 March - 4 April, 1995. Authors are grateful to M.M. Berk, B. Liibkert-Alcamo, B. Metz and R.J. Swart for reviewing this manuscript.
145 A p p e n d i x 1. O v e r v i e w of t h e IMAGE 2.0 m o d e l ES~ROV- ~NDVSrRY S~VmH ii .... i .......... i i.......... ~
i ,.,-=., L__J
J
EL
ouu, A r ~ s p , E~.~:~AN sYstEM
Input
' IP | " G~oo~ho~o G ~ I
"-* ~A
!
.oooo;
h " ~X,~ Zonal |
I
L gases
cl
.
L.ControlPolicies
.J
=~
TERRESTRIAL ENV1RONMENT SYSTEM
J
~~lj aiospher e
.O~an
"GreenhouseGase~
- J "Ozone-.re/ated | L gases ___j
! !X \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
UF~ba=s
J
i!.;;,~.......~......
La,d u= ! .........:~"~ .']~ ~'::
rEm'~~
]
/ "~176 L gases
/ J
"~ r
NewLandUse
LPattern
F Ao~u,u~
I Impac~ / Ecosystemand
-I
LOther Risks
Figure 9. Box Diagram of the IMAGN 2.0 Model. Each box represents a submode]. The IMAGE 2.0 model is a multi-disciplinary, integrated model designed to simulate the dynamics of the global society-biosphere-climate system. The objectives of the model are to investigate linkages and feedbacks in the system, and to evaluate consequences of climate policies. Dynamic calculations are performed from year 1970 to 2100, with a spatial scale ranging from grid (0.50 x 0.50 latitude-longitude) to world regional level, depending on the sub-model. The model consists of three fully linked subsystems: Energy-Industry, Terrestrial Environment, and Atmosphere-Ocean. The Energy-Industry models compute the emissions of greenhouse gases in 13 world regions as a function of energy consumption and industrial production. End use energy consumption is computed from various economic/demographic driving forces. The Terrestrial Environment models simulate the changes in global land cover on a grid-scale based on climatic and economic factors, and the flux of carbon dioxide and other greenhouse gases from the biosphere to the atmosphere. The Atmosphere-Ocean models compute the buildup of greenhouse gases in the atmosphere and the resulting zonal-average temperature and precipitation patterns. The fully linked model has been tested against data from 1970 to 1990, and after calibration can reproduce the following observed trends: regional energy consumption and energy-related emissions, terrestrial flux of carbon dioxide and
146 emissions of greenhouse gases, concentrations of greenhouse gases in the atmosphere, and t r a n s f o r m a t i o n of land cover. The model can also s i m u l a t e current zonal average surface and vertical temperatures. For f u r t h e r i n f o r m a t i o n consult: Alcamo, J. (Editor), 1994. I M A G E 2.0: Integrated Modeling of Global Climate Change. Kluwer Academic Publishers, Dordrecht, 318 pp. Also published as Special Issue of Water, Air and Soil Pollution, 1994. Volume 76, Nos 1-2.
147
Table 1. Regional I m p a c t s of Climate Change. Region !
Area With Substantial D e c r e a s e in W h e a t Yielda
Area With S u b s t a n t i a l Decrease in Millet Yieldb (% of C u r r e n t Millet Areas)
(% of C u r r e n t W h e a t Ares)
Areas of N a t u r e Reserves With C h a n g e in Potential Vegetation Due to Climate Changec (% of C u r r e n t N a t u r e R e s e r v e Areas)
1
2
3
1
2
3
1
2
3
Global
32
24
16
37
21
8
44
17
9
Canada
68
62
48
-
-
-
66
34
16
USA
59
52
43
65
52
33
68
31
16
Latin America
26
6
2
53
27
1
32
13
9
Africa
16
11
3
46
19
3
42
12
7
OECD Europe
46
37
27
38
38
38
67
38
26
Eastern Europe
35
31
21
43
44
30
58
43
27
CIS
10
9
5
18
16
11
48
23
9
Middle East
14
11
6
6
4
1
16
6
3
India + S. Asia
2
1
1
40
15
2
43
9
4
China + C.P. Asia
43
21
3
22
20
12
37
18
10
E a s t Asia
-
-
-
4
2
0
31
12
8
45
20
11
52
46
3
56
14
8
Oceania
N O T E S for TABLE 1 = Baseline Scenario 2 = Stabilization of CO2 at 450 ppm in 2100. 3 = Stabilization of CO2 at 350 ppm in 2150. a I n d i c a t e s t h e a r e a s t h a t e x p e r i e n c e a " s u b s t a n t i a l " d e c r e a s e in t h e p o t e n t i a l r a i n f e d p r o d u c t i v i t y of w h e a t over t h e period of t h e s i m u l a t i o n , 1990 to 2100. S u b s t a n t i a l is defined as a d e c r e a s e of 0.5 t/haJyr or more. This a m o u n t s to a roughly 10% change in the current level of potential rainfed productivity in c u r r e n t
148 wheat-growing areas. For comparison, the current n e t yield of wheat is substantially lower-- 2.6 t/ha/yr, globally averaged. (Agrostat PC, FAO, Rome, Computerized Information Series no 1, October, 1992). Note that impacts on only current wheat growing areas are shown. New areas might be become productive for wheat under climate change. bIndicates the areas t h a t experience a "substantial" decrease in the potential rainfed productivity of millet over the period of the simulation, 1990 to 2100. Substantial is defined as a decrease of 0.25 t/ha/yr or more. This threshold is set lower than wheat because millet is grown more often than wheat by subsistence farmers who obtain low net yields. Indeed the current net yield of millet (0.8 t~a/yr globally averaged, FAO, 1992, op cit.) is substantially lower than that of wheat. Hence, a smaller change in potential productivity for millet is of importance. eIndicated are areas in which the potential vegetation is estimated to change because of climate change over the period of simulation, 1990 to 2100. The potential vegetation calculations employ a global vegetation model, BIOME (Prentice, I.C., Cramer, W., Harrison, S.P., Leemans, R., Monserud, R.A. and Solomon, A.M., 1992. A global biome model based on plant physiology and dominance, soil properties and climate. Journal of Biogeography, 19: 117-134.), which is embedded in IMAGE 2.0 (Leemans, R. and van den Born, G.J., 1994. Determining the potential global distribution of natural vegetation, crops and agricultural productivity. Water, Air and Soil Pollution, 76: 133-161.) ENDNOTES
1 The IMAGE 2.0 model used for calculations in this report is fully documented in Alcamo, J. (Editor), 1994a. IMAGE 2.0: Integrated Modeling of Global Climate Change. Kluwer Academic Publishers, Dordrecht. Also published as Special Issue of Water, Air and Soil Pollution, 1994. Volume 76, Nos 1-2. 2 The baseline scenario is based on the Conventional Wisdom scenario documented in: Alcamo, J., van den Born, G.J., Bouwman, A.F., de Haan, B., Klein Goldewijk, K., Klepper, O., Leemans, R., Olivier, J.A., de Vries, B., van der Woerd, H. and van den Wijngaard, R., 1994b. Modeling the global society-biosphere-climate system, Part 2: computed scenarios. Water, Air and Soil Pollution, 76: 37-78. This scenario takes population and economic growth assumptions from the intermediate emissions scenario (IS92a) of the IPCC (1992). The population assumptions correspond to median estimates of the U.N. F u r t h e r assumptions of the Conventional Wisdom scenario are given in Alcamo, et al., Ibid.
Leggett, J., Pepper, W.J., and Swart, R.J., 1992. Emission scenarios for the IPCC: an update, in Houghton, J.T., Callander, B.A., and S.K. Varney (eds) Climate Change 1992: Supplement to the IPCC 1990 Assessment. Cambridge University Press, Cambridge, pp.71-95. 3
4 Enting, I.G., Wigley, T.M.L. and Heimann, M., 1994. Future emissions and concentrations of carbon dioxide. Technical Paper No. 31., CSIRO, Australian
149 Division of Atmospheric Research, Mordialloc, Australia. 5 The IPCC Working Group I exercise o n C 0 2 stabilization (Enting, et al., 1994, op cit.) did not specify the trend of non-CO2 gases. Therefore, we make the following assumptions: (i) CFC emissions are assumed to be phased out according to international agreements as interpreted in the intermediate IPCC emission scenarios (Leggett, 1992, op.cit.); (ii) other greenhouse gas concentrations (pX) are a s s u m e d to have a similar historical and future p a t h w a y of CO2 gas concentrations: pX(t) - pX(1990) pX(1990) - pX(1900)
pCO2(t) - pCO2(1990) pCO2(1990) - pCO2 (1900)
6 For these calculations it was assumed that sinks of greenhouse gases would not be artificially enhanced by large geoengineering projects such as pumping CO 2 t o low levels of the ocean. Also, it was assumed that land use emissions would not be reduced. 7
Enting, et al., 1994, op cit.
8 See, for example, Conventional Wisdom scenario of IMAGE 2.0, Alcamo, et al., 1994b, op cit., or IPCC reference scenarios in Leggett, et al., 1992, op cit. 9 Johannson, T., Kelly, H., Reddy, A. and Williams, R. (eds), 1993, Renewable Energy, Island Press, Washington.
10 Kassler, P., 1994. Energy for development. Selected paper, Shell International Petroleum Company, Shell Centre, London. 11 Ishitani, H. and Johanssson, T., 1995. Energy supply mitigation options. In: R.T. Watson and R. Moss (Editors), IPCC Working Group II: Impacts, Adaptation, and Mitigation. Cambridge University Press, Cambridge, (in review). 12 One of many important questions regarding these scenarios is whether there will be adequate land to provide the biofuels specified in these scenarios. Calculations with the IMAGE 2.0 model indicate that there could be spatial limitations in some regions (Alcamo, et al., 1994b, op. cit.) 13 Moreover, even if greenhouse gases were immediately stabilized, we still expect some climate change due to the historical build-up of greenhouse gases in the atmosphere, and because of the momentum of the climate system. Hence, a certain amount of climate impacts may be very difficult to avoid.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
155
ASSESSMENT REPORT ON NRP SUBTHEME "ATMOSPHERIC
PROCESSES
& UV-B RADIATION"
R. Guicherit TNO Institute of Environmental Sciences P.O. Box 6011 2600 JA Delft The Netherlands
With contributions by: H. ten Brink W. Ruijgrok, M. Vosbeek M. Allaart, R. van Dorland, A.J. Feijt, F. Kuik, A.C.A.P. van Lammeren, E. van Meijgaard, P. Stammes, G.H.L. Verver F.C. van Duyl, H.J. Lindeboom, R. Osinga J.P. Beck, J. Bordewijk, W.A.J. van Pul, H. Reinen, E. Schlamann, H. Slaper, D.P.J. Swart, T.H.P. The, D.L. Veenstra L. Dijkhuizen, W.W.C. Gieskens, T.A. Hansen, M.J.E.C. van der Maarel, P. Quest, J. Stefels B. Bregman, J. Vil~-Guerau de Arellano A. Baart, R. Bosman, P.J.H. Builtjes, P. Esser, R. Guicherit, K.J.M. Kramer R.L.J. Kwint, M. Roemer
ECN, Netherlands Energy Research Foundation, Petten KEMA, Environmental Services, Arnhem KNMI, Royal Netherlands Meteorological Institute, De Bilt NIOZ, Netherlands Institute for Sea Research, Texel RIVM, National Institute of Public Health and Environment, Bilthoven RUG, University of Groningen RUU, University of Utrecht TNO, Institute for Environmental Sciences, Delft
156
Contents Abstract 1. 0
0
0
Introduction Clouds, aerosols and radiation 2.1 Formation and air/sea exchange of dimethylsulphide (DMS) from marine sources 2.2 The role of aerosol of anthropogenic origin in the radiation balance of the atmosphere 2.3 Clouds-radiation-hydrologic interactions in a limited-area model 2.4 The effect of clouds on UV radiation and photodissociation rates in the troposphere The role of atmospheric ozone in global change 3.1 Introduction 3.2 Dutch participation in the international network for the detection of stratospheric change (NDSC) 3.3 Spectral UV radiation measurements and correlation with atmospheric parameters 3.4 Determination of the UV-B climate in the Netherlands: high resolution spectral measurements, monitoring and modelling from the perspective of risk-analysis 3.5 STREAM: Stratosphere Troposphere Experiment, study by Aircraft Measurements Tropospheric budget of non-CO2 greenhouse gases 4.1 Introduction 4.2 GLOMAC: Subproject of EUROTRAC related to climate modelling of KNMI 4.3 Application of 2-D global models 4.4 Global modelling of atmosphere trace gases-application of a global three dimensional model 4.5 Continental ozone issues; monitoring of trace gases, data analysis and modelling of ozone over Europe "Dutch contribution to The EUROTRAC-TOR project"
5.
Acknowledgements
6.
References
157 ABSTRACT The a t m o s p h e r e is a very complex, open, dynamic and multi-causal relation s y s t e m in which non-linear processes and feedback m e c h a n i s m s play an important role. Research within this subtheme in NRP-1 focused on uncertainties in our understanding of three issues i.o. the role of clouds and aerosols on the radiation budget the role of atmospheric ozone in global change and the effect of atmospheric change on I_W-B climatology tropospheric budgets of non CO2 greenhouse gases These issues are being dealt with in 12 projects. On average good progress has been made in m a n y of the projects and through t h e r e are some weaknesses, the overall quality of science and technological d e v e l o p m e n t s is good and in some cases even excellent in comparison to international standards. Links with international programmes such as IGBP, CEC, EUREKA/EUROTRAC are well established. 1.
INTRODUCTION
M a n y different gases can interfere with the earth's radiation budget. Some are highly stable and have residence times of decades or even a century or more. The most i m p o r t a n t of these gases are w a t e r vapour (H20); carbon dioxide (CO2), m e t h a n e (CH4), n i t r o u s oxide (N20), ozone (03) a n d t h e ( H ) C F C ' s (hydrogen-containing chlorofluorocarbons). O t h e r atmospheric constituents, including certain aerosols and notably clouds, m a y also affect radiation regimes. Most scientists agree t h a t this m a y have significant effect on regional climates. A model of global warming simulates the effects of various policy strategies on the built-up of greenhouse gases in the atmosphere (see Figure 1.1). First the built-up of radiation influencing constituents is linked to emission scenarios, which are based on assumptions about future population levels, global and regional energy resources, energy and material use, land-use changes (e.g. due to deforestation, regrowth and biomass burning), agricultural activity, industrial activity, certain policy d e p e n d e n t a s s u m p t i o n s about taxes, income and price elasticities of demand, and other economic factors. In Table 1.1 an overview of activities and emissions of radiatively active compounds are given.
158
Input & assumptions Productions & emissions
Demographics; Technical & Macro-economics; Policy etc. Anthropogenic / Natural (biogenic) Atmospheric (retention) models
Atmospheric concentrations
N20, C02, (H)CFC's, 03, CH4, aerosols (SO2]DMS) Clouds, H20-vapour; radiation Equilibrium radiative forcing models
$
Radiative effects
$
Estimated global warming
Figure 1.1 Schematic structure of the model of global warming In the second stage of analysis, atmospheric retention models are used to simulate chemical and biophysical processes by which the relevant emitted compounds are removed from the atmosphere, resulting in projections of their future concentrations. Table 1.1 Anthropogenic activities and emissions of radiatively active compounds
1:
2:
3: 4:
Fossil-fuel combustion a: CO2 emission (infrared IR trapping) b: CH4 emission by natural gas leakage (IR trapping and 03 changes, which on its turn leads to changes in I_W absorption and IR trapping) c: NOx emission (alters 03) & Carbonaceous soot emission (efficient solar absorption) e: SO2-sulfate emission (solar reflection and IR trapping) f: VOC/CO emission (alters 03) Land-use changes a: Deforestation (releases CO2 and increases albedo) b: Regrowth (absorbs CO2 and decreases albedo) c: Biomass burning (releases CO2, NOx and aerosols) Agricultural activity a: Release of CH4 (IR trapping and changes 03) b: Release of N20 (IR trapping and changes 03) Industrial activity a: Release of CFC's and their substitutes (IR trapping and 03 destruction) b: Release of SF6, CF4 and other ultra-long lived gases (IR trapping virtually forever) c: Release of VOC (03 changes)
159 In the third stage, radiative effects of the projected concentration changes are estimated and translated into equilibrium global and regional radiative forcing, which may be translated into temperature changes. The effects of the radiatively active constituents will not register immediately as a change in surface temperature. The oceans large thermal mass will cause a lag in warming effects. Nonetheless radiative active constituents will cause an eventual or "equilibrium" warming after some time, perhaps several decades after a certain atmospheric concentration has been reached. Since the industrial revolution an additional global average radiative forcing, due to an increase in atmospheric CO2, N20, CH4, stratospheric water vapour, CFC's and tropospheric 03 of 2.4 W/m2 has been calculated in this way, which is equivalent to a temperature increase of about 0.7 K The s u b t h e m e "Atmospheric processes and UV-B radiation" focuses on atmospheric processes. The atmosphere is a very complex, open, dynamic and multi-causal relation system in which nonlinear processes and feedback m e c h a n i s m s play an i m p o r t a n t role. One should note t h a t the feedback mechanisms themselves might influence the emissions in a direct way. The major problems we are faced with can be summarized as follows: 9 Radiative forcing by gases is a function of their concentration. The relation between emissions and concentration is usually not a linear one, but is determined by chemical processes in the atmosphere. The chemistry of the atmosphere, however, will change if the composition of the atmosphere, as is the case, changes. The consequence e.g. being t h a t changes in the atmosphere's methane concentration since the turn of the century does not reflect changes in its emission pattern. This means that for projections of atmospheric concentrations of radiatively active gases for different emission scenarios, one should also take into account the chemical processes occurring in the atmosphere. 9 Aerosols may change the earth's albedo in a direct way through absorption and backscatter of solar radiation and indirectly as condensation nuclei in cloud formation. Aerosol cloud condensation nuclei (CCN) may namely increase cloud droplet concentration and cloud reflectance (albedo) of incoming solar radiation. Atmospheric aerosol particles of concern are both biogenically derived from dimethylsulphide (DMS) oxidation and by anthropogenic activities, particularly sulphates from SO2 emissions, organic condensates and soot from biomass burning. Recent applications of coupled atmospheric chemical/radiative transfer models, by utilizing empirical aerosol scattering properties, have shown that anthropogenically derived sulphate aerosols cause clear-sky climatic forcing, that when averaged over the globe is comparable in magnitude, but opposite in sign, to forcing by CO2. Unlike CO2, anthropogenically derived aerosol particles are not uniformly distributed over the globe but are mainly found over industrialized areas in the Northern Hemisphere. As a matter of fact the present day aerosol forcing in many regions of the Northern Hemisphere, as an annual average, may offset the combined greenhouse effect of CO2; CH4; N20; CFC's, and 03 (Figure 1.2).
160
Radiation
Sulphate haze CCN Greenhouse gas trapping
Emissions CH 4 N20 CO 2 (H)CFC's
VOC NO x CO Halons
.. -lP"
'V....
I l
... - - " l ......
"
1'
SO 2 gas
t
DMS
Figure 1.2 Interference of radiation with gases, aerosols and clouds The effect of clouds is even more complex. Depending on cloud physics and cloud dynamics, clouds may exert a positive or negative feedback in radiative forcing. It is assumed, and this certainly holds for low level stratos- and s t r a t o c u m u l u s clouds over the oceans, t h a t the average net effect is a negative one. Cloud statistics are very sensitive to minor changes in atmospheric circulation patterns and the hydrological cycle. Also the n u m b e r of condensation nuclei may play a role, especially over the oceanic areas of the Southern Hemisphere and to a lesser extent also over oceanic areas of the N o r t h e r n Hemisphere. At the moment it is not clear in w h a t way cloud statistics will react to climate change. The uncertainties projected increase in average global t e m p e r a t u r e (ranging between 2-5 K) for doubling the atmospheric CO2 concentration is merely due to the way clouds are treated in the climate models.
161 The role of atmospheric ozone needs special attention in global change. Depletion of s t r a t o s p h e r i c ozone m a y alter the UV-B climatology, while an increase in t r o p o s p h e r i c ozone m a y have serious adverse effects on the biosphere. Tropospheric ozone is not emitted, it is formed in the atmosphere by chemical reactions involving compounds such as NOx, CO and VOC, which are called ozone precursors. The role of tropospheric ozone in climate change is significant. A complicated factor being, t h a t the magnitude of the O3-forcing effect is height dependent with a maximum around the tropopause and an "opposite" effect above 30 km. Since the effect of precursor emissions and atmospheric chemical processes on tropospheric ozone levels depend on varying regional atmospheric conditions, it is difficult to predict future global changes in tropospheric ozone concentrations accurately. This also holds for changes in stratospheri~ ozone depletion. Although the ozone depletion substances, due to their long atmospheric residence times, are distributed evenly over the globe, there exists large regional differences in the depletion of the ozone layer, notably in the antarctic and arctic regions, over the tropics and at medium latitudes. One may state that next to the hydrological cycle and the direct and indirect effects of tropospheric aerosol, changes in the ozone column density distribution pose the largest uncertainty in model calculations of climate forcing due to anthropogenic induced changes in the composition of the atmosphere. Related to stratospheric ozone depletion are, as was mentioned before, possible changes in the UV-B climatology. This needs special attention since UV-B plays an i m p o r t a n t role in the chemistry of the atmosphere and may enhance u r b a n photochemical smog and UV-B may also exhibit negative effects on the biosphere. The UV-B climatology is dependent on many factors of which the most important are: 9 Stratospheric ozone destruction. 9 Changes in tropospheric ozone. 9 Scattering by aerosols and interaction with clouds. Insight in these processes is crucial to predict changes in the I.W-B climatology to be expected near the surface of the earth. The NRP s u b t h e m e "Atmospheric processes and UV-B radiation" is aimed at s t u d y i n g the aforementioned processes. Changes in the composition of the atmosphere on a global scale and related changes in radiative forcing may have far reaching social-economic consequences especially with regard to preventive and adaptive measures to be taken. For this reason projects belonging to this theme, cannot predominantly be qualified as basic science, but are at the same time policy oriented. The a s s e s s m e n t report will focus on uncertainties in our u n d e r s t a n d i n g of 3 priority issues i.e.: I. Clouds, aerosols and radiation II. The role of atmospheric ozone in global change III. Tropospheric budgets of non-CO2 greenhouse gases The projects to be assessed and the priority issue they are linked with, are listed Table 1.2.
162 Table 1.2 List of projects in the NRP subtheme on "Atmospheric processes and UV-B radiation" Title
Projectleader
Clouds, aerosols and radiation Formation and air/sea exchange of dimethylsulphide R.Guicherit (DMS) from marine sources
Number
850026
The role of aerosols of anthropogenic origin in the radiation balance of the atmosphere
H.M. ten Brink
852066
Clouds-radiation-hydrological interactions in a limited area model
A. van Lammeren
851058
The effect of clouds on ultraviolet radiation and photodissociation rate in the troposphere
H. van Dop
850018
The role of atmospheric ozone in global change Dutch participation in the international network for D.P.J. Swart the detection of stratospheric change (NDSC)
850024
H. Kelder
852088
H. Slaper Determination of the UV-B climate in The Netherlands: high resolution spectral measurements, monitoring and modelling from perspective of risk-analysis
852087
Spectral UV radiation measurements and correlation with atmospheric parameters
P.J.H. Builtjes
VvA205
Budget studies Climate modelling/Global Modelling of Atmospheric Chemistry (GLOMAC)
H. Kelder
851050
Application of 2-D global models
M.G.M. Roemer
852072
Global modelling of atmospheric trace gases "Application of a global three dimensional model"
J.P. Beck
852070
Continental ozone issues: monitoring of trace gases, J.P. Beck data analysis and modelling of ozone over Europe "Dutch contribution to the EUROTRAC-TOR project"
852094
Stratosphere Troposphere Experiment: Study by aircraft measurement
163 2.
C L O U D S , A E R O S O L S AND R A D I A T I O N
Aerosol particles play an important role in the radiation budget of the atmosphere because of t h e i r direct interaction (absorption and scattering) of solar and t e r r e s t r i a l radiation, as well as t h r o u g h their influence on cloud formation processes. Aerosol particles have a lifetime of a few weeks in the troposphere and occur in highly variable concentrations. A large proportion of the particles, which are of interest for the radiation balance and for cloud processes are derived from n a t u r a l sources, a n t h r o p o g e n i c gaseous s u l p h u r emissions, emissions of carbonaceous and organic particles, and biomass burning. It is complicated to determine the direct effect of aerosols. This is due to the fact t h a t the effect is dependent on the aerosol absorption to backscattering ratio, surface albedo, total aerosol optical depth and solar elevation. Their is, however, general agreement, that anthropogenically generated sulphate aerosols will reduce solar irradiance, leading to a net negative change in radiative forcing and thus regionally offsetting the effect of w a r m i n g due to increased concentration of greenhouse gases. The global radiation (energy) balance is very sensitive to cloud albedo, particularly for marine stratus clouds which cover a substantial part (about 25%) of the earth. Cloud albedo itself is sensitive to changes in droplet n u m b e r concentration. The droplet n u m b e r depends on the concentration of cloud condensation nuclei (CCN) which in turn depends on condensation nuclei (CN) or on the aerosol concentration. The ability of an aerosol particle to act as a CCN under low supersaturation found in clouds depends on its size and its chemical composition (notably its w a t e r solubility and substances t h a t influence surface tension). There is a significant non-linearity in the effect on cloud formation and cloud microphysics of changes in CCN concentration; depending on the starting CCN concentration. The effect is far more pronounced in a r e a s (such as clean oceanic sites) where low aerosol concentrations prevail, compared to more polluted areas, e.g. areas over the continents which are strongly influenced by anthropogenic emissions. The hypothesis by Charlson et al (1987) of a connection between climate and p h y t o p l a n k t o n activity in ocean surface waters is based on the fact t h a t CCN concentrations in air over oceans far from land are low, t h a t CCN available in clean maritime air are composed almost totally of sulphate particles, and t h a t this sulphate originates almost entirely from emissions of dimethylsulphide (DMS) from the ocean surface (see Figure 2.1). This subtheme comprises 4 projects.
164
O
indirect._ Radiation ~
Cloud condensation nuclei I'
+
'~xter
Sulfate aerosol I'
budget
Global temperature Climate feedbacks
+
S~ 2 +
+ or-?
DMS
I
Atmosphere
+
Ocean
/
DMSP~
I Phyt~176
I
CO 2 / S042-
H2S / CH 4
Figure 2.1 Proposed feedback cycle between climate and marine DMS production (adapted from M.O. Andreae, 1990) 2.1 F o r m a t i o n a n d a i r / s e a e x c h a n g e of d i m e t h y l s u l p h i d e (DMS) f r o m marine sources
K.J.M. Kramerl, A. Baartl, L. Dijkhuizen2, R. Guicherit (coordinator)l, W.W.C. Gieskes2, T.A. Hansen2, R.L.J. Kwint 1, H.J. Lindeboom3, R. Osinga3, P. Quist2, J. Stefels2, M.J.E.C. van der Maarel2 and F.C. van Duyl3 1 TNO, Institute of Environmental Sciences (IMW) P.O. Box 6011, 2600 JA Delft University of Groningen, Department Microbiology, P.O. Box 14, 9750 AA Haren 3 Netherlands Institute for Sea Research (NIOZ), P.O. Box 59, 1790 AB Den Burg 2
Introduction
Global climate and the "greenhouse effect" can be influenced by several feedback mechanisms. The present project focuses on dimethyl sulphide (DMS) as a likely candidate for important negative feedback processes. DMS is the most important biogenic precursor of non-seasalt sulphate (NSS-SO4) which is a major source of
165 cloud condensation nuclei (CCN). DMS production in marine environments is linked to p h y t o p l a n k t o n a n d m a c r o - a l g a e ; it is formed from t h e p r e c u r s o r B-dimethylsulphoniopropionate (DMSP), which is believed to be an osmoregulating or cryoprotecting agent. It is i m p o r t a n t to realise, t h a t the ultimate flux of DMS to the atmosphere is dependent on m a n y biological and physical/chemical factors. The production of DMSP by algae is subject to species abundance and composition, which is a function of the environmental conditions. The conversion of DMSP to DMS by algae and microorganisms, the degradation of DMS and DMSP by bacteria, loss of compounds due to sedimentation, photochemical processes or escape to the atmosphere, all processes are intricately linked to each other. The various paths of formation and degradation of DMSP and DMS, and their relative importance are presented in Figure 2.2a, where the state of the art at the start of the project is depicted. Quantification of the DMS emitted to the atmosphere has been found difficult as large temporal and geographic differences occur. For a better estimation of the fluxes, a proper u n d e r s t a n d i n g of the (micro)biological processes in the w a t e r column related to the biogeochemical cycle of DMS(P) is essential. Although the possible effects upon the world climate will to a large extent be a function of open ocean processes, coastal waters (and sediments) may play an i m p o r t a n t role on a regional scale, not in the last place because of its enhanced algal productivity. If we consider the algal species Phaeocystis sp. and Emiliania huxleyi as typical and i m p o r t a n t representatives for coastal and open ocean waters respectively, Figure 2.3 illustrates schematically the DMS(P) cycles t h a t play a role in the various marine compartments. The objectives of the present project are: 9 To assess these processes and rates of biological production. 9 To assess the processes of degradation and t r a n s f o r m a t i o n in the w a t e r column and the role of sedimentation therein. 9 To estimate the fluxes from water to the atmosphere.
166
Air Water; sediments
CO2, sulfate
DMSO
~,~. ~~
DMSP (in algae) ~ I I
~
photochemical
algae: lyase?
aerobic
bacterial egradation
anaerobic ~i~ ~ DMS ~ CH4, CO2, sulfide (+ acrylate) bact. degr.
excretion sedi- " ~ , m e n - ~si~nescence . ~
tation
/~
oxidation ~~ bacterial ~~/red
bacterial
degradation
r MMPA, MPA rDMSP DMSP (in algae) autolysis (free) bacterial bacterial degradation degradation
Figure 2.2a The understanding of major DMS pathways before the start of the NRP/DMS project
167
Air Water; sediments
| CO 2, sulfate
DMSO ~ ~
DMSP (in algae) ~ I
QI
I
,!,
algae: lyase? ._, ~,)
~1~ anaerobic ~ DMS ~ CH 4, CO 2, sulfide (+ acrylate) bact. degr. Q
sedi-~~ excretion men"%~enescence tation O ~grazing
(in algae)
"
aerobic bacterial egradation
"
autolysis ( ~ bacterial degradation
|
I I " algal: lyase + I I bacterial degradation
(free)
bacterial degradation
Figure 2.2b Our understanding of the major DMS pathways including results of the NRP/DMS project
c'~;&--~-cc~-, --__~ . . . . . . .
rivers-)~N+P intertical mats
I I
I
I T N O (field + lab)
i TNO
I
~ phaeocystis --)~DMS(P) ~ S O 4 , .... I ~ " -- ..
n emilania --~DMS(P) ~
~,1.-~_~
"---
o s/P)
coastal zone
c"T~_L~ coN:, ~__.-
.
~ DMS(P) ~
SO4n
ocean + northern North Sea
Figure 2.3 The sulphur cycle in different parts of the marine environment
SO4n
168 To investigate the different parts of the DMS(P) cycle, experiments were performed under well defined laboratory conditions, in large experimental enclosures (mesocosms) and in the natural environment (Dutch coastal waters). In this project research focussed on species typical for the Dutch coastal environment because it was considered that although the rates may not be comparable to those in oceanic systems, the processes as such will probably be analogous in both systems. Within the project a close cooperation was established between different marine scientific disciplines, in order to be able to tackle the various routes of production/elimination and the process rates involved. The project is a combined scientific effort of the: 9 University of Groningen, Department of Marine Biology and Department of Microbiology. 9 TNO Institute of Environmental Sciences, Department of Biology, and Department of Environmental Chemistry. 9 Netherlands Institute for Sea Research, Department of Applied Scientific Research. In the following sections the DMS(P) production/elimination paths will be discussed in the framework of the activities performed within the present project. References will be made to the numbered routes (paths) in Figure 2.2b, which represents the situation as we understand it now.
Laboratory studies
Phytoplankton. DMSP is produced intracellularly by phytoplankton. Two mechanisms are to be considered for the formation of DMS, either direct (through algal lyase) or via DMSP that is released into solution (by excretion, senescence and/or predation). Until now, little attention has been paid to the influence of environmental stress factors on the DMSP content of algal cells, and the possible role of algae in the conversion of DMSP to DMS (paths 1 and 2). As it is often suggested that bacteria are most important in this conversion (see below), evidence for an important role of phytoplankton itself was never presented. As Phaeocystis sp. is considered an important producer of DMSP in coastal waters, this species was investigated in detail in laboratory studies. It is often hypothesized t ha t DMSP production in algae is stimulated under nitrogen limited conditions, when DMSP substitutes for other- nitrogen containing - osmolytes. This was tested by inducing osmolyte production in axenic Phaeocystis sp. cultures at different salinities (from 25 to 45x10-3) and different N:P ratios. Although DMSP per cell did increase with salinity, indicating its function as an osmolyte, no significant difference was found between the N and P limited cultures. In Phaeocystis sp. cultures, that were free of bacteria (axenic) the transformation of extracelluar DMSP to DMS and acrylate was measured (path 2). It was proven t h a t this conversion could be correlated with an enzymatic (lyase) reaction associated with the cells, which could be inhibited by heating. The lyase activity in
169 living cultures increased with temperature: at 5 ~ activity was 50% of the activity at 20 ~ During the growth phase of an axenic Phaeocystis sp. culture leakage of DMSP or DMS was very low. DMS production started in early stationary phase, when only small a m o u n t s of dissolved DMSP appeared to be present, indicating a rapid conversion of DMSP by the algal lyase. In a completely lysed culture, approximately 75 % of total DMSP was found to have been converted to DMS. In experiments with crude extracts of Phaeocystis sp. cells, the DMSP-lyase activity exhibited an alkaline optimum, which is in contrast to results from experiments with other organisms as reported in the literature. This points into the direction of a species specific enzyme.
Bacteria. Once DMSP or DMS are liberated in the w a t e r column, bacterial d e g r a d a t i o n / t r a n s f o r m a t i o n m a y occur. Previous studies on the microbial metabolism of DMSP had shown that there are two pathways for degradation of DMSP. The first p a t h w a y involves a cleavage to dimethyl sulphide (DMS) and a c r y l a t e (path 4). The second p a t h w a y involves the d e m e t h y l a t i o n to 3-S-methylmercaptopropionate (MMPA) and subsequently to 3-mercaptopropionate (MPA) (path 3). DMS may be further degraded to CO2 and SO42- (path 5) under aerobic conditions. F r o m a v a r i e t y of m a r i n e sources ( p h y t o p l a n k t o n cultures, s e d i m e n t s , macro-algae and mesocosms) bacteria capable of degradation DMS or DMSP in aerobic environments could be isolated and characterized during the present project. Common characteristics of these strains are the ability to utilize a broad substrate spectrum, their motility and their coccoid rod-like morphology. Apart from aerobic conversion of DMS(P), anaerobic degradation of both DMS and DMSP is possible. With a few exceptions, anaerobic processes are not considered important for the water column. However, anoxic sediments can be a b u n d a n t in many coastal waters, where they may influence the cycling of DMS(P). Under anoxic conditions, DMS may be converted by sulphate-reducing bacteria or by m e t h a n o g e n s to m e t h a n e (path 6). The degradation step from DMSP via MMPA and MPA (path 3), can be followed by a conversion to m e t h a n e by m e t h a n o g e n s . Except for D M S - m e t a b o l i z i n g m e t h a n o g e n s , no anaerobic microorganisms t h a t are responsible for one of the above mentioned conversions had been isolated from anoxic marine sediments. During this study it was found that demethylation of DMSP to MMPA in anoxic W a d d e n s e a sediments is coupled to sulphate reduction. A pure culture of a DMSP-demethylating sulphate-reducing bacterium, which was identified on basis of physiological characteristics and positive hybridisation with genus-specific molecular RNA probes as a Desulfobacterium sp, was isolated from the sediment. Certain other Desulfobacterium species were also found to demethylate DMSP. It is, however, not yet clear w h a t the ratio of DMSP cleavage over DMSP d e m e t h y l a t i o n is at n a t u r a l (anoxic) concentrations of DMSP. Recently a DMSP-cleaving anaerobic bacterium was isolated which ferments acrylate to propionate and presumably CO2 but which is not able to utilize DMS. This strain
170 showed a high DMSP-lyase activity (approximately 10 ~imol/min.mg protein) and a rather high Km value for DMSP (150 ~tM). The methanogens Methanosarcina sp strain MTP4 and Methanosarcina acetivorans were found to be able to convert MMPA to MPA and methane during growth. Methanogenic conversion of MMPA to MPA was only found in anoxic Waddensea sediment when certain antibiotics, acting against bacteria but not methanogens, were used. Under non-inhibited conditions MMPA was rapidly converted to methanethiol, which was subsequently converted to methane. It was concluded that conversion of MMPA in anoxic marine sediments directly or indirectly results in the formation of methane. The anaerobic metabolism of DMS was also studied in anoxic Waddensea sediment. In this sediment DMS appeared to be converted to methane, as had been described for other anoxic marine sediments. The conversions of DMS to methane seems to be restricted to Methanosarcina species. The population of DMS-metabolizing methanogens appears to be in the same range as the total population of methylotrophic (i.e. methanol- and trimethylamine-utilizing) methanogens (1-6x106 cells per gram dry weight). Both pathways indicate that methane can be produced under anaerobic conditions; there may thus exist a positive flux of methane to the atmosphere. Thus, in anoxic marine environments, not only the counteractive effect of global warming through DMS may take place, but also the production of the potent greenhouse gas methane. Mesocosm studies Apart from the direct production of DMS by algae, DMSP may be introduced into the water column by various other mechanisms (path 1). Living algae may excrete DMSP, the DMSP may be liberated when the algal cells die (senescence), or zooplankton may release DMS(P) during or after digestion of the algal cells (predation, sloppy feeding). Particulate DMSP may be transported out of the surface layers by sedimentation.
In order to study the role of phytoplankton and zooplankton (and their interaction) in the production of DMSP and the release of DMS to the water column (and eventually to the atmosphere), three series of large scale experimental systems (pelagic mesocosms) were used to study the development of phytoplankton blooms as function of the presence of zooplankton, different nutrient regimes, species composition and varying general environmental characteristics. Measurements on densities and activities of aerobic DMS(P) metabolizing bacteria were included. In one mesocosm experiment the deposition of particulate m a t t e r (containing DMSPpart) was quantified. The first two mesocosm experiments showed that the DMS is generally released in the water column directly after the peak of the phytoplankton bloom, during the senescence phase. This was observed for a diatom bloom. For the bloom of Phaeocystis sp. in the last mesocosm experiment, a maximum DMS production was found during the stationary phase; no direct correlation was found with chlorophyll-a. The decline of the Phaeocystis bloom did not lead to an increased DMS concentration. However, but just before the release of DMS a dramatic drop in cellular DMSPpart was observed. The latter species produced substantially
171 higher amounts of DMS(P) than the diatom species, which confirmed earlier findings reported in the literature. Due to boundary effects, the duration of the chlorophyll-a peak appeared to be compressed in time in our mesocosm studies, as was the peak of [DMS]water. It appeared that the change in DMS concentration could occur very fast. In a matter of days the [DMS]water can change by orders of magnitude. Both increase and decrease of [DMS]water showed this dynamic character. In m e s o c o s m - s y s t e m s with a p p a r e n t l y similar characteristics (both in chlorophyll-a, nutrients, plankton assemblages) the production of DMS did not always follow the same pattern. Considering the results of the last mesocosm experiment the role of Phaeocystis may have been underestimated in the first two experiments. Another reason could be that the importance of the microbiological activity is not fully understood. In order to study the potential effect of zooplankton upon the production of DMS, in a number of mesocosm experiments the zooplankton was inhibited or removed. A positive influence of zooplankton e.g. of grazing (path 1) upon the release of DMS into the watercolumn could not be found, however. This could indicate t h a t zooplankton does not play an important role in the DMS(P) dynamics. However, as the time axis of the phytoplankton bloom in the mesocosm experiments is compressed, a mismatching of life stages of algae and zooplankton may have biased these experiments to some extent. During this last experiment, deposition of suspended particulate m a t t e r was studied (path 7). At least in the mesocosms, sedimentation is an important sink for biomass of Phaeocystis sp. The sedimentation of Phaeocystis cells showed a good correlation with Phaeocystis cell numbers in the water column, suggesting that the sinking rate of Phaeocystis cells was rather constant. The daily sedimentation of living cells being nearly equal to the standing stock at that time. At the decline of the Phaeocystis bloom in the mesocosm, sedimentation accounts for only 50 % to the observed loss in biomass. This strongly suggests that cell lysis in the water column must have been important at this time. Although this cell lysis does increase the DMSPdiss concentration, this does not lead to a peak in the DMS concentration: possibly the production and consumption are of a similar magnitude. The importance of bacteria in the conversion of DMSP and DMS was studied in the same mesocosm experiments. The DMSP-degrading population consisted of a variety of microorganisms. Bacterial populations are strongly stimulated by the collapse of an algal bloom, as their substrate becomes abundant from the decaying algal cells. The DMS-utilizing bacterial population reacted upon the increase of DMS with a delay of about a week. The DMS assimilating ability seemed to need some induction. The bacterial activity found in the mesocosm systems, 10-1000 ~tmol DMS per mg protein per day, was in the range of activities of laboratory cultures. The biological turnover rate ranged between and 0.2 and 0.6 days, whereas the turnover by DMS effiux ranged between 2 and 5 days. In these experiments, the contribution of the bacteria to the loss of DMS was calculated to be approximately
172 90% of the total DMS loss. This figure should be taken with some care, as only few observations were made. Nevertheless, the importance of this process should not be underestimated. Although limited data on the turnover of DMSP exist, they indicate a key role for the demethylation of DMSP via MMPA (path 3) rather than the cleavage of DMSP (path 4). This means that only a fraction of the DMSPwater seems to contribute to the formation of DMS. In benthic mesocosm studies carried out separately, addition to the sediment of fresh algal material (dominated by Phaeocystis sp., path 7) resulted in a strong (benthic) microbial response after 2 days, which indicated that freshly deposited algae become subject to rapid bacterial degradation. In shallow waters, benthic processes are therefore likely to have a strong influence on the processes in the water column in the pelagic mesocosms. Since m a n y Phaeocystis cells reach the mesocosm floor intact, there is also a continuous downward transport of considerable amounts of particulate DMSP. Sedimentation of DMSP may significantly contribute to the production of DMS in sheltered, shallow systems like the mesocosms, as a result of autolysis or bacterial d e g r a d a t i o n on the bottom (path 8). Despite the relatively high d o w n w a r d transport of DMSP, and the potentially fast transformation to DMS, there was no evidence that DMS in the water column correlated with the sedimentation rate. In deep, oceanic systems, path 6 may be a sink for DMSPpart. Since the rates of bacterial transformation were not sufficiently well determined, it seems premature to base budget studies on these results. Another u n k n o w n box in such budget studies forms the potential formation of dimethylsulphoxide (DMSO) by photochemical oxidation and/or bacterial activity (path 9). This compound has worldwide received little attention, due to the problematic analysis. By closing of one mesocosm system, the actual flux of DMS over the water-atmosphere interface could be determined. The m e a s u r e d flux of DMS to the atmosphere (path 10), under constant wind speed conditions (1.3 m/s) in one mesocosm system, confirmed t h a t the m a x i m u m efflux of DMS from the waterphase to the atmosphere took place at the same period of the development of the phytoplankton bloom, linked to the (elevated) water concentrations. The measured flux of DMS (at this low wind speed) agreed well with the calculated flux according to the Liss-Slater model (see next Section).
Field o b s e r v a t i o n s The laboratory observation t h a t the DMSP-lyase activity could result from a species specific enzyme, was confirmed by field experiments. During the spring bloom of 1993, a DMSP-lyase-assay was applied to n a t u r a l s e a w a t e r samples off the Dutch coast. A very good correlation was observed between the DMSP-lyase activity and Phaeocystis sp. n u m b e r s , the most abundant species found during the cruise (r2 = 0.9660, n = 23). No correlation was found with either one of the other species. From these results a potential DMS production via route 2 by Phaeocystis sp. cells in the field was calculated. It appeared t h a t DMS production rates were in the same order of magnitude as the
173 total abiotic loss factors (fluxes of DMS to the atmosphere and photochemical degradation). In order to try to link the processes in the mesocosms to the natural environment, water samples were collected regularly in the Marsdiep tidal channel over a period of 1.5 years. M e a s u r e m e n t s of DMS(P) and other organic sulphur compounds in the coastal waters of the Marsdiep tidal inlet, showed that also here only during or shortly after the phytoplankton peak the majority of the DMS was released. In only one week the [DMS]water increased or decreased five to ten fold. M a x i m u m concentrations were up to 20 nM, as compared to 450 nM in the mesocosms in the same period. Other sulphur compounds, like carbonylsulphide (COS) and dimethyldisulphide (DMDS) were found to change not much with the time of the year, w i t h concentrations s u b s t a n t i a l l y lower t h a n those of DMS (for COS and DMDS respectively: 1.2 nM and 0.35 nM in the field, comparable to 1 nM and 0.4 nM found in the mesocosms). Maximum concentrations of particulate DMSP observed, were about 1500 nM in the Marsdiep, which is substantially less than those found in the mesocosms (7500 nM). The ratio of p a r t i c u l a t e DMSP/Chl-a, however, was 20 nmol/~g in the M a r s d i e p which compared well with the 10-60 nmol/~g in the mesocosm experiment. The Marsdiep experiment (and the last mesocosm experiment) showed, t h a t from the total a m o u n t of DMSP produced by the phytoplankton, only 5 to 10% can be detected as DMS in the watercolumn. The sea to air transfer-velocity of DMS has been experimentally determined. This p a r a m e t e r is of major importance in calculating the DMS flux from the DMS concentration in the w a t e r (flux = concentration in water x transfer velocity). Liss and Slater showed that the transfer velocity (of CO2) primarily to be dependant on the wind velocity. Their empirical relations are being applied to DMS as well. To our knowledge there have been no report of the experimental determination of the sea-air transfer velocity of DMS in the open literature. In this study the transfer velocity of DMS from water to the atmosphere has been determined by two different methods : 9 Enclosure (at low, artificial wind velocity of 1.3 m/s). 9 Concentration vs. height gradient method (field measurements, wind velocities over a range from nearly 0 up to 11 m/s). Good agreement was found between the Liss-Slater relations and the experimental data at lower wind velocities (up to 4 m/s). At higher velocities the Liss Slater relations u n d e r e s t i m a t e s the transport velocity up to 20 %., assuming, however, t h a t the collected w a t e r in the field observations contained a r e p r e s e n t a t i v e a m o u n t of DMS for the surface w a t e r s t h a t d e t e r m i n e d the a t m o s p h e r i c concentrations.
Climate modelling The influence of the sea to air t r a n s p o r t of DMS upon the climate has been estimated by mathematical modelling.
174 In the atmosphere DMS is converted to primarily sulphate and methanesulfonic acid (MSA). The climate effect has been calculated using a global zonal averaged 2D atmospheric chemistry and transport model. This global model has a resolution of 10 x latitude and 20 layers (of 500m and l k m height) in a vertical dimension from 0 to 16 km. Transport is described by using seasonally averaged data for advection and eddy-diffusion and also for DMS emissions (literature data). The data base contained over 130 different chemical compounds and over 200 different chemical reactions. The climate effect has been calculated from the direct climate effect of sulfate aerosols, and the effect of DMS emissions has been compared to t h a t of anthropogenic SO2 emissions. The radiative forcing resulting from DMS emissions amounts to a global annual averaged value of-0.4 W/m2. It was shown, that at northern hemisphere mid-latitudes, the radiative forcing by sulfate aerosols is governed more by anthropogenic sulfurdioxide (SO2) emissions than DMS emissions. DMS, however, still accounts for 20 to 30 % of the annual averaged radiative forcing of sulfate aerosols. At the other latitudes and especially on the southern hemisphere, DMS is the major precursor of sulfate aerosols and the relative contribution of DMS emissions to the direct radiative forcing was calculated ranging from 50-99 %. Conclusion Many routes of formation/degradation of DMS(P) and related compounds were studied in this project. Figure 2.2a presents the situation as we understood it at the start of the project; Figure 2.2b summarizes our present understanding, based on our experimental results. Bold arrows indicate that these paths are considered important. It became possible to get a better insight in the production and fate of DMSP and DMS, and the relative importance of the DMS flux to the atmosphere and bacterial turnover as the main loss factors for DMS during an algal bloom. The strong relation with the development of algal blooms was confirmed. It was found t h a t the efflux of DMS to the atmosphere was definitely episodic in character: closely connected to (only) the occurrence of an algal bloom, fast increases and decreases of DMS in the water column were observed in all experiments. The importance of the processes, as indicated by the arrows in Figure 2.2b, may temporarily be enhanced due to physical effect: wind speed and turbulence. This emphasizes the significance of the processes that may occur over short time periods. Because not all results fit easily into one model, and because the importance of the bacterial compartment could only in a late stage of the project be confirmed, it seems too early to quantify the rates of the various processes. Based on the availability of DMSPpart, the DMSwater, and the calculated and measured DMS fluxes, it was calculated that only part of the DMSP produced is transformed into DMS, and that only part of this DMS produced actually escapes to the atmosphere. This implies t h a t minor changes in the bacterial and phytoplankton population density or composition, could have pronounced effects on the global S budget, with all of its consequences.
175
Assessment This project has given insight in some important processes governing the oceanic S-cycle. A major finding is, that only part of the DMSP produced, is transformed into DMS and that only a small part of this DMS actually escapes to the atmosphere. This implies that minor changes in the bacterial and phytoplankton population density or composition, could have pronounced effects on the global sulphur budget, with all its consequences. For the transfer of DMS a good agreement was found between the Liss-Slater relations and the experimental data. It was also found that the global annual radiative forcing due t.o DMS derived aerosols amounts to -0.4 W/m2. It was further found that at northern midlatitudes DMS accounts for 20-30% of the radiative forcing due to sulphate aerosols. For the other latitudes and especially over the southern hemisphere DMS is the dominant precursor of sulphate aerosols. The indirect effects due to cloud formation are much h a r d e r to deal with, no attempts are made. Although progress in understanding the oceanic S-cycle has been achieved, the state of the art is such, that at his moment no predictions can be made, with respect to expected changes in DMS emissions as a result of expected environmental changes, and the consequence thereof on climate change. 2.2 The role of aerosol of anthropogenic origin in the radiation balance of the a t m o s p h e r e H.M. ten Brink ECN, P.O. Box 1, 1755ZG Petten, The Netherlands
Direct effect The present study aims at assessing the influence of anthropogenic aerosol particles on the solar radiative flux in Europe. Aerosol particles reflect shortwave radiation and absorb little infrared terrestrial radiation. They thus exert a cooling forcing over surfaces with low albedo, the "direct" forcing effect. Aerosol particles originate both from natural and anthropogenic sources. The anthropogenic particles form an (extra) forcing factor introduced by mankind. They mostly result from fossil fuel combustion. Estimates for the forcing by the anthropogenic aerosols indicate a regional forcing in the heavily polluted regions of Europe of up to 10 W.m-2 with an uncertainty of the same order of magnitude. This means that, after the forcing by clouds, the forcing effect of aerosols is the second largest climatic forcing term. The forcing by aerosols is a regional effect, because of the limited residence time of the particles in the atmosphere. Better quantification of the "direct" aerosol effect in Europe is studied by ECN in a separate European project, of which the progress report is in press. A main preliminary conclusion is that nitrate and, possibly, carbonaceous aerosol are more important than sulfate in the local "direct" effect. In the above mentioned estimates sulfate was used as the only anthropogenic aerosol component. The actual local forcing is therefore higher due to nitrate and carbonaceous aerosol. These aerosol components are not directly emitted but almost exclusively formed in the atmosphere from gaseous emissions, which makes source apportionment difficult.
176
I n d i r e c t effect In de indirect effect, anthropogenic aerosols act as a radiative cooling forcing via clouds. Aerosol particles serve as the nuclei on which the clouds in the atmosphere form. In the present study, ECN assesses the influence of anthropogenic aerosol particles on cloud structure. Anthropogenic aerosol particles act as extra cloud nuclei, in addition to the n a t u r a l nuclei. The extra cloud droplets change the microstructure of the clouds and thus their radiative transfer as well as their lifetime. Simple radiative models show t h a t this leads to an increase in the reflectivity of the clouds. The influence on the infrared absorption is, however, small. Thus anthropogenic aerosol particles exert an "indirect" radiative cooling effect. Sensitivity studies show that the "indirect" effect is most i m p o r t a n t in marine stratus near polluted continents.
Cloud c h a m b e r In a first approach to assess the effect of anthropogenic aerosols on clouds the differences in the microphysics of clouds formed in clean and polluted marine air were investigated. This is done by artificial cloud formation using a cloud chamber, drawing in ambient air and comparing the number of cloud droplets formed. The site of the cloud chamber is ideally located to study marine clouds, since it is situated at the coast of the North Sea in The Netherlands. In arctic North-West airflows only natural aerosols occur and extra anthropogenic aerosol is present when the air travels more southerly over the UK. In the cloud chamber the supersaturation can be controlled in the range of 0.1 to 0.3% s u p e r s a t u r a t i o n values, which correspond to the s u p e r s a t u r a t i o n s in the marine s t r a t u s of interest. The particle size and number-concentration of the ingoing aerosol and the unactivated aerosol (particles not grown to droplets) are also compared. First, laboratory generated aerosols were used for testing the p e r f o r m a n c e of the chamber. Aerosol particles w i t h a size l a r g e r t h a n approximately 0.07 ~tm were found to grow into cloud droplets. This "critical" size was in agreement with the calculated size for the supersaturations used. From these tests it was concluded that the chamber has the proper characteristics for simulating marine stratus. Results The n u m b e r of cloud droplets in the polluted air is higher by a factor of three compared to that in the clean air. This is less than the increase in particle number. A substantial fraction of the particles larger t h a n the critical size do not grow, presumably since they are not soluble, see below. In order to compare results obtained with the cloud chamber with actual clouds, ECN cooperates with one of the leading i n s t i t u t e s (Brookhaven N a t i o n a l L a b o r a t o r y ) in the ARM p r o g r a m of US-DoE, in the e v a l u a t i o n of the microstructure of clean and polluted marine stratus off the coast of Nova Scotia. ECN has also participated in a recent cloud study near the English west coast. In special campaigns, droplets and aerosol particles were collected for chemical analysis. Unfortunately these campaigns were on days with continental air. In the polluted continental air, the number concentration of particles is about twenty times higher than that in the clean marine air, but the number of cloud droplets is
177 not proportionally larger. Droplet concentrations were typically 1100 per cm-3, independent of the aerosol concentration. This is indicative of a saturation effect of the cloud n u m b e r in continental air. This phenomenon is in part explained by the fact t h a t particles with the proper size are non-soluble and therefore cannot act as cloud nuclei. M e a s u r e m e n t of the actual amount of non-soluble particles is very difficult, because most cloud nuclei are small and thus contain very little mass. Preliminary conclusions1 2 9 The local "direct" aerosol effect can be assessed by m e a s u r e m e n t s and is thus easier to quantify t h a n the "indirect" effect. 9 Nitrate seems of more importance for the local direct effect t h a n sulfate. 9 Non-soluble, presumably carbonaceous, particles form a central factor, both in the direct and the indirect effect.
Assessment The project is one of the so called "late starters". Progress is questionable. No substantial new results are as yet available. The use of a cloud chamber seems promising. At this m o m e n t it cannot be judged if the goals of the project will be met. 2.3 C l o u d s - r a d i a t i o n - h y d r o l o g i c i n t e r a c t i o n s in a l i m i t e d - a r e a m o d e l
A.C.A.P. van Lammeren, A.J. Feijt, R. van Dorland, E. van Meijgaard, P. Stammes KNMI, Royal Netherlands Meterological Institute, P.O. Box 201, 3730 AE De Bilt
Introduction Clouds play an important role in our climate. Clouds produce precipitation which is an essential ingredient of the hydrological cycle. Clouds modify the earth-radiation budget. Thin cirrus clouds have a warming effect while low clouds have a distinct cooling effect ( R a m a n a t h a n , 1989). Clouds dominate the vertical t r a n s p o r t of energy, m o m e n t u m and trace gases in the free troposphere. Despite t h e i r importance, clouds are represented only rudimentary in climate as well as weather forecast models. It appears t h a t the model r e p r e s e n t a t i o n of clouds in climate models has a major impact on model predictions for climate change. Cess et al. (1989) showed t h a t cloud feedback is a major source of u n c e r t a i n t y in model responses to climate forcing. There are two main reasons why the uncertainties with respect to clouds are so large. The first reason is t h a t cloud processes are extremely complicated. A proper
1. In the p r e s e n t s u m m a r y the comments by the cluster coordinator on presentation of 19 May 1994 are incorporated. These comments centred on m a g n i t u d e a n d the u n c e r t a i n t y of the aerosol forcing r e l a t i v e to m a g n i t u d e / u n c e r t a i n t y of the forcing by clouds and on the values for anthropogenic forcing in recent literature.
the the the the
2. In a special investigation, which was not part of the European project, the local "direct" aerosol effect in November 1993 was studied. The data indicate a local daytime radiative forcing by the aerosol in polluted air of 20 W.m2.
178 representation of clouds requires the parameterization of subgrid processes both on t h e m a c r o s c a l e (cm - km) and on the microscale ( 100 jam fractions in which the Phaeocystis sp. colonies were trapped. DMSP-lyase activity showed a
243 very good con'elation with Phaeocystis sp. numbers (r2=0.9660, n=23), but no conelation with any other abundant species, nor with total diatom numbers, total diatom biovolume or total protein. DISCUSSION In the literature, the conversion of DMSP to DMS has mostly been attributed to bacterial activity. Our results show, however, that the alga Phaeocystis sp. has a very active DMSPlyase, specific for this species, which is potentially responsible for the conversion of DMSP to DMS during the early spring bloom off the Dutch coast. DMS production rates by Phaeocystis can be calculated for these waters, using the production rates measured in axenic cultures, Phaeocvstis numbers and dissolved DMSP concentrations found during the cruise, and a mean depth of the n'fixing layer of 5 m. In the northern part of the study area, values ranged from 47 to 131 btlnol l-n-2 day -1. We have compared these production rates with the main abiotic loss factors. Loss by air-sea exchange was estimated to be 16.6 btmol m -2 day-i; photochemical oxidation of DMS to DMSO is comparable with the flux to the atmosphere. Total abiotic loss rates can therefore be estimated to be approximately 30 l.tmol m -2 day -1. This is in the same range as DMS production by Phaeocystis; indeed a 1.5 to 4.5 times overproduction of DMS can be calculated, potentially available for bacterial consumption. Considering the conservative estimates of the pm'ameters used, production by Phaeoo, stis may even be higher. Several field studies have shown lm'ge seasonal vm'iations of DMS in the southern North Sea, with a maximum in fi'ont of the Dutch coast dm'ing the Phaeoo'stis bloom. Ore" study has made it plausible that Phaeocvstis itself plays an important role in this production of DMS. References: Andreae, M.O. (1990). Ocean-atmosphere interactions in the global biogeochemical sulfur cycle. Mar. Chem. 30:1-29 Bates, T.S., Kiene, R.P., Wolfe, G.V., Matrai, P.A., Chavez, F.P., Buck, K.R., Blomquist, B.W., Cuhel, R.L. (1994). The cycle of sulfur in surface seawater of the northeast Pacific. J. Geoph. Res. 99 (C4): 7835-7843 Charlson, R.J., Lovelock, J.E., Andreae, M.O., Warren, S.G. (1987). Oceanic phytoplankton, atmospheric sulfur, cloud albedo and climate. Natme 326:655-661 Charlson, R.J., Wigley, T.M.L. (1994). Sulfate aerosol and climate change. Scientific American 270 (2): 28-35 Keller, M.D., Bellows, W.K., Guillard, R.R.L. (1989). Dimethyl sulfide production in marine phytoplankton. In: Saltzman, E.S., Cooper, W.J. (eds.) Biogenic sulfur in the environment. ACS Symp. Set. 393, Washington DC. p. 167-182 Malin, G., Turner, S.M., Liss, P.S. (1992). Sulfur: the plankton/climate connection. J. Phycol. 28:590-597 Stefels, J., van Boekel, W.H.M. (1993). Production of DMS fi'om dissolved DMSP in axenic cultures of the marine phytoplankton species Phaeocystis sp. Mm. Ecol. Prog. Ser. 97: 11-18
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
245
Clouds-Radiation-Hydrologic interactions in a limited-area model A.C.A.P. van Lammeren, A.J. Feijt, R. van Dorland, E. van Meijgaard, P. Stammes, A.P. van Ulden
Royal Netherlands Meteorological Institute (KNMI) P.O. Box 201, 3730 AE De Bilt
Abstract In this project work has been directed towards the improvement of the knowledge on clouds and the way they influence our climate. The activities include measurement of cloud properties on a regional scale (120x120 km:), analysis of global satellite datasets and the development of a model environment to enhance the regional data analysis and the improvement of parametrizations of clouds and radiation.
1. Introduction
Clouds play an important role in our climate. They produce precipitation which is an essential ingredient of the hydrological cycle. Clouds modify the earth-radiation budget. Thin cirrus clouds have a warming effect while low clouds have a distinct cooling effect [1]. Clouds dominate the vertical transport of energy, momentum and trace gasses in the free troposphere. Despite their importance, clouds are represented only rudimentary in climate as well as weather forecast models. It appears that the model representation of clouds in climate models has a major impact on model predictions for climate change. Cess et al. [2] showed that cloud feed-back is a major source of uncertainty in model responses to climate forcing. There are two main reasons why the uncertainties with respect to clouds are so large. The first reason is that cloud processes are extremely complicated. A proper represenation of clouds requires the parameterization of subgrid processe both on the macro-scale ( c m - km) and on the microscale ('.'_
-/,-///.'/////'//////////..///.'.'//.'/.'//I'//,'////.'///.',""
"/////,//"/.
",'////
-
~_
.,.,,.,.v..,..,_, ",'/
-////-'//"//
~I~YY;~?~`:;~::~;~Z~;~Z~(~:~:~:~?Z4~Z9;:9Z~?:~ZZ~Z~Z~Z~9~~:~:j/i
~3000
4000
~~
~ ~~.F~~
~w ~ ~ ~ ~ ~T~~
~Y~ ~ ~ - ~
~~
~C~F ~ ~ ~ Z ~ ~ ~ ~ ~
:~ ~ Z ~ ~ ~ ~ ~
L','.'.',',~ . ","U.,Z,,~',"T',"T..~"/.~'. "%.~,","U.,~, ..'~,'~,'.
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :9::::::
-20
0
20
40
IIc ~_ _
60
- , .~',~ ~ "
,
~';';'.',',',',','),: ~ l I :'T'.'~,','," .'Z,,., ~
::::::::;:: :::::;:':::::;:::l::
80
deg N Figure 3" Difference in annual-mean meridional overturning stream function between a run with daily and one with monthly forcing, both with MLM. Contour interval is 1 Sv, negative values are stippled.
346
80'm - ' - - ~ ~ ,-~>--~-'............
~
.---+=-~r . . . . - ~- =--I-/,, = ) ~ ' 7I.~~ %i . ~ - - ~ - ~ ~~~ '
70
,~
s::: ~2~i~,~ .....
,'~. . . . .
::~:: :: ,:,-~ .
~ 40 30 20
Figure 6: A time sequence of the absolute dynamic height in the Agulhas Retroflection at weekly intervals determined by applying the method of (Feron et al., 1994) to Geosat altimeter height observations. In this sequence the actual shedding of large Agulhas rings south of Africa can be observed. Analysis of the 3-year observational period shows that 18 similar ring-shedding events occurred. These rings carry a large amount of heat into the South Atlantic, and thus establish one of the interbasin links in the global heat and fresh water circulation. Time is relative to 8 november 1986 and the contour interval is 10 cm. Observing and Forecasting System. In this project a few fundamental steps have been taken toward developing such a system. The first part of the present study has revealed the sensitivity of a new eddy-resolving model of the Antarctic Circumpolar Current (ACC) to its external and internal parameters. The ACC is most sensitive to the vertical density stratification and the wind stress profile (Walsteijn [33]). Depending on the wind stress profile the ACC consists of one or more jets. When wind stress has a significant curl, an ACC jet emerges that is forced towards the latitude where the curl crosses zero, i.e., where the wind stress is maximal. An alternative or second jet occurs if the wind stress is large at latitudes south of Drake Passage. Then the volume transport increases significantly as does its sensitivity to the stratification. The time-mean transport of simulations with a large domain and broad wind stress profile (e.g., Fig. 4) is of the same order as the observed value (cf. Nowlin and Klinck [29]). The transport history of the model shows realistic time scales, such as mild modulations with periods on the order of 6 months to several years. However, barotropic fluctuations with a period of roughly 25 days have an amplitude which is about three times larger than corresponding fluctuations in observations. Using a more realistic topography (e.g., by including small-scale "bottom roughness") would probably reduce the
355 magnitude of these fluctuations to more realistic levels. The dynamics of ACC jets are determined by transient eddies and the density stratification. Eddy activity is strongly affected by changes in stratification (Walsteijn [33]). The strength of the reverse feedback is still unknown, i.e., it is unclear to which degree the eddy heat transport is coupled with the stratification and thermohaline overturning. This is an area of active research. A full description of (sensitivities of) the ACC variability and transports requires both eddy-resolution and more complete thermodynamics than is present in a quasi-geostrophic model. For future climate research it is, therefore, important to extend the present sensitivity study to a system of primitive equations. The possibility to synoptically study the ocean's eddy field and related mean circulation from satellite altimeter observations has been verified and explored in the second part of this study. Progress has been made concerning the generation mechanism, formation rates, trajectories, translation speeds, and lifetime/dissipation rates of eddies in the Southern Ocean western boundary currents. Over the 3 year observational period of Geosat approximately 18 large rings pinched-off from the Agulhas Current, south of Africa. The majority of these rings reaches the South Atlantic. Just after being formed Agulhas rings have the largest translation speed, approximately 8 cm/s. When they move out of the highly active Agulhas retroflection area their translation speed reduces to approximately 4 cm/s. The associated volume transport on a yearly basis is at least 7 • 106m3/s, approximately half of the total exchange between these two oceans. We therefore conclude that Agulhas rings contribute significantly to the Indian-South Atlantic connection and the associated heat and fresh water flux. Van Ballegooyen et al. [1] estimated from a combined hydrographic-altimeter study that the net heat fluxes (300 Wm -2) and evaporative losses (1 cm/day) to the atmosphere due to eddies within the Agulhas region are appreciable larger than the summer climatological means for this region. Their volume flux 6 . 3 - 7.3 • is consistent with other estimates. Similarly, they estimate fluxes of heat and salt into the Atlantic Ocean via the Agulhas eddy field of 0.045 PW and 78 • 1012 kg per year, respectively. In the concluding part of this work, which is still ongoing, we try to better determine how the eddy field interacts with and modifies the mean circulation and how it is coupled to the deeper circulation. Coupling with numerical models will lead to improved estimates of meridional and cross frontal fluxes. First results are promising and show that it appears to be possible to improve the estimates of the mean sea surface dynamic topography by using satellite altimeter observations of horizontal eddy momentum and vorticity fluxes (Feron et al. [15]). Partly due to the success of Geosat, new altimeters are now operational. With the successful European Remote Sensing Satellite (ERS1) and Topex/Poseidon new altimeter data is becoming available with increased absolute accuracy and precision (Haagmans et al. [21]). Aside from major improvements in the new altimeter observations, the continuity of observations and their accurate analysis is a high priority for future climate studies aiming at Global Ocean and Climate Forecasting. Acknowledgment. This investigation was supported by the Dutch National Research Programme on Global Air Pollution and Climate Change, projectnumber 850025, and the Stichting Ruimte Onderzoek Nederland (SRON). Computations were performed on the CRAY Y-MP and CRAY C90 at the Academic Computing Centre (SARA), Amsterdam, The Netherlands. Use of these supercomputer facilities
356 was sponsored by the Stichting Nationale Computerfaciliteiten (National Computing Facilities Foundation, NCF), with financial support from the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (Netherlands Organization for Scientific Research, NWO). Delft University of Technology (section for Space Research and Technology and Faculty of Geodesy) is acknowledged for the altimeter data processing.
References [1] Ballegooyen, R.C. van, M.L. Grfindlingh, and J.R.E. Lutjeharms, J. of Geophys. Res., 99, 1405314070, 1994. [2] Baker, D.J., Jr., J. Mar. Res., 40, suppl., 21-26 (1982). [3] Broecker, W.S., Oceanography, 4, 79-89, 1991. [4] Boyle, E.A. and L.D. Keigwin, Nature, 330, 35-40, 1987. [5] Boudra, D.B. and W.P.M. De Ruijter, Deep Sea Res., 33, 447-482, 1986. [6] Cheney, R.E. and J.G. Marsh, EOS Trans, AGU 62(45), 743-752, 1981. [7] Dansgaard W., J.W.C. White, and S.J. Johnson, Nature, 339, 532-533, 1989. [8] De Ruijter, W.P.M., J. Phys. Oceanogr., 12, 361-373, 1982. [9] Douglas, B.C. and R.E. Cheney, J. of Geophys. Res., 95, 2833-2836, 1990. [10] Drijfhout, S.S., PhD thesis, Utrecht University, 1992. [11] Feron, R.C.V., W.P.M. De Ruijter, and D. Oskam, J. Geophys. Res., 97, 9467-9477, 1992. [12] Feron, R.C.V., Change, 11, 6-7, 1992. [13] Feron, R.C.V., Satellite Altimetry in Geodesy and Oceanography, Springer-Verlag, Berlin, 1992. [14] Feron, R.C.V., accepted in J. Geophys. Res., 1994. [15] Feron, R.C.V., W.P.M. de Ruijter, and P.J. van Leeuwen, Submitted to J. Geophys. Res., 1994. [16] Feron, R.C.V., PhD thesis, Utrecht University, 1994. [17] Godfrey, J.S., Geophys. Astrophys. Fluid Dynamics, 45, 89-112, 1989. [18] Gordon, A.L., J. Geophys. Res., 91, 5037-5046, 1986. [19] Gordon, A.L., J.R.E. Lutjeharms, and M.L. Griindlingh, Deep Sea Res., 34, 565-599, 1987. [20] Gordon, A.L. and W.F. Haxby, J. Geophys. Res., 95, 3117-3125, 1990. [21] Haagmans, H.N., M.C. Naeije, and R.C.V. Feron, Geodetical In]o Magazine, 5, Nov-Dec 1993 [22] Harvey, L.D.D., Quaternary Science Reviews, 8, 137-149, 1989. [23] Johnson, G.C. and H.L. Bryden, Deep-Sea Research, 36, 39-53, 1989. [24] Lutjeharms, J.R.E., W.P.M. De Ruijter, and R.G. Peterson, Deep Sea Res., 39, 1791-1807, 1992. [25] Marshall, J., D. Olbers, H. Ross, and D. Wolf-Gladrow, J. Phys. Oceanogr., 23, 465-487, 1993. [26] McWilliams, J.C., W.R. Holland, and J.H.S. Chow, Dyn. Atmos. Oceans, 2, 213-291, 1978. [27] Naeije, M.C., Wakker K.F., R. Scharroo, and B.A.C. Ambrosius, ISPRS, J. Photogramm. Rein. Sensing, 47, 347-368, 1992. [28] Nerem, R.S., B.D. Tapley, and C.K. Shum, J. Geophys. Res., 95, 3163-3179, 1990. [29] Nowlin, W.D., Jr. and J.M. Klinck, Rev. Geophys. Space Phys., 24, 469-491, 1986. [30] Rapp, R.H. and Y.M. Wang, Geophys. J. Int., 117, 511-528, 1994. [31] Straub, D.N., J. Phys. Oceanogr., 23, 776-782, 1993. [32] Wakker, K.F., R.C.A. Zandbergen, M.C. Naeije, and B.A.C. Ambrosius, J. Geophys. Res., 95, 2991-3006, 1990. [33] Walsteijn, F.H., In preparation, 1994. [34] Webb, D.J., P.D. Killworth, A.C. Coward, and S.R. Thompson, Natural Environment Research Council, 1991.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
Determination of A R G O S drifters.
NE.
the
Atlantic
357
current
field
with
L.Otto, H.M. Van Aken and R.X. de Koster. Netherlands Institute for Sea Research, P.O.Box 59, 1790 AB den Burg, Texel, The Netherlands. The patterns of near-surface currents in the ocean as we know them today only give a very general picture, because they are based upon data that and calculations that are not adequate in every respect. Regional and temporal variability cannot be resolved in sufficient detail and this is for instance reflected in uncertainties in the estimates of the transport of mass and heat in the ocean. Yet these estimates are important for the assessment of the role of the oceans in the climate system. For example, in the northern part of the North Atlantic, north of latitude 53 ~, the near-surface circulation follows an anticyclonic pattern (the Sub-Arctic Gyre), with northward flow in the eastern part, and southward flow in the west. But how much of the water flows into the Norwegian Sea beyond the Scotland-Iceland Ridges, and along what routes, and how much of it turns westward south of Iceland, is still a matter of dispute. As the Sub-Arctic Gyre is an important link in the thermohaline ocean circulation (the "Conveyor Belt") better information is required for a realistic modelling of the oceanic part of the climate system.
,F o,._ ~
Gyre
f .... IJ /
50 N
i
60 W Fig.1.
/
~
,
ql
". . . . . . . . .
.---
4o N
]
Sub_Trropica I
Gy e )
"
i
40 W
i
20 W
/
0
General pattern of the surface circulation of the NE. Atlantic and the "DUTCH-WARP" study area.
358
Since about a decade the use of satellite-followed drifters offers a new possibility for observing ocean currents. The principle is that drifters that are released in the ocean can be followed regularly from satellites, and the analysis of the tracks thus obtained gives information on the mean current pattern, and the variability in time, the occurrence of eddies, etc. In the framework of the "DUTCH-WARP" programme ("Deep and Upper Transport, Circulation and Hydography, WOCE Atlantic Research Programme"), aiming at a better description of the circulation of that part of the Atlantic, a series of drifter observations was initiated, that has been supported by the VvA-3 programme. In the years 1990-1993 in total 19 drifters were released from the RV "Tyro" and the weathership "Cumulus" in the NE Atlantic. The drifters were drogued at 15 m, and were followed over periods between 43 and 365 days. In total the data cover some 10 drifter-years. The mean drift velocity for the area is typically of the order of 2 cm/s, to the northeast. However, the tracks reveal an important effect of the submarine topography and the main thermohaline structure on the regional current pattern. During the summer mean westward flow is observed over the western part of the Iceland Basin, and northeastward flow over most of the eastern parts. Over the Rockall Plateau the currents are variable and smaller. It is also interesting to compare the drift observed across the WOCE AR-7E section (IrelandGreenland) with the hydrographic structure observed during "DUTCHWARP" 1991. As for many ocean areas the eddy kinetic energy in the area is much higher than the mean kinetic energy. This means that the role of eddies in the transport of heat and salt cannot be neglected. An interesting result is that there is a marked difference between the high levels of eddy kinetic energy over a deep (> 2000 m) region as the Iceland Basin, and the much lower levels over the shallower (< 1000 m) parts of the Rockall Plateau. The results found here show that transport and exchange of water over parts of this area can be quite different. As there are indications that the convection over the Rockall Plateau is an important mechanism in the formation of the so-called "Mode Water", the transport and exchange between this area and the surrounding waters is an important point in the air-sea exchange of the NE Atlantic region. In the coming years drifter data will be used to obtain improved maps of surface currents. The programme reported here is a contribution to this international effort.
"Jewwns u! 8(s/wo) ssew l!un Jed AfJeue o!leU!>l Appe eoetJns to uo!lnq!als!a 00
I.ooo.,,r It-
k te ,' ~~9~
o0 ~
~ #
~,
~."
'.~~x
,.'M'
'Y
,'(
~
.,
/-"
i~
'
,
i "
"
.',,.
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ii::::::::f::::::::;? ,-7" :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: .; ::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::: .....
~
X~i::~:,~i~-~,~; -
,.;>;
,..:;;
/ ./ ....
. - .... /
. . . . . . ,,-
o
~
./
?:, .
!;
/,
o
/'i
Y.
;')
,, i..;.. 5 , ' ,,
"(epnl!BUOlo8 x epnl!lelo L) u!q Jed 866 ~-066k elep ~euJwns LUOJt pe^!Jep lueJsno eoe~Jns u e e ~ o
i'tOQ9N ~
J, , ,
I o9-I
i
I" J
...... ' ~ ~ ' ~
?
/~--7
*
/,.-'
0s
6S'~
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
361
Repeated XBT sections in the framework of WOCE T.F. de Bruin, L.Otto, S. Ober, R.X de Koster, H.M. van Aken Netherlands Institute for Sea Research (NIOZ) P.O. Box 59, 1790 AB Den Burg, The Netherlands The upper layers of the ocean have an important role in the coupled oceanatmosphere system. They exchange heat with the atmosphere on seasonal and interannual time scales and the meridional transport a substantial part in the global heat budget is via the upper layers of the oceans. In the observational programmes of T O G A and W O C E special attention is given to a co-ordinated system of regular observations of the thermal structure of the upper ocean by m e a n s of XBT (eXpendable BathyThermograph) measurements. XBT observations can be made from ships underway. The XBT sonde consists of a thermistor probe that is launched from the ship and that sinks with a well-known falling speed, transmitting its temperature signal to the recorder on board via a thin unwinding copper wire. As a result a temperature-depth registration is obtained down to depths of about 400, 700 or 1800 m (depending on the type of probe). When completely unwound the wire breaks. The advantage of this method is that data can be obtained by "ships of opportunity" without interfering with their normal duties.
TABLE 1. Overview of XBT measurements in the joint Navy-NIOZ project. Royal Netherlands Navy frigate
Period
Number of XBT's Total
T7
T5
Banckert Piet Heyn
14-05-91 / 24-05-91 25-05-91 / 04-06-91
85 82
85 82
0 0
Kortenaer Banckert
05-11-91 / 13-11-91 17-11-91 / 26-11-91
113 117
94 94
19 23
Philips van Almonde Kortenaer
19-05-92/ 29-05-92 31-05-92 / 09-06-92
114 104
94 78
20 26
Callenburgh Philips van Almonde
03-11-92 / 12-11-92 16-11-92/ 24-11-92
115 108
93 91
22 17
Bloys van Treslong Callenburgh
25-05-93/ 01-06-93 07-06-93 / 15-06-93
95 106
75 91
22 15
Karel Doorman Bloys van Treslong
09-11-93 / 18-11-93 21-11-93/ 30-11-93
111 116
88 92
23 24
Willem van der Z a a n Karel Doorman
17-05-94/ 23-05-94 04-06-94 / 08-06-94
92 61
72 61
20 0
Karel Doorman Willem van der Z a a n
22-11-94 / 30-11-94 05-12-94/ 13-12-94
114 125
91 104
23 21
1658
1385
273
Total:
362
Naval ships normally make XBT observations in connection with anti-submarine programmes, but as a rule not at high spatial resolution. The programme reported here is a co-operation between the Royal Netherlands Navy and the Netherlands Institute of Sea Research, with support from the VvA-3 programme. The regular relieving of the frigates stationed at Curas offers an opportunity to carry out XBT observations (T-7 probes, down to 700 m) on the route English Channel-Antilles (the WOCE AX-5 section) at 30 miles intervals, twice in spring and in fall. This spatial density ("highdensity mode") resolves much of the meso-scale structure, while the time schedule gives an opportunity to investigate seasonal variability in heat content and heat transport. Fig. 1. MAP OF XBT MEASUREMENTS
MAY 1991- JUNE 1994
50 40 -1l-.rr" O 3O z uJ a D !- 20
~
...-,, ........
.
...........e e ~
I
I
oee .......:.:.o.
. . . . . .. . . . . .
~
r--
9 .. ~ "'r
...., .,~..-~ -
f
I-< "-J 10
-0
. ....:':
i
,f"
I
-80
,~,'t'..~~
L
"~
-60
I
-40
.
I
-20
0
LONGITUDE (WEST) Since the beginning of the programme in 1991 16 sections have been accomplished (see Table 1). Figure 1 shows the tracks followed. Although for operational reasons the routes followed by the ships may differ, most of them are sufficiently close together to permit comparison and analysis of variability in time. Figure 2 shows a typical spring and autumn section. Features that can be recognized are the stratified (spring) and vertically mixed (autumn) conditions northeast of longitude 25~ the frontal structure near the Azores near 35~ the thickness of the 18 ~ mode water in the Sargasso Sea and the shallow thermocline in the Antilles area, southwest of 50~ In addition to the T-7 probes intermittently T-5 probes, that go down to 1800 m are launched along the NE part of the section. These observations (not represented on the results shown here) show the extent of the Mediterranean outflow. Preliminary results are published as NIOZ data reports [1]. A programme like this has its potential in its continuation over many years. Similar programmes should make part of a future ocean component of the GCOS (Global Climate Observing System). 1.
T.F.
de Bruin, R.X. de Koster, S.Ober and L. Otto, 1992.
Netherlands XBT programme along the WOCE AX-5 section. NIOZ Data-report 1992-3.
363
Temperature section
Hr. Ms. Banckert
May 1991
,oo 200
1 U
300 r
~- 400 a
500 600
7oot
.
I
'
I
-65
'
-60
;
-55
-50
'
-45
'
-40
I
I
-35
-30
-25
I
I
-
-20
-15
-10
Longitude (~ Temperature section 26
0
Hr. Ms. Karel Doorman
24
'
2
'
'
November 1993
22
0
'
11
/
(
.. fT-x 4"]}./-'x
100 200 ._. 300 E r
~- 400 a
500 600 700 '
I
I
-60
-55
'
I
-50
'
I
-45
'
'
-40 -35 Longitude (~
'
'
-30
I
-25
'
I
-20
'
I
I
-15
-10
Figure 2. Typical spring (top) and fall (bottom) temperature contourplots. The main difference between these plots is the clear presence of a well mixed surface layer of about 70 meters during the fall. The sharp frontal structure at about 35 ~ is the Azores Front. The cold water at 60 - 65 ~ is Antarctic Intermediate Water (AAIW), transported northwards along the South-American east coast into the Caribbean Sea.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
The ASGASEX
365
program
W.A.Oost Royal Netherlands Meteorological Institute (KNMI), De Biit, the Netherlands. Abstract There is a difference of sometimes more than an order of magnitude in the transfer velocity of CO2 at the sea surface if data based on concentration measurements are compared with direct flux measurements with the eddy correlation technique. To explain this discrepancy we performed the international VIERS-I and ASGASEX (for Air Sea GAS EXchange) experiments at the research platform Meetpost Noordwijk off the Dutch coast. The data indicate that the solution of the controversy may lie in the presence of an unexpected vertical concentration (fugacity) gradient of CO2 in water. Acting on this result another experiment (ASGAMAGE) is planned for 1996 to confirm or refute this conclusion. On that occasion the fluxes of a number of other trace gases beside CO2 will be measured with more methods and the CO2 profile will be monitored over the full water column. 1.THE PROBLEM The oceans play a crucial r61e in the carbon dioxide balance of the earth. Over 71)% of the earth's surface is covered with water. The CO2 transport (or CO2 flux) between the oceans and the atmosphere is not well known. Two types of methods have been used to measure CO2 fluxes: those based on chemical concentration measurements, and those using the eddy correlation technique. The chemical methods all need one or more assumptions that are possibly not always fulfilled and they require measurement times in excess of 24hrs. The eddy correlation method measures fluxes directly without such assumptions and within about half an hour, but suffers from a lack of accuracy, because the instruments used are functioning at the very limit of what's technically possible. The differences between the results of the two types of methods, however, are sometimes orders of magnitude, which is way outside the expected uncertainties [1]. 2.METtlOD
AND RESULTS
The 1990 VIERS-1 experiment and the 1993 experiment of the Air Sea GAS EXchange (ASGASEX) project, both supported by NOP, were designed to establish the feasibility of COz eddy correlation measurements from a relatively bulky platform and to explain the differences just mentioned by comparing the results of the two types of methods. The VIERS-I results showed that COz eddy correlation flux measurements were possible at Meetpost Noordwijk (MPN), a research platform 9km off the Dutch coast. ASGASEX'93 took place in September 1993 on the same platform. A large number
366 of environmental parameters, were measured to compare both types of methods.
Fig. 1 Research platform Meetpost Noordwijk The A S G A S E X '93 CO2 flux data, plotted as a function of dCO2 (the difference between pCO2, the concentration in the water, and the concentration in the air) showed no relationship at all. Plotted versus the wind speed, however, they showed a fairly systematic behavior, indicating trustworthy data. Looking more closely into the data we found a correlation of pCO2 with the tide. pCO2 had been measured using a pump, fixed to the platform. The depth from which the water was brought up therefore changed with the tide, so the correlation could indicate a vertical pCO2 gradient at the level of the pump. 120 ]"[-:~::::::.......................................................................... i:i:i:i:::!~:i:ii:i: ::i+:i:i:i::i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:::i:i:i:iiiii:i::: i iii~:iii:i:ii:i~iiiiiii!iiiiiii~:iii:!i~iiiiiiiiiii!iiiiiiii~ii~iiiiiii!iiiii!i:i!i!ii iiiiiii 600
,ooii i~!i ii~iiiiiiil ~!
~ili
~
i
i!
i~
8o i',iiiiiiiiiiil ~!ii!',iiiiiiiiiiiiiiiiiiiiiii!iiiiiiiiii ~ii~iii?,!,'iliii~i~,-~ili!iiiiiii~:iiiiiiiilili~i::!iiiiiiiiiii!iiiiiiU"~!iiill i!i 550 i!iiiiii!i',ii!!iiii,'~i~,iii'ii!iiiliiiiiiiii',i~iiii iii~iii!ii~' ii~i/~i i,!i!iii!i!ii!iiili~i!i!iii,'i,'ii~,!i~!!i ,'iliiliiiiiii,'i~i!i~' ,~i 'ii,'i~i ~o iiiiiii~!',ii!!!iii!i~,i:i~' ii,'iii!iiiii~,~~!?~ii!iiiiiiiiii i:ilili!',!ii!ii!'~.i ii!ii!ii!',iiiii~!i !illi!,'i:,!i!i!i,'iii~' ii~/ :~:ili!:J:iiill iii!:,i:.iiiii:-:.iiiU: "ii!::i!ii::::ii:~i ~ ::!!::!ii~.::~iiii!!ii::i:~i i.~ ~iiiii::!::!:#.ilI~
,o~ ~ ~ o iiiii'ii !iii?
iliiii~iill iii!ii!ii!i'
~!~i~i ~ ",iiiil !iiiiiiii
iiii~#:iiii~:i::i i::i:/ji:::/j::i:~!iiiiii:~!iiiii ii:.
500
i~!i~
~!iiii'i:ilili!iii,i~! 9 ~ii i!i~,ii iii!iii',!ii: ii~ii!i
- o iiii!i:i.iiiiiiiii::i~:: ~..... ..ii iiiiiii!i::,, iiiiiiii .i~iiiiiiiiiiiiiiiiiiii:~!iiill~ ii!iiii i::i::i::i::iii!~iiiili:::.ii ................. ~i:!:i~i!!iiiiiiiiiiiii~ii; ~ii::i!::i::i!i!!ii ::i:::::.i ::i!:.!i
!i,ili~!! 'i~!i',iiii~,~ 9 iiiiiiiiiiiiiiiii i i i'~i i i i!ii i!i~i ~i~i~ ~:~:~'
21.5
22.5
o ~o.
:::!ii::iiiii!iii~::i 400 .....:::i:~:i~ii::~!:::.i!
-80 !:igi!~!~!i~!iiiii!ii::!ii::il;: :~i~5~i!::!i::~!gi1)i!~ii~!ig!!i::)==!g{~iii~)~i!i~:::~:.:::!?::~i~!i!~i!i;ig:!::;iigi:.~ii:: 20.5
450
23.5
350
Fig.2 pCO2 (see text) and the tide (heavy single line) versus time Our measurements also indicated that the wind speed affects the COz concentration in the water. This "wind effect" may be assumed to be absent at wind speed zero.
367 Extrapolating to that value we therefore find the "deep" concentration difference actuaily driving the CO2 transport, in this case 272#atm. 3oo i~iii!i:iiiiiiiliiiiiiii!!iiiiii!!iiiiii 250
!I
i ~ii i!i i i!i i iilii!~i ~;:~iiiiiii!i!~ii i i!i i i i !i !i i i i i i i ii i i ii~ii!i i i !i i i i i i ~i ii il i!i!i i i i!iii~l i~i i ~ i !i !i i i !i i i i i i i i i i i i ~!ii iiiiiiilii!i 200
c~
iiiliiiilli! i i i i i i;i;i iiiiiii;iii~.!~!iiiiiiiiiiiii~F ilii2?
!ii!i!iliiiiiiiiliiii!i!!ii i il i iiiiil;iill!ii
iii
-
::::::::::,:::::
0
...-::::::::~:::::==================================================== . . . . ::::::::::~: -
,:::::~::
iiii ii iiii !i:i i isi~i!iiisi!iss/iss:::::::::::::.:.:.:.
0
"...........
4
::,i:i
:-
8
::
.:.:::,::::::::-~((i:~s~:~s~s~ssii!siiis:~s~s~ ....
12
iiii
:~:",:s~si
~rssl
16
wind speed (m/s)
Fig.3 CO2 concentration difference (sea-air) against wind speed. With this value of the concentration difference we find a good correspondence with parametrisations for the CO2 flux based on concentration measurements (here the one of Wanninkhof [2]). This in principle could solve the controversy mentioned in [1] and so pave the way for improved accuracy in the measurement of COz transport over the oceans.
0.s ~i~ii!~ii~!ii~i~i~i!!~iii~ii~i!i!ii~i!i;;~ii~i;iii!i!~ii~iiiiii~iii~iiii~iiiiiii~i~!;i!i~i~i!ii~i~i~!i~ii~!wa"~"kh~f ~oael w th opco2.272 .,mat., ~iiiiiii!ii!iiiii!iiii!iiiil iil;iii!iiii!i;;;I
i 0!i!i!!!i!!!!!i!!!!! 0o -0.5
-1
0
5
10
15
20
wind speed (m/s)
Fig.4 Measured flux values (open squares) compared with the results of the Wanninkhof parametrization. If, as the ASGASEX'93 results suggest, the CO2 concentration (fugacity) in water can vary with depth in the first few meters below the surface, the use of the CO2 concentration close to the surface when comparing the flux methods - as has been done so far! - will lead to spurious values for the transport coefficients. This insight can explain the differences between the two types of methods and may lead to a greatly improved reliability and accuracy of CO2 transport between sea and air.
368 3. F U R T H E R
NEEDS
The ASGASEX '93 results require confirmation. The crucial data were obtained almost by accident and their interpretation is not unequivocal. More information is needed on the presence or absence of a COz gradient. To improve the accuracy of gas flux data over the sea we furthermore need better and generally applicable measurement methods. These methods must be tested and compared to assess their value. The eddy correlation method, as used presently, can measure the transport of only a single gas simultaneously and these measurements can so far only be done from a stable platform i.e. near the coast. However, due to the high quality of present-day motion sensing devices, it has become possible to allow with sufficient accuracy for the movements of e.g. a ship. Another new technique, the eddy accumulation method, holds promise of simultaneous and direct measurement of the fluxes of a number of gases. This would bring the direct measurement of gas fluxes on the high seas within reach. 4. T H E F U T U R E :
ASGAMAGE
In 1996 there will be another experiment at Meetpost Noordwijk, called ASGAMAGE (MAGE, for Marine Aerosol and Gase Exchange, is a working group of the IGBP IGAC (International Global Atmospheric Chemistry) program. There will be two five week measurement periods (one starting in May, the other in October). Several eddy correlation systems for COz fluxes will be functioning simultaneously and one or two eddy accumulation systems will also be present. One or two ships will make measurements in the MPN area to detect horizontal concentration gradients, one of them may be used for an attempt to perform eddy transport measurements from on board. Fluxes will be measured not only of COz, but also of DMS (with the eddy correlation technique!), CH4 and N20. 5. R E F E R E N C E S 1 W.S.Broecker et al, J. Geophys.Res. 91, (1986), 10517-10527. 2 R.Wanninkhof, J.Geophys.Res.. 97, (1992), 7373-7382.
Institutes (in alphabetical order) participating in A S G A S E X / A S G A M A G E : Bedford Institute of Oceanography (BIO) Dalhousie University Institute of Oceanographic Sciences (IOS) Riso National Laboratory Royal Netherlands Meteorological Institute (KNMI) Netherlands Institute for Sea Research (NIOZ) Southampton University TNO-Physics and Electronics Laboratory (TNO-FEL) University College Galway
Canada Canada UK Denmark the Netherlands the Netherlands UK the Netherlands Ireland
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
369
E1 Nifio and mixed-layer processes M a r c v a n E i j k I en G e r r i t B u r g e r s Royal Netherlands Meteorological Institute, Division of Oceanographic Research, P.O. Box 201, NL-3730 AE De Bilt, Netherlands Abstract E1 Nifio, the dominant form of interannual variability of climate, is caused by interactions between the tropical ocean and atmosphere which exchange heat and momentum. How the tropical ocean responds to atmospheric forcings depends sensitively on its mixing properties. So the nature of the E1 Nifio depends on the oceanic mixing properties too. The present description of mixing processes in ocean general circulation models is not fully satisfactory. Parameters are tuned to the best representation of E1 Nifio without much regard how well they represent small scale processes or not. Even then the models have difficulties in simulating both the background state and E1 Nifio's correctly. We want to use ideas from atmospheric general circulation models, in particular from non-local boundary-layer model schemes, and apply them to ocean models. So far, this has been hampered by a lack of data, but we hope that recent experiments like TOGACOARE will provide enough new information.
1. I N T R O D U C T I O N El Nifio, Spanish for the Christ child, has historically been associated with a weak, warm current appearing along the coast of Ecuador and Peru annually around Christmas, replacing the usual cold waters of the Peru current. Nowadays, the name El Nifio tends to be used for a much larger scale phenomenon that occurs not annually, but every three to seven years in which the normally cold waters over the entire eastern Pacific Ocean show a dramatic warming. Also, very large anomalies in the oceanic and atmospheric circulations and in the global weather are associated with these changes in the equatorial sea surface temperature. E1 Nifio is the dominant form of interannual variability of climate. Severe effects occur all over the world, like droughts in Australia, and reduced fishing near the coast of Peru. At the time of this conference, it looks like that a new E1 Nifio is about to start. One of indicators is the low pressure difference anomaly between Darwin and Tahiti over the last half year, another indicator is that the sea surface temperature (SST) around the date line is higher than normal and that this area of warm SST is moving westwards. 1Supported by the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO), project VvA-2-250.
370 2. O C E A N - A T M O S P H E R E
INTERACTIONS
The E1 Nifio phenomenon (Philander, 1990) is caused by interactions between the tropical ocean and atmosphere which exchange heat and momentum. The atmosphere influences the ocean mainly through the stress exerted by the surface winds, and to a lesser extent through the heat flux and the precipation and evaporation. The ocean in turn influences the atmosphere through the sea surface temperature. During an E1 Nifio, weakened easterlies cause warm western Pacific water to flood over the cold eastern water and to slow down the upwelling of cold water at the coast of South America. The barrier between warm surface waters and cold deep ocean waters, the thermocline, which normally slopes upward from west to east, becomes more fiat. This also contributes to the temperature rise of the waters in the east. The shift in sea surface temperature is accompanied by a shift of the major rain zone to the east. Related adjustments in the atmosphere cause a further weakening of the easterlies in the central Pacific. In this way all the coupled influences amplify each other, until eventually one can speak of a full-blown E1 Nifio.
3. F R I C T I O N A N D M I X I N G IN T H E T R O P I C S Friction plays a more important role in the dynamics of the tropical ocean-atmosphere system than in mid-latitude systems. This is because at the equator pressure gradients can be balanced only by frictional forces, while at higher latitudes there is the possibility of a balance between the Coriolis force and the pressure force. Indeed, strong ocean currents are found along the equator. In some places, the currents have such a strong horizontal or vertical shear that mixing is much enhanced. Horizontal friction is the main limiting mechanism for equatorial currents (Crawford 1982). Vertical mixing also affects the strength of the current, but, more important, controls the shape of the thermocline (Pacanowski and Philander, 1981). In a large region in the Western Pacific, there is a deep (about 200m), mixed layer of warm water (around 30 C). The heat which can be stored in this deep warm pool plays an essental role in triggering and maintaining E1 Nino events. The evolution of the deep warm pool depends on mixed-layer processes. All this has been confirmed by numerous studies, including some where KNMI was participating (M. Latif et al., 1994). At the equator, in the Pacific, a strong eastward countercurrent, the Equatorial Undercurrent, flows below the westwards flowing surface waters. The core of this current follows approximately the thermocline. In the high-shear regions around the undercurrent, mixing is substantially enhanced. The dynamics of the undercurrent is quite sensitive to this mixing. General Circulation Models without a proper representation of the undercurrent have difficulties in making good SST simulations. Mixing properties affect both the E1 Nifio's, and the background mean annual cycle. In general, ocean general circulation models (OGCM's) have difficulties in reproducing both good E1 Nifio's and a good mean annual cycle. But this is necessary if one wants to study how global climate change might affect the character of E1 Nifio.
371 4. M I X I N G F O R M U L A T I O N IN O C E A N M O D E L S The present description of mixing processes in OGGM's is not fully satisfactory. The usual mixing formulation for the shear around the undercurrent (Pacanowski and Philander, 1981) is rather ad hoc and parameters are tuned to the best representation of El Nifio without much regard how well they represent small scale processes. Even then the models have difficulties in simulating both the background state and E1 Nifio's correctly. For the surface mixed layer, mixed-layer schemes of the Niiler-Kraus (1977) type often are used, but they were originally developed for mid-latitude situations, and whether they can simulate the deep warm pool in the West-Pacific and the surface waters above the undercurrent is questionable.
5. N E W M I X I N G S C H E M E S Models of the atmospheric boundary layer, both stable and unstable, are more versatile and seem to have reached a more advanced state of development (Holtslag and Nieuwstadt 1986). Holtslag and Moeng (1991) have found a way to parameterize countergradient heat transport in the convective boundary layer. A module based on this parameterization was made for the NCAR Community Climate Model by Holtslag and Boville (1993). We want to apply these ideas, in particular the non-local schemes, to the oceanic boundary layer. First a 1-D model for a column of water will be made which can simulate well the detailed evolution, including day-night contrast and how the mixed layer responds to a shower. So far, testing such models has been hampered by a lack of data, but we hope to be able to do that with data from recent experiments like TOGA-COARE. Next a parameterization will be devised which is suitable for implementation in an OGCM with a resolution of the order of 15m in the first 150m, and the consequences for E1 Nifio will be investigated.
6. R E F E R E N C E S
Crawford, W.R. 1982: Pacific equatorial turbulence. J. Phys. Ocean. 12, 1137-1149. Holtslag, A.A.M., and Nieuwstadt, F.T.M. 1986: Scaling the atmospheric boundary layer. Bound.-Layer Meteor. 36, 201-209. Holtslag, A.A.M., and Moeng, C.-H. 1991: Eddy diffusivity and countergradient heat transport in the convective atmospheric boundary layer. J. Atmos. Sc.48, 16901698. Holtslag, A.A.M., and Boville, B.A. 1993: Local vs. non-local boundary-layer diffusion in a global climate model. J. Clim. 6, 1825-1842. Latif, M., T. Stockdale, J. Wolff, G. Burgers, E. Maier-Reimer, M.M. Junge, K. Arpe and L. Bengtsson 1994: Climatology and variability in the ECHO coupled GCM. Tellus 46A, 351-366. Niiler, P.P. and Kraus, E.B. 1977: One dimensional models of the upper ocean. In: Mod-
372 elling and prediction of the upper layers of the ocean, E.B. Kraus, ed., Pergamon Press, 143-177. Pacanowski, R.C. and Philander, S.G.H. 1981: Parameterization of vertical mixing in numerical models of the tropical oceans. J. Phys. Ocean. 11, 1443-1451. Philander, S.G.H., 1990: El Nifio, La Nifia and the Southern Oscillation. Academic Press, 291 pp. Yin, F.L., and E.S. Sarachik 1993: Dynamics and Heat Balance of Steady Equatorial Undercurrents. J. Phys. Ocean. 23, 1647-1669.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
375
LAND ICE AND SEA L E V E L CHANGE I n s t i t u t e for Marine and Atmospheric Research, Utrecht University in co-operation with the University of A m s t e r d a m and the Free University (Amsterdam) sponsored by - Dutch National Research P r o g r a m m e on Global Air Pollution and Climate Change - N e t h e r l a n d s Organization for Scientific Research (NWO) - N e t h e r l a n d s Antarctic Research P r o g r a m m e - E u r o p e a n Commission (ENVIRONMENT) Abstract Sea-level change is an important issue in the greenhouse problem. All workers agree t h a t predictions made so far have a high degree of uncertainty. Comparable contributions to this u n c e r t a i n t y come from limited knowledge of future emissions of greenhouse gases, different opinions concerning the response of the climate system, and inadequate knowledge of the sensitivity of land ice to climate change. The goal of this project is to improve estimates of the contribution of land ice masses to sea-level change in the coming 150 years.
MASS B ~ C E
OF THE GREENLAND ICE S H E E T
The mass balance of the Greenland ice sheet has been studied with an energy balance model. The mass balance is generated from climatological input. Data from several field experiments have been used to improve the p a r a m e t e r i z a t i o n of energy transfers between atmosphere and surface. The grid resolution is 20 km. The picture below shows the calculated surface mass balance for the "reference case". Sensitivity tests reveal that a 1K warming implies a 0.30 mm/year sea-level rise a 1K warming (+ dP) implies a 0.21 mm/year sea-level rise a 10% increase in cloudiness implies a 0.02 mm/yr sea-level drop a 0.02 decrease in albedo implies a 0.17 mm/year sea-level rise [dP is a change in precipitation in proportion to saturation vapour pressure]
376
m of water equivalent per year (nonlinear scale) +2.0
+1.0
+ 0.5
+ 0.25
+o.o
- 2.0
Conclusion:
Greenland ice sheet, +IK: 0.21 mm/year sea-level RISE (best estimate) G L A C I E R S AND SMALL ICE CAPS A model has been designed t h a t s i m u l a t e s m a s s b a l a n c e profiles on glaciers. It has been tested on 12 glaciers for which good observations exist. After careful calibration a large n u m b e r of sensitivity tests have been carried out. There appears to be a significant correlation between glacier sensitivity and precipitation regime. The figure below shows the result for an experiment with u n i f o r m 1K w a r m i n g and an i n c r e a s e in p r e c i p i t a t i o n p r o p o r t i o n a l to s a t u r a t i o n vapour pressure of the air. The m e a n loss of ice, averaged over the entire glacier, is shown for the 12 glaciers studied.
377
1 " ~>, E
0.8
.,_....
0.6
.o_ ,,.,,_. 0
0.4
__o t--
0.2 E ~
0
1
0
2
3 4 5 annual precipitation (m)
,!
I
6
7
,,
8
Extrapolation of this result to all glaciers and small ice caps outside Greenland and Antarctica yields a sea-level rise of 0.46 mm/yr for a uniform 1K warming (this includes increasing precipitation). This is about half the value of the 1.2 + 0.6 mm/yr used in IPCC-1990. The difference is due to an earlier overestimation of glacier sensitivity in the dry subpolar regions, where a large amount of glaciers and small ice caps are located (see below). 100
~E o
o
80
!
I
i
I
100 Qlacier reqions
~, .-
60
o
~
4O
@
20 0 -6o
-30
0
30 latitude
60
90
(")
Conclusion: Glaciers and ice caps, +IK: 0.46 mm/year sea-level RISE (best estimate)
SNOW ACCUMULATION ON THE ANTARCTIC ICE SHEET A simple meteorological model has been developed to simulate the temperature and precipitation distribution over the Antarctic continent. The model is two-dimensional (vertical plane, see figure below), and has 4 layers: stratosphere, troposphere, boundary layer and surface of the ice sheet. It has a detailed radiation scheme for short and long wave radiation.
378 The k a t a b a t i c outflow is explicitly calculated and drives the circulation over the ice sheet. The b o u n d a r y layer has two shear zones: one at the ice sheet surface a n d one at the top of the b o u n d a r y layer, where significant e n t r a i n m e n t t a k e s place. B o u n d a r y layer depth is a prognostic variable.
stratosphere (only radiation)
troposphere D~E
:: .... humid a ir
OCE/N Moisture is brought to the ice sheet by the r e t u r n flow in the free troposphere. Precipitation occurs because of cooling of the air (due to uplift and a negative r a d i a t i o n balance). The moisture budget at the surface has four contributions: - precipitation - riming - evaporation - divergence of snow drift W h e n r u n with a p p r o p r i a t e b o u n d a r y conditions (annual m e a n insolation a n d t e m p e r a t u r e at the ocean boundary), the model gives a satisfactory s i m u l a t i o n of the meridional profiles of t e m p e r a t u r e and a c c u m u l a t i o n on the Antarctic ice sheet (annual m e a n state). In case of a w a r m e r climate, snow a c c u m u l a t i o n increases because the "moisture pump" intensifies. The increase is p a r t l y c o m p e n s a t e d by l a r g e r evaporation on the steep slopes of the ice sheet, however. For a uniform 1K w a r m i n g , the model predicts an increase in snow a c c u m u l a t i o n t h a t is equivalent to a 0.27 m m / y e a r sea-level drop.
Conclusion: Antarctic ice sheet, +IK: 0.27 m m / y e a r sea-level DROP (best estimate)
379 PAPERS FROM THIS PROJECT (printed or accepted, status November 1994)
R S W van de Wal, J Oerlemans and J C van der Hage (1991): A study of ablation variations on the tongue of Hintereisferner, Austria. Journal of Glaciology 38, 319-324. J Oerlemans (1992): Climate sensitivity of glaciers in southern Norway: application of an energy-balance model to Nigardsbreen, Hellstugubreen and Alfotbreen. Journal of Glaciology 38, 223-232. J Oerlemans and J P F Fortuin (1992): Sensitivity of glaciers and small ice caps to greenhouse warming. Science 258, 115-117. 4. J Oerlemans and H F Vugts (1992): A meteorological experiment in the melting zone of the Greenland ice sheet. Bulletin of the American Meteorological Society 74, 355-365. J P F Fortuin and J Oerlemans (1993): An axi-symmetric atmospheric model to simulate the mass balance and temperatue distribution over the Antarctic ice sheet. Z. Gletscherk. Glazialgeol. 26, 31-56. J Oerlemans (1993): Modelling of glacier mass balance. In: Ice in the Climate System (ed. W R Peltier), NATA ASI Series, Vol. 1-12 (Springer), 101-116. M R van den Broeke, P G Duynkerke and J Oerlemans (1994): The observed katabatic flow at the edge of the Greenland ice sheet during GIMEX-91. Global and Planetary Change 9, 3-15. P G Duynkerke and M R van den Broeke (1994). Surface energy balance and katabatic flow over glacier and tundra during GIMEX-91. Global and Planetary Change 9, 17-28. .
R S W van de Wal and A J Russell (1994): A comparison of energy balance calculations, measured ablation and meltwater runoff near Scndre Strcmi~ord, West Greenland. Global and Planetary Change 9, 29-38.
10 A Meesters, E Henneken, N J Bink, H F Vugts and F Cannemeijer (1994): Simulation of the atmospheric circulation near the Greenland ice margin. Global and Planetary Change 9, 53-67.
380 11 E Henneken, N J Bink, H F Vugts, F Cannemeijer and A Meesters (1994): A case study of the daily energy balance at the VU-GIMEX camp. Global and Planetary Change 9, 69-78. 12 W Greuell and T Konzelmann (1994): Numerical modelling of the energy balance and the englacial temperature of the Greenland ice sheet. Calculations for the ETH-Camp location (West-Greenland, 1155 m a.s.1.). Global and Planetary Change 9, 91-114. 13 R S W van de Wal and J Oerlemans (1994): An energy balance model for the Greenland ice sheet. Global and Planetary Change 9, 115-131. 14 T Konzelmann, R S W van de Wal, W Greuell, R Bintanja, E A C Henneken and A Abe-Ouchi (1993): Parameterization of global and longwave incoming radiation for the Greenland ice sheet. Global and Planetary Change 9, 69-78. 15 J Oerlemans (1994)" Quantifying global warming from the retreat of glaciers. Science 264, 243-245. 16 M R van den Broeke, P G Duynkerke and E A C Henneken (1994): Heat, m o m e n t u m and moisture budgets of the katabatic layer over the melting zone of the West-Greenland ice sheet in summer, Boundary-Layer Meteorology, in press. 17 A G C A Meesters (1994): Dependence of the energy balance of the Greenland ice sheet on climate change: influence of katabatic wind and tundra. Quarterly Journal of the Royal Meteorological Society 120, 491-517. 18 F G M van Tatenhove, C Roelfsema, G Blommers, A van Voorden (1995): Change in position and altitude of a small outlet glacier during the period 1943-1992, Leverett glacier, West Greenland. Annals of Glaciology, in press. 19 F G M van Tatenhove and O B Olesen (1995) Ground temperature and related permafrost characteristics in west Greenland. Permafrost and Periglacial Processes, in press About 12 additional papers have been submitted
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
381
Stresses in the lithosphere caused by glacial loads P. Johnston and S. Cloetingh
Faculteit der Aardwetenschappen, Vrije Universiteit, De Boelelaan 1085, 1081 HV, Amsterdam
Abstract Simple elastic theory shows that horizontal stresses caused by a surface load which has a similar wavelength to the flexural wavelength of the lithosphere exceeds the vertical stress by several times. The enhancement factor is very sensitive to the wavelength. For glacial timescales, the wavelength of the British ice sheet was close to the flexural wavelength, while the Fennoscandian one was much larger. For models which allow time-dependent glacio-isostatic rebound, it is inferred that the stresses due to both the British and Fennoscandian ice loads were large until the end of glaciation, but very small by the present.
1
T w o - l a y e r e l a s t i c over fluid fiat E a r t h m o d e l
For a uniform, incompressible elastic layer overlying an inviscid fluid halfspace, the deformation in response to surface loads of arbitrary wavelengths can be calculated analytically. To determine the response to more general loads in two dimensions or with axial symmetry, the load may be expressed as a sum of Fourier or Hankel components and the total response is the sum of responses to each component. The continuum version of Newton's second law is (eg [5, 8]) , (o)
t!~ )+tp,juj),~z3,3
p(A)g(O)
~ -p
(0) ~(A)
~
=0
(1)
where t is the Cauchy stress tensor, p is the pressure (= --tkk/3), p the density and g the gravity. The superscripts (0), (5) and (A) indicate the initial, material incremental and local incremental fields respectively. If perturbations of the gravitational field are ignored, and incompressibility assumed, the last two terms do not enter the equation. Calculations have been made for the deformation and stress caused by a load of given wavelength for a 65 km thick elastic incompressible plate of density 3000 kg/m 3 and shear modulus 4 • 1010 Pa overlying an inviscid half-space. The elastic thickness is chosen to coincide with estimates of elastic thickness determined from postglacial rebound in the British Isles [4]. The fiexural wavelength of the elastic layer in the thin-plate approximation is A] -- 27c(Eh3/(12(1
-
l]2)pg))
1/4 --
660 km
(2)
where E is the Young's modulus, h is the thickness of the elastic plate and ~ is Poisson's ratio for the elastic layer. In Figure 1, the dimensionless horizontal stress is plotted versus wavelength of the load and depth within the elastic layer. The vertical stress for the same load is unity at the surface and
382
attenuates with depth, more sharply for short wavelengths than long wavelengths. The maximum amplification of the horizontal stress occurs at wavelengths close to the flexural rigidity where the bending of the lithosphere is greatest.
0
~
.....
"
. ....................
,
9
,.,
, 9............ -..,
.- .
... ...
9. ~
,.......................
,.
.................
:
"~..:
.....
~
oe,
Figure 1: Horizontal and vertical stress as a function of wavelength and depth within the lithosphere normalised by the weight of the load. The maximum response factor for the horizontal stress is close to the flexural wavelength of the lithosphere.
1.1
Results for axisymmetric
ice l o a d s w i t h e l l i p t i c p r o f i l e
The deformation and stress field has been calculated for two ice sheet models which :represent approximately the British ice sheet and Fennoscandian ice sheet at the last glacial maximum. Of particular interest is the maximum shear stress, which is equal to half the difference between the maximum and minimum principal components of stress. In this geometry, they will always be the radial horizontal stress and vertical stress. The vertical stress is constrained to be equal to the load at the surface and for ice loads of moderate to large lateral extent, there is little variation with depth within the lithosphere. The horizontal stress is much more dependent on the horizontal extent of the load compared with the flexural wavelength of the lithosphere. Because the smaller ice sheet (diameter 660 kin) is much closer in lateral extent to the flexural wavelength, it produces larger stresses, despite being just over half the thickness. The maximum shear stress is related to the likelihood of seismicity and faulting occurring. If the shear stress is in excess of 10 MPa, then pre-existing faults may be re-activated [3]. Because the horizontal stress is much larger in magnitude than the vertical stress, it is the main contributor to the maximum shear stress. In Figure 2, we compare the radial stress predicted for two ice sheets of elliptic profile, one with radius 330 km to model the British ice sheet and the other with radius 1000 km for the Fennoscandian ice sheet.
2
Spherical Maxwell viscoelastic model
The deformation of the lithosphere has been calculated using the full equation (1) above, without any approximations, a spherical Earth model, and an elastic lithosphere overlying a Maxwell
383
g 3000
i 2000 .o
1000 0
o
0
i
20
[ "
i
I i~.,/
I
I
I
i
I
o ...................... i ................. ~ ................. i .......................................... i ..........................................
a=
.qp
60
0
.... I
I
I
I
0
500
1000
1500
2000
0
500
1000
Radial distance (km)
Figure 2: Horizontal radial stress (MPa) as a function of distance from the centre of the load and depth within the lithosphere for an axisymmetric ice load with elliptic semi-profile for two ice sheets of different lateral extent. The maximum compressional horizontal stress occurs at the centre of the ice sheet with a small amount of extension outside the edge of the ice sheet. viscoelastic mantle, with seismically determined elastic properties [2] and mantle viscosities determined from fitting relative sea-level observations [4]. A glaciation/deglaciation cycle has been used to approximate the growth and decay of the Fennoscandian ice sheet. Figure 3 shows the maximum stress difference at the end of deglaciation and at the present. Because the Maxwell viscoelastic theology behaves elastically on short timescales and viscously on long timescales, the effective flexural wavelength is time-dependent but with a lower bound of about 660 km as in the model above. Therefore, the maximum stress difference is somewhat smaller than in Figure 2. The values of stress are quite strongly dependent on the elastic properties of the various layers as seen by the sharp variation in stress at 15 and 25 km depth. Because most of the postglacial rebound is complete by the present, there remains little residual stress difference near the surface in the model. However, at the end of the glaciation, the stresses are still large enough to cause seismicity. This is consistent with observations of late glacial faulting [6] and the predominance of the NW-SE regional stress field in Fennoscandia [7] rather than a radial pattern which would be caused by postglacial rebound.
3
Conclusions
The maximum shear stress in the lithosphere has been calculated using the same models which fit relative sea-level observations. The calculations indicate that the stresses were large enough to cause faulting during and after the end of deglaciation, but the residual stress today is probably too small to be observed in comparison with the prevailing NW-SE pattern in Europe due to ridge push from the Mid-Atlantic ridge. The stresses at glacial maximum may have been larger for the British Isles than for Fennoscandia. The results of the modelling are consistent with
384
0 I
,
I
--U
I
I
I
I
IL ~
I
I
I
I
I
L
40 60 0
500
1000 ' 1500 Distance (km)
2000
0
' 500
' 1000 ' 1500 ' 2000 Distance (km)
Figure 3: Maximum shear stress (MPa) for the Fennoscandian ice sheet at the end of deglaciation (left) and at the present (right) for an axisymmetric ice sheet with maximum radius of 1000 km at 18 thousand years before present (kaBP) and finished melting at 8 kaBP. observations of late glacial faulting in both Fennoscandia [6] and Great Britain [1], and with the observed stress field and seismicity pattern in Scandinavia today [7].
References [1] C. A. Davenport, P. S. Ringrose, A. Becker, P. Hancock, and C. Fenton. Geological investigations of late and post glacial earthquake activity in Scotland. In S. Gregersen and P. W. Basham, editors, Earthquakes at North-Atlantic Passive Margins: Neotectonics and PostglaciaI Rebound, pages 175-194. Kluwer, Dordrecht, 1989. [2] A. M. Dziewonski and D. L. Anderson. Preliminary reference Earth model. Phys. Earth Planet. Inter., 25:297-356, 1981. [3] A. C. Johnston. The effect of large ice sheets on earthquake genesis. In S. Gregersen and P. W. Basham, editors, Earthquakes at North-Atlantic Passive Margins: Neotectonics and PostglaciaI Rebound, pages 581-599. Kluwer, Dordrecht, 1989. [4] K. Lambeck. Glacial rebound of the British Isles. II. A high resolution, high-precision model. Geophys. J. Int., 115:960-990, 1993. [5] L. E. Malvern. Introduction to the mechanics of a continuous medium. Prentice-Hall, Inc., New Jersey, 1969. [6] R. Muir Wood. Extraordinary deglaciation reverse faulting in northern Fennoscandia. In S. Gregersen and P. W. Basham, editors, Earthquakes at North-Atlantic Passive Margins: Neotectonics and PostgIacial Rebound, pages 141-173. Kluwer, Dordrecht, 1989. [7] B. Miiller, M. L. Zoback, K. Fuchs, L. Mastin, S. Gregersen, N. Pavoni, O. Stephansson, and C. Ljunggren. Regional patterns of tectonic stress in Europe. J. Geophys. Res., 97:1178311804, 1992. [8] D. Wolf. Viscoelastodynamics of a stratified, compressible planet: incremental field equations and short- and long-time asymptotes. Geophys. J. Int., 104:401-417, 1991.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
385
The response of permafrost ecosystems to climate change Eduard A. Koster a a Department of Physical Geography, University of Utrecht, P.O.Box 80.115, 3508 TC Utrecht, The Netherlands
Abstract Permafrost areas are extremely sensitive to change as permanently frozen ground is directly temperature dependent. The heat exchange interactions of climate and permafrost in the socalled "buffer layer" are highly complicated, as ground temperatures are strongly influenced by local factors (snow cover thickness and duration, vegetation, organic layer and soil characteristics), which are interrelated with climate. Variations in these variables may either enhance or counteract each other, which makes it difficult to predict the accumulated effect of all changes. However, several simulation experiments indicate large shifts of permafrost boundaries due to a temperature increase, resulting in extensive permafrost degradation (thermokarst). Geothermal profiles of the upper 100-200 metres of permafrost, which yield a temporally integrated record of air temperature changes in the past decades to centuries, show significant changes. However, the quantitative relationships between permafrost degradation and biogeochemical processes, including the generation or uptake of carbon dioxide and methane are still largely unknown. 1. INTRODUCTION Approximately 25 % of the land surface of the Northern Hemisphere is underlain by permafrost. A major part of this huge area is designated as discontinuous permafrost (approx. 17.3 million square kms), the southern boundary of which roughly coincides with a mean annual air temperature of-1 to -2~ Near its southern boundary it occurs in isolated patches or islands and is sometimes referred to as sporadic permafrost. Approximately north of the -6 to -8~ isotherm continuous permafrost (approx. 7.6 million square kms) occurs. Moreover, an area of approximately 2.3 million square kms, mainly at lower latitudes, is covered by Alpine or mountain permafrost. Permafrost areas will be among the most heavily affected parts of the world in the event of accelerated future warming [1, 2, 3, 4]. The objectives of this review paper are: 1) to emphasize the complex interrelations in the atmosphere-"buffer layer"-permafrost system, 2) to summarize permafrost response to past and future temperature changes and 3) to indicate the uncertainties with respect to permafrost ecosystems as sources or sinks of carbon dioxide and methane. 2. HEAT EXCHANGE AND THE ACTIVE LAYER In permafrost areas several types of temperatures are defined (Fig.l), depending on where they are measured [1, 2]. The mean annual air temperature (MAAT) usually is several degrees lower than the mean annual ground temperature (MAGT), the latter being defined as the ground temperature at a depth where temperature fluctuates by less than 0.1~ per year. Above this depth the ground is subjected to strong seasonal fluctuations. Nevertheless, mean annual ground surface temperature (MAGST) can be deduced by upward extrapolation of the geothermal gradient, provided the measured gradient has achieved equilibrium and there are no recent climatic changes. Extrapolation of the geothermal gradient downwards will lead to an approximation of the depth of the permafrost base. Where the geothermal heat flow is
386 constant, the geothermal gradient is inversely proportional to conductivity. The thermal conductivity in its turn strongly varies depending on soil properties and sediment texture. The water or ice content is especially critical. The geothermal gradient in different types of sediment ranges from about l~ for sandy, relatively ice-rich material (high conductivity ~ low gradient ~ thick permafrost) to about 1~ for fine-grained, relatively ice-poor material (low conductivity - high gradient ~ thin permafrost).
/ MAAT 2 M(A)SST^ MAGSTu MAPST2
/ 5~'~.*~
,a,er
] [ [ ]f'J
" f - ~ l ~J~-~ X I~ -"
~**]~r'~..~ Unfrozen ground
, Tmean
. / ".~" " ;~t1~ betweenbaseof .'~" :'~l~l~Jll p3t~Vm~IpYs~rtae&bl
4 MAGT 6 8 10 12 14 16 20 2
ItHH-----/~~--iliiiila~mepi
IIIlll ,IIIIIL I IIL~
P2eertm[ IaJ fl lrlOJ lsl tl ~u
+ !
IIIIIII
I[l l IJlllJlll[l
I (5-1bin)
JJJlJJJ~Unfrozenground
,
withinpermafrost
IlflllJ ~J~L"~ mafrotUnfr~176 bellowpe
120 122
-;o
-~
o
Temperature
~ (~
,'o
Figure 1. Permafrost terminology and ground temperature profile. The heat exchange interactions of climate and permafrost through the active layer are strongly influenced by the vegetation cover, the (seasonal) snow cover thickness and duration, the organic soil (if present) and the mineral soil. Variations in these variables may either enhance or counteract each other [1, 2]. The snow cover thickness and duration probably is the most important factor due to its insulating properties. The conductive capacity of dry new snow is 1/5 of that of old compacted snow or dry sand, or 1/20 of that of wet sand or 1/25 of that of ice. Generally, a snow cover keeps the ground warmer, as it can heat up in summer, but is hampered in cooling down in winter. Snow cover thickness and duration depend to a large extent on the presence and nature of vegetation. Vegetation mainly has an effect upon surface temperatures by shading, thus cooling the ground. Moreover, vegetation prevents radiation back into the sky at night, and the soil is dried by evapotranspiration and this decreases its heat capacity. The organic layer strongly influences ground temperatures by its differential insulation capacities. In summer conductivity of a dry peat layer is extremely low (< 0.1 WmK). In winter, however, the organic layer freezes leading to conductivities many times higher (> 1.0 WmK), and consequently the ground cools off. To a lesser extent the properties of the mineral soil also influence the conductivity. Both changes in snow regime and in vegetation will determine the moisture condition of the soil and thereby the thermal conductivity of the materials.
387 3. HISTORIC CHANGES IN G E O T H E R M A L R E G I M E The above-mentioned uncertainties notwithstanding, in principle a rise in MAAT and consequently in MAGST will have the following effects. Firstly, the thickness of the active layer will increase. Secondly, the temperature profile within the permafrost will adjust itself to the new MAGST. The rate with which this happens depends on the thermal conductivity of the permafrost and the ice content of the ground. Response times of the active layer are in the order of years to tens of years. Eventually permafrost will decrease in thickness. During Pleistocene glacial episodes the permafrost area was probably twice as extensive as the present-day extent, whereas during the "climatic optimum" of the Holocene the southern limit of discontinuous permafrost in the Soviet Arctic was up to 600km north of its present position. In historic times significant changes in permafrost zonation have also been documented; e.g. in the southern part of the discontinuous permafrost zone in Manitoba (Canada) the southern limit has shifted northwards over the past 150-200 years and the areal extent of permafrost terrain has diminished strongly. In the Mackenzie Valley (Arctic Canada) MAGT values increased by about 3~ during a recent warm period (late 1800s to the 1940s) and have since decreased about I~ resulting in a shift in the continuous-discontinuous permafrost boundary of several hundreds of kms. Thus, the analysis of permafrost temperature as a function of depth appears to yield an integrated record of air temperature changes in the past. This has been well-documented by temperature profiles obtained from boreholes in the Alaskan Arctic Coastal Plain [5]. A vast number of these temperature profiles shows a distinct curvature towards higher temperature near the surface. The exact onset of warming seems to vary between locations, but they all indicate a warming in the range of 1.5-3~ during the last century, which seems to be in agreement with a similar trend in air temperatures as shown by regional weather records. Time series of annual permafrost ground temperatures in shallow drill holes ( 15~ These systems become rare at T > 18~ but then convective showers become increasingly active and the mean amounts rise again. KNMI Scenario 1 is based on Figure 1. It is obtained by transforming the precipitation amounts on wet days. The procedure is as follows: 1) apply a GCM-predicted change in seasonal mean temperature to all observed daily temperatures T; 2) determine for each wet day the resulting relative change in the mean precipitation amount R from Figure 1; 3) multiply the observed daily amounts by the calculated relative changes (multiplying factors).
393
De Bilt (1906-1981 ) '
I
'
I
'
I
'
I
/'
r ~" E rr
6 i
v
c0
E
4-
t--
._o
§
._o. ft.
I
2-
. / ~ 1
0 -2(
m
I -10
10 I
,
I 0
,
I 10
J
I 20
,
Temperature T (~
Figure 1. Mean precipitation amounts R (dots) at surface air temperature T class intervals of 2~ for wet days at De Bilt (1906-1981). The smooth curve represents the fitted regression relation; the error bars the standard deviations of the means. After Buishand and Klein Tank (in press).
In Scenario 1 only the effect of a prescribed atmospheric warming is taken into account. The relative changes in Steps 2 and 3 assume implicitely that the atmospheric circulation changes according to its present-day dependence on T. More flexible scenarios can be obtained by prescribing also a change in the atmospheric circulation. In Scenario 2 the daily mean surface air pressure P is included in the analysis for this purpose. Figure 2 presents the relation between R, T and P for wet days at De Bilt. The procedure for transforming precipitation amounts on wet days into a consistent scenario (Scenario 2) for the case of both a prescribed warming and a prescribed change in surface air pressure (atmospheric circulation) is similar to that for Scenario 1, but now Figure 2 is used instead of Figure 1. In Scenario 1, it is assumed that the number and the sequence of wet and dry days in the future climate time series remains the same as in the observed record. Since the occurrence of a wet day is linked to the atmospheric circulation, Scenario 2 with a systematic change in P must account for a change in the sequence of wet and dry days. This was done as follows: 1) assign probabilities of rain to each day using Figure 3; 2) compute for each season the change in the number of wet days from these probabilities in the present-day and future climate; 3a) if the number of wet days in a season decreases by n: assign n wet days in the series as dry on the basis of their probability of rain, e.g. using a Monte-Carlo method. 3b) if the number of wet days in a season increases by n: assign n dry days in the series as wet on the basis of their probability of rain and determine their amounts using Figure 2.
394
De Bilt (1906-1981) lo5o
.
,
.
,
9
9
,
De Bilt (1961-1990) .
,
:
.
9
~
......... i .......... ! ..........
.t:: ,.i.
r "~
i .......
- ......... ! ..................
J..........
! ..........
. 7 5 . . . . . . . . . . -.. ......... ! .......... i ......... : . . . . . . . . . i ......... i ......... i .......... i .......... I :: i i : : : low solar radiation .............................................................
ii,,i
}
liiiii iiii:iii ii =:iiiiii:.. iiiii:ii i,, i iiii ,, i i....,'.......... ,iiii: .
~
~..........i~.........~i..........!.........:..........:.........i .........i ........i..........t
,
*2-4
........... .!.......... !....... ~::~.:.:J;i :a:)~i::~.!...'i .............
.0-2 960
, -20
I -10
,
I 0
,
I
,
10
I 20
,
, 30
960
Temperature T (~
990
1020
I 1050
Surface air pressure P (hPa)
Figure 2. Mean precipitation amounts R (dots) at surface air temperature T class intervals of 2~ and surface air pressure P class intervals of 6 hPa for wet days at De Bilt (1906-1981). The isolines represent the theoretical mean amounts from a fitted regression model. After Buishand and Klein Tank (in press).
Figure 3. Fitted logistic regression relations for the probability of rain at De Bilt (19611990).
3. E X A M P L E F O R DE BILT The daily precipitation amounts in the 1961-1990 record of De Bilt were transformed by the above methods. The prescribed changes in T and P were taken from the Canadian Climate Centre G C M predictions of large-scale changes in the seasonal means (Table 1; 2xCO2 - l xCO2 experiment). Figures 4 and 5 illustrate the effects of the transformation on the July 1962 precipitation data for Scenarios 1 and 2, respectively. The multiplying factors (Scenarios 1 and 2) and the probabilities of rain (Scenario 2) are also shown. Note that in Scenario 2 two wet days (6 and 21 July 1962) were assigned dry. Table 1 Predictions of the large-scale changes in the seasonal mean temperature and surface air pressure over Western Europe for the 2xCO2 Canadian Climate Centre GCM experiment.
AT(~ AP (hPa)
Winter
Spring
+3.0 -3.4
+2.3 - 1.1
Summer
Autumn
+3.7 +0.3
+3.4 -0.1
395 D e Bilt
AT=+3.7~ 40
f2.00 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
~
~0 .................................................................. t~.~0 ~ 9 Multiplying factor
8
~
~
a
95. 1 o ................................................................... .... ,,m-.._
.5o
:-Ji:::f
........................
0
o.9,
5
3
13
9
7
11
17 15
21 19
25
23
(total 69 mm) M Scenario 1 (total 80 mm)
0.00
1
mObserved
29
27
31
J u l y 1962
Figure 4. Precipitation amounts in the observed and transformed July 1962 month at De Bilt for Scenario 1. The solid squares represent the multiplying factors. D e Bilt
ZXT=+3.7~
AP=+0.3 hPa
c'-
40
-2.00
E E -'30 t-"
---
~__ ~ o
:3
v
1.50 ~
E t~ 20
~t
c-, lO
ii] "
F O
1.00
t-" O
0 0
"0 O
.n.,
O
11t 50
-~-
~L_
13..
0
~0.00 1
5
3
13
9
7
11
17 15
21 19
25 23
29 27
o Probability of rain 9 Multiplying factor mObserved (total 69 mm) Scenario 2
0
(total 74 mm) m.
31
:3
J u l y 1962
Figure 5. Precipitation amounts in the observed and transformed July 1962 month at De Bilt for Scenario 2. The solid squares represent the multiplying factors and the open squares the probabilities of rain.
396 The largest precipitation changes in the two scenarios occur in winter (+20% and +44% for Scenarios 1 and 2, respectively) and the smallest in autumn (+5% for Scenario 1) or summer (+8% for Scenario 2). The annual mean amount increases by 10% in Scenario 1 and 20% in Scenario 2. These values differ considerably from the precipitation changes as predicted directly by the GCM itself (+28% in winter, -26% in summer and no change on average over the year).
4. APPLICABILITY OF THE SCENARIOS
An attractive feature of the KNMI approach is that the scenarios and their updates can easily be implemented by impact groups. The transformed daily series have a realistic variability on daily as well as longer time scales. Extreme case scenarios for sensitivity studies can be constructed from past (extreme) episodes. Scenarios in the form of monthly, seasonal or annual time series can be obtained directly from the transformed daily series. The scenarios facilitate integration of NRP climate change impact studies. After consulting the potential users a follow-up project is planned to construct a wider range of scenarios in which changes in air humidity and solar radiation are included.
5. REFERENCES
Buishand, T.A. and A.M.G. Klein Tank, (in press). Regression model for generating time series of daily precipitation amounts for climate change impact studies. Stochastic Hydrology and Hydraulics. Giorgi, F. and L.O. Mearns, 1991. Approaches to the simulation of regional climate change: A review. Reviews of Geophysics, 29, 191-216. Klein Tank, A.M.G. and T.A. Buishand, 1993. Modelling daily precipitation as a function of temperature for climate change impact studies. KNMI Scientific Report WR 93-02, De Bilt. Klein Tank, A.M.G. and T.A. Buishand, 1995. Transformation of precipitation time series for climate change impact studies. KNMI Scientific Report WR 95-01, De Bilt. Wilks, D.S., 1992. Adapting stochastic weather generation algorithms for climate change studies. Clim. Change, 22, 67-84.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
397
Climate Change Scenarios for Great Britain and Europe M.Hulme, E.M.Barrow, O.Brown, D.Conway, T.Jiang, P.D.Jones and C.Turney
Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, United Kingdom
1. I N T R O D U C T I O N The objective of the work described here was to develop future climate change scenarios for Great Britain and for Europe related to global emissions of greenhouse gases. These scenarios were to be used by a variety of ecosystem and hydrological modellers in a research project titled 'Landscape Dynamics and Climate Change', a project sponsored by the UK Natural Environment Research Council (NERC) under their TIGER (Terrestrial Initiative in Global Environmental Research) programme. The work on the scenarios is now complete and the project as a whole will report its findings during 1995. This paper describes three stages to the scenario construction: the construction of gridded baseline climatologies for 196190 using station observations; the construction of the patterns of future climate change using results from General Circulation Model (GCM) experiments; and linking the previous two steps to generate estimates of future climate for specified decades in the future. At all stages work progressed at two spatial resolutions - a 10km resolution for Great Britain and a 0.5 ~ latitude/longitude resolution for Europe.
2. T H E 1961-90 C L I M A T O L O G I E S Since GCMs are generally not regarded as accurate enough to provide useful descriptions of current climate at local or regional scales, one of the essential components of any future climate scenario is an adequate description of the current climatology of the region of interest based on observed data. Mean monthly climatologies were therefore constructed for the two TIGER domains (Great Britain and Europe) for the following surface climate variables: mean, minimum and maximum temperature, precipitation and raindays, sunshine hours, vapour pressure, wind speed and ground frost days. These climatologies used station data for the period 1961-90 collected from National Meteorological Agencies (NMAs) across the region. The distributions of European stations for which 1961-90 data were obtained are shown in Figure 1 for temperature. The interpolation of the station data to the respective grids used partial thin-plate splines as developed by Mike Hutchinson from the Australian National University. Since elevation was one of the predictor variables, three climate surfaces were produced for each variable reflecting the 'minimum', mean and 'maximum' elevation within each 10km or 0.5 ~ cell. Month-by-month anomalies on these grids for the period 1961 to 1990 were also calculated for the variables mean temperature and precipitation.
398
80 70 .__
....
Z
o v
9
60
-
-
".E'
-
el) na
~ e..,-
~,
50
-
-
-
"~ ""
..... . :
~ .. "
9
.'"." : . . .
::
-
I
I
-20
(
" *
~
*.
0
"
"
" ~
""
"
"
{;
9 9
,.
"
9
"
-
"
7 -
~'-'.-:,'
.=
"':,\ "
'
"
" 9
.~' 42
"
-
~'
9
~
.
. , , ~ . . . . . . . . :
..... t'-"---'~-__.~-.-- 9
"
,
-10
"
-
,~ "~-;[.'~.~',~,_..~.,~-,..,Ci.~:..~.~..~.,.9 '.,.% _,o.)/..
"
9 9
9
'
";" "." .'" 9 I '..
..
~%..:-,.,.
j ?
""
'~
-
1
" ."
I,**~'k,
.---.r " ' - ~ . " ' " " ; . . . . .
:2-..~,.t, ".~X" . ~ - ~
,,,:,.:::.:-:..~*.~;';,~,),.-',
"= 9~
~
-
/
m/,'z.-" ',
m'/ . / ~ ""- 7 . ~ , ' , ~'1 ' 9
,.,
,"
-
--0.0~
---~
.___ ~ ,
1.0
___
.
~Z o
9
_o.os.--:~~-.q~
5-
~,,.o
0.6 --;.~..-:_::6:~-5-:~--
_ . . . . . . .
O.
_.
i: = 9
;:r.
0.4
-
""
,'"
. .........
0.2 O.Oo.~_~6"---7 :
~
.
~
..........
; o.~o 8.0
, 16.0
24.0
32.."
Distance (km) Figure 3. Contour plots of simulated sensible heat flux over the surfaces depicted in Fig. 1. Upper panel: Lx = 2 km; middle panel: Lx = 8 km; lower panel: Lx = 32 km.
435 Isolines are tilted in the direction of the horizontal wind (from the left to the fight) indicating the influence of one surface on the other. Furthermore, a larger part of the PBL is directly affected by a new surface if the patch size becomes larger, that is, if Lx increases. For Lx = 2 km the influence of the inhomogeneities is felt in the lower 40% (300-400 m) of the PBL only. Above this level, the PBL is able to average out the effects of the individual patches. If Lx is 8 km, the influence of the inhomogeneities extends almost to the top of the PBL. For Lx - 32 km the entire PBL is affected. This scale is in fact the regional scale at which feedback between the PBL and the surface becomes significant [10]. Furthermore, meso-scale circulations around inhomogeneities may become important at this scale [11]. The present model is not able to simulate such features and therefore, results for Lx -- 32 km should be interpreted with caution. Average fluxes over the inhomogeneous region are shown in Fig. 4. Lx has hardly any influence on the averages. This suggests that, in the absence of meso-scale circulations, the flux profiles can be calculated using a single set of surface characteristics, independent of the scale of the inhomogeneities. Such an approach has been tested here. An average surface was created by taking a=0.5"0.1+0.5"0.2--0.15 and rs=1./(0. 5/100 + 0.5/50)=67 s/m. An effective roughness length, z0.~, was calculated using the concept of blending height [12-14], by taking [13]:
t 1 ~) l t 05 ln(/b/Zo,
ln(/b/Zol)
05 l
(1)
ln(l b/z02)
1.2 - ~ , , .
1.2
9-" 1.0
,~1.0 o1.,~
.~
,.~ 0.8 o
o 0.6
........-..,
"~ 0.4
o~..,
"..............
0.8 K,, .... "'-...
0.6
"\
:,,.
~ 0.4,
@
NxN
Z o.2
Z
\
oo~ 9
0
, 0.1
""".... 0.2 0.3 0.4 0.5 0.6 0.7 Momentum flux (m2/s 2) .
:,
.
9
O.Od.1
0.0
0.1
0.2
0.3
Sensible heat flux (Krn/s)
Figure 4. Profiles of momentum flux (left) and heat flux (fight), averaged over the inhomogeneous part, over homogeneous forest and grassland, and over the surface with "average" characteristics (see text). Lines with squares: Lx-2 km; Lines with crosses: Lx-8 km; Lines with hourglasses: Lx=32 km; Solid line: average surface; Dotted lines: homogeneous forest; Dashed lines: grassland.
436 where z01 and z02 denote the roughness length of forest and grassland, and 1b is the blending height. According to (1), z0.e depends rather strongly on lb for lb100 m. Thus, the results given above suggest lb>100 m. Therefore, we used the parameterization for 1b suggested by Claussen [14], according to which lb= 0.7z0(Lx/z0) 4/5, leading to lb = 249 m and 746 m for Lx = 2 and 8 km, respectively. Note that these values are in reasonable agreement with the results presented in Figs. 2 and 3. For Lx = 32 we obtain lb=2240 m, which would suggest that the entire PBL is strongly affected by the inhomogeneity. However, this value of Lx may be associated with the occurrence of meso-scale circulations, in which case the concept of blending height breaks down [14]. Using the values of 1b given above leads to z0.e=0.35 m. An additional run was made with the inhomogeneous area replaced by a homogeneous area with a---0.15, r~=67 s/m and z0=0.35 m. Results are also shown in Fig. 4. A satisfactory representation of the average momentum flux is obtained, but the average heat flux is somewhat underestimated. The latter feature corresponds to a slight overestimation of the latent heat flux (not shown here). This result can perhaps be improved if an alternative way to calculate average surface resistance and perhaps albedo would be used. Also, the calculation of roughness length for scalars might have to be adjusted.
5. R E F E R E N C E S
3 4 5 6 7 8
H.F. Vugts, A.F.G. Jacobs and W. Klaassen, This Volume (1995). J.P. Nieveen, C.M.J. Jacobs and A.F.G. Jacobs, This Volume (1995). R.S. Rao, J.C. Wyngaard and O.R. Cot6, Boundary-Layer Meteorol. 7 (1974), 331. R.S. Rao, J.C. Wyngaard and O.R. Cot6, J. Atm. Sci., 31 (1974), 738. J.C. Wyngaard, Boundary-Layer Meteor., 9 (1975), 441. J.C. Wyngaard & O.R. Cot6, Boundary-Layer Meteorol. 7 (1974), 289. J.L. Monteith, Symp. Soc. Exp. Biol., 19 (1965), 205. A.J. Dyer and B.B. Hicks, Q. J. R. Meteorol. Soc. 96 (1970), 715.
9
J.R. Garrat & B.B. Hicks, Q. J. R. Meteorol. Soc., 99 (1973), 680.
10 11 12 13 14
C.M.J. Jacobs and H.A.R. de Bruin, J. of Clim. 5 (1992), 683.
1
2
J.P. Pinty, P. Mascart, E. Richard and R. Rosset, J. Appl. Meteorol. 29 (1989), 976. J. Wieringa, Q. J. R. Meteorol. Soc. 112 (1986), 867. P. Mason, Q. J. R. Meteorol. Soc. 114 (1988), 399. M. Claussen, Atmos. Environ. 24A (1990), 1349.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
437
Sensible and latent heat flux over natural bog vegetation R.S. Singh a, J.P. Nieveen b, C.M.J. Jacobs b and A.F.G. Jacobs b "Central Arid Zone Research Institute, Jodhpur-342 003, India. Presently working as Research Fellow at Department of Meteorology, Duivendaal 2, 6701 AP Wageningen, The Netherlands bDepartment of Meteorology, Duivendaal 2, 6701 AP Wageningen, The Netherlands
Abstract Goudriaan's crop model is tested for its ability to describe the energy balance of a natural bog vegetation. Data used to drive the model were collected during the summer of 1994. Model results were compared with field data obtained by means of the eddy correlation technique. The present study indicates that the model results are satisfactory as both sensible and latent heat fluxes are overestimated by 6% only. Furthermore, sensitivity analyses have shown that the model results are sensitive to the soil surface resistance to evaporation. Hence, there is a need to incorporate the soil surface resistance to evaporation as a function of time.
1. I N T R O D U C T I O N Many physical and biological models have been developed to simulate the heat and mass exchange between arable crops and the atmosphere. Most of them were validated for different field crops in the past. Application of these crop models to natural vegetation will further identify areas of their strength and weaknesses under varying soil, vegetation, and climatic conditions. With this aim, in the present paper, Goudriaan's crop model [1] is tested for its suitability to describe the physical processes of a ~atural vegetation from a bog region.
2. MODEL, MATERIALS AND METHODS The model fomaulated by Goudriaan [1] simulates the micro-weather within crops as a function of vegetation, soil, and boundary conditions. Feedback of vegetation and soil on their environment is included in the model. The partitioning of the absorbed radiant energy into sensible heat, latent heat, and photosynthesis is calculated using the energy balance equations [2, 3]. For details about the model, the readers are referred to Goudriaan's monograph [1]. The study area is located in a bog region of the northern part of The Netherlands. The dominating type of natural vegetation is grass locally known as Pijpestrootje (Molinia Caerulea) growing in irregular humps over the region. The soil of the area is waterlogged peat mainly formed due to decomposition of dead organic matter. The canopy and the boundary weather observations were carried out under the SLIMM project [4, 5] during the summer 1994. Two kinds of meteorological data were collected. An input set, to drive the model, includes wind speed, vapour pressure, air temperature at 4.0 m (reference level) above the soil surface and global as well as net radiation above the vegetation. In another set, to compare the model results, fluxes of sensible and latent heat were measured by the eddy correlation technique. The grass and soil physical parameters were taken either from the measured data from the field or from the available literature. Some major input parameters used to drive the model for the standard run are shown in Table 1.
438 Table 1 Grass and soil characteristics used in the model Parameters
Unit
Grass height Leaf area index (LAI) Average leaf width Surface roughness (z 0) Internal regulatory CO2 concentration External CO2 concentration Thermal conductivity of the top soil layer Volumetric heat capacity of the top soil layer Water stress in the soil Soil surface resistance to evaporation (RESS)
m m2n1-2 m m vpm vpm Wm-'K -~ Jm-~K-1 bar sin-'
Value 0.75 1.50 0.005 0.07 240.0 350.0 0.08 2.4* 106 -0.1 500.0
Most of the area of the bog region was covered by the humps. The top layer of the humps, which constitute dead organic (grass) matter, was partially dry during the period of simulation. This partially dry layer of the dead grass on the humps reduced the water loss from the soil through evaporation. Therefore soil surface resistance to evaporation (RESS) was taken arbitrarily 500 sm ~ as input for the model. Measured data of the soil thermal conductivity, using the nonsteady-state probe method [6], were also used in this study. Further, a sensitivity analysis was done to judge the irnportance of the input soil and grass variables for the behaviour of the model. The model was run at different input values of the RESS: at 0 sm -1 to represent wet soil surface [1] and at 250 sin", which could be possible due to occurrence of rainfall and dewfall over the bog region. Besictes these the model was also run at different input values of LAI to see its influence in comparison to RESS for the same magnitude of variation.
3. RESULTS AND DISCUSSION The model was run for two consecutive days (August 30 and 31, 1994). The weather conditions used as input to drive the model, are shown in Figure 1. The simulated energy balance over natural vegetation is presented in Figure 2. The temporal variation of measured and simulated fluxes of sensible and latent heat are shown in Figures 3 and 4. The root mean square error (RMS) of the simulated sensible heat flux equals 13.9 Wm -2. The RMS is calculated using equation (1) in which H,,,~ and }-1~,are the nleasured and simulated
RMS=
/
i=i
N
/
sensible heat fluxes, respectively, at half an hourly intervals i, and N=96 is the number of half an hourly measurements. Similarly the RMS of the simulated latent heat flux equals 20.9 Wm -2. The Regression line forced through the orioine indicated that both sensible and latent heat fluxes were overestimated by about 6% (Figure 5 and 6). The results of the sensitivity analysis are presented in Table 2. It was found that soil surface resistance to evaporation, RESS has much influence on latent and sensible heat fluxes [7].
439 244
5
L~ t
TA.....
/..%
450
4 c~
.............
.E~ 2 5 0
S
144 ,-" ,-/,
;~.~,
versus simulated sensible heat flux (H,). The linear regression line forced through the origin is ~ = 1.06H m (N = 96).
0
Figure versus linear origin
6. Measured latent heat flux (LE,,) simulated latent heat flux (LEs). The regression line forced through the is LE.~ = 1.06LE m (N = 96).
440
The 50% variation in RESS value is quite possible under peat soil situations of the bog region. Also, unlike grass parameters, the value of the RESS even can vary significantly within a day due to occurrence of rainfall or dewfall. Therefore, it is likely to get an improve model results with time variable of RESS. Obviously, an increase in LAI of grass caused a substantial increase in CO2 assimilation. Table 2 Influence of some parameters on daily fluxes. The change with respect to the standard run is indicated by the arrow Variables
Unit
DNCO2 DNRAD DLHF DSHF DLHFB DSHFB DSOILF DNCO2 DNRAD DLHF DSHF
kg CO2ha -1 106jn1-2 106jm -2 106jm -2 106jm -z 106jm "2
106jm -2 : : : :
Standard
RESS
run
500-+0
500-->250
1.5-->3.0
1.5--+2.25
189.0 7.50 4.69 2.04 1.51 1.30 0.51
193.0 7.50 7.78 -0.75 5.29 -1.47 0.20
190.0 7.50 5.39 1.40 2.35 0.68 0.44
228.0 7.50 5.41 1.34 1.11 0.21 0.44
218.0 7.50 5.14 1.59 1.31 0.59 0.47
Net CO2 assimilation Net radiation Latent heat flux above canopy Sensible heat flux above canopy
LAI
DLHFB DSHFB DSOILF
: Latent heat flux at bottom : Sensible heat flux at bottom : Soil heat flux
4. S U M M A R Y AND CONCLUSIONS Goudriaan's model simulated sensible and latent heat flux reasonably well over the natural grass in the bog region. It is likely that an incorrect value of soil surface resistance to evaporation may lead to an error in the simulated fluxes. Therefore, incorporation of soil surface resistance to evaporation as a function of time may further improve the model results, particularly in a case of a longer period of simulation.
5. A C K N O W L E D G E M E N T S R.S. Singh has been supported by CEC, Brussels and DST, New Delhi.
6. R E F E R E N C E S 1 2 3 4 5 6 7
H.L. Penman, Proceedings of Royal Soc. of America, 193 (1948) 120. J.L. Monteith, Principles of environmental physics. Edward Arnold, London, 1973. J. Goudriaan, Simulation Monographs, Pudoc, Wageningen, The Netherlands, 1977. H.F. Vugts, A.F.G. Jacobs and W. Klaassen, This Volume (1994). J.P. Nieveen, C.M.J. Jacobs and A.F.G. Jacobs, This Volume (1994). W. van Loon, I. van Haneghem and J. Schenk, Int. J. Heat Mass Transfer, 32 (1989) 1473. R.S. Singh and A.F.G. Jacobs, Neth. J. Agric. Sci., Submitted (1994).
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
441
SLIMM-PROJECT
H. Vugts", A.F.G. Jacobs b, W. Klaassen ~ a) Free University of Amsterdam (VUA), Department of Meteorology, De Boelelaan 1085, NL 1081 HV Amsterdam, The Netherlands b) Wageningen Agricultural University (WAU), Department of Meteorology, Duivendaal 2, NL 6701 AP Wageningen, The Netherlands c) State University of Groningen (RUG), Department of Physical Geography, Kerklaan 30, NL 9751 NN Haren, The Netherlands
1. INTRODUCTION
An important aim of BAHC (Biological Aspects of the Hydrological Cycle) is to find the average atmospheric exchange at the gridscale of General Circulation Models (GCM's). To achieve this goal, scaling-up rules must be developed to average local observations. At an intermediate scale of landscapes, scaling-up rules depend on advection in the atmospheric surface layer above individual patches. The available scaling-up rules at this scale show considerable scatter and have hardly been validated. Moreover, recent studies indicate that scaling-up rules may strongly underestimate the influence of surface heterogeneities on the average landscape exchange. The general aim of the SLIMM project (Surface Layer Integration Measurement and Modelling) is to determine the soil-vegetation-atmosphere exchange of momentum, heat (sensible and latent) and carbon dioxide at the landscape scale (...10 km). At this scale most landscapes are inhomogeneous. Average fluxes at the landscape scale are at present simply estimated by direct averaging of the fluxes of the elements, or, by estimating average values using the concept of blending height. The method of using blending height is to be extended and tested in a heterogeneous landscape. The resulting method should be the first step in scalingup local observations to areal averages for GCM's. To achieve the general aim of the SLIMM project, an intensive cooperation has been started between three Dutch universities (Amsterdam, Wageningen and Groningen. In this cooperation an intensive two years' measurement programme is carried out over an inhomogeneous terrain. Moreover, various computer simulations have been initiated in which the experimental evidence is used to validate these models.
442 2. SITE CONDITIONS
The region from Norg to Fochtelo~rveen in the north of The Netherlands has been selected as experimental site for collecting data. This location has been indicated in Figure 1.
Foc hte
ar ea
loot
t
Figure 1.Fochtelo~rveen location
............. ! ili!:ii:i!:!ii ~i
in The Netherlands.
The whole area of interest can be subdivided into three sub-sites; a forest area (Groningen), a bog area (Wageningen) and an area mainly consisting of arable land (Amsterdam). Each subregion has a principle investigation group notated in brackets. The forest site is a combined coniferous/deciduous forest located about 2 km NE from the natural bog site. The agricultural site consists of grass and berry bushes and is situated about 2 km NE from the forest location. This specific region has been selected for the following main reasons: 1) The region has a marked surface heterogeneity within the 10 km scale, our scale of interest. 2) This region is far away from major surface heterogenieties like land-sea interfaces. 3) In the same area a second hydrological experiment is executed in the same period which allows an intensive cooperation with other groups. 4) The region is situated close to one of the universities (Groningen) which guarantees supplementary manpower during the experimental period. 5) The Fochtelo~rveen area is the largest bog relict in the Netherlands and is extremely useful to study the exchange of greenhouse gases like CO2 and CH4.
443
3. TIME SCHEDULE
The main observations will take place continuously in a two year's period (1994 - 1995) above all the three subregions. The radiation fluxes, the turbulent fluxes of heat, mass (water vapour and carbon dioxide) and momentum and the soil fluxes of heat will be measured continuously. Moreover, during this period, the vegetation and soil characteristics of all three areas will be monitored like, for example, the Leaf Area Index (LAI), foliage area distribution, ratio between dead and living material and soil water content. In addition to the continuous measurement programme, at least three Intensive Field Experiments (IFE's) are planned to investigate areal variations, local advection and regional averages in more detail. One of the IFE's will be focused on the change in surface conditions around the bog-forest interface. Here special attention will be focused on the within-forest and up-wind flow field and the static pressure [1, 2] around the bog-forest interface. During the IFE's but also incidentally during the continuous measurement period, the following observations of landscape averages will be executed: 1) Boundary layer observations of 3-Dim wind and structure parameters by using a SODAR. 2) Boundary layer observations of the temperature by using a so-called RASS system.
4. MODELLING
The observation results will be used to validate various existing models and, if necessary, to extend these models with the goal to develop advection rules for application in meso-scale models. For example the models of Klaassen [3] and the model of Meesters [4] will be used. Extending local advection to the 10 km scale implies that the influence of multiple step changes will be analyzed as well. Primarily a smooth-rough-smooth transition will be studied with smaller variations within these elements. For this study the second order model of Kroon [5], and the extended model of Rao [6] will be used. In day-time, the planetary boundary layer (PBL) has a height of about 1000 m and has horizontal variations of temperature and windspeed that can be neglected at the 10 km scale. At night and possibly after rainy periods, however, a much shallower boundary layer occurs. This means for the boundary layer that
444
can influence the integration rules of the land surface-atmosphere exchange and will be investigated. Existing, but not yet calibrated, up-scaling rules are based on the concept of a blending height. At this height, the local variations are thought to merge into a regional average. In literature [7, 8,9 ] a first estimate of this height shows a variation of an order of magnitude. The height of blending might be estimated from elevated measurements at different locations in the landscape and is expected to relate to the scale of heterogeneity and atmospheric stability. Attention will be focused on a technique to arrive at an accurate measure for this height. 5. FUNDING
The SLIMM research programme is part of the Dutch National Research Programme on Global Air Pollution and Climate Change. 6. REFERENCES
[1] Jacobs, A.F.G., 1984: Static pressure around a thin barrier. Archiv Meteor. Geoph. Bioclim., Ser. B35: 127-135. [2] Jacobs, A.F.G., Van Boxel, J.H. and Shaw, R.H., 1992: The dependence of within-canopy stratification parameters on within-canopy turbulence properties. Boundary-Layer Meteorology, 58: 247-256. [3] Klaassen, W., 1992: Average fluxes from heterogeneous vegetated regions. Boundary-Layer Meteorology, 58: 329-354. [4] Meesters, A., 1991: Thermally-forced meso-scale circulation in tidal areas. PhD thesis, Free University Amsterdam, pp 180. [5] Kroon, J.L.M., 1985: Profile derived fluxes above heterogeneous terrain: a numerical approach. PhD. thesis, Agric. Univ. Wageningen, the Netherlands. pp 159. [6] Jacobs, C.M.J., Nieveen, J.P. and Jacobs, A.F.G., 1995: Fluxes over nonuniform vegetation: a numerical study. This volume. [7] Wieringa, J., Roughness-dependent geographical interpolation of surface wind speed averages. Quart. J. Royal Meteorol. Soc., 112: 867-889. [8] Mason, P.J., 1988: The formation of areally averaged roughness lengths. Quart. J. Royal Meteorol. Soc., 114: 399-420. [9] Claussen, M., 1991: Estimation of areally-averaged surface fluxes. BoundaryLayer Meteorology, 54: 387-410.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
445
Exchange processes of a natural bog vegetation; SLIMM measurements J.P. Nieveen, C.M.J. Jacobs and A.F.G. Jacobs Department of Meteorology, Wageningen Agricultural University, Duivendaal 2, 6701 AP Wageningen, The Netherlands
Abstract The Surface Layer Integration Measurement Modelling project (SLIMM)is set up to determine the soil-vegetation-atmosphere exchange of momentum, heat, water vapor and carbon dioxide at a regional scale. Starting point of the joint experiment is that the exchange processes above a heterogeneous landscape is not a simple weighed sum of these processes above the homogeneous parts of the composing elements. Therefore an experiment is set up during which the exchange processes of the different components are measured.
1. INTRODUCTION The experimental area is in the region from Appelscha to Norg in the north of the Netherlands. The site of the Wageningen Agricultural University (WAU)is located at the prevailing windward end of the experimental area, above a bog landscape type. The continuous experiment takes place during a two year's period (1994-1995). During this period flux measurements, representative for the bog area, will be performed as good as possible on a routine basis. In addition to the continuous measurements, three more detailed so called Intensive Field Experiments (IFE) are planned. The main goal of the IFE's is to provide insight into the mechanisms that are responsible for the areal fluxes from the complex bog area as for the total fluxes of the whole measurement site.
446
The contribution of the WAU to the SLIMM project include a continuous monitoring of the surface fluxes of radiation and profile measurements of temperature and wind speed. Second, simultaneous monitoring momentum, heat, water vapor and carbon dioxide fluxes during the two year's period. Third, temperature, water vapor and carbon dioxide fluctuations and mean water vapor and carbon dioxide concentration will be measured, and four, a detailed description of the temporal changes of the architecture of the vegetation and soil features. These points will be discussed here.
2. STANDARD METEOROLOGICAL MEASUREMENTS Two masts are placed at a representative location within the bog area (20 meters apart). To reduce shading effects, the radiation sensors are spread out over the two masts. All components of the radiation balance are measured using radiometers (Kipp & Zonen, CM 5) for short-wave incoming and reflected radiation (Albedo), a Funk net radiometer (Middleton) and a pyrgeometer (Kipp & Zonen, CG1) for Iongwave radiation. Figure 1, shows an example of the measured radiative fluxes.
8ool 700]
o,,.",/"".. ',
6001 ~" 500E 400v X Z3
300 200100-
O-100
Figure 1
0
300 600 900 1200150018002100 Time (GUT) Radiative fluxes measured at the 3rd of August 1994. Rn = Net, Sin-- Short wave incoming, Sour = Shortwave outgoing, Lin = Longwave incoming and Lout = Longwave outgoing radiation.
447
Apart from the radiation measurements, the following standard meteorological quantities will be measured: dry and wet bulb temperatures (aspirated Pt-100 psychrometers) at three heights as well as wind speeds at the same heights (sensitive cup anemometers, length constant l m), at four depths soil temperature using Pt-100 thermometers (0.05, 0.2, 0.5 en 1 m) and soil heat flux using a TNO transducer (WS 31 CP). A wind vane is used to measure the mean wind direction. The profile measurement of temperature and wind enable, by using various meteorological techniques (for example: the aerodynamic and Bowen ratio energy budget approach), to make assessment of fluxes of momentum, heat and water vapor.
3. SURFACE FLUXES To make a comparison of the surface fluxes calculated with the Bowen ratio or aerodynamic technique and direct measurements of the surface fluxes, an eddycorrelation system has been installed. The system consists of an ultrasonic anemometer/thermometer (Gill Instruments Ltd.), a fast response thermometer and a CO 2/H20 infrared gas analyzer (LiCor, Li 6262). This technique also allows us to add two relatively new techniques to the measurement program, namely: 1. The standard deviation or fluctuation technique [1] for heat, water vapor and carbon dioxide. 2. The structure parameter method [2] for heat, water vapor and carbon dioxide. Figure 2 shows the surface fluxes as measured with the eddy-correlation system and the net radiometer. Large errors can occur when the covariances are directly calculated from the measurements, so several corrections should be carried out to obtain the correct values (eg. Axis rotation and frequency response correction). These corrections are important, and should be carefully looked at in this experiment [3,4].
4. CARBON DIOXIDE The increase of carbon dioxide is the main cause for global warming, possibly resulting in climate change. The global carbon dioxide cycle is only partly understood [5], as the exchange between the atmosphere and oceans and vegetated
448
soils is still poorly quantified. A main aim of the SLIMM project is to obtain further insight into COa-exchange processes. 600
~
500 40004
E O0 X
300-
H
fX
~~,,,'; LE
/
200-
~\i"."., ~ ,.',;i~ :
,,.,
. :.,. ,, ~ ,
,
/ ..d..'"~ W"",v i~\ /.I....
m
I.L
100
-100
an
i i i 1 1 1 1 1 1 1 1 1
0
iiii
iii
|111111
i .....,.f,, ,", ',\
iii
ii11]
i l l i l
iii
iiii
i
300 600 900 1200150018002100 Time (GMT)
Figure 2:
Measured sensible heat (H) and latent heat (LE) flux and net radiation (Rn) on the 3rd of August 1994.
Carbon dioxide is a so-called spore gas. This means that measuring the COa flux density very serious errors emerge for which correction should be carried out. When CO 2 fluctuations are directly correlated to the vertical wind component fluctuations by eddy-correlation, density fluctuations affect the vertical velocity fluctuations. Density fluctuations are caused by heat and water vapor quantities. To be able to carry out the corrections properly, together with the CO2 flux the fluxes of heat and water vapor have to be measured. A so-called fast suction technique is used for taking air samples. These air samples are directly analyzed in the field for their water vapor and CO 2 components and the results are correlated with the vertical wind component near the sampling intake at 7 meters above the soil surface. Density corrections or Webb-corrections are applied later [6]. The CO2 concentration and flux will be measured all year round, so some estimates of the soil contribution to the total flux could be made.
5. SURFACE CHARACTERISTICS Throughout the growing season the state of the vegetation will be monitored.
449
This means that at representative locations within the bog area the leaf area index (LAI) and the vertical distribution will be estimated. Moreover estimates will be made about the horizontal variability. For the LAI measurements two techniques will be used; first, a direct method where leaf area is measured optically (PC hand scanner), second an indirect technique where the extinction of direct irradiation within the canopy is measured (Delta-T sunfleck ceptometer). By selecting a representative measurement site for the meteorological station, the obtained data will provide adequate information about two important surface characteristics: the roughness length, z o and the displacement height, d and their courses during the experiment. Moreover estimates will be made of the roughness length for heat, Zo,, for this characteristic can differ much from the z o for momentum. At various locations in the bog area, soil samples will be taken to obtain assessment of important soil parameters like: thermal conductivity, moisture content, soil composition and heat capacity.
6. C O N C L U S I O N S Because the measurements have just started, but will continue for an other year no hard results can be shown here. Some of the results are used for canopy simulation models and will be used in a numerical boundary layer study. It's clear that the eddy correlation method is not a very straight forward method but significant corrections should be applied. Especially when CO 2 fluxes are measured, the Webb corrections are of major importance. Attention should be given to soil and vegetation cycle and their contribution to and influence on the surface fluxes.
7. R E F E R E N C E S
[1] [2] [3]
De Bruin, H.A.R., 1994; Boundary-Layer Meteorology, 68, pp. 427-432. Kohsiek, W, 1982; Boundary-Layer Meteorology, 24, pp.89-107.
[4]
Moore, C.J., 1986; Boundary-Layer Meteorology, 37, pp. 17-35. MCMillen, R.T., 1988; Boundary-Layer Meteorology, 43, pp. 231-245.
[5]
Watson, R.T., Rodhe, H., Oeschger, H and Siegenthaler, U., 1990; In: Climate change: the IPCC scientific assessment, Cambridge.
450 [6]
Webb, E.K., Pearman, G.I. and Leuning, R., 1980 Quart. J. Royal Meteorol. Soc., 106, pp 85-100.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
453
ASSESSMENT REPORT ON NRP SUBTHEME "GREENHOUSE
GASES"
SOURCES AND SINKS OF CO2 CH4 AND N20, DATABASES AND SOCIO-ECONOMIC CAUSES
J.J.M. Berdowskil A.F. Bouwman2 W.M. Kieskamp3 J. Slanina3
1Netherlands Organization for Applied Scientific Research (TNO-MW) P.O.Box 6011, 2600 JA Delft, The Netherlands 2National Institute of Public Health and Environmental Protection (RIVM) P.O.Box 1, 3720 BA Bilthoven, The Netherlands 3Netherlands Energy Research Foundation (ECN) P.O.Box 1, 1755 ZG Petten, The Netherlands
With contributions by: H.G. van Faassen, P.J. Kuikman
W. M. Kieskamp W. Ruijgrok, M. Vosbeek, H. Spoelstra G.M.J. Mohren N.H. Batjes, E.M. Bridges, C.R. Oldeman
AB-DLO, Research Institute for Agrobiology and Soil Fertility, Wageningen ECN, Netherlands Energy Research Foundation, Petten KEMA Environmental Services, Arnhem IBN-DLO, Institute for Forestry and Nature Research, Wageningen ISRIC, International Soil Reference and Information Centre, Wageningen
454 J.A.M. de Bont, H. Denier van der Gon, A. van Dasselaar, B.O.M. Dirks, J. Goudriaan, H.J. Heipieper, P. Hofschreuder, P. Leffelaar, J. Lelieveld, S.W.M. Kengen, J.C. Koops, O. Oenema, R. Segers, A.J.M. Stams, D. van Veenhuysen, G. Velthof
LUW, Wageningen Agricultural University
C.G.M. Klein Goldewijk, C. Kroeze, RIVM, National Institute of Public R. Leemans, C.W.M. van der Maas, Health and Environmental J.G. van Minnen, J.G.J. Olivier, W.L.M. Smeets, Protection, Bilthoven R.J. Swart J. Oonk, J.I. Walpot
H.P. Baars, J. Baas, H.S.M.A. Diederen, J.H. Duyzer, J.C.Th. Hollander J.G. de Beer, A.P.C. Faaij
TNO-M&E, Institute for Applied Scientific Research Environmental and Energy Technology, Apeldoorn TNO-MW, Institute for Environmental Sciences, Delft RUU, University of Utrecht
455
Contents Abstract 0
0
0
0
0
0
7.
Introduction 1.1. Aim and organization 1.2 Assessment of the uncertainties in sources and sinks of greenhouse gases at the time of the start of NRP 1.3 Assessment of the developments in the knowledge on sources and sinks of greenhouse gases during the course of NRP phase I Carbon d i o x i d e (CO2) 2.1 Overview of the C02 cluster 2.2 Methodology 2.3 Results 2.4 F u t u r e research M e t h a n e (CH4) 3.1 Preparatory studies and organization 3.2 Methods 3.3 Results 3.4 Integration of results 3.5 F u t u r e research needs N i t r o u s o x i d e (N20) 4.1 Preparatory studies and organization 4.2 Methods 4.3 Results 4.4 Integration of results 4.5 F u t u r e research needs Emission database development 5.1 World Inventory of Soil Emission potentials (WISE) 5.2 Emission Database for Global Atmospheric Research (EDGAR) 5.3 F u t u r e research needs with respect to database development Socio-economic causes 6.1 Methods 6.2 Results 6.3 F u t u r e research needs R e f e r e n c e s and p u b l i c a t i o n s
456 ABSTRACT The aim of the subtheme Greenhouse gases of the Dutch National Research programme on (NRP) is to quantify the sources and sinks of the major greenhouse gases to enable estimates of the future atmospheric concentration. The major part of the projects in this theme is focused on the Dutch situation, but the results can be extrapolated countries or regions. The information gained will be used for Dutch policy decisions regarding abatement of greenhouse gases. S e c t i o n 1 deals with the aim and organization of Causes of climate change, and relates the scope to increased awareness of uncertainties in sources and sinks of greenhouse gases: at the start of the National Research Programme the general consensus of the scientific community was t h a t these uncertainties were not extreme large, it is nowadays accepted that these uncertainties are larger t h a n assumed before. The aim the Cluster CO 2 ( S e c t i o n 2) was devoted to study the exchange between t e r r e s t r i a l ecosystems and the atmosphere to gain more knowledge of the "fertilization" flux. The research was mainly focused on the development of a CO2 exchange model for grassland describing diurnal and seasonal fluxes, and on the validation of this local scale model on a regional and national scale. In both the clusters CH4 and N20 (respectively S e c t i o n 3 and S e c t i o n 4) anthropogenic and biogenic sources were studied. Major criteria to study sources were the source strength, but also the uncertainty in the source estimate and the potential emission reduction, all projected on the Dutch situation. Exception were the projects on CH4 emission from rice fields, and the sea/air exchange of N20 in oceans; expertise was available in The Netherlands to carry out these studies. As in the sub-theme CO2 the study of processes in grasslands was given a high priority in the sub themes CH4 and N20 in order to quantify emission the mentioned greenhouse gases. Moreover, in the CH4-sub theme projects were performed to evaluate and validate the strength of various sources. The two remaining clusters (limited in extend) were aimed at the development of emission databases and geographic quantification of soil processes controlling greenhouse gas fluxes (cluster Database Development, S e c t i o n 5), and on national inventories (cluster Socio-economic Causes, S e c t i o n 6). In the framework of the first cluster two databases were developed, one was the World Inventory of Soil Emission potentials (WISE), a global gridded database of the primary soil factors controlling soil greenhouse gas emissions, and the other was Emission Database for Global Atmospheric Research (EDGAR) aimed to describe the processes as land use, energy consumption etc, which control the emissions of greenhouse gases and other air pollutants. The goal of the other sub theme was to develop and apply methodologies to compile national inventories of greenhouse gas emissions in The Netherlands, focused on the compounds CH4 and N20.
457 1.
INTRODUCTION
1.1 Aim and organization
The scope of this theme was the a s s e s s m e n t of the causes of Dutch National Research P r o g r a m m e Global Air Pollution and Climate Change (NRP). This obviously includes the cycle of the greenhouse gases but also the description of anthropogenic activities resulting in changes in atmospheric concentrations. The aim of the causes of climate change was to provide information needed to quantify the sources and sinks of the major greenhouse gases in order to enable more accurate e s t i m a t e s of future atmospheric concentrations. While most of the programmes within National Research Programme on were defined in a bottom-up process, it was decided to organise the research on cycles of greenhouse gases in a different way. First, a first order analysis of the uncertainty in the cycles of CO2 (carbon dioxide), CH4 (methane), and N 2 0 (nitrous oxide) was made by a small group of Dutch scientists, active in this field. The conclusions of this analysis were t h a t the main u n c e r t a i n t y in the cycle of CO2 was caused by insufficient information on i m p o r t a n t sinks such as u p t a k e by oceans and t e r r e s t r i a l ecosystems. More knowledge on the level of mechanistic descriptions was clearly needed, but the process of integrating locally derived information regarding important sinks of CO2 up to the level of relevant descriptions of CO2 exchange on regional, continental, or global scale also introduce large uncertainties. The same situation was observed r e g a r d i n g the emissions of methane. Also in this case local m e a s u r e m e n t s are extrapolated to regional and global scale, introducing large errors. For both CO2 and CH4 the scaling problem, this is generalisation of local m e a s u r e m e n t s to regional and (sub)continental scale, was seen as a serious problem. In the case of N20 the state of knowledge was worse, information on the mechanisms of e.g. emission of N20 during nitrification or denitrification was not really available. Second, it decided t h a t coherent programs would be formulated for each of these greenhouse gases, which should address the mentioned m a i n problems. These programs should be formulated in such a way t h a t on one h a n d typical Dutch aspects of the cycle of greenhouse gases would be emphasised and t h a t on the other hand the information gained in the research would constitute a worthwhile contribution to international programs, primarily within the scope of IGAC. Based on these considerations the following clusters of projects were developed: s t u d y of the CO2 exchange between grasslands and the atmosphere, and development of methodology to validate CO2 exchange models for larger areas; investigation of CH4 emissions of selected sources, thought to contribute significantly and with a large margin of error, and development of methodology to validate CH4 exchange models for larger areas; quantification of N 2 0 emissions from fossil sources and i m p o r t a n t biogenic systems, in particular from grasslands, sewage t r e a t m e n t systems and from freshwater and marine systems.
458 Table 1.1 List of projects in the NRP subtheme "Greenhouse gases" by clusters Title
Project leade r
Numbe r
Carbon dioxide (C02) ecosystem studies, model development and validation The seasonal cycle of the CO2 exchange between J. Goudriaan 852062 atmosphere and vegetated surfaces Quantification of carbon fluxes in grassland
P.J. Kuikman
852063
A feasibility study for aircraft based flux measurements of CO2
W. Ruijgrok
852065
The development of a geographically explicit dynamic carbon cycle model
R Leemans
852067
Quantification of carbon fluxes in Dutch forests (Part 1: Desk study)
G.M.J. Mohren
852071
Determining relative importance of sources and sinks of carbon dioxide using carbon isotope measurements
W.M. Kieskamp
852076
Measurement of the exchange of C02 between the atmosphere and a grassland
W. Ruijgrok
853116
R.J. Nielen
850008
Methane (CH4) biogenic sources, fossil sources Quantification of methane emissions due to natural gas losses and petroleum production The influence of soil parameters on the production and emission of methane in/by wet rice paddies
N. van Breemen 850009
Greenhouse gases from landfills in the Netherlands
C. Verschut
850023
Methane formation by anaerobic consortia in organic grassland soils
A.J.M. Stams
853120
Programming study for methane research Validation of source strengths of atmospheric methane using carbon isotope ratios.
J.J.M. Berdowski 852068 W.M. Kieskamp 852097
Quantification of methane emissions in the exploration and production of natural gas and petroleum The Netherlands
J. Oonk
853104
459 Measurement study landfill gas production emission and recovery
J. Oonk
853105
Effects of grassland management on the emission of CH4 from grassland on peat soils
O. Oenema
853121
The methane consumption by indigenous grassland micro flora
J.A.M. de Bont
853122
From methane formation and oxidation to methane fluxes in organic grassland soils: modelling
P.A. Leffelaar
853123
Evaluation and validation of the CH4 emissions in the Netherlands and contributions from various sources
J.C.Th. Hollander 853124
Determination of emissions of methane in rural areas
P. Hofschreuder
853125
N20 emission from fossil fuel combustion in power plants
H. Spoelstra
850006
Preliminary study on N20 flux measurements
H.S.M. Diederen 850012
Investigation of the contribution of traffic to N20 emissions both now and in the future
J. Baas
850030
Effects of nitrogen fertilization and grazing on the N20 emission from grassland.
O. Oenema
852073
Factors influencing the ratio N ~ 2 0 as nitrate is removed from the soil by denitrification The emission of N20 from grassland
P.A. Leffelaar
852074
H.G.v. Faassen
852078
Modelling of soil emissions of nitrous oxide for global studies
A.F. Bouwman
852079
Measurement of atmospheric emissions of N20 from biogenous surface sources in general and grasslandecosystems in particular
J.H. Duyzer
852096
Nitrous Oxide (N20) biogenic sources, fossil sources
Database development emission database, geographic quantification of soil controlling gas fluxes Global emission database
J.J.M. Berdowski 850032
Geographic quantification of soil factors and soil processes that control fluxes of greenhouse gases (Currently used acronym: WISE, World Inventory of Soil factors and processes that control Emissions of greenhouse gases)
E.M. Bridges / N.H. Batjes
851039
460 Emission Database for Global Atmospheric Research (EDGAR); Phase 2: data collection and implementation
J.G.J. Olivier
851060
R.J. Swart
850019
Socio-economic causes national inventory, policy analysis Social causes of the greenhouse effect and emissions inventories
The information gained in these clustered projects should contribute to existing or future emission data bases. In the CO2 cluster, a specific project was formulated to e n s u r e t h a t the results of this cluster was t r a n s f e r r e d to the EDGAR and IMAGE data base (See Annex 2 for acronyms). In the CH4 and N20 clusters this t r a n s f e r was regulated in a less formal way, mainly because the estimate was at the time of the start of this work, t h a t insufficient information would be available after 3 years to perform this task rigorously. These clusters started fairly late in the NRP phase-I, so only some two and a halve years were available for these studies, in stead of five for most of the other NRP-I projects. 1.2 A s s e s s m e n t of t h e u n c e r t a i n t i e s in s o u r c e s a n d s i n k s o f g r e e n h o u s e g a s e s at t h e t i m e of t h e start of N R P At the time of the s t a r t of NRP-I (1989-1990) general consensus existed in the scientific community as expressed in the 1992 IPCC report, on the uncertainties in the predictions of climatic changes. It was widely accepted t h a t it was possible to predict future concentrations of greenhouse gases, with a considerable margin of errors of course. An important problem was perceived in the translation of these changes in concentrations of the radiative active gases in changes in the radiative balance of the earth. But the main problem, as perceived in t h a t period (as reflected in the discussions during the Chamrousse conference in 1989) were the effects of these shifts in the radiative budgets expressed in t e r m s of possible climatic changes.
The Carbon cycle T h a t the above mentioned consensus existed regarding the Carbon cycle was r e m a r k a b l e , in view of the fact t h a t already m a n y observations were available which indicated t h a t the uncertainty in sources and sinks of greenhouse gases including CO2 were much larger than thus far assumed. This was made very clear during a n u m b e r of scientific meetings, e.g. in the proceedings of the IUPAC workshop 'Assessment of uncertainties in the projected concentrations of carbon dioxide in the atmosphere' (Slanina et al., 1991). In these proceedings it is stated t h a t the uncertainties in the most important sinks of CO2 are very large. The estimates of uptake by the oceans vary from 1-2.5 Pg C per year, with a value of b e t w e e n 1-2 Pg as the most probable range. The increase of CO2 in the atmosphere will induce enhanced growth of vegetation and part of the carbon, fixed in this m a n n e r , will be present in the form of enlarged root systems. A certain fraction of the extra carbon, present in roots, will remain in the soil after the decay of the vegetation and be stabilised for periods between 50 and 500 years. Estimates of the value of this so-called 'fertilization' flux range from i to 3 Pg C y-1
461 (Table 1.2) (Goudriaan, 1989 and Tans et al., 1990). These uncertainties have a decisive influence on future environmental policy decisions (Slanina et al., 1991). If a large fertilization effect exists, abatement measures are feasible. If uptake by the oceans and fertilization effect are in the range of the lower estimates, stabilisation of the atmospheric CO2 concentration is very difficult. If we assume a n n u a l fluxes of 6 Pg C, 1.5 Pg C, 2 Pg C, and of 2.5 Pg C, for respectively fossil fuel, landuse changes, oceanic uptake, and terrestrial fertilization, the difference of 3 Pg C between sinks and sources accounts for the accumulation of CO2 in the atmosphere. If we are able to stop the deforestation and induce increase of forests (corresponding to -0.2 Pg C y-l), the difference between sinks and sources would be in the order of 1.3 Pg C, corresponding to approximately 25% reduction of the emissions by fossil fuels. Measures directed to stop deforestation and to optimise agricultural production are probably cheaper than reductions in the order of 60% or more of emissions by fossil fuel. The relatively low agricultural productivity per unit land area of the countries containing the large tropical forests (0.1 to 0.2 of the potential production per unit of land area, to be compared with 0.6 to 1.1 for E u r o p e a n countries) leave room for such a policy. A combination of different measures, reduction of the use of fossil fuel, reforestation, and optimisation of a g r i c u l t u r a l activities could be effective in this case and will leave room for extension of emissions by the developing countries.In the case t h a t no fertilization effect exists and the uptake by oceans is only moderate, a very different picture emerges. A reduction of at least 60% of the emissions of fossil fuel could be necessary to stabilise the present CO2 concentrations in the atmosphere. Any increase in the emissions of the third world countries would ask an even more s t r i n g e n t emission a b a t e m e n t in order to prevent a f u r t h e r increase of the a t m o s p h e r i c CO2 concentration. In the worst case even the most s t r i n g e n t emission reductions in the industrial countries could be insufficient to counteract increasing emissions in the developing countries. One would be tempted to invest in adaptation strategies r a t h e r than in abatement of emissions if this latter situation proves to be true. It is clear t h a t the uncertainties in the sinks of CO2 will enormously influence future political developments and t h a t the reduction of these uncertainties is of prime importance. The conclusions of the IUPAC workshop made very clear that predictions of future CO2 concentrations had far larger uncertainties as was assumed until then, and t h a t additional research on the role of terrestrial systems on the CO2 budget was urgently needed.
462 Table 1.2 Sources and sinks of C O 2 (IPCC, 1990) Flux (Tg y-l)
Source Fossil fuel combustion Deforestation/landuse
5.4 + 0.5 1.6 + 1.0 n
Sink U p t a k e by oceans "Terrestrial fertilization"
2.0 + 0.8 1.6 + 1.4
Atmospheric increase
3.4
+
0.2
Sources and sinks of CH4 The same situation as for C O 2 existed, in essence, regarding the sources and sinks of CH4. In Table 1.3 the estimates are given of the source strength for the most i m p o r t a n t m e t h a n e emissions, as presented by IPCC in 1990 (IPCC, 1990). It was felt t h a t these estimates were, of course with a degree of uncertainty, fairly well established. The fact t h a t the emissions of methane were nearly equal to the sum of m e t h a n e oxidized in photo-chemical reactions and the a m o u n t tied in with increasing atmospheric concentrations, was regarded as an objective proof for this assessment. The conclusions of the IUPAC workshop on uncertainties of in the projected concentrations of m e t h a n e in the atmosphere (Slanina et al., 1994) among other scientific meetings, were very different: the latest reports (Slanina et al., 1994) about the emissions of N o r t h e r n Wetlands, indicated t h a t these emissions could be substantially lower as formerly assumed. The emissions of Northern wetlands were extrapolated to be in the order of 20 Tg y-1 instead of 80 Tg y-1 ; an unexpected conclusion of the workshop was that the possibility exists t h a t India, one of the main rice growing countries, contributes no more t h a n 7% of global emissions from rice crops, because most Indian paddy crop is t a k e n from irrigated field and only a small portion from water-logged fields (Slanina et al., 1994); the u n c e r t a i n t y in atmospheric oxidation is so large (in the order of 40% according the estimates of Calvert, 1994) t h a t this m e c h a n i s m cannot be used to check our emission inventories; the emissions of landfills could be much larger as previously assumed, based on the first results of m e a s u r e m e n t s in Canada, China and other countries (Slanina et al., 1994). In these proceedings it is very clearly concluded t h a t the extrapolations of very local m e a s u r e m e n t s to emission fluxes on regional, continental and global scale is probably one of the m a i n sources of errors and uncertainties. The specific recommendations is made to develop methods and m e a s u r e m e n t strategies for specific v a l i d a t i o n m e a s u r e m e n t s to e v a l u a t e emissions on regional and
463 (sub)continental scale. These validation m e a s u r e m e n t s are required to check whether the measurements of methane emissions carried out on very small scale, have been extrapolated correctly in the past to global dimensions. This assessment of the sources and sinks of methane makes very clear t h a t the state of knowledge at that time did not provide a suitable scientific fundament for predictions of future development and hence for abatement policies. This a s s e s s m e n t already mentions that the trends are changing (Khalil et al., 1994). For an extended period a exponential growth, in the order of about 1% per year has been observed. This trend has abruptly changed and is much less t h a n formerly observed (Steele et al., 1992). Rigorous explanations were not offered, a clear proof of the lack of knowledge in this area. Table 1.3 Sources and sinks of CH4 (IPCC, 1990) Flux (Tg y-l)
Range (Tg y-l)
Source Natural Wetlands Rice paddies Enteric fermentation Gas drilling, venting, transmission Biomass burning Termites Landfills Coal mining Ocean Freshwaters CH4 hydrates destablilization
115 110 80 45 40 40 40 35 10 5 5
(100-200) (25-170) (65-100) (25-50) (20-80) ( 10-100) (20-70) (19-50) (5-20) (1-25) (0-100)
30 500
(14-45) (400-600)
44
(40-48)
Sink Removal by soils Reaction with OH in the atmosphere
Atmospheric increase
S o u r c e s a n d s i n k s of N20 The assessment of the sources of nitrous oxide was at the time different from the other gases. It was assumed t h a t all estimates of sources were quite uncertain. Tropical forest soils were regarded as the single most important source of nitrous oxide to the atmosphere. N20 is also emitted by a large number of smaller sources, such as biomass burning, agricultural activities leading to nitrification and denitrification processes and specialised industrial processes (Table 1.4). The conclusion was that most of these sources were very difficult to evaluate. As a consequence the uncertainties in the emission estimates were estimated to be large. In Europe, the production of N20 by agricultural systems with high loads of
464 nitrogen, the emissions of electricity generation plants and the exhausts of cars equipped with catalysts were seen as major sources. The high e s t i m a t e s of emissions of electricity generation plants were caused by artifacts in sampling and analysis of flue gases, as was proven in the first stages of NRP (Spoelstra, 1992). Table 1.4 Sources and sinks of N20 (IPCC, 1990) Flux (Tg y-l)
Source Ocean, estuaries Fertilizer (including ground water) Soils (tropical forest) (temperate forest) Combustion Biomass burning
1.4- 2.6 0.01 - 2.2 2.2 - 3.7 0.7- 1.5 0.1 - 0.3 0.02 - 0.2
Sink Removal by soil Stratospheric loss
? 7 - 13
Atmospheric increase
3 -4.5
1.3 A s s e s s m e n t o f t h e d e v e l o p m e n t s in t h e k n o w l e d g e o n s o u r c e s a n d s i n k s of g r e e n h o u s e s gases d u r i n g the course of N R P p h a s e I During the last 5 years, the period of NRP phase I, the scientific community has accepted t h a t the uncertainties in sources and sinks of greenhouse gases are much larger as assumed in 1989. The warnings, exemplified by the proceedings of the IUPAC workshops on the uncertainties of in the projected concentrations of carbon dioxide and methane in the atmosphere (Slanina et al., 1991 and 1994) that these uncertainties are a major in future predictions, have now widely be accepted. That a large uncertainties exist has been made clear by the fact t h a t the trends in the concentration of greenhouse gases in the atmosphere have changed drastically and t h a t no reasonable explanation can be provided for these changes in the trends. So the present state of affairs can be summed up as follows: The fair amount of research on sources and sinks of greenhouse gases, carried out internationally during this period of 5 years, has led to the conclusion that the u n c e r t a i n t y in these sources and sinks is much larger t h a n formerly assumed. This may seem a r a t h e r negative conclusion, but it must be born in mind t h a t a proper evaluation of the state of affairs is essential to formulate effective research in the future to remedy this problem. The long-term average growth rate of atmospheric CO2 concentration has increased since the s t a r t of the m e a s u r e m e n t s at M a u n a Loa. This rate was about 0.8 ppmv, 1.3 ppmv, and 1.6 ppmv for respectively the 1960s, the 1970s, and the 1980s. Systematic higher CO2 concentration growth rates have been observed during the years 1988-90, which exceeded the level of 2.0
465 ppmv y-l, while in the subsequent years (1991, 1992, 1993) very low growth rates have been observed, in the order of 0.6 ppmv y-1. Indications exist, based on the most recent data, that the trend is returning towards long-term g r o w t h rates. It m u s t be kept in mind t h a t the a b r u p t decrease in atmospheric CO2 growth rate in the period 1991-1993 exceeds any previous variation in the existing time series of atmospheric CO2 concentration.
co~ The research in this area has expanded, leading to an increase in knowledge of the oceanic and t e r r e s t r i a l sinks of the carbon dioxide. New insights have been obtained on the problem of "unidentified" terrestrial sink (indicated by Tans et al., 1990 and the IUPAC report Slanina et al., 1994, as "fertilization effect") to specific processes. The uptake of CO2 by terrestrial systems is governed, most likely by two important processes: 1) Changes in land use. The present estimate is t h a t the net emissions by changes in land use total 1.1 Pg with an uncertainty of 1.2 Pg. This net emission flux consists of the sum of emissions by tropical sources (1.6 Pg) minus a mid-latitude uptake due to forestation, etc. of about 0.5 Pg, according to recent estimates. 2) The existence of the CO 2 fertilization is increasingly accepted as an important sink. But there is increasing evidence that different interactions play a role. The direct CO2-enhanced plant growth could provide a sink with a strength of 0.5-2.0 Pg C y-1. E n h a n c e d supply of nutrients, e.g. by t r a n s p o r t and deposition of sulphur and nitrogen compounds leads to the so-called nitrogen fertilization, which could contribute to an uptake of 0.2-1.0 Pg C y-1. As a result, the future effects of CO2 are difficult to predict. Climatic change could have, on a global scale lead to a net uptake, equivalent to 0-1.0 Pg C y-1. The resulting picture is that indeed the knowledge of these processes has been increased. In 1989 sinks such as the "fertilization effect" were still hotly debated. But the uncertainty is still very large and remains in boundaries as indicated by the IUPAC report. This is the most i m p o r t a n t reason, why no consistent explanations can be offered for the variations in the yearly trends, observed in the last decade. The debate between scientists, who contribute most of the terrestrial sink to changes in land use, and those who claim that fertilization effects are the main cause, is still raging, see articles of Tans et al., (1990) and Smith et al., (1992). It is very clear that no reliable predictions of future CO2 concentrations can be made and t h a t the development of optimal strategies for a b a t e m e n t is severely hindered, until these questions are resolved.
CIt4 Essentially the same situation exists for CH4 as described for CO2. The changes in the trends for CH4 have in fact been more pronounced as observed for CO2. Results from different networks indicate that the globally averaged growth rates for m e t h a n e have declined from approximately 20 ppbv y-1 in the period 1979-1980 to 13 ppbv y-1 in 1983, to 10 ppbv y-1 in 1990 and to about 5 ppbv y-1 in 1992 (Steele et al., 1992 and Khalil et al., 1993d en 1993e). The trend in southern hemisphere has halved and the increase in 1991-1992 in the northern hemisphere was close to zero (Dlugokencky et al., 1992). The cause of this change
466 in methane growth rates is unknown and still a matter of speculation. (Khalil et al., 1992c and Steele et al., 1992), (Dlugokencky et al., 1992 and (Dlugokencky et al.) A wide range of explanations are given: decreasing CH4 emissions from the former Soviet Union; lowering of lowered the t e m p e r a t u r e of the northern wetlands and thereby decreasing methane emissions; indirectly caused by emission of the Pinatubo Volcano; lowering of the w a t e r table increases the thickness of the layer over which m e t h a n e oxidation can take place, so northern wetlands appear to be more sensitive to changes in moisture than temperature; t e r m i n a t i o n of the one-to-one correlation between m e t h a n e emissions and growth of the global population, as result of lack of suitable areas for rice cultivation or cattle raising; increasing OH-radical concentrations, caused by increasing UV-B radiation, could have shortened the life time of CH4. This wide range of hypotheses demonstrates quite clearly the lack of information on the s t r e n g t h and the variability of sources of m e t h a n e and this situation is acknowledged widely within the scientific community.
N2,0 The development in views about the emissions of N20 have been slightly different compared to the other two gases mentioned. U n c e r t a i n t y exists about the atmospheric concentration of N20 in the pre-industrial period. Estimates range from 260 to about 290 ppbv, compared with a concentration of 310 ppbv in 1993. So, the estimates of the yearly trend show a considerable uncertainty (Khalil et al., 1992C) Recent results indicate that the trend of N20 has been smaller in the last years t h a n the average of the last two decades and decreased from about 0.8 ppbv to 0.6 ppbv. The general opinion in the scientific community is that N20 plays only a minor role in the changes of the radiative balance of the e a r t h and t h a t the increase in concentration has been much less spectacular compared to the other greenhouse gases. Questions are raised however on the impact of new industrial processes (large scale introduction of catalytic devices for cars and catalytic NOx reduction in industry), of changes in agricultural practice (large scale application of nitrogen fertilizers) or changes in the water tables of wetlands and agricultural areas on future N20 emissions. The developments of the last five years can be s u m m e d up in the following statements: the considerable a m o u n t of new information has m a d e clear t h a t the uncertainties in sources and sinks of greenhouse gases were much larger than assumed in the recent past; major causes for the uncertainties in sinks and sources of greenhouse gases have been identified. This increased knowledge will provide the necessary fundament for effective research in the future. -
467 This s u m m a r y of the recent results of could give the impression t h a t less risk for climatic change is present, compared with a few years ago, as the concentrations of greenhouse gases are increasing slower as expected. This would be a very d a n g e r o u s assumption. As the changes to lower t r e n d s in the a t m o s p h e r i c concentrations of greenhouse gases are not well understood it is impossible to indicate w h e t h e r this situation will last. To the contrary, it is very well possible t h a t the t r e n d s could change in u p w a r d direction very fast by the impact of changes in industrial and agricultural practice. A certain lack of knowledge regarding the contribution of different sources of greenhouse gases is not a problem in the first stages of abatement policies. A wide range of so-called no regret options, which will not only reduce the emission of g r e e n h o u s e gases but also contribute to a b a t e m e n t of other e n v i r o n m e n t a l problems, is available and environmental policies can been adapted accordingly. This state of knowledge, however, is not a good basis for developing a b a t e m e n t policies over longer periods. For this reason a better understanding of the sources and sinks of greenhouse gases must have a high priority on the scientific agenda.
2.
C A R B O N D I O X I D E (CO2)
2.1 Overview
o f t h e CO2 c l u s t e r
The considerations in chapter 1 led to the decision t h a t additional research on sources and sinks of CO2 should be directed to the role of terrestrial ecosystems in the CO2 cycle, a p a r t from the already on-going activities on the exchange of CO2 between the oceans and the atmosphere. Reports on the impact of the "fertilization flux" t h a t were published at t h a t time, h a d m a d e clear t h a t the exchange between t e r r e s t r i a l ecosystems and the atmosphere was very important and m u s t be known better in order to be able to model the CO2 cycle and to predict future CO2 concentrations in the atmosphere. It was proposed to study the exchange of CO2 between g r a s s l a n d s and the atmosphere for the following reasons: The Netherlands are to a large extend covered by grasslands. Pastures are a major component in European land use and the amount of grassland has been considerably extended on a global scale during recent decades; grasslands exhibit the same behaviour as forests as far as the fertilization effect is concerned: net primary production increases, allocation to roots as well as losses into soil, and potentially more C will be stored as organic m a t t e r with large residence time. F a r m l a n d s are amongst the most productive ecosystems (in terms of net photosynthesis). The soils of grasslands contain generally large amounts of carbon, and the carbon content of soils increases at higher concentration of CO2 in the atmosphere. less knowledge was available on the exchange of CO2 between grasslands and the atmosphere compared to forest ecosystems; expertise was available in The Netherlands. It was decided to develop a coherent program, dedicated to formulate and validate a improved model simulating the exchange of CO2 between grassland and the
468 atmosphere. This model encompass diurnal to seasonal fluxes, and the exchange of carbon between the soils of grasslands and the atmosphere. Better knowledge on the gross exchange of carbon between grasslands and the atmosphere is urgently needed to: assess the effects of changes in land use on the global carbon cycle; u n d e r s t a n d short time to yearly trends of CO2 concentrations in the atmosphere. (Analysis of these trends are essential tools to u n d e r s t a n d the CO2 cycle); assess the potential contribution of fertilization effect to sequestering of carbon in soils. -
-
In order to reach this goal the following activities were incorporated in a number of projects:
1)
Development of an improved model describing the exchange of CO2 between grassland and the atmosphere (LUW-TPE, project no. 852062). This model will provide a good description of diurnal and seasonal fluxes, and will include the exchange of carbon between the soils of grasslands and the atmosphere.
2)
Measurements of the exchange flux of CO 2 over the most important types of grasslands (soil) in The Netherlands (KEMA, project no. 853116; ECN, project no. 852076). It was conceived that the different kinds of soil of grasslands (clay and peat) would have a strong effect on these exchange fluxes. The results of these measurements would be used to parameterize better models and to validate them on a local scale (ECN, project no. 852076).
3)
I n v e s t i g a t i o n of the fertilization effect on g r a s s l a n d by m e a n s of pulse-labelling by 14C02 (IB-DLO -at present AB-DLO-, project no. 852063). Grass is exposed to 14C02 during short periods and the distribution of 14C between different parts of the vegetation and the soil is determined and fluxes of carbon are calculated.
4)
M e a s u r e m e n t s of the exchange flux of C02 over larger areas of grassland, using eddy-correlation, gradient m e a s u r e m e n t s at higher elevation (ECN, project no. 852076), and eddy-correlation m e a s u r e m e n t s from aircraft (KEMA, project no. 852065).
5)
M e a s u r e m e n t of changes in CO 2 concentration and isotopic composition (13C/12C and 14C/12C) at an altitude of 200 meters on a tower, with the objective to obtain regional validation of sources and sinks of CO2 with emphasis on the role of terrestrial systems (ECN, project no. 8520786). The isotope ratios are dependent on sources and exchange processes.
An co-ordination group monitored the progress of the different projects and facilitated that coherent results could be obtained. The sub-theme CO2 encompasses two additional projects. One was a desk study to assess the role of (Dutch) forests as apart of the carbon cycle (IBN-DLO, project no. 852071), while the other project was on the development of a geographically
469 explicit dynamic carbon cycle model t h a t will be incorporated in more complex, integrated models as IMAGE 2.0 (RIVM, project no. 852067).
2.2 Methodology Measuring CO2 concentrations with sufficient precision and accuracy does not present severe problems in the present state of methodology development. Several monitors, based on IR absorption methods, are commercially available. They can m e a s u r e atmospheric concentration with an accuracy of b e t t e r t h a n 1 ppmv. Accuracy is d e p e n d e n t on the quality of calibration and quality a s s u r a n c e s t a n d a r d s , but generally an accuracy of 1 ppmv or better is attainable without major problems. The situation is different when flux measurements must be applied in the field to study in detail the exchange fluxes of CO2 between grasslands and the atmosphere. Three available methods to measure gas exchange between the atmosphere and biosphere have been used in the described projects, i) enclosure, ii) eddy-correlation, and iii) gradient.
Enclosure methods A box is placed over vegetation, water, or soil. Air is pumped through the enclosure and the difference in concentrations measured at inlet and outlet is used to assess deposition or emission rates. Enclosure methods suffer from two problems: one, the enclosure can alter the behaviour of vegetation or soil, and two, the deposition or e m i s s i o n m e a s u r e m e n t is e x t r e m e l y local. The a d v a n t a g e of e n c l o s u r e m e a s u r e m e n t s is t h a t the present state of i n s t r u m e n t a t i o n can be applied in nearly all studies. In m a n y cases enclosure methods have to be used as no other alternative is available. In view of the extreme local effects and as alternatives are available for CO2, it was decided to not to apply enclosure techniques for the s t a n d a r d CO2 exchange m e a s u r e m e n t s . Box m e a s u r e m e n t s are applied in this cluster to investigate the exchange of CO2 between atmosphere and grass with exclusion of soil respiration. In order to quantify the CO2 flux by grass (in contrast to the integrated CO2 flux: grass + soil) an enclosure system was developed which m e a s u r e s continuously the CO2 flux u n d e r conditions of overpressure. This overpressure prohibits the exchange of CO2 between the soil and the atmosphere. The m e a s u r e d CO 2 flUX is, under these conditions, related to net CO2 assimilation of a grass canopy u n d e r field conditions - one of the two components of the integrated CO2 flux. Moisture content, t e m p e r a t u r e and CO2 concentration of the circulating air are regulated to avoid the already mentioned artifacts caused by deviation in the box from local conditions.
Eddy-correlation Eddy-correlation m e a s u r e m e n t s are based on the covariance of fluctuations in a m b i e n t concentrations and vertical windspeed. T u r b u l e n t t r a n s p o r t in the a t m o s p h e r e t a k e s place by eddies. In eddy-correlation m e a s u r e m e n t s the difference in concentration of the investigated compound is measured with a time resolution of 1 to 10 hertz in air moving downward to the surface and moving u p w a r d from the surface. The upward moving air has been in contact with the surface and the concentration has altered due to exchange at the surface. As eddy correlation measures directly, it is very often the preferred method. The regarding speed and precision of the instrumentation for eddy correlation are very often so
470 extreme t h a t the method cannot be used for m a n y trace gases. However, the method can be applied for CO2, and was used in the CO2 cluster. Gradient measurements
Depletion or emissions of pollutants at the surface results in a g r a d i e n t in concentration. Air concentrations of compounds, temperature, and windspeed are m e a s u r e d at different heights over the surface. F r o m these g r a d i e n t s the turbulence of the atmosphere is derived and the fluxes can be calculated. The problem with the gradient method is t h a t a high precision is required of the m e a s u r e m e n t method, as the concentration gradients are often in the range of a few percent of the atmospheric concentration. The i n s t r u m e n t a t i o n for CO2 m e a s u r e m e n t s can fulfil these requirements and gradient m e a s u r e m e n t s have been applied in two projects in the CO2 cluster to measure fluxes. Micrometeorological methods (eddy-correlation and g r a d i e n t m e a s u r e m e n t ) enables to estimate fluxes over a certain area as function of the height of the measurements. If local exchanges are studied, gradient measurements are carried out at heights between 1 and 5 m. The integrated exchange flux over an area of some hundreds of square meters is characterized this way. To study the exchange over an area of a few hectares, the gradient is measured between 1 and 20 m. The precision of flux measurements are limited, typically a precision in the order of 20 % can be reached in most cases. The consequence is that direct m e a s u r e m e n t s of the fertilization flux is not possible. This can be illustrated easily if the total exchange flux of terrestrial ecosystems with the atmosphere, in the order of 100 Pg C y-1 is compared with a high estimate of the fertilization flux of 2 Pg C. The quality of flux measurements was validated in an intercomparison experiment, organized by ECN. The participants of the CO2 cluster ECN and KEMA, and also KNMI and TNO, the latter two institutes are engaged in flux m e a s u r e m e n t s at sea, took part in this one-week experiment in November 1993 at Cabauw, The N e t h e r l a n d s . Two methods were used: the eddy-correlation technique and the gradient technique. Unfortunately, the t e m p e r a t u r e was about 0 ~ and reduced the flux of CO2 considerably. Although the small magnitude of the fluxes makes comparison difficult, it seems t h a t the gradient method tends to result in larger fluxes compared to the eddy-correlation method. Moreover, this experiment made clear t h a t there was a considerable differences between the calibration standards t h a t were used. This implies t h a t direct comparison of absolute concentration values of different set-ups as at the Cabauw experiment can only be achieved by inter-calibration of the standards. Validation of exchange fluxes over larger regional areas (The N e t h e r l a n d s and surroundings) were investigated by two different methods, a dynamic method (aircraft measurements) and a static method (tower measurements). Eddy-correlation m e a s u r e m e n t s in aircraft were applied as a method to obtain integrated exchange fluxes over larger areas. Variations in CO2 concentration and isotopic composition were m e a s u r e d at an tower (200 m) at Cabauw, The Netherlands. Uptake and emission of CO2 over large areas changes not only the CO2 concentrations, but also the isotopic composition. Emissions of fossil fuel contains no 14C and the 13C-12C ratio is dependent on the sources of CO2. The
471 concentrations were measured with a non dispersive infrared spectrometer with an accuracy better t h a n 0.1%. Working standards were calibrated against so called NOAA station standards. An wet a n n u l a r rotating denuder, filled with a N a O H solution, was used to extract CO2 from the air quantitatively. In the laboratory the formed carbonate was isolated (using barium chloride), stored, and before analysis re-converted into CO2. The 13C/12C ratio and the 14C/12C ratio were determined at E C N and the U n i v e r s i t y of U t r e c h t respectively. Meteorological d a t a were provided by the Royal N e t h e r l a n d s Meteorological I n s t i t u t e (KNMI). An 2-dimensional 2-compartment mesoscale transport model was developed at ECN. D a t a on the spatial distribution as well as descriptions of CO2 exchange (both biogenic and anthropogenic) will be used as i n p u t - p a r a m e t e r s to model the observed CO2 concentration and carbon isotopes. To quantify the potential fertilization effect directly, gross a n n u a l carbon flows were estimated in grasslands with 14C pulse labelling. 14CO2 was supplied to grass plants as a single pulse (1-2 hours in a plastic bag covering the plants growing within a soil column) and subsequently the distribution within the plant and soil c o m p a r t m e n t s was m e a s u r e d after a 21-days period in which carbon allocation was completed. This labelling was repeated on 13 representative moments during the growing season. Moreover, the decomposition of shoots and roots and the remaining carbon in soil organic matter was estimated by adding uniformly labelled dead shoots to planted soils in the field and by leaving pulse-labelled plants in the field and m e a s u r i n g the dynamics of the r e m a i n i n g carbon over 18 m o n t h s following the addition and labelling of shoots and roots, respectively. The fate of carbon compounds t h a t are exuded from living roots within 21 days after being a s s i m i l a t e d was followed by adding 'model-rhizodeposits' and m e a s u r i n g the remaining carbon. 2.3 R e s u l t s
Overview o f the results As the cluster of projects started late (some were stated at the end of 1992, others in the middle of 1993) only initial results are available. The most important results are summarized below:
1)
A detailed model is developed to describe the exchange of CO 2. This new model provides information on exchange fluxes with a resolution of 30 m i n u t e s ( F i g u r e 2.1), and thus be used for mechanistic studies. The process to incorporate this model in IMAGE has been started.
2)
Exchange fluxes with a time resolution of I hour or better have been measured over meadows on clay, and peat soils. The results have been transferred to the modellers. The flux measurements are of good quality and provide a good basis for the parameterization.
3)
Gradient m e a s u r e m e n t of exchange fluxes on a scale of hectares have been carried out near Cabauw by measuring concentrations at altitudes between 1 and 10 m over clay/peat soil. A first comparison between model results and actual m e a s u r e m e n t s is given in Figure 2.1. The difference between observed and calculated values are most probably caused by oxidation of peat in the soil
472 regulated by the water table. The combination of better models and good flux measurements enables the study of these important phenomena.
4)
Eddy-correlation measurements of CO2 exchange using an aircraft have been tested. The resolution of the method is insufficient to be useful under the prevailing conditions in The Netherlands.
5)
The experiments with labelled 14C02 have been carried out and the distribution of carbon has been measured. The interpretation of the results is in progress.
6)
Regional validation of models, describing the exchange of CO2 between vegetation and atmosphere, by monitoring variations in CO2 concentrations and isotopic composition at an altitude of 200 m is not possible yet. Distribution of sources and probably very large homogeneous processes like u p tak e in oceans can be studied by this method, but a more detailed assessment will be difficult in view of the limitation in the present models.
Detailed results 1) A dynamic simulation model was developed to calculate the CO 2 flux related to net CO2 assimilation of a grass canopy. The existing carbon cycle model WCCM2 (Goudriaan, 1989) operates with annual time steps and does not consider the precise seasonal and diurnal pattern of CO2 exchange. This study has incorporated these cycles while retaining the final result of the net annual exchange rate. To this end an existing simulation model for crop growth (SUCROS) has been utilized as a basis, in combination with other models for carbon dynamics in the soil (CENTURY). The model first generates the diurnal cycle to obtain the net diurnal assimilation rate, and diurnal soil respiration. These diurnal rates follow a seasonal cycle and are integrated to generate a net annual uptake. The net annual uptake of the above ground vegetation is called the Net Primary Productivity. Factors such as green soil cover, progress in the growing season on basis of accumulated temperature, soil wetness, partitioning of assimilates between plant organs, root dynamic are considered. Respiration rate of plant and soil have been modelled on basis of temperature, biomass and growth rate. The model has the potential to drive a 3-D model for atmospheric CO2 content, first to generate a diurnal cycle in the vertical profile, second to obtain net CO2 exchange rates of a region on a seasonal basis.
A first comparison of the model calculations (grass component) with the i n t e g r a t e d CO2 flux m e a s u r e m e n t s (grass and soil organic m a t t e r components) at C a ba uw (The N e t h e r l a n d s ) , in c o m b i n a t i o n w i t h environmental conditions is presented in Figure 2.1, in which the measured and calculated (grass component - potential) CO2 flux for Cabauw is given (Period March 18 to 20, 1993). LAI2 and LAI4 represent different leaf area index used in the model. The results indicate larger emission fluxes for CO2 as calculated by the model. The difference between calculated and measured CO2 fluxes may well be attributed to oxidation of soil organic matter, as discussed in the section dealing with the integrated flux measurements performed in
473 Cabauw. The combination of modelling activities and experimental approach provides a very useful tool to develop mechanistic descriptions of exchange processes between soil and vegetation and the atmosphere. The development of a model for the calculation of the CO2 flux related to oxidation of soil organic matter was initiated. 0.4 I 0.2
9
y
9~
- urea9
:..:..
:ram "
.
m
. _ _ _ _ . _ I9149149149149149149 9 mn
0
,,,, ..........
~ -0.2
~,,,i, l l l l l l l ..... ~
-d ~
d o
c~ ,~ , d ' ~
mr-
-
g
................... d
cq d
c~ ,~
T,,,IIIIIIIIIIIIIIII',, d
~o ~ 9
,o
~0 n
--
-1,2
--
-1.4
--
~m" ~ 9
" _9
c~ d
~
,~
9
d
~o~ 9
o
0
9
cq ~
,d ~
r"-mr--
d ~
o~
~
9
o 9
===o
[]
9
9
9 009
9 =.,
9 nn
n
9 9 O
-1
9
9
~
-~ -0.6 - -
-0.8 - -
:
W"m,= ,, e e
....... , ....... ,,, ..... ,,~,,,, I l I',',',',', I l I I',',',',',,*,I l IIl III,,
N ~: ,d,,,~ ~
n o
0
9
O O
e
9
~176 o E -0.4 o~
9 e
~
o 0 9
9
9 m ~0
mm
9
O, No O9 O
0
: e9
[]
~176
[] "
9
9
9
."
9
daytime (GMT+ I) .
.
I " measured
Figure
2.1
.
.
.
n calculated (LAI 2)
.
I
* calculated (LAI 4) t I
Measured (integrated) and calculated (grasscomponent-potential) CO2 flux on the experimental site Cabauw (The Netherlands) form March 18 to March 20, 1993
2)
The project on the development of a geographically explicit dynamic carbon cycle model has been fully integrated within the IMAGE 2 projects and data, experience and personnel and results were exchanged and model developments were immediately implemented in the IMAGE 2 framework. A geographic explicit C cycle model has been developed based upon a series of global databases with topography, soil, climate and land cover characteristics. The model identifies globally different land cover types, each of which is divided into its appropriate compartments for C storage and dynamics. The model is driven by Net Primary Production, that is a function of local climate, soil and land use. Several feedback processes are implemented in a mechanistic manner. Sensitivity analysis was carried out and some specific applications to analyze the importance of different feedback processes and the influence of a transient dynamics vegetation response. Large regional differences were obtained for different feedback processes. For example, CO2 fertilization was the dominant feedback in tropical regions, while the temperature response on growth and respiration became dominant in boreal regions. Although changes through feedbacks processes are important determinant of C cycle properties,
474 changes in land use will probably more dominant in the near future. The model is appropriate to also assess the impact of land use change on the C cycle. C o m p o n e n t s of the project have now been reviewed twice d u r i n g the international IMAGE review meetings and adjustments in the approach have been added after recommendations of the review committee. Together with the P o t s d a m I n s t i t u t e for Climate I m p a c t Research an improved version of the IIASA Climate database has been developed. This d a t a b a s e (CLIMATE: Cramer, L e e m a n s I n t e r p o l a t e d Meteorology for Applications in Terrestrial Ecology) now forms the basis for several global modelling efforts (e.g. IGBP-GAIM). The structure of the modelling approach and its main databases used within IMAGE 2.0 Terrestrial Environment Subsystem is accepted by IGBP-GCTE as a valid contribution to their core-project research.
3)
Local scale exchange fluxes with a time resolution of I hour or better have been m e a s u r e d over grasslands on clay and peat soils using both the eddy correlation method and the gradient method. Eddy-correlation was performed at Zegveld (grassland over peat soil), gradient m e a s u r e m e n t s at Lelystad (grass on clay) and at Cabauw. The latter m e a s u r e m e n t s were carried out between 1 and 10 m, so integrated fluxes were measured over several km2 of surface. The soil at Cabauw (peat covered with a layer of typically 20 cm of clay) is not as well characterized as is the case in Zegveld and Lelystad. The sites at Zegveld and Lelystad are p a r t of experimental farms, so soil characteristics and all agricultural treatments are very well documented. In contrast, the meadows around Cabauw are commercially farmed, so little information is available about any application of fertilizer, grazing, pasture, etc. The basic idea was to use the d a t a from Zegveld and Lelystad to p a r a m e t e r i z e the models and carry out validation by m e a n s of the flux m e a s u r e m e n t s at Cabauw over an integrated area. As all the infrastructure at C a b a u w was a l r e a d y in condition, it was possible to e v a l u a t e the m e a s u r e m e n t s of fluxes, convert the results in a format which is suitable for the modellers and transfer them in a very short period. A nearly complete data set of fluxes for Cabauw with a time resolution of 30 minutes for the period March 1993 to March 1994 has been transferred to the modellers and is used for p a r a m e t r i s a t i o n and validation (Figure 2.2). Examination of the results indicate t h a t the fluxes are measured with, for this kind of measurements, a very good precision, in the order of 10% relative. Evaluation and validation of the raw data at Zegveld and Lelystad has not been completed yet, as time was needed to e s t a b l i s h the n e c e s s a r y i n s t r u m e n t a t i o n , to carry out the actual m e a s u r e m e n t s , and to perform intercomparison m e a s u r e m e n t s in order to compare the results form both locations etc. Results, as far as available, have been t r a n s f e r r e d to the modellers. The evaluation of the total data set of both sites is n e a r l y completed, so they will be transferred to the modellers in the near future. The flux m e a s u r e m e n t s performed at the Cabauw site are of such a quality t h a t they indeed provide a good basis for the intended parameterisation and validation. The comparisons of models and experimental results, as reported later in this section, is therefor based on the results of Cabauw only.
475 Normally flux exchange measurements are carried out during relatively short periods. The only m e a s u r e m e n t s performed over longer periods were m e a s u r e m e n t s of dry deposition of NH3 and SO2, carried out by ECN and RIVM over forests and grassland. These m e a s u r e m e n t s were based on gradient methods. The experience at Zegveld, where eddy-correlation was applied, has led to the conclusion t h a t gradient methods are much more suitable for m e a s u r i n g over long periods t h a n eddy-correlation based methodology.
4)
CO2 fluxes on km2 scale have been estimated by measuring concentration gradients up till 10 m altitude at Cabauw, The Netherlands. The results are available to validate the exchange models. The measurement of the exchange of CO2 between grass-on-peat and the atmosphere showed that fluxes caused by photosynthesis as well as soil respiration due to oxidation of peat can be assessed (Figure 2.2). A net uptake only takes place in March, April and May, while during all other months a net emission occurred. The total net emission for the period March 1993 to March 1994 was calculated to be about 3000 kg C ha-1. Two sources contributing to the soil respiration are oxidation of peat layers and animal waste. The CO2 emission due to animal waste was estimated at approximately 600 kg C ha-1 y-1 (about 20%) of the net estimated emission. This leaves 2400 kg C ha-1 y-1 for the oxidation of soil organic m a t t e r (e.g. peat) and is of the same order as the potential emission estimated for the oxidation of "shallow-drained" peat soil (about 2300 kg C ha-1 y-l, Wolff). This is approximately 2.5% and 4% of the total anthropogenic CO2 emission in the N e t h e r l a n d s respectively with and without emission from animal waste. Because the peat layer at Cabauw is covered with clay, it is expected t h a t emissions for grass-on-peat will be larger due to the larger amount of peat that can be oxidized. This type of m e a s u r e m e n t , in combination with models describing the exchange of CO2 between the atmosphere and grass and soils, is in principle capable to provide the necessary information on uptake or loss of carbon as a function of changes in land use.
476 Emission of CO~ at Cabauw March'93 to February'94 200
1200
150
I900
_~ 100
600
~UD .~
O
~
~
50
3oo
~ o~
O "" ~
~
0
0
-5o
-300
< -100 Z -150 -200
=
E
-6oo
........... i
1 i
J
E ~
[m Cumulative Emission [ [] Monthly Net Diurnal Emission
-900 -1200
Figure 2.2 Emissions of CO2 at Cabauw (The Netherlands) from March 1993 to February 1994
5)
The instrumentation has been developed and tested to m e a s u r e C O 2 exchange fluxes by eddy correlation m e a s u r e m e n t s using aircraft. Airborne flux measurements provide information on fluxes over very large areas, depending on height and on large scale geographical distribution of these fluxes. Two measuring flights were performed to test an airborne eddy correlation system developed in the Netherlands. The results of these tests show that the accuracy of the system is insufficient to be useful to support model development under the prevailing conditions in the Netherlands. Fluxes are measured with a precision of 30 to 50% relative and parametrisation and validation of models require data of a better quality. The generally very inhomogeneous landscape of The Netherlands is a severe handicap for aircraft measurements, even if they are carried out at a minimum altitude to reduce the surface area which is observed. It was concluded that the airborne flux measurements were not feasible for application in the Netherlands taking a number of considerations into account: inaccuracy of the measurements, the heterogeneous character of the Dutch landscape and the relatively high costs of this type of measurements.
6)
The experiments with labelled 1 4 C 0 2 have been carried out and the distribution of carbon has been measured. The interpretation of the results is in progress. Preliminary conclusion is that in peat soiis decomposition (mg C added per mg soil C) is twice as fast as in sand and clay, and results in 5-10 times higher incorporation of carbon in biomass and microbial products. Structural organic carbon was retained more in clay (80%) than in sand (58%) whereas the extract in clay was decomposed as fast as in sand. The latter is
477 surprising given the generally higher retention capacity of clay soils than of sandy soils. Care should be taken since these data concern similar additions to all soils wh er eas the carbon input (total C, s t r u c t u r a l and soluble rhizodeposits) to soil might differ between the soils studied.
7)
Regional validation of models, describing the exchange of CO 2 between vegetation and atmosphere, by monitoring variations in CO2 concentrations and isotopic composition at an altitude of 200 m has not been possible yet, although a mesoscale transport model was developed. Also "mesoscale" m e a s u r e m e n t of CO2 concentration were performed and carbon isotopes ratios were determined. Main problem in this project was the availability of detailed databases to be used as input for the model. Also good model descriptions of CO2 exchange were lacking. Both problems will be (partly) solved in the near future. Trajectory analysis show that in periods with enhanced C 0 2 concentrations air was transported over the continent i.e. industrialized areas, especially over G e r m a n y (South East). These periods mostly occurred in winter. The concentrations of CO2, and both the carbon isotopes were also strongly correlated during these periods indication that the main source for CO2 was combustion of fossil fuel. By use of the carbon isotopes the relative contribution of the anthropogenic sources can be estimated. An example of such an estimate is given for December 24, 1992. The concentration and 14C changed from 380 to 425 ppmv, and from 113 to 107 pmC (percent modern carbon) respectively. The air was coming from the north of Germany during this period. The anthropogenic contribution of approximately 50 % was calculated using the increase in concentration and the decrease in 14C. It is assumed here that the CO2 exchange by vegetation in this winter period was negligible, and consequently, soil respiration and litter decomposition are the only biogenic sources for CO2. These sources were assumed to contribute for the other 50% in the change in concentration. The 13C of the emitted CO 2 (anthropogenic and biogenic) was calculated on -24 promilles. The absolute emission of anthropogenic CO2 was estimated using a emission inventory and the trajectory of that particular day. An anthropogenic flux contributing to the change in atmospheric CO2 was estimated to be 9 g CO2 m2 d-1. Using the calculated ratio between anthropogenic and biogenic sources (50%/50%) a biogenic flux of 9 gram CO2 m-2 d-1 is calculated. As comparison the estimated flux of the grasslands in the surroundings of the site Cabauw is 2 g CO2 m-2 d-1. This value for the flux is calculated for the November 1992, when the assimilatory and respiratory processes of the vegetation were still active. (In the period November 1993 to December 1993, estimated emissions by grasslands range from 1 to 10 gram CO2 m-2 d-l). A better estimate of the anthropogenic emission of CO2 will be obtained in the near future with better description of spatial distributions of sources combined with the transport model.
478 8)
The present stock of carbon in living biomass, litter and stable humus and the annual accumulation of carbon in stems for fifteen forest types has been quantified from inventory data on growth and standing volume, and forest soil information in combination with literature data on forest biomass. The forrest area in The Netherlands is about 330000 ha, mainly young plantations of conifers. The present standing volume is 170 m3 ha-1 and the average volume increment was 9.0 m3 ha-1 y-1 over the period 1984 to 1989. At present approximately 63.7 Tg C is stored in the entire forest, including dead organic matter in the forrest soil. About 60% of the carbon is stored in the humus of the soil compartment. The average carbon stock in the stable h u m u s is approximately 110 Mg C ha-l, whereas only 60 Mg C ha-1 and 20 Mg C ha-1 is contained in respectively the living biomass and the litter layer. About 0.66 Tg C of atmospheric carbon is stored annually (by means of stem volume increment). About 50% of the annual storage is harvest each year. This implies that the Dutch forests act as a sink with a strength of approximately 0.33 Tg C y-1. The nett accumulation for the whole forest area amounts at present about 1 Mg C ha-1 y-1. The current sink acting of the Dutch forest can most likely be explained by the fact that the forests are young and still in building phase. However, this sink is not always as strong as reported here. The latest forest inventory reported an average annual volume increment of 7.8 m3 ha-1 y-1. The net storage rate as reported here, decreases correspondingly. The presented results therefore, depend very much on year tot year variation in growth of forest caused by climatic variability. The net annual sequestration probably varies in between 0.2 and 0.4 Tg C y-1. According to the investigators long rotation with species as oak, beech, and Douglas-fir are most suitable for long-term storage.
2.4 F u t u r e r e s e a r c h The uncertainties in the estimates of the so-called fertilization flux is still very high. Another problem is the uptake and loss of carbon by changes in land use. The m e a s u r e m e n t s of NRP I have indicated t h a t a substantial amount of CO2 (equivalent to 3000 kg C ha-1 y-i) is emitted by peaty soils induced by lowering of the water table. It is to be expected that the reverse process, uptake of carbon by peat formation will take place if the water level in peaty meadows is at lower depth than present. The plans to restore the former conditions in many areas in The Netherlands of very high water tables have clear consequences regarding CO2 emissions. Emissions due to peat oxidation will be stopped and an enhanced uptake of CO2 will take place not only due to the fertilization effect but also due to peat formation. This situation indicates that is very important to improve our knowledge regarding the exchange of CO2 between the atmosphere and ecosystems which are able to sequester large amount of carbon in their soils. Intensified research on the exchange of CO2 between grass lands and the atmosphere, including fertilization effect, peat formation and peat destruction has a high priority in this respect.
479
3.
M E T H A N E (CH4)
3.1 Preparation studies and organization The programming of the CH4 cluster has been based on three preparatory studies. One study presented estimates of the CH4 emissions and their uncertainty ranges for The Netherlands based on literature (Born et al., 1991). The other two studies (Leffelaar et al., 1991 and Diederen, 1992) discuss priorities and criteria for the CH4 research in The Netherlands.
Inventory of Dutch CH4 emissions A first a t t e m p t to quantify Dutch CH4 emissions and to estimate ranges of uncertainties was made by Van den Born et al. (1991). The major conclusions of this work are p r e s e n t e d in Table 3.1. This inventory indicated t h a t enteric fermentation, landfills, the oil and gas industry, and organic soils cover about 90% of the total national m e t h a n e emissions. On a global scale these sources are relatively less important, covering about 47% of the global m e t h a n e emissions. Sources like rice paddies, biomass burning and coal mining are i m p o r t a n t on a global scale, but are absent or minor sources in the Netherlands. The large uncertainties in emission estimates are large, both on a national and global scale.
Criteria and priorities Starting point of the research program on methane within the NRP was to achieve a significant reduction of uncertainties in knowledge on important emission sources (Leffelaar et al., 1991 and Diederen, 1992). Important criteria developed for the planning of the programme were the relative importance of the sources, the ranges of uncertainty and the availability of specific expertise on emission sources. The source s t r e n g t h of m e t h a n e from enteric fermentation in The N e t h e r l a n d s is relatively well known. It was therefore decided not to plan any research activities on this item, although it is the largest source on a national scale. The uncertainty of this source in developing countries, however, is large, but the specific expertise on the differing diet situation and the impact on physiology was too poor to plan a research project. Rice paddies are of no importance on a national scale, but contribute significantly on a global scale. As the appropriate expertise on this subject was available, it was decided to formulate a research project on this topic. Furthermore, it was decided to plan research on the emissions from landfills, the oil and gas industry, and organic soils. Also, a n u m b e r of research projects were formulated to validate local emission m e a s u r e m e n t s , and to extrapolate the information to a larger spatial scale.
480
Table 3.1 Relative contribution (%) of national and global sources tot CH4 emissions in The Netherlands in 1989/1990 and the world (Van den Bom et al., 1991) Source Animals - enteric fermentation Landfills Oil & gas industry/distribution Wetlands/organic soils Ocean/coastal waters Freshwater Animal waste Waste Water t r e a t m e n t Rice paddies Termites Biomass burning Coal mining Other Emission range (weight units)
Netherlands
Globe
40 27 16 7 4 2 2 0.3 NE NE 2
13 7 8 19 2 1 6 4 18 7 9 6 1
710-1230 Gg CH4 y-1
310-990 Tg CH4 y-1
NE: Not Estimated, not zero; -: Not applicable
Organization of the CH4 cluster The overview of the coherence between the research projects within the cluster is presented in Figure 3.1. Three relative important national sources were studied: the oil and gas industry, the landfills and the organic grassland soils. The organic grassland-soil related projects aimed at understanding the processes of m e t h a n e formation and consumption in the organic soils. The results of experiments and m e a s u r e m e n t s are integrated in a model of the methane flux from the soil to the atmosphere. The following projects are part of the CH4 cluster: B. Biogenic sources
BRP. BMF. BMC. BGM. BMMF.
Soil parameters controlling methane production and emission from rice paddies (LUW; project no. 850009) Methane formation by anaerobic consortia in organic grassland soil (LUW; project no. 853120) Methane consumption by indigenous grassland microflora. (LUW; project no. 853122) Effects of grassland m a n a g e m e n t on the emission of m e t h a n e from grassland on peat soils (LUW; project no. 853121) Modelling methane fluxes from and to grass covered peat soils (LUW; project no. 853123)
481 A. Anthropogenic sources ALl.
AL2. AOG1. AOG2.
Greenhouse gases from landfills in The Netherlands. (TNO-ME; project no. 850023) Landfill gas formation, emission and recovery in The N e t h e r l a n d s (TNO-ME; project no. 853105) Q u a n t i f i c a t i o n of CH4 emissions due to n a t u r a l gas losses and petroleum production (TNO-ME; project no. 850008) Quantification of methane emissions in the exploration and production of natural gas and petroleum in The Netherlands. (TNO-ME; project no. 853104)
V. Evaluation and validation
VCI. VEV. VUA.
Validation of source strengths of atmospheric CH4 using carbon isotope ratios (ECN; project no. 852097)) Evaluation and validation of the CH4 emissions in The Netherlands and contributions from various sources ( TNO-MW; project no. 853124) Methane emission of the Amsterdam urban area (LUW; project no. 853125)
VALIDATION
OIL & GAS INDUSTRY
LANDFILLS
ORGANIC SOILS INTEGRATION FLUX MODELLING F O R M A T I O N
C O N S U M P T I O N
M A N A G E M E N T
Figure 3.1 Schematic overview of the research of the projects within the m e t h a n e cluster of the NRP in The Netherlands 3.2 M e t h o d s Biogenic sources
Rice paddies (BRP). In this project the impact of various soil related parameters on the CH4 emission from wetland rice fields was studied. Methane fluxes from wetland rice fields in the Philippines were monitored with a closed chamber
482 technique as described by Schfitz et al. (1989) during two wet seasons (1991 and 1992) and one dry season (1992). The effects of soil-sulphate, soil-salinity, and organic m a n u r e on CH4 emission were studied in experiments where gypsum, salt and green m a n u r e were added respectively. The effect of a calcareous soil was studied by a comparison with a non-calcareous soil. Methane oxidation in the rhizosphere was studied using a specific inhibitor of m e t h a n e oxidising bacteria. The research was done in close co-operation with a project of the I n t e r n a t i o n a l Rice Research Institute (IRRI) in the Philippines which aims at collecting base-line CH4 emission data from Asian rice fields.
The integrated CH4 grassland projects. The main aims of the CH4 research projects on grassland are the understanding and quantification of m e t h a n e formation and consumption in grassland on peat soils, and of the net fluxes of m e t h a n e between soil and atmosphere by experiments and simulation modelling. Four different scales are distinguished, i.e. (i) micro organisms in pure culture studies; (ii) batch experiments with homogenised soils; (iii) intact soil columns; (iv) field scale. The modelling aims to inter-relate the data obtained from the different scales. Grasslands cover more t h a n 35% of the total surface area in the Netherlands, of which 32% is on peat soils. The study sites are located in the major peat area of the w e s t e r n p a r t of the N e t h e r l a n d s , around Zegveld (52~ 4~ The p r e d o m i n a n t l y eutrophic peat originates from sedges, reeds and wood, and generally have a clayey top-layer. Maximum peat depth is about 6 m. The organic m a t t e r content ranges from about 40% in the top 10 cm to about 90% below a depth of 60 cm, generally. Soil pH ranges between 3.5 and 5.0. A lowered m e a n ground w a t e r level, fertilizer application and removal of the grass crop via grazing and mowing are the major measures t h a t take place on intensively m a n a g e d grassland. On extensively managed grassland the vegetation is cut once a year in summer. The studied sites include both intensively managed, drained grassland and extensively managed natural grasslands. On intensively managed grassland, located at Zegveld, two typical sites have been chosen, i.e. site '8B' with a m e a n ground w a t e r level of 30 cm and site 'Bos 6' with a mean ground water level of 60 cm. Next to the effect of ground w a t e r level, effects of fertilizer application and grazing versus mowing on net exchanges of CH4 between peat soil and atmosphere are investigated. On extensively managed grassland, three typical sites have been chosen in the Nieuwkoopse Plassen area with mean ground w a t e r levels of 5, 10 and 15 cm. Data on ground water level, soil and air temperatures, soil water filled pore spaces, soil nitrate contents and net CH4 fluxes have been monitored on a weekly basis from September 1993 onwards. M e a s u r e m e n t s will continue till about August 1995.
Methane formation grassland soils (BMF). Soil profiles were t a k e n from the two Zegveld grassland sites, with water tables of 30 cm and 60 cm below surface. The soil profiles were sectioned, t a k e n to the laboratory in sealed plastic bags, and stored at 4~ Inside an anaerobic glove box soil samples (20 g wet weight) from each section were transferred to 300 ml serum bottles, containing 40 ml of anoxic
483 distilled water. The stoppered bottles were incubated under a N2 atmosphere (50 kPa overpressure) at 15~ in the dark. The initial pH of the suspended soils ranged from approximately 4.8 to 5.5. At certain time intervals samples were taken from the head space as well as the liquid phase and analyzed for gases (CH4, CO2, H2) and fatty acids or alcohols, respectively.
Methane consumption grassland soils (BMC). Soil samples from different depths (0-5, 5-10, 10-20, 30-40 cm) were taken from Zegveld. To investigate the kinetics of methane oxidation of these different depths the soil was placed in bath cultures in 300 ml flasks with gas tight septa and incubated with 1, 10, 100 and 10,000 ppmv methane, respectively, in artificial air with 1% (v/v) CO2 (Bender et al., 1992). For the enrichment of methanotrophic bacteria with different affinities for methane, soil (100 g) was incubated in a system receiving a continuous gas-flow of 4 ml/min containing methane at 4 different concentrations (1, 10, 100 and 10,000 ppmv).
Grassland management (BGM). Net CH4 emissions from grassland on peat soils in The Netherlands have been monitored with vented closed flux chamber (Hutchinson et al., 1981) from September 1993 onwards. Monitoring will continue in 1994 and 1995. At Zegveld, intensively managed grassland on peat soil with a mean ground water level of 30 cm and intensively managed grassland on peat soil with a mean ground water level of 60 cm have been investigated. Also, on both Zegveld sites the effects of nitrogen fertilization and grazing versus mowing on net CH4 emissions have been investigated. Finally, CH4 fluxes from three extensively managed grasslands at Nieuwkoop have been measured as well. Modelling methane fluxes grassland soils (BMMF). This project started in September 1993 and will last for 4 years. It aims at developing a process model for methane fluxes to and from organic grassland soils. Water dynamics at Zegveld will be obtained to be used as input for a gas transport model for Zegveld. With this gas transport model oxygen dynamics and methane transport will be described. The oxygen dynamics will be used as input for the methane production model. Results of the integrated model of methane production (results from project BMF), consumption (results from project BMC) and transport will be compared with field fluxes measured in project BGM. This comparison in combination with a sensitivity analysis of the model will show which aspects need most attention in further research. The model for methane production has already been developed. It calculates the dynamics of biomass, acetate, and methane formation. Moreover, it describes experimental data from the BMF project quite well. In this model, besides oxygen, a time lag is incorporated before methane production can start. This lag period could, in a later stage, be specified in terms of the presence of electron acceptors like nitrate and sulphate. If we succeed in incorporating all major processes in a realistic way, it will be possible to test two hypothesis: in-situ methane emission is low, because production is limited by the short duration of the anaerobic periods during wet periods; during dry periods, methane uptake by the soil is controlled by m e t h a n e t r a n s p o r t from the atmosphere to the methanotrophs.
484 Anthropogenic sources
Landfills - emission measurements (ALl). This preparatory study compared two m e a s u r e m e n t methods at three landfills from J u n e until S e p t e m b e r 1991 (Verschut et al., 1991) a dynamic closed-chamber method Balfour et al., 1987, Reinhart et al., 1992) measuring concentration differences between air entering and leaving a closed chamber system (10 m2; 8-10 replicates per landfill; during >_24h) and a micrometeorological method (Fowler et al., 1989) measuring concentration gradients and wind velocity along a pole of 10 m height situated for about two weeks at the centre of a landfill, combined with vertical and horizontal flux calculations and wind speed profiles. Landfills- gas formation, emission and recovery (AL2). To improve the reliability of the emission quantification from landfills the methane material balance (Emission = Formation - Oxidation - Recovery; no accumulation assumed) is investigated. Landfill gas formation is determined in two ways: by emission m e a s u r e m e n t s at various landfill sites (micrometeorological method Oonk et al., 1995) and adding the amount of landfill gas recovered and oxidized to the amount emitted; by evaluating recovery efficiency from the results of recovery projects in relation with landfill geometry, composition of the top liner system and the lay-out of the recovery system - landfill gas formation is subsequently obtained as the product of the a m o u n t of landfill gas recovered and the recovery efficiency. Oxidation data used are from literature (UK-DoE, 1993, US-EPA, 1990, Oonk, 1993). The formation of landfill gas is subsequently modelled, by correlating the formation to waste composition, age and amount of waste landfilled. Natural gas losses and petroleum production (AOG1). A first estimate of CH4 emissions due to the production and treatment, the high pressure transport, the distribution and consumption of natural gas and the production and transportation of petroleum in The Netherlands was made by an engineering approach (Nielen, 1991). This estimate was based on available information of m e a s u r e m e n t s , emission factors and production and consumption data for 1989. Exploration and production of natural gas and petroleum (AOG2). Three types of m e t h a n e emissions related to oil and gas production were examined. The continuous emissions due to leakages of systems used and from off-gases of various gas t r e a t m e n t installations have been quantified by an engineering approach using: emission factors, material and energy balances; knowledge on maintenance and testing procedures; information from NOGEPA on amounts of associated gas produced; information on process equipment, compressors, turbines etc. The irregular emissions due to periodic tests and maintenance of installations were quantified partly as a result of an engineering approach as described above, partly by a combination of measurements and dispersion modelling. The third type of m e t h a n e emissions, the incidental emissions due to failures of devices also are estimated by the combination of m e a s u r e m e n t s and dispersion modelling. The a m b i e n t CH4 concentration at Kollumerwaard, located N o r t h - w e s t of the Groningen gas field, which has been registered permanently since July 1991, was screened on indications of events elevating the background concentration. This was done in combination with information on meteorology, potential source location and distribution calculations by the Plume model.
485 Valuation and validation
Validation by carbon isotope ratios (VCI). At two sites a t m o s p h e r i c CH4 concentration has been monitored continuously (Eisma et al., 1995 and Kieskamp et al., 1995): the 200 m meteorological tower at Cabauw and the 'Vuurtoren' island near Amsterdam. The latter to examine the emissions of the Amsterdam area (see VUA). Carbon isotopic analysis in atmospheric CH4 t a k e n at Cabauw has been performed as well. However, due to interference from near-by 14CH4 emissions of Pressurised Water Reactors (PWR) 14C could not be used as a tracer for fossil and biogenic CH4 in Europe. Instead, from the 14CH4 record at Cabauw, emission factors from PWRs have been determined. A laboratory intercomparison was organized for ambient CH4 m e a s u r e m e n t s exercise, with ECN, TNO, KEMA and LUW as participants. Results obtained for Cabauw are interpreted by analysis of air mass trajectories, meteorological d a t a and the application of a L a g r a n g i a n t r a n s p o r t model in c o m b i n a t i o n w i t h CH4 emission i n v e n t o r i e s and c o m p a r i s o n of isotopic m e a s u r e m e n t s with characteristic isotopic values of CH4 sources. Also, the CH4 concentration records m e a s u r e d simultaneously at Cabauw and D u r g e r d a m are compared. As a first approximation a GIS application has been developed in which the excess CH4 concentration at Cabauw (> 1.75 ppmv) is related to the area over which the air mass was t r a n s p o r t e d (using 36 h backwards air trajectories at Cabauw and assuming a constant, 1000 m mixing layer height). The total area covered by the air trajectory was calculated as a reverse plume. The CH4 emission flux required for explanation of the observed excess CH4 was s u b s e q u e n t l y assigned to all points in a 0.1 ~ x 0.1 ~ grid over Europe. Methane emission from w a t e r surfaces was a s s u m e d to be zero w i t h exception of the oil-and gas-production sites on the North Sea. Evaluation and validation from various sources (VEV). This project validates present knowledge of CH4 emissions by comparing measured concentrations with t h o s e c a l c u l a t e d by a t m o s p h e r i c dispersion models from e m i s s i o n a n d m e t e o r o l o g i c a l data. C o n t i n u o u s m e a s u r i n g is p e r f o r m e d by u s i n g gas c h r o m a t o g r a p h y on two monitoring sites, namely, Arnhem and K o l l u m e r w a a r d from 1990 and 1991 onwards, respectively. Dispersion modelling has not s t a r t e d yet. Urban area (VUA). The u r b a n m e t h a n e emissions have been quantified by modelling methane air concentrations and comparison of calculated and measured imission concentrations. M e a s u r e d concentrations were obtained from E C N (project VCI, E i s m a et al., 1995) from the m e t h a n e m o n i t o r i n g station at Vuurtoreneiland, about 2 km east of Amsterdam. Methane emissions estimates from the urban area were based on literature data Amstel et al., 1993 Veldt et al., 1993) road traffic, the natural gas distribution network, and industrial sources. The D a n i s h OML model was selected for calculation of immission concentrations (Lofstom et al., 1992). This Gaussian plume model has a preprocessor to calculate dispersion height, atmospheric stability and turbulent mixing from synoptic and balloon meteorological measurements. The synoptic m e a s u r e m e n t s were obtained from Schiphol Airport, the balloon data from De Bilt. Emission data and dispersion data were fed to the OML model to calculate immission concentrations for the
486 continuous methane monitoring station. The calculated emissions were compared with data collected at the monitoring site. To obtain an estimate of the increase in ambient methane concentrations due to the urban plume, methane concentrations of Cabauw (some 40 km south of Amsterdam) (data obtained from ECN) were subtracted from the Vuurtoreneiland data. 3.3 R e s u l t s Biogenic sources
Rice paddies (BRP) Soil-sulphate. The methane emission from plots amended with 6.66x103 kg ha-1 gypsum (CaSO4) was reduced by 55-70% compared to non-amended plots (Figure 3.2). The reduced CH4 emission upon gypsum application was most likely due to inhibition of methanogenesis by sulphate-reducing bacteria. Observed SO42concentrations in the soil solution of gypsum-amended plots were well above minimum concentrations reported in the literature for successful competition of sulphate-reducing bacteria with methanogens. The data indicate t h a t CH4 emissions from rice grown on high-sulphate containing soils or gypsum-amended soils is low compared to non or low-sulphate containing soils. However, fertilization of rice fields with (NH4)2SO4 will not necessarily result in lower CH4 emissions because the amounts of sulphate added are relatively low. Soil-salinity. Rice is often grown on saline soils. To investigate whether the presence of salinity results in lower CH4 emissions NaC1 salt was added to a rice field. Pore water EC increased to about 4 dS m-l, which caused a reduction by 25% only in CH4 emission. It was shown that the CH4 production in the salt-amended field was strongly reduced compared to the control field (Table 3.2). However, CH4 oxidation in the salt-amended plot was even more inhibited than CH4 production. This resulted in about equal net CH4 fluxes from both salt-amended plots and non-amended plots. The data illustrate the importance of knowledge of both CH4 production and CH4 oxidation when estimating CH4 emission and show that a reduction in CH4 production does not necessarily lead to reduced CH4 emissions.
487 Methane emission (mg.m-2.day -1 ) 5,000
/~
J G. Manure
G. Manure + Gypsum
4,000
3,000
2,000
1,000
0
2
3
4
5
6
7
8
Days after transplanting Methane emission (mg.m-2day-1) 1,200 1,000 800 600 400 200
0
I 20
i 40
i 60
I 80
100
Days after transplanting Figure 3.2 The impact of Gypsum application on methane emission from wet rice fields during the wet season in 1992 at The Philippines Calcareous soils. CH4 emissions from rice grown on a calcareous soil were higher t h a n from a non-calcareous soil. The seasonal p a t t e r n of CH4 emission differed, with a more pronounced emission peak early in the season, probably due to the favourable pH for CH4 production in the calcareous soil. The difference in emission between the two soil types was no longer observed when the fields were fertilized with green manure, indicating that the "soil"-factor may be overruled by the input of organic matter. Application of organic manure. Application of green m a n u r e s t i m u l a t e d CH4 emissions. CH4 emission was highest during the first half of the growing season in plots t h a t received more t h a n 11x103 kg ha-1 of green m a n u r e . Ebullition contributes significantly to total CH4 transport, if rice fields receive high inputs of organic matter. The impact of organic m a n u r e on CH4 emissions, at different locations of the world, can be described by a dose-response curve if CH4 emission from the organically amended plots is expressed relative to CH4 emission from mineral fertilizer t r e a t m e n t s . Such an approach may prove to be useful when e s t i m a t i n g CH4 emissions from larger regions if information on the amounts of organic m a n u r e used in the region becomes available.
488 Plant-mediated gas transport. Plant mediated CH4 transport was shown to be described by diffusion only. The results combined with data from the literature suggest that the rate limiting step in plant-mediated methane transport is diffusion of CH4 across the root/shoot junction. CH4_ oxidation in the rice rhizosphere. CH4 oxidation in the rice rhizosphere was studied using methyl fluoride, a specific inhibitor of methane oxidising bacteria. CH4 oxidation in the rice rhizosphere depended on the growth stage of the rice plant and becomes much less important when the rice plant reaches the ripening stage. Therefore seasonal patterns of CH4 emission in rice fields do not only depend on changes in CH4 production but also on changes in CH4 oxidation. These findings indicate that methanotrophs do not oxidise a constant percentage of the CH4 produced throughout the growing season. Table 3.2 Average methane flux from triplicate soil cores of rice fields with and without salt amendment, during anaerobic and aerobic incubation, and percentage CH4 oxidized Sampling date a) Flux)
CH4 flux (nmol cm-2 h-l) anaerobic no salt salt
76 DAT 11.02 2,75 96 DAT 24.53 7.04 110 DATb) 20.82 7.42 a) b)
CH4 oxidized (nmol cm-2 h-l)
CH4 oxidized (% of anaerobic
aerobic no salt
salt
no salt
salt
no salt
1.31 3.29 2.75
1.32 2.17 2.99
9.71 21.24 18.07
1.43 4.87 4.43
88 88 87
salt 52 70 60
DAT = days after transplanting 110 DAT =1 week after harvest
Methane formation grassland soils (BMF) Formation of CH4__Methane formation was observed almost exclusively in the upper 10 cm of the soil, with the upper 5 cm of the soil being most active. Below 10 cm, methane formation decreased drastically. The compact peat layers, below 30 cm did not show any methane formation (Figure 3.3). In the upper two soil layers (0-5 cm, 5-10 cm) formation of CH4 was exponential indicating that there was no substrate limitation. The difference observed for both layers therefore probably results from a difference in the number of methanogens originally present. A doubling time of approximately 2-3 days could be determined. In the top soil methane formation started immediately. The highest rate of CH4 formation was reached after approximately 40 days for the low water table soil type (0.41 mmol 1-1; 0.042 mg CH4 g-1 dry soil d-l). Ultimately, CH4 formation more or less paralleled CO2 formation.
489
0
0 1: Eb Acetate ~
10
20
Concentration 30
..~E]I
(mmol]l) 40
I
~
,~ CH
,
........
"~"
9 ~-~ ................................................;~-.:.:.: ...............................................
-d
-g
-6O
" ..... ~ . .............................................................................................................. ,
x::
:
121 -80
-I~ - - ~ ..................................................................................................................
G
- [ I.}x 1 ................. ................................................................................................... g
,
9[ ~ , ,
-[
~ .....................................................................................................................
Figure 3.3 The relationship between formation rates of CH4, C O 2 , acetate and soil depth of various soil slurries after incubation under a N2 atmosphere. All concentrations expressed in mmol 1-1. Data obtained for the low water table series (60 cm) Formation of CO2-- A l s o , the formation of CO was highest in the top soil (0 - 10 cm). The initial rate of CO2 formation amounted up to 1.3 mmol.l-1 d-1 (0.36 mg CO2 g-1 dry soil d-l). CO2 formation showed no lag-phase; the rate of CO2 formation decreased in time. After 40 days a small increase was observed again, due to an increased activity of aceticlastic methanogens that produce CO2 and CH4 from acetate. Formation of acetate. Analysis of the soil suspensions for fatty acids showed that the two upper layers (0-5 cm, 5-10 cm) produce considerable amounts of acetate. The top layer produced up to 4.8 mM acetate, which was rapidly degraded by methanogens after 35 days. Small amounts of propionate and butyrate were formed as well.
Methane consumption grassland soils (BMC) Methane oxidation. All 4 applied concentrations were biologically degraded by this type of grassland soil. The highest oxidative activities, especially for lower concentrations (1-100 ppmv), were observed between 5 and 20 cm. One reason for the lower activity in the highest depth (0-5 cm) may be that this region was very wet at the time of sampling, resulting in hampered gas diffusion. The time course of methane degradation is plotted in Figure 3.4 for the initial concentrations of both
490 10,000 (A) and 1 (B) ppmv methane. A correlation between CH4-concentration and degradation rates was observed (0.19 nmol and 1.9 ~tmol g-1 dry soil d-1 for 1 and 10,000 ppmv, respectively) and, most important, it is demonstrated that this soil acts as a sink for methane even at concentrations well below 1 ppmv.
0 ~"
1.00
E Q.
0.90
vo .
0.80
._
0.70
24 ' ~ f
48
72
96
120
r
9
r
9
144
168
0.60 c:
0.50
c'-
0.40
0
0.30 0.20 O.lO
E
~
,"~Y'~'~l~l-
.
9
o.oo 0
24
48
72
96
time -e-
0-5 cm
-l-
120
144
168
(h)
5-10 cm - T -
10-20 c
---
30-40 c
Figure 3.4 Degradation rate of methane in samples from organic grassland soil from different depths in batch cultures at initial concentrations of 10,000 ppmv (A) an 1 ppmv (B) CH4
Identification of methanotrophic bacteria. A decrease of the effiux concentration was observed after 14 days of incubation in the columns incubated with 10,000 ppmv. Here the degradation rate (caused by microbial growth) increased to 19 ~tmol g-1 dry soil d-1 within 7 days. In the column incubated with 100 ppmv the increase was observed after 35 days of incubation. At lower concentrations (1, 10 ppmv) the efflux concentrations remained constant for 40 days, but then these methane concentrations were also degraded. The main goal of this part of work will be to isolate and identify the strains of methanotrophic bacteria which are responsible for the degradation of atmospheric methane concentrations.
Grassland management (BGM) The effect of ground water table. The results presented here are based on observations until July 1994. Grassland with a high ground water level and a relatively thin aerobic layer is expected to show more CH4 emission and/or less immission than grassland with a low ground water level and a relatively thick aerobic layer. However, the site with the relatively high ground water level showed equal or only slightly higher net CH4 emissions than the site with the relatively low
491 ground water level during the measuring period. Net CH4 fluxes were low in the period October 1993 - July 1994, in general less than 0.1 mg CH4 m-2 d-1. The effect of N-fertilization and mowing/grazing. Nitrogen fertilization could decrease the consumption of CH4. Mowing or grazing could effect CH4 emissions by influencing the amount of organic material that is added to the soil annually. However, there were no clear differences between the t r e a t m e n t s during the measuring period. The effect of m a n a g e m e n t intensity. Differences between the different sites were quite large (Figure. 3.5) as were the spatial variations at each of the sites. The site with the lowest CH4 emission had a somewhat lower ground water level t h a n the other sites. In the period J a n u a r y - June 1994, CH4 emission ranged from 0 to 185 mg CH4 m-2 d-1. CH4 fluxes were much higher at the Nieuwkoop area t h a n at Zegveld.
Modelling methane fluxes grassland soils (BMMF) (See before) Anthropogenic sources
Landfills- emission measurements (ALl) D y n a m i c closed-chamber m e a s u r e m e n t s . The spatial variation in m e t h a n e emissions proved to be large on each landfill: differences over a factor of 1000 were m e a s u r e d (e.g. 0.09 g m-2 d-1 to 225 g m-2 d-1 for one landfill and 7.2 g m-2 d-1 to 3150 g m-2 d-1 for another). It is clear t h a t emissions are s p a t i a l l y very inhomogeneous and t h a t local coincidental factors determine the emission rate through the top-layer. So, the m e a s u r e m e n t data are not representative for the total landfill area. Micrometeorological measurements. Depending on the type and age of the landfill, the amount of waste available, the height of the waste tip, the efficiency of landfill gas recovery and the composition of the top layer emissions were m e a s u r e d ranging from 2 to 150 g m-2 d-1. The method as such yielded quite reliable and uniform emission data. Decay rate. A waste decay rate constant (k) of k = 0.1 could be calculated on the basis of these emission measurements, combined with data on amounts of waste, landfill composition and age. This implicates a half-life of 7 years.
492
Nieuwkoopse Plassen, 1994 mg CH4 per m2 per day 200
150
100
50 /r,
Jan
Febr ,,n
Koole
March -
DBZ
April ~
May
June
Brampjesgat
Figure 3.5 Time course of mean CH4 fluxes (mg CH4 m-2 d-l) at three different sites in the Nieuwkoop area
L a n d f i l l s - g a s f o r m a t i o n , e m i s s i o n a n d r e c o v e r y (AL2)
Landfill gas formation. A first - very preliminary - result of the modelling of landfill gas formation by correlating the formation to waste composition, age and amount landfiUed is: {2t =~ 1,87 Q Co kl e-kit where, Dutch (0.094 waste;
Co = amount of degradable organic carbon in the waste in kg ton-1 (the m e a n value for Co is 112 kg ton-l; kl = rate constant of biodegradation y-l); Q = amount of waste landfilled in ton; t = time after dumping of the a t = formation of landfill gas in m3 y-1 (with a mean methane content of 57
vol%); ~ = 0.58 (formation factor).
Landfill gas recovery. In 1993 about 124 million m3 of landfill gas was extracted; 85 million m3 was utilized. In 1992 m e t h a n e emission reduction was 57 Gg (Adviescentrum Stortgas, 1994). During the exploitation period of the landfill, landfill gas formation increases with increasing amounts of waste in place. After closure of the waste tip, landfill gas formation gradually declines. Landfill gas recovery normally starts when the landfill is closed. The effectiveness of recovery is increased when a top-liner system is applied. Normally this is done 5 years after closure of the landfill. The environmental impact of landfill gas recovery is closely connected to its integral recovery efficiency, being the ratio of formation and recovery throughout the years. High integral efficiencies can only be obtained,
493 when landfill gas recovery starts during the exploitation period of the landfill. The technology for doing this is available, and proves to be very cheap, when landfill gas recovery is reckoned in the design phase of the landfill (J. Oonk 1993 and 1994). Emission estimates. Landfill gas emissions have been estimated using the material balance as described in 3.2.2. The CH4 emission estimates from landfills in The Netherlands range from 400-500 Gg y-1. Uncertainties in these estimates are due to uncertainties in the amounts of waste landfilled, amounts of m e t h a n e formed per ton of waste and amounts of methane oxidized in the top-soil of the landfill.
N a t u r a l gas losses and petroleum production (AOG1) The results of this study are summarized in Table 3.3. Total CH4 emissions have been estimated between 127 and 220 Gg y-1. At this point it m u s t be emphasised t h a t none of these estimates have been based on actual measurements, which are required for more accurate quantification.
Exploration and production of natural gas and petroleum (AOG2) The project has not been finalized yet. Preliminary results of the engineering study are partly in a qualitative form still. Table 3.4 presents the sources and source strengths of methane in the oil and natural gas exploration and production. Table 3.3 Methane emissions from natural gas losses and petroleum production in The Netherlands in 1989 (Gg y-l) Sector Natural gas Production/treatment High pressure transport Distribution Consumption Petroleum Production Total
Methane emission
40-70 6.5 65-79 15-30 1-35 127-220
494 T a b l e 3.4 Sources of m e t h a n e emission in the oil a n d n a t u r a l gas exploration a n d production
Sources
Strenghtl
Emission during exploration - drilling - well t e s t s
minor moderate/major
Emission during exploitation of natural gas continuous - vents - flares - e x h a u s t g a s e s of t u r b i n e s - e x h a u s t gases of reciproking engines - e x h a u s t g a s e s of f u r n a c e s - chronic leaks in production - chronic leaks in g a t h e r i n g a n d t r a n s p o r t - glycol d e h y d r a t i o n - t r e a t m e n t of f o r m a t i o n w a t e r - u s e of p n e u m a t i c devices - condensate treatment - condensate storage - p u r g e gas d r o m v e n t i n g s y s t e m s
major moderate minor moderate minor major minor major moderate moderate/major moderate minor moderate
non-continuous
- m a i n t e n a n c e in production - m a i n t e n a n c e of g a t h e r i n g a n d t r a n s p o r t pipelines - incidents a n d accidents in production - incidents a n d accidents pipelines
minor minor moderate minor
Emission due to exploration of oil continuous
- flaring of associated gas - e x h a u s t gases of reciproking engines - t r e a t m e n t of production w a t e r
minor minor minor
non-continuous
- n o n - e x h a u s t engine emissions
minor
Abandoned phase - chronic leaks from a b a n d o n e d wells
none
1minor: ...~.~:~-~ ;
5 CO 2
+
7 H20 + 2 N 2 + energy.
The reduction of nitrate (NO 3- ) in heterotrophic denitrification occurs stepwise; N20 is one of the intermediate compounds that can be either further reduced to N 2 or transported to the soil surface and eventually the atmosphere: N O 3 - - - > N O 2 - - - > NO ~ >
N20 m > N2"
The extent to which N20 is further reduced principally depends on the aeration of the soil, the amount of nitrate present and the residence time of N20 in a reduced soil portion. The process-based denitrification model accounting for these phenomena of [6] will be used as a starting point for model development. It essentially describes the dynamics of several (nitrogen) compounds as a result of the biological processes performed by two groups of heterotrophic bacteria: one group of strict aerobes, using only oxygen as electron acceptor, and one group of denitrifying bacteria that use either oxygen or, under anaerobic conditions, nitrate, nitrite and nitrous oxide as electron acceptors. Mineralisation and immobilisation are described in a very simplified manner. 2.1.2. N i t r i f i c a t i o n Nitrous oxide production in relatively dry soils, i.e. at moisture contents below field capacity, is generally attributed to nitrification [7]. The overall reaction of nitrification is given as:
621 NH4 + + 2 02 - - > NO3- + 2 H + + H20 + energy. In our opinion, however, N20 production attributed to nitrification should also be classified as N20 production in denitrification, since it is essentially due to denitrification of intermediate products in nitrification. Nitrification is performed by autotrophs as well as by heterotrophs, of which autotrophs are the most important. In fact, nitrification takes place in two steps. First ammonium (NH4 +) is oxidized to nitrite (NO2-) (ammonium oxidation). Next nitrite is oxidised to nitrate (nitrite oxidation). Nitrous oxide production from nitrification is ascribed to two processes [5]: 1. nitrifier denitrification: ammonium oxidisers use NO 2- as an alternative electron acceptor when 02 is (locally, temporarily) limiting and produce N20 (in fact this process is pure denitrification which occurs in nitrifying organisms), 2. a type of chemodenitrification: chemical decomposition of intermediates between NH4 + and NO2-, or NO 2- itself, to N20. We are not aware of any existing process-based model of nitrifier denitrification. However, this process could be described analogously to the description of denitrification by [6]. The second process can probably be described by elementary chemical reaction kinetics as soon as the reactions involved are identified.
2 . 2 . Transport of nitrous oxide within the soil and at the soil-atmosphere interface Several mechanisms of gas transport in soils have been distinguished [8]. Ordinary diffusion is the most important in the continuous gas phase of an unsaturated soil [9]. We assume that our system can be described as a one-dimensional system, basically governed by the reaction diffusion equation (Fick's Second Law plus production/consumption): iOc= 0 {Deff0__cc} + S , Ot ~)z iOz where:
(1)
c = nitrous oxide concentration, t = time, z = depth (z increasing with depth, z=0 at the surface), D etf = (modelled) effective diffusion coefficient, S = volumetric nitrous oxide production or consumption strength.
A more refined description of gas transport in the soil can be found in [ 10]. The equation describing nitrous transport at the soil-atmosphere interface is a special case of equation 1.
3. C A L I B R A T I O N AND V A L I D A T I O N P R O C E D U R E The model that will be developed has to be calibrated and validated. For this purpose, a twoyear experiment is performed in the Wageningen Rhizolab [2]. Regularly, nitrous oxide fluxes and belowground profiles of the soil moisture content and several gases and nutrients were determined on 4 grassland plots on a sandy soil. A portion of the results will be used for calibration. Results for the other dates will be used for validation. For validation, probably also results from other experiments, like the experiments of the Nutrient Management Institute (NMI; Velthof et al., this volume), will be used.
622 4. P O T E N T I A L REDUCTION STRATEGIES Experiments suggest that nitrous oxide emission from grassland soils could be reduced by applying suitable management practices like split application of nitrate fertilisers during dry periods [11, 12]. The simulation model could be very helpful to explore the possibilities to reduce emissions by measures like this.
5. ACKNOWLEDGEMENTS We gratefully acknowledge the support of J E Hofman, M H van den Bergh, A M van Dam, the staff of the Wageningen Rhizolab, the Nutrient Management Institute (NMI) and the members of discussion group 4 of the C.T. de Wit Graduate School Production Ecology during the research and preparation of this paper. It is part of the integrated N20 grassland research project in which also participate the NMI, and the Research Institute for Agrobiology and Soil Fertility (AB-DLO), Wageningen and Haren, The Netherlands. This Project is financially supported by the Dutch National Research Program on Global Air Pollution and Climate Change (Project 852074).
6. REFERENCES
1 A.F. Bouwman, In A. F. Bouwman (ed.), Soils and the Greenhouse Effect, John Wiley, Chichester (1990) 61. 2 C.A. Langeveld, P.A. Leffelaar and J.Goudriaan, Submitted to the Proceedings of the 8th Nitrogen Workshop, Ghent, Belgium, 5-8 September 1994. 3 S.C. Van de Geijn, J. Vos, J. Groenwold, J. Goudriaan and P.A. Leffelaar, Plant and Soil, 161 (1994) 275. 4 E.A. Davidson, In J.E. Rogers and W.B. Whitman (eds.), Microbial production and consumption of greenhouse gases: methane, nitrogen oxides, and halomethanes, American Society for Microbiology, Washington, D.C. ( 1991) 219 5 T. Granli and O.C. Beckman, Norwegian J. Agric. Sci., 12(Supplement) (1994), 1. 6 P.A. Leffelaar and W. Wessel, Soil Sci., 146 (1988) 335. 7 D.W. Bergstrom, M. Tenuta and E.G. Beauchamp 1994, Biology and Fertility of Soils 18 (1994) 1. 8 E.A. Mason and A.P. Malinauskas, Gas transport in porous media: The dusty-gas model, Elsevier, Amsterdam, 1983. 9 D.B. Jaynes and A.S. Rogowski, Soil Sci. Soc. Am. J. 47 (1983) 425. 10 P.A. Leffelaar, Soil Sci. 143(1987)79-91. 11 O. Van Cleemput, A. Vermoesen, C.J. De Groot and K. Van Ryckeghem, In J. Van Ham, L.J.H.M. Janssen and R.J.Swart (eds.), Non-CO2 greenhouse gases. Why and how to control?, Kluwer Academic Publishers, Dordrecht (1994) 145. 12 C.J. De Groot, A. Vermoesen and O. Van Cleemput, In J. Van Ham, L.J.H.M. Janssen and R.J.Swart (eds.), Non-CO2 greenhouse gases. Why and how to control?, Kluwer Academic Publishers, Dordrecht (1994), 183.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
623
Measurements of the atmospheric emission of N20 from biogenic sources in general and by grassland ecosystems in particular Jan Duyzer IMW TNO, P.O. Box 6011, 2600 JA Delft, The Netherlands Abstract
The project is part of the 'Integrated N 2 0 grassland project'. The project carried out at TNO aims to determine the atmospheric emissions of N20 from biogenic surface sources in the Netherlands. The following activities were part of the project: 9 determination of N20 emissions from grassland ~ comparison of methods to measure N 2 0 emissions in the field 9 determination of N20 emissions from arable land ~ determination of N20 emissions from fresh water systems and coastal waters 9 organisation of a round robin to compare measurements of N20 concentrations in air 1. INTRODUCTION
The objective of this study was to measure the N20 emission from relevant sources in the Netherlands with an emphasis on grassland. Other ecosystems studied include agricultural land, sea water an fresh water bodies. It was intended to measure these fluxes on an ecosystem scale ie. using the aerodynamic gradient method wherever possible. In other cases the dynamic enclosure technique was to be employed. This project was part of the so called 'Integrated N 2 0 grassland project' in which several Dutch groups (including RIVM, NMI, LUW and AB-DLO) participated. In the framework of this project a round robin was organised in which a comparison was made between the methods that are used in the programme to measure N 2 0 concentrations in air [ 1]. 2. THEORY AND METHODS
Dynamic enclosures designed by TNO [2] and NMI design static enclosures [3] and the aerodynamic gradient method were used. The dynamic enclosures cover a surface of 20 x 100 cm whereas the static enclosure cover a surface with a diameter of 20 cm.
624 In the gradient method the flux is derived from the concentration gradient of the trace gas in the air above the surface and measurements of the turbulence intensity. The method is described in detail in [4]. 3. RESULTS 3.1
Comparison studies
In two comparison campaigns the enclosure methods and aerodynamic methods were applied simultaneously. The aim of this experiment was specifically to investigate the spatial distribution of N 2 0 sources on a field scale. To this purpose 42 static enclosures were operated continuously by the NMI [3]. The enclosures were placed along a transect between the two dikes bordering the site at distances of one meter. Close to the dikes the observed fluxes were smaller, often around 0.5 mg N/m2/hr whereas in other areas some chambers showed fluxes as high as 10 mg/m2/hr. More than 80 % of the variation in N 2 0 fluxes could be attributed to variations in soil moisture and nitrate concentrations in the soil. The equipment used for the gradient method was located in such a way that the flux determination was expected to be representative for the area studied using the enclosures. Figure 1 shows the results of the experiment. The flux estimated using this method varied with time between 0 and 2.5 mg N/m2/hr. At night fluxes showed much less variation and varied between 0.5 and 1 mg N/m2/hr. During the day fluxes as high as 2.5 mg N/m2/hr were observed but the variation was much stronger. The dynamic enclosures showed fluxes with a median value of around 1 mg/m2/hr. During or after rain the direction of the flux indicated deposition, i.e. flux of N 2 0 towards the surface. These phenomena were also observed with the enclosures. 2.5
I
!
1.5
o E 0r z 0.5 z
~ "--4
0.9 -0.8 -0.7 -0.6
ra'nl
-0.5 r._
-0.4
!i
,i
o
-0.3 -0.2 -0.1 0
I1
"0.5
-1.5
1800
()
6()0
1:;:;00
1800
()
6()0
1200
1800
Time (hour)
Figure 1 Fluxes of N 2 0 observed during the inter comparison experiment over peat grassland at Zegveld in June 1993
625
In a second experiment in November the static enclosures and the gradient method were used. With the 42 static enclosures, this time located in a 6X7 meter grid showed fluxes varying between zero and 1.3 with an average of 0.25 mg N/m2/hr. With the gradient method hourly averaged fluxes varying between 0.15 and 0.7 mg/m2/hr were observed. The results of these experiments are not easy to generalize. The results of the first phase study were confirmed showing that static chamber results can show large variation probably linked with the scale of variation of relevant soil parameters and processes. It seems justified to conclude that on a day scale the fluxes of N20 measured using the static chambers lie within a factor of 2 to 3 from fluxes measured with the gradient method. 3.2
Continuous measurements of fluxes from grassland
On the peat grassland site at Zegveld N20 fluxes were determined every day from April 1993 to June 1994. To this purpose samples were taken every afternoon. Figure 2 shows the monthly averaged emissions calculated from all observations from April 1993 to June 1994 expressed in kg N/ha/year. The average emission is roughly 10 kg N/ha over this period with peaks in the summer. In general two peaks are found midsummer and November 1993. The December period was quite wet. The annual fertilizer input on the site was roughly 100 kg N/ha.
2~I
T
150
c"
..r
I
100-
50-
g ,,...
-~
o-
-50
-I00
-
-
.
Figure 2 3.4
.
.
.
.
.
g
g
g
~;
g
~;
Median emission of N 2 0 observed over peat grassland at Zegveld.
Emissions of N20 from coastal waters and freshwater systems.
A campaign was carried out along the Dutch Waddensea. To estimate the flux the concentration of N20 was determined at two heights 0.3 and 4 m above the sea surface. The N20 concentration in the sea was high compared to open sea values ie. around 1000 ng/1. The observed emission from the sea was equivalent to 2 kg N20 N/ha/yr. If this emission rate would be observed all year the Dutch Waddensea would emit roughly 500 ton N20 N per year.
626 In April and May several campaigns were carried out over the freshwater lake Ketelmeer. This lake was chosen because of its high nitrate concentrations. In one case significant gradients were detected with relatively high fluxes equivalent to 3 kg N/ha/yr. In those cases high N20 concentrations of more than 2000 ng/1 were observed in the water phase. Further water samples were taken from sweet as well as salt water bodies in the Netherlands. The results indicate a quite strong relation between nitrate and ammonium content on one hand and N 2 0 concentration on the other. This result is shown in Figure 3 which shows the N 2 0 concentration observed as a function of the ammonium and nitrate content for several salt and flesh water systems in the Netherlands. 9
.-.
fresh
1600
1600
1400
1400
1200
~.
salt
1200
c-
vc
0 c~
o
1000
O
z
1000
z
800
800
600
600
O9
0
0
Figure 3 3.5
2
4
6 8 1'0 1'2 1'4 NO 3 (mg/I)
6
0
0'.1
0'.2 NH 4 (rng/I)
013
0.4
The concentration of N 2 0 as a function of Nitrate and ammonium content
Emissions from arable land
Measurements of the emission rates of N20 from arable land were started in spring 1994. Two sites were chosen. One set of experiments was carried out at the agricultural experimental station at Nagele in the Noordoostpolder on a sea clay soil. The other experiments were 9 . carried out over a sandy soil in Vredepeel. Measurements were carried out over potatoes and onions over the clay soil and over peas, potatoes and grassland on a sandy soil. Measurements using two enclosures were carried out over bare soils and during crop growth. Around two weeks after fertilizer application a maximum in the emission rate is found. The emission from the onions upon which N fertilizer was applied was around a factor of three lower. 4. REFERENCES
1. Verhagen, H.L.M. and J.C.Th. Hollander (1994). TNO-MW report R 94/273. 2.
Baas, J. H.S.M.A. Diederen (1992). IMW-TNO report R 92/211. Delft, The Netherlands.
3.
Velthof, G.L., O.Oenema (1994). Draft report C 94.193 Nutrient Management Institute Wageningen, the Netherlands
4.
Duyzer, J.H. (1994). Draft scientific report. Emission of N20 from biogenic sources.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
627
Effects of nitrogen fertilization and grazing on the emission of nitrous oxide from grassland G.L. Velthof, A.B. Brader and O. Oenema NMI, Department of Soil Science and Plant Nutrition, Wageningen Agricultural University, P.O. Box 8005, 6700 EC, The Netherlands.
Abstract In the Netherlands, managed grasslands are potentially a large source of nitrous oxide (N20), because of the large nitrogen (N) input and the relatively high groundwater levels. To provide insight into the major factors that contribute to N20 emission from grassland and to provide quantitative N20 emission rates, a monitoring study was carried out on four sites, during March 1992 to March 1994. Fluxes of N2O increased after N fertilizer application and grazing, especially during wet conditions. Fluxes were higher from peat soils than from sand and clay soils. Fluxes were low during the winter periods. Total N20 losses were 2 to 4.5 times higher on grassland fertilized with 160-460 kg N ha 1 yr 1 than on unfertilized grassland. Losses from grazed grasslands were 1.5 to 3.5 times higher than losses from mown grassland. This study shows that management practice of grassland and soil type are major factors controlling N20 emission from grasslands.
1. I N T R O D U C T I O N
On a global scale, soils are a major source of nitrous oxide (N2O). In soils, N2O is produced during the microbial processes nitrification and denitrification, primarily controlled by the availability of mineral nitrogen (N), oxygen (02) and mineralizable carbon (C) in the soil [1]. In the Netherlands, intensively managed grasslands are possibly a large source of N20 because grasslands cover 30 % of the total surface area, the N input via fertilizer and animal excretions is high and many soils are relatively wet due to the shallow groundwater level. Furthermore, about 30% of the grasslands are situated on peat soils. Due to the high contents of organic N and C and due to the shallow groundwater levels, it is expected that especially grasslands on peat soils are a major source of N20. In the present study, the effects of N fertilization, grazing, and soil type on N20 emission from grasslands were investigated. The aim was to provide insight into the major factors that contribute to N20 emission from managed grassland and to provide quantitative N20 emission rates, obtained from field measurements.
628 2. M A T E R I A L S A N D M E T H O D S Fluxes of N20 were measured weekly in the period March 1992 to March 1994 on four contrasting grassland sites in the Netherlands, namely on a sand soil in Heino, a clay soil in Lelystad and two peat soils in Zegveld [2]. Peat soil I had a mean groundwater level of 35 cm below soil surface and peat soil II had a mean groundwater level of 50 cm below soil surface. Perennial ryegrass (Lolium perenne L.) was the dominant grass species in all swards. At each site, the experiments had a complete randomized block design, with three treatments in three replicates. The plots were 2.5 x 20 m. The treatments were mown grassland without N fertilizer applications, mown and N fertilized grassland and predominantly grazed and N fertilized grassland. Fertilizer N was applied as calcium ammonium nitrate (CAN), in six or seven dressings during the growing season. The economic optimum application rates of N fertilizer were assessed by using an interactive fertilization system based on a combination of modelling and measuring soil mineral N and N uptake. The application rates for the grazed grasslands were equal to those of the mown grasslands. Total N input by urine and dung of the grazing cattle was calculated using standard calculation procedures [3]. Fluxes of N20 were measured using vented closed flux chambers made of PVC cylinders with an internal diameter of 20 cm and height of 15 cm. Concentration of N20 in the headspace of the flux chambers was measured in the field at 0, 10, 20 and 30 minutes after closing, using a photo-acoustic spectroscopic infra-red gas analyzer, directly attached to the flux chambers. All fluxes were measured in six replicates. Mean N20 fluxes were calculated as arithmetic means and total N20 losses were calculated by integration of the mean N20 fluxes over time [2].
3. R E S U L T S A N D D I S C U S S I O N Fluxes of N20 increased after application of N fertilizer and grazing. The N20 flux pattern of fertilized grassland depicted in Figure 1 is typical for N20 fluxes from grasslands fertilized in several N dressings. This pattern is mainly due to fluctuations in mineral N content in the soil, caused by a combination of fertilizer N application, N uptake by the grass roots and microbial biomass and N losses by leaching, denitrification and volatilization. The magnitude of the N20 flux after N application was also dependent on soil moisture content, because during dry periods the effect of N application on N20 loss was much smaller than during wet periods. During the winter periods, fluxes were much lower than during the growing seasons, probably due to the low temperatures and low mineral N contents in the soil (not shown). Generally, both flux magnitude and duration were higher for the peat soils t h a n for the sand and clay soils and those from peat soil II were higher than those from peat soil I. The higher N20 fluxes from the peat soils than the sand and clay soils were likely due to higher organic C and N contents, higher denitrification potentials and higher groundwater levels of the peat soils (data not shown).
629 N20 flux, kg N ha-1 day -1 0.6
l lLll i
1 1[ 1
0.4
-'-'Unfertilized + mown - V - ' N fertilized + rnown ~'N
fertilized + g r a z e d
0.2-
0.0-
M'A'M'J 'J ' A ' S ' O I N ' D ' J ~F ' M ' A ' M ' J ' J ~A'S'O'N'D'J 'FfM ~
1992
!
Month
1993
i 1994
Figure 1. Time course of N20 flux from grassland on peat soil I, for the three m a n a g e m e n t practices. Arrows indicate time of N application and grazing. Dotted arrows indicate grazing without N application. For all sites, both the order in total N input via fertilizer, dung and urine, and the order in total NzO losses was: unfertilized and mown < N fertilized and mown < N fertilized and grazed grasslands (Figures 2A and 2B). Total NzO losses were 2 to 4.5 times higher on N fertilized grassland t h a n on unfertilized grassland and 1.5 to 3.5 times higher on grazed grassland t h a n on mown grassland. Losses from the peat soils were higher t h a n from the sand and clay soils, for all treatments. Remarkably, N20 losses from peat II, the soil with lowest N input, were highest, indicating the large effect of soil type on NzO losses. On the sand soil, 1.0% of the fertilizer N applied on mown grassland and 1.5% of the urine and dung N deposited on grazed grassland was lost as N20, during the two year period. This was 0.9 and 3.3% on the clay soil, 1.9 and 2.3% on peat soil I and 3.9 and 9.8% on peat soil II, respectively. The higher grazing derived losses t h a n fertilizer derived losses, in terms of % of the N input, suggests t h a t the effect of grazing on N20 losses was not only an effect of the higher N input. Several mechanisms may have contributed to the high grazing derived N20 losses: - the much higher nitrate contents of grazed grasslands t h a n mown grasslands (data not shown) may have increased N20 emission more t h a n proportionally, because N20 becomes a more important end product of denitrification at increasing nitrate contents [1]; - compaction of the soil by treading of the grazing cattle decreases 02 diffusion into the soil and may enhance production of N20; - high ammonia concentrations in urine patchess may inhibit nitrification, which may enhance production of N20; - urine and dung contain mineralizable C, which may increase denitrification rate.
630 In conclusion, this study shows that both soil type and practice of grassland management are major factors controlling N20 losses from grasslands. The high grazing derived N20 losses indicate that, besides N fertilizer application, also grazing has to be considered in N20 budget studies.
N input, kg N ha -1 year -1
N20 loss, kg N ha -1 year -1
A
B
I
800
4O
600
30
400
2O
|
"m m !
2ooi o
Sand
Clay
Peat l
] Unfertilized + mown
Peat II
0 _ 7/ Sand
N fertilized + mown
~ / Clay
, Peat l
/c Peat II
N fertilized + grazed
Figure 2. Total annual N input via fertilizer, urine and dung (A), and total annual N20 losses (B), for all sites and treatments. Averages of two years. 4. A C K N O W L E D G E M E N T S
The authors thank ROC Heino, ROC Zegveld, Waiboerhoeve in Lelystad and the colleagues of the Research Station for Cattle, Sheep and Horse Husbandry in Lelystad for their support. This investigation was supported financially by the Dutch National Research Program on Global Air Pollution and Climate Change. 5. R E F E R E N C E S
1 E.A. Davidson, Fluxes of nitrous oxide and nitric oxide from terrrestial ecosystems, p. 219-235 In: J.E. Rogers and W.B. Whitman, Microbial production and consumption of greenhouse gases: methane, nitrogen oxides, and halomethanes, American Society for Microbiology, Washington D.C (1991). 2 G.L. Velthof and O. Oenema, Nitrous oxide fluxes from grassland in the Netherlands: II. Effects of soil type, nitrogen fertilizer application and grazing. Submitted to European Journal of Soil Science. 3 D.W. Bussink, Relationships between ammonia volatilization and nitrogen application rate, intake and excretion of herbage nitrogen by cattle on grazed swards, Fertilizer Research 38 (1994), 111-121.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
631
Modelling the emission of dinitrogen oxide from mown and from grazed grassland J. Bril, H.G. van Faassen and H. Klein Gunnewiek Institute for Agrobiological and Soil Fertility Research, P.O. Box 129, 9750 AC Haren, The Netherlands
Abstract The integrated model SONICG was developed to simulate the emission of dinitrogen oxide (N20) from grassland soil. The model comprises modules on soil physico-chemical conditions and processes, and on soil microbial carbon and nitrogen turnover and makes use of data from an existing model on grass development. Some typical model results are shown for the daily emission of N20 from mown and from grazed grassland, with special attention for effects of urine. 1. INTRODUCTION
Based on literature data on N20 emission, intensively managed grazed grasslands were expected to be a major biogenic source of N20 emission in The Netherlands. A high contribution of intensively managed grassland to N20 emission might be explained from the low efficiency of high nitrogen (N) inputs in dairy farming. Furthermore, high concentrations of mineral N in urine spots in grazed grassland form active centres for N loss by nitrification-denitrification. To get a better understanding of the complex relationships that result in N20 emission from grazed grassland the model SONICG -Simulation Of the Nitrogen Cycle in Grassland soil- was developed. 2. THE SIMULATION MODEL SONICG
SONICG mechanistically simulates the relevant soil physical, chemical and biological processes in grassland soil layerwise. The model considers 30 layers of 2.5 cm, each with its own properties. Above the soil a gaslayer of 2.5 cm is present in the model as a kind of gasbuffer between the soil and the atmosphere. Figure 1 shows schematically the nitrogen cycle processes that play a central role in the model SONICG: Nitrification and Denitrification, including the production and reduction of N20. A separate M.I.T. module simulates mineralizationimmobilization-turnover of organic matter. Inputs of organic matter in the form of dead grass litter and roots, as well the uptake of mineral N by the grass are derived from a separate mechanistic model on grass development (Verberne, 1992).
632
I INPUTDATA b HYDROLOGY MODULE
GRASS development model (Verberne) I Roots ..........I N~46~EC_!.... ~
~
NH3~NH~
11
.
.
.
.
.
.
'
.
.
.
.
I I __.~ 5 _._s~ water pha_se _
I-
M.I.T. MODULE 1 =Nitrification 2= Denitrification 3= Mineralization 4=Immobilization 5 = Plant uptake 6=Cation exchange 7=water/gas exchange
Urine Fertilizer Manure
Figure 1. Nitrogen transformations and exchanges modelled in SONICG The order of the different modules and processes in a timestep (one day) of the simulation is shown in Figure 2: input data are read from a hydrology module (daily weather data; soil data, including water-filled pore space fractions for each layer) as well as from a grass growth model (organic C and N inputs from and potential uptake of mineral N by the grass). Next the temperature profile of the soil is calculated. Thus the actual process rates of M.I.T., nitrification and denitrification can be calculated next, including temperature and moisture effects. At the field scale soil water-filled pore space is a major rate controlling factor for the production and reduction of N20 (Groffman, 1991). Figures 3 and 4 show how M.I.T., nitrification and denitrification depend on WFPS in the model. After the M.I.T. module the actual plant uptake of mineral N is calculated. Then nitrification rates-production of N20 and nitrate- are calculated, followed by denitrification rates -nitrate and N20 reduction. Next gas transport and transport of dissolved substances are calculated. As a final steps chemical equilibrium calculations are made for the following components: H20, H§ Ca2§ K§ NH4§ NO3-, N20, CO32, CI-, CEC (Cation Exchange Capacity) and inert gas (all gasses except the explicitly modelled gasses CO2, NH3, and N20). For each layer cation exchange, complexation in solution, precipitation/dissolution of calcite and the exchange of gasses between the soil water- and gasphase are considered. As the main parameter of interest here SONICG can calculate the daily emission of N20 from the soil surface, as well as that of CO2 and NH3. Escape of gasses with drainwater at the lower boundary of the soil is also calculated. For details on SONICG see the endreport of NOP-project 852078 (Van Faassen and Bril, 1994).
633
~
Re t t i v e 1.0
t
i np u t
READ
1
,,,._ ,...-
1,2
1
PLANT
,,.._
1, 3, 5, 6._~
1
pp.2- 8
~
M.I.T.
~
ID 6, 7, 9
UPTAKE
~
ID~ 9
0.4
7, 9
DEN I TR I F I CAT ION
~
ID~ 7, g
~
ID~ 9
~
lp~ 9
TRANSPORT
D I SS.
TRANSPORT
EQUI L I BR I A
.
,
.
0
~
GAS
0.2
o
N I TR I F I CAT I ON
1 1 1
0.8
0.6 3, 5, 9
TEMPERATURE
,,.._ y
1~2,3
1 1 1
~
Rate
,2
.
.
.
0,4
.
.
,
0,6
Water-Filled
0.8
Pore
,0
Space
Figure 3. Dependance of organic matter decay and denitrification rates on soil water-filled pore space.
b Re
at
i ve
rate
1.0
OUTPUT t
=
t
+
1
F
0.8
0.6
0.4
012
o
I
0
I
012
I
9
i
o,,q Water-
F i I Ied
I
0.6 Pore
9
I
I
0.8 Space
Figure 2. Flow chart of SONICG
Figure 4. Dependance of nitrification, nitrate and N20-production rates on soil water-filled pore space.
I
1.0
634
3. SIMULATED EMISSION OF N20 FROM MOWN AND GRAZED GRASSLAND
As an example of SONICG, mown grassland was compared with grazed grassland, fertilized with 480 and 360 kg of N per ha per year, respectively. To simulate grazed grassland, at least two situations have to be considered: urine spots and area unaffected by urine. Grazed grassland gets less fertilizer N than mown grassland, but in urine spots an excess of mineral N will be present over a long period. Urine spots were simulated to get additional N from urine equivalent to 420 kg per ha. To get an overall picture of grazed grassland, simulation results of urine spots and of area without urine have to be added on an areal basis. Large differences are found in the simulated emission of N20 from mown grassland and from urine spots in grazed grassland. As shown in Figure 5, several months passed between the main N20 emission and the deposition of urine.
N 2 0 flux in kg N ha -1 day -1 0.70.60.5
with urine ~ t b . .
0.4 Re
0.3 urine 420 kg N/ha 0.2
0.1 _ 0
J~L '
0
'
I 91
'
'
I 182
'
'
I 273
'
'
I 364
Day of the year
Figure 5. Simulated daily emission of N20 from mown grassland and from urine spots in grazed grassland. Urine was deposited on day 140. 4. REFERENCES
1 E. Verberne, 1992. Simulation of the nitrogen and water balance in a system of grassland and soil. Nota 258. IB-DLO, Haren. 2 P.M. Groffman, 1991. Ecology of nitrification and denitrification in soil at scales relevant to atmospheric chemistry. In: J.E. Rogers & W.B. Whitman, eds., Microbial production and consumption of greenhouse gases, Am. Soc. Microbiol., p.201-218. 3 H.G. van Faassen & J. Bril, 1994. Modelling N20 emission from grazed grassland. Endreport of NOP-project 852078. AB-DLO, Haren.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
635
Emission of greenhouse gases from wastewater treatment processes J.G. Bruins, H.D. Oostergo & M.A. Brinkhorst BKH Consulting Engineers, P.O. Box 5094, 2600 GB Delft, The Netherlands
Abstract
Biological wastewater treatment processes are a source of emission of the greenhouse gases CO2, CH4 and N20 into the atmosphere. Studies were carried out to quantify the emissions of these gases from the municipal wastewater treatment plants in The Netherlands, which in 1987 handle a total waste load of 17,049,000 population equivalents. On the basis of detailed calculations for the different treatment methods and assumptions for the formation of N20 the following emissions were calculated: CO2 - 880 million kg/yr (0.5% of the total CO2-emission in The Netherlands), CH412.5 million kg/yr (1% of the total CH4-emission) and N20 - 829,000 kg N per year (0.9% of the total N20-emission). Measurements of N20-emissions from 2 low load activated sludge plants were carried out. First results indicated that the N20 formation and emission amount to 0.1% of the total N-load with the influent to the wastewater treatment plants. 1. INTRODUCTION Biodegradation processes in soil and water are an important source of emission of greenhouse gases into the atmosphere. Municipal wastewater treatment plants incorporate various types of processes, which involve the biodegradation of organic matter and biological reduction and oxidation of nitrogen compounds. In general terms a biological wastewater treatment system involves the following processes, causing emission of greenhouse gases: - aeration for bio-oxidation of organics and of nitrogen compounds (nitrification), which results in emission of CO2 and possibly N20; CH4 and NH 3 may also be emitted in these processes when process conditions are not optimal; anaerobic digestion of wastewater treatment sludge with production of biogas, mainly consisting of CH4 and CO2; the biogas is mostly used as an energy source in wastewater treatment plants, which positively effects the energy balance and hence the CO2-emission from the wastewater treatment plant; reduction of nitrate (denitrification) with formation of N2 and N20; disposal of wastewater treatment sludge, e.g. by use in agriculture, dumping at solid waste disposal sites or incineration, with the possible formation of CO2, CH4 and N20;
636 - biodegradation
(aerobic and anaerobic) of residuals (organic and nitrogen compounds) discharged with treatment plant effluent into surface water, with formation of COs, CH4, N2, NH3 and N20; - consumption of external energy resources, in particular for aeration, causing the emission of COs. In the period 1990-1994 several studies were carried out with the objective to quantify the emissions of greenhouse gases from the municipal wastewater treatment plants in The Netherlands. The results of these studies are summarized here. 2. E M I S S I O N OF G R E E N H O U S E GASES FROM THE NETHERLANDS
WASTEWATER TREATMENT PLANTS In 1987 a number of 491 municipal wastewater treatment plants was in operation in The Netherlands treating a total organic waste load of 886,210 tonnes COD (equal to 17,049,000 population equivalents). In 1987 the most applied methods for treatment of municipal wastewaters were respectively: Plant type
Organic waste load (population equivalents)
Trickling filters Aeration tanks (high load activated sludge) Oxidation tanks (low load) Carrousel (low load) Other Total
1,259,000 6,111,000 833,000 3,537,000 4,457,000 17,049,000
For each type of wastewater treatment systems the emissions of greenhouse gases were calculated on the basis of the overall treatment efficiencies of the different systems and the consumption of external energy resources. The total calculated CO2-emission from the wastewater treatment plants in The Netherlands equals about 880 million kg CO2 per year (about 0.5% of the total estimated COs-emission in The Netherlands). The specific emission equals 1.2 kg CO2 per kg COD removed. These emissions originates from the following sources: Source of COs
Aerobic biodegradation Use of external energy sources Use of biogas Wastage of biogas Biodegradation of organic residues after discharge Sludge disposal
Percentage
48 27 8 1.5 8.5 7
C H 4 in wastewater treatment processes is generated by anaerobic sludge digestion and anaerobic decomposition of sludge disposed of to solid waste dumping sites. The total calculated CH4-emission from wastewater treatment processes amounts to about 12.5 million kg C H 4 per year (1987), equivalent to about 1% of the total CH4-emission in
637 The Netherlands. The specific emission equals 17 g CH4 per kg COD removed. Most of the digestion gas, generated in the Netherlands wastewater treatment plants, is utilized as a source of energy, but some CH 4 is wasted or flared off. The CH4-emission into the air originates for 20% from wastage of digestion gas and for 80% from anaerobic decomposition of sludge after disposal. N20 in wastewater treatment processes is produced in nitrification and denitrification processes. Research results indicate that N20 is formed in the nitrification process as a result of non-optimal process conditions, and that in denitrification always formation of N20 takes place in conjunction with formation of N 2. The proportion between the quantities N 2 and N20 depends on the process conditions and the presence of particular micro-organisms. Literature data on N20emissions from wastewater treatment processes show large differences, varying from 0.01 to 6% of the total N-load to the wastewater treatment plant, being converted into N20. For calculation of the N20-emission from the municipal wastewater treatment plants in The Netherlands the assumption was made that by nitrification 0.3% of the N-load in the influent minus the N-load in the effluent is converted into N20 and that 0.3% of the nitrate-N, formed by nitrification, is converted into N20. On the basis of these assumptions the total NzO-emission from the Netherlands wastewater treatment plants would amount to 330 t N per year. On the basis that 1% of the residual N, discharged with the effluent, is converted into N20 , it was calculated that the N20emission from surface waters as a result of effluent discharge amounts to 415 t N per year. On the assumption that 1% of the N in sludge, disposed of to agriculture and solid waste disposal sites, it was calculated that the N20-emission from sludge disposal amounts to 84 t N per year. The estimated N20-emission from wastewater treatment processes equals about 0.9% of the total estimated N20-emission in The Netherlands.
3. MEASUREMENT OF N20-EMISSION FROM WASTEWATER TREATMENT PROCESSES In 1994 indicative measurements regarding the production and emission of N20 by two municipal wastewater treatment plants in The Netherlands were carried out. The photo-acoustic method for NzO-analysis was applied. The measurements were carried out at the wastewater treatment plant in Capelle aan de IJssel, a carrousel plant with a covered aeration circuit andsimultaneous nitrification and denitrification, and at the wastewater treatment plant in Alblasserdam, a Schreiber type plant with separated nitrification- and denitrification compartments. Samples of the air above the nitrification and denitrification compartments were taken according to a standardized method. The analyses were carried out over a period of about six hours, during which air samples were analyzed at intervals of six minutes. On the basis of the analysis results it was calculated that at the plant in Capelle aan de IJssel 0.006% of the Nload in the influent was emitted as N20 and 0.07% at the Alblasserdam plant.
638 4. R E D U C T I O N OF THE E M I S S I O N OF G R E E N H O U S E GASES F R O M WASTEWATER TREATMENT P R O C E S S E S
Measures for reduction of the emission of greenhouse gases from wastewater treatment processes should be considered in view of future effluent discharge quality standards, which gradually become stricter especially for the concentrations of N- and P-compounds. In The Netherlands the new limit values for discharge of effluent into surface waters are 10 mg N-total/1 and 1 mg P-total/1. Such values can only be realized in highly efficient wastewater treatment plants with special process steps for nitrification/denitrification and P-removal (chemical or biological). For this purpose usually ultra low-load activated sludge systems, with aerobic sludge mineralization, are applied, since these systems are the most suitable for nitrification/denitrification and incorporation of biological P-removal processes. However these low load systems are also characterized by the highest specific CO2-emission (kg CO2/kg COD removed), due to the high consumption of external energy. Hence optimization of these wastewater treatment processes should be considered in order to reduce the use of external energy resources. Measures to this effect may include: - incorporation of less energy consuming oxidation methods, e.g. trickling filters or rotating biological contactors in the wastewater treatment process; - application of anaerobic sludge digestion, with efficient use of the digestion gas, where possible. Very little is known about the exact mechanisms for formation of NzO in biological wastewater treatment processes and there are not many representative data on measurements of the N20-emission from wastewater treatment processes. Research data indicate that the emission of NzO can be reduced by creating optimum process conditions for nitrification and denitrification. 5. R E F E R E N C E S
1 Emission of greenhouse gases from wastewater treatment plants (in Dutch) Ministry of Housing, Spatial Planning and Environment BKH Consulting Engineers November 1990 2 Study regarding the formation of N20 in wastewater treatment plants (in Dutch) National Research Programme on Global Air Pollution and Climate Change (N.O.P) BKH Consulting Engineers June 1994
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
N20 E M I S S I O N S
FROM COMBUSTION
639
PROCESSES
H.Spoelstra KEMA Nederland
B.V.,
P.O.Box
9035,
6800
ET Arnhem,
Netherlands
Abstract
N20 e m i s s i o n s from the D u t c h p o w e r plants, c h e m i c a l industry, oil r e f i n e r y and a w a s t e i n c i n e r a t i o n p l a n t w e r e d e t e r m i n e d . D u r i n g s a m p l i n g s p e c i a l p r e c a u t i o n s were t a k e n in o r d e r to a v o i d the well k n o w n artifacts. N20 e m i s s i o n s at the level of a m b i e n t air c o n c e n t r a t i o n s of 0.3 ppm(v) w e r e d e t e r m i n e d at p o w e r p l a n t s f i r e d w i t h coal, oil and gas, r e f i n e r y f u r n a c e s fired w i t h a v a r i e t y of d i f f e r e n t fuels and at c h e m i c a l industry f u r n a c e s fired w i t h a v a r i e t y of h i g h c a l o r i f i c fuels. Low N20 c o n c e n t r a t i o n s w e r e d e t e r m i n e d in flue g a s e s from g a s t u r b i nes w i t h o u t a d d i t i o n a l f i r i n g in a s u b s e q u e n t b o i l e r and also at a w a s t e i n c i n e r a t i o n plant.
I.
INTRODUCTION
In 1987 EPA e s t i m a t e d that the N20 c o n c e n t r a t i o n in flue g a s e s from coal c o m b u s t i o n a m o u n t e d 20-25% of the NOx c o n c e n tration. For gas a v a l u e of 3-7% was e s t i m a t e d [i]. In 1988 it became c l e a r that t h e s e high v a l u e s w e r e due to s a m p l i n g artifacts [2]. For r e l i a b l e e s t i m a t i o n of the N20 e m i s s i o n s from p o w e r p l a n t s a m e t h o d was d e v e l o p e d in w h i c h t h e s e samp l i n g a r t i f a c t s w e r e avoided. T h e n N20 c o n c e n t r a t i o n s in the flue gases of d i f f e r e n t p o w e r p l a n t s were m e a s u r e d . Secondly the N20 e m i s s i o n s of s e v e r a l o t h e r c o m b u s t i o n p r o c e s s e s w e r e determined. This last part of the w o r k was c a r r i e d out and f i n a n c e d w i t h i n the f r a m e w o r k of the N a t i o n a l R e s e a r c h Prog r a m m a on G l o b a l Air P o l l u t i o n and C l i m a t e Change.
2.
DEVELOPMENT
OF
AN
ARTIFACT
FREE
SAMPLING
METHOD
A m e t h o d was c h o s e n in w h i c h g r a b s a m p l e s w e r e t a k e n from flue gases w h i c h w e r e a n a l y s e d on a g a s c h r o m a t o g r a p h a f t e r some time. D u r i n g a series of e x t e n s i v e tests c a r r i e d out w i t h flue gases from a p i l o t p l a n t gas b u r n e r in w h i c h a d d i t i o n a l g a s e s c o u l d be a d d e d an o p t i m a l m e t h o d was achieved. The SO2 in the flue gases was r e m o v e d w i t h a s e r i e s of w a s h i n g b o t t l e s f i l l e d w i t h a H202 solution. Then the flue g a s e s w e r e d r i e d by means of a p e r m e a t i o n dryer and s t o r e d in g l a s s sampling bottles under overpressure (0.5 bar). By a n a l y z i n g the flue gas d i r e c t l y and a f t e r s e v e r a l p e r i o d s of s t o r a g e no i n c r e a s e
640
in N20 c o n c e n t r a t i o n s (up to one week) c o u l d be detected. This was also the case d u r i n g the e m i s s i o n m e a s u r e m e n t s in w h i c h the s a m p l e s w e r e a n a l y z e d as soon as p o s s i b l e (mostly w i t h i n 24 hours) and as a c h e c k also a f t e r 3-7 days. The tests s h o w e d also t h a t s t o r a g e in s t a i n l e s s steel c i l i n d e r s in some c a s e s give a rise in N20 c o n c e n t r a t i o n .
3. N20 E M I S S I O N S
FROM ELECTRICITY GENERATION
M e a s u r e m e n t s w e r e c a r r i e d out at a v a r i e t y of p o w e r p l a n t s f i r e d w i t h coal, gas and oil. A l s o the i n f l u e n c e of load, f i r i n g m e t h o d and flue gas c l e a n i n g e q u i p m e n t was i n v e s t i g a ted. 3.1. Coal f i r e d p o w e r p l a n t s M e a s u r e m e n t s w e r e c a r r i e d out at six p o w e r p l a n t s w i t h a c a p a c i t y in the range b e t w e e n 115 and 650 MW. N20 c o n c e n t r a t i o n s w e r e b e l o w 0.2 ppm for t h r e e b o i l e r s w i t h a c o n v e n t i o n a l f i r i n g m e t h o d at full load (90 - 96%). For t w o - s t a g e c o m b u s t i on at t h r e e o t h e r b o i l e r s (180 - 520 MW) N20 c o n c e n t r a t i o n s r a n g e d from b e l o w 0.2 ppm to 0.4 • 0.2 ppm in the load r a n g e b e t w e e n 45 - 95%. No e n h a n c e m e n t of N20 c o n c e n t r a t i o n s was found at two b o i l e r s (650 and 520 MW) e q u i p p e d w i t h a flue gas d e s u l p h u r i z a t i o n p l a n t ( c o n c e n t r a t i o n s 1E+8 m >IE+7 [] >1E+6
~
>lE+5 [E >lE+4 []
'
'
'~
'
i
. . . . .
,
'
'
'1
'
'
|
'
i
'
'
U |
'
'
'
............
yl
>1E+3 I~1 >IE+Z ffl >1E+1 [ ] >IE+9 V]
IE+8 >IE+7 >IE+6 >IE+5 >IE+4 >IE+3 >IE+Z >IE+I >IE+8 ~
o
mm m ~ [%] uop,0npaEI
o
:
,
,
K~ l
,i
,~
~ g
'
.s
t'~
-,--,
>
5 3 5 3
t'~
0
.,.,
>
t'~
.t)
~
.Q
>
0
"0
> < o
0
u_
u_
.," ,
2 at low and high temperatures. KNMI-2 explicitly keeps the daily surface air pressure P unchanged.
0
I
I I
9
I
i
-20
i
-10
i
I
i
0
I
10
i
i
20
i
30
T (~ Figure 2.1 Mean precipitation amounts at t e m p e r a t u r e T class intervals of 2~ for wet days (threshold 0.1 ram) at De Bilt (1906-1981). The figure is based on 15897 wet days (57% of the total n u m b e r of days). The n u m b e r of wet days in a t e m p e r a t u r e interval is 2104 for the T= 6~ class and decreases to about 10 at the extreme t e m p e r a t u r e s . The error bars indicate the estimated s t a n d a r d deviations of the means. The smooth curve represents the fitted regression relation (Equation 1)
846 2.4 C o n c l u s i o n s Precipitation scenarios with a time resolution of one day can be obtained from the empirical relation between observed precipitation amounts and temperature. The corresponding change in seasonal precipitation compares well with GCM-based scenarios, apart from the summer. The scenario can be refined by taking pressure into account as predictor, to account for systematic changes in circulation in case there are clear indications of these in GCM output. The scenarios obtained in these ways are meteorological consistent and provide a plausible description of extremes. Extension to other regions in Europe requires study of the local time series to find the geographical dependence in the results of Eq. 1.
3.
LAND USE SCENARIOS
B. P a r m e t R i j k s w a t e r s t a a t , I n s t i t u t e of I n l a n d W a t e r M a n a g e m e n t and W a s t e W a t e r T r e a t m e n t , RIZA P.O.Box 9072, 6800 ET Arnhem, The Netherlands Abstract Land use is an important p a r a m e t e r in hydrological and morphological processes. Climate change can induce changes in land use because the production and w a t e r use of crops is influenced. In the framework of a project of the I n t e r n a t i o n a l Commission for the Hydrology of the Rhine Basin, land use scenarios have been developed for the Rhine area. Besides climate change, autonomous developments were t a k e n into account, since these determine for a major p a r t the land use changes. A biophysical classification system has been designed and in combination with a crop simulation model geo-referenced information on land use potentials under present and possible future conditions is generated. The influence of climate change is mainly positive, the production increases. Autonomous developments were expressed in a Central Projection with a Plus and a Minus variant. In the Central Projection about one million hectare (10%) is vacated and comes available for other purposes t h a n agriculture or urban land. In the Minus v a r i a n t this is 3 million and in the Plus variant zero. Changed climate adds 0.2 million hectare to this, because less land is required due to the higher production. 3.1 I n t r o d u c t i o n Land use determines interception of precipitation, influences the ratio between i n f i l t r a t i o n a n d surface r u n o f f and d e t e r m i n e s to a l a r g e e x t e n t t h e evapotranspiration. It is therefore an important p a r a m e t e r in hydrological and morphological processes. An increased CO2-content and associated climate change might induce changes in land use, since growth and evapotranspiration of plants are influenced, see also Section 4 and 5. For n a t u r a l vegetation this could m e a n t h a t existing ecosystems move, alter in t h e i r species composition or even completely disappear. For agricultural crops the most i m p o r t a n t aspect is t h a t crop production may increase. Furthermore cropping patterns can change and new v a r i e t i e s can be introduced, t h a t cannot be grown u n d e r p r e s e n t climate
847 conditions. W h e t h e r changed climate conditions lead to changes in land use as described above, depends for a major part on economic, political, demographic and technical, socalled autonomous developments. As there are large uncertainties, both with respect to climate change and to autonomous developments, possible changes in land use have to be expressed in alternative scenarios. In the project 'Influence of climate change on the discharge of the river Rhine', that is coordinated by the International Commission for the Hydrology of the Rhine Basin (CHR), also the effects of land use changes are considered. Land use scenarios taking into account the effects of climate change in combination with autonomous developments were not available and have been developed as part of the CHR-project. The study has been carried out by the Winand Staring Centre at the request of The I n s t i t u t e of Inland Water M a n a g e m e n t and Waste W a t e r Treatment. In this chapter the methodology and the results of the study are presented. 3.2 M e t h o d To d e t e r m i n e the possible impacts of climate change on crop production a preliminary study was carried out (Wolf en van Diepen, 1991). The study showed t h a t the effects of a doubling of the CO2-concentration and an increase in temperature are mainly positive. Most crops grown in Western Europe are of the socalled C3 type, for which the CO2-concentration is sub-optimal. An increase in CO2 acts as a fertilizer and the assimilation rate increases. For socalled C4 crops, of which maize is the only important representative, the CO2-concentration is optimal and the increase in assimilation rate does not occur. An increase in t e m p e r a t u r e enhances the CO2 growth stimulation and increases production where temperature conditions are sub-optimal. Besides production, CO2 influences the water use efficiency. With higher CO2-concentrations, the s t o m a t a of crops have to be opened less to take up the same ammount of CO2. The water loss per s t o m a t a is less. For the overall water use of crops the increase in production counterbalances for a part the increase in water use efficiency, because the leaf surface increases.
An important conclusion of the preliminary study is that m o r e CO2, an increase in temperature and a small change in precipitation during the growing season, does not bring about limitations and even improves the circumstances for the cultivation of presently grown crops. Moreover possibilities for other crops arise. Climate change itself will however not directly generate changes in land use in the Rhine Basin. Although the changed climate boundary conditions will play a role, land use changes will be determined by autonomous developments. A farmer will only grow another crop if it is economically more profitable. It follows t h a t the autonomous developments are very important with respect to changes in land use. The study to land use scenarios for the Rhine basin was therefore divided in two parts. A biophysical and a socio-economic part. The target period is around the mid of next century when, according to the Business as Usual emission scenario of IPCC, the CO2-concentration has doubled. A best guess climate scenario for this period was derived from Kwadijk (1993). The scenario assumes an increase in t e m p e r a t u r e of 1.5~ in s u m m e r and 2~ in winter. Precipitation r e m a i n s unchanged during summer and increases with 10% during winter.
848 The biophysical part is aimed at assessing the effects of a doubling of the CO2concentration and a changed climate on crop production, crop w a t e r use and cropping calendar (Roetter, 1994; Roetter en van Diepen, 1994). The specific aim is to give geo-referenced information on land use potentials under present and possible future conditions. To cover the regional differences in climate and soil in the Rhine basin, a biophysical classification system has been developed. The changes in potential (optimal supply of water, nutrients and pesticides) and water limited yields (optimal use of nutrients and pesticides) and water use of agricultural crops have been investigated using a crop growth simulation model. Simulation results for present and possible future climate were combined into changes in land suitability and attainable yields in the Rhine Basin. The socio-economic part examines the influence of autonomous developments on land use and combines this with the results from the biophysical p a r t into scenarios or projections (Veeneklaas et al, 1994). A Central Projection describes the long-term tendency in land use and is based on secular historic trends, f u n d a m e n t a l scientific and technical principles and basic assumptions. Secular t r e n d s have been used to u n d e r p i n q u a n t i t a t i v e s t a t e m e n t s about f u t u r e developments. Scientific and technical restrictions refer mainly to a t t a i n a b l e agricultural production levels and land suitability and follow from the biophysical part. The basic assumptions are the most controversial. By referring to other studies on future developments they can be made plausible to a greater or lesser degree. In case of great u n c e r t a i n t y a Plus v a r i a n t and a Minus v a r i a n t is constructed. For the Plus variant maximum, and for the Minus variant m i n i m u m u r b a n and agricultural claims on land are assumed. The socio-economic part results in two types of land use projections. For unchanged and changed conditions a Central Projection is constructed with, if necessary a Plus and a Minus variant. 3.3 R e s u l t s
B i o p h y s i c a l p a r t ; changes in l a n d use p o t e n t i a l s A biophysical classification system containing the elements climate and soils and adapted for present and possible future conditions was not available for the Rhine basin and had to be developed. First a bioclimatic classification was designed, which was combined with a soil classification and integrated in a Geographical Information System (GIS) (Roetter, 1994). Climatic, agroclimatic and agroecological m a p s show t h a t a n n u a l m e a n t e m p e r a t u r e , precipitation and annual t e m p e r a t u r e amplitude are the m a i n factors to describe the regional differentiation of agricultural crops and n a t u r a l vegetation. The bioclimatic classification system was based on meteorological data for 53 stations and a digitized altitude map. Regression equations were derived between meteorological variables as dependent variables and combinations of altitude, longitude and latitude as independent variables. Based on the regression analysis and known classification systems the set-up of the bioclimatic system for the Rhine basin is based on: 1) annual mean temperature (seven classes); 2) annual mean temperature amplitude (four classes); 3) a n n u a l mean temperature of the coldest month (five classes); 4) annual mean precipitation (five classes).
849 The first three levels are based on regression equations and the fourth level is based on a digitized precipitation map. The equations have been derived both for p r e s e n t and possible future conditions. They are i m p l e m e n t e d in a GIS and consequently bioclimatic d a t a surfaces can be easily obtained. In total 700 combinations are possible, but only 90 occur at present in the Rhine Basin of which 25 have a surface area larger t h a n 1%. With the a s s u m e d scenario, the climate becomes more maritime/less continental and warmer. The soil suitability classification was based on a digitized soil map. Soil mapping units were clustered in four soil suitability groups based among others on slope class, soil texture, depth, moisture retention characteristics and soil genesis. Biophysical types were generated with the GIS by combining the bioclimatic types with soil suitability groups. This was done for present and possible f u t u r e conditions. It has been assumed that a change in climate does not affect the soil characteristics used in defining suitability groups. With the crop growth model WOFOST, potential and water-limited yields have been computed for seven major crops in the Rhine basin; w i n t e r wheat, silage maize, barley, oil seed, potato, sugar beet and rye grass, for present and possible future conditions (Roetter en van Diepen, 1994). Computations were carried out with meteorological data from 18 weather stations, representing the predominant base-line climatic types, and for two soil types representing the soil moisture and retention characteristics of the soil suitability groups. The crop characteristics were adapted for future conditions according to state of the art knowledge. In line with the preliminary study, the simulations with WOFOST showed that, in general, production increases. U n d e r water-limited situations, besides the CO2-fertilizer effect, the increased w a t e r use efficiency causes the production to increase. For the group of soils with an available w a t e r capacity of 70 mm, the average production for the Rhine area increased for winter wheat with 40%, of rye-grass with 33%, of sugar beets with 25% and of silage maize with 12%. It can be derived from the simulations t h a t soil and t e r r a i n characteristics in combination with a change in mean annual temperature are the main determining factors with respect to land suitability. Based on these criteria, five land suitability classes were defined: Very high, high, moderate, m a r g i n a l and unsuitable. If climate changes according to the described best guess scenario, the a r e a l percentages of land suitability classes change as described in table 3.1. The class "very high" increases from 1.3 to 38.6%. The percentages of the other classes decrease. The a s s u m e d climate change has a positive effect on the overall suitability of land for cultivation of current crops and tree species. S o c i o - e c o n o m i c p a r t ; l a n d use p r o j e c t i o n s Starting point for the land use projections is the present land use in the Rhine basin. The Rhine basin has been divided into 13 regions based on the NUTS-1 division of the European Union (EU). Land use was derived from statistics. Half of the total area of the Rhine basin is used for agriculture and about one third is covered with forest. The basin is densely populated with about 55 million people, consequently a relatively large share, 11%, is built-up land. The Central Projection is based on secular trends in the past, other surveys of the future and basic a s s u m p t i o n s including technical and scientific restrictions
850
(Veeneklaas et al., 1994). Looking at past secular trends in land use, it seems t h a t we enter a period of contraction of the agricultural area. This is founded on the ongoing productivity increases and s t a g n a t i n g demand following from the low expected population growth. Furthermore, there are m a n y parallels with other historic periods of contraction. A decline in agricultural area is also the outcome of other surveys of future land use. The rate of decline in these surveys depends on the scenario assumptions, for example free trade versus protected markets. Table 3.1 Areal percentages of land suitability classes for unchanged and changed climate conditions (Roetter en van Diepen, 1994) Land suitability class
Percentage of total area (%) Unchanged climate
Very high High Moderate Marginal Unsuitable
1.3 28.1 41.8 8.8 20.0
Change (%)
Changed climate 38.6 3.7 37.3 0.7 19.7
+ 37.3 - 24.4 - 4.5 - 8.1 - 0.3
Basic assumptions in the Central Projection for urban land use are that population growth is marginal, but the amount of urban land per i n h a b i t a n t will increase, although at a slower rate t h a n during the last 40 years. For agriculture it is a s s u m e d t h a t technical progress will go on and t h a t regional differences in ratio between actual and water limited production will level out. Around the mid of next century yield levels will have reached 90% of the water limited yield in all regions. The common m a r k e t of agricultural products within the EU will remain. Because food r e m a i n s a strategic good a completely free m a r k e t will not develop. Consequently, world trade in agricultural products will not expand dramatically and protection of own markets for food will not disappear. For agricultural production stricter environmental regulations are expected, which will however not prevent approaching the water-limited yields. F u r t h e r m o r e it is assumed t h a t in the long r u n there will be a tendency to grow crops in those parts of the Rhine Basin t h a t have the highest yields. A certain degree of diversification within the regions will however remain. To construct the projections a hierarchical scheme is applied. U r b a n land needs and n a t u r e claims as defined in national policy plans have the highest priority. Second in line are agricultural land requirements and the lowest priority is given to forest and other land use. This hierarchy is based on the price of land paid by the different categories. For agriculture a second hierarchical scheme is nested, based on the profitability and the required quality of land; Horticulture and p e r m a n e n t crops, root crops, cereals and, with the lowest priority, grassland and fodder crops. For the Rhine basin as a whole the changes are listed in table 3.2. The basic assumptions of the Central Projection result in an increase in urban land use. The
851 Plus v a r i a n t assumes an increased population growth and more u r b a n sprawl and results in a larger increase of u r b a n land use. In the Minus v a r i a n t it decreases because of a decrease in population and lower land claims per i n h a b i t a n t . N a t u r e conservation claimed by policy plans has the same position in the h i e r a r c h y as u r b a n land use. In the N e t h e r l a n d s explicit claims have been f o r m u l a t e d in the N a t u r e Policy Plan of about 10% of the agricultural area. The a r e a used for agriculture decreases in all projections. With changed climate conditions this decrease is even larger, because production levels increase a n d hence, less land is needed. The m a i n decrease is found for cereals. Outside the E U production costs are lower and f u r t h e r m o r e the physical production conditions of the R h i n e b a s i n w i t h i n t h e E U are not optimal for cereal production. The production will therefore partly shift to outside the Rhine Basin. Next in line are potatoes. For this crop strong competition is expected with E a s t e r n Europe. Only for beets a small increase in area is expected for the Central Projection and the Plus Variant, for u n c h a n g e d climate. This is mainly caused by an increase in the production of fodder beets, t h a t will be used in cattle feed in line with a development of more self-sufficiency in dairy farming. The changes in a r e a of u r b a n and a g r i c u l t u r a l land use can differ for the 13 d i s t i n g u i s h e d regions. For the region N e d e r l a n d - O o s t ( N e t h e r l a n d s - E a s t ) for example u r b a n land use increases with 37% in the Central Projection. If n a t u r e reserves are included the increase is 72%. Agriculture decreases w i t h 16% for unchanged and with 21% for changed climate conditions. In the Minus v a r i a n t the agricultural land use decreases with 35%. Besides grassland, the acreage of cereals and potato decreases.
Table 3.2 Changes in areas of u r b a n and agricultural land use for u n c h a n g e d and changed climate conditions, for three v a r i a n t s with respect to the basic a s s u m p t i o n s , for the decade 2040-2050, in million ha and percentages (Veeneklaas et al., 1994) Central Projection Land use
unchanged
Agriculture -
Urban
Urban+ agriculture
Plus v a r i a n t
Minus variant
changed unchanged
changed unchanged
changed
1.57
-
1.83
- 2.67
- 2.84
- 1.26
-
20%
-
24%
-
-
-
-
34%
37%
16%
1.52 20%
+ 0.68 32%
+ 0.68 + 32%
- 0.18 - 9%
- 0.18 - 9%
+ 1.39 +66%
+ 1.39 + 66%
-
-
1.15
-
-
-
0.13
-
-
12%
- 29%
+ 1%
-
-
0.89
9%
2.85
3.02
-31%
0.13 1%
In the C e n t r a l Projection about one million hectare would become available for other use, in the Minus v a r i a n t this is 3 million hectare and in the Plus v a r i a n t no s u b s t a n t i a l s u r p l u s w o u l d be available. C h a n g e d c l i m a t e conditions add
852 approximately 0.2 million hectare. In Germany and the French part of the Rhine basin large parts will be vacated, mainly the areas were presently cereals are grown. The vacated areas could be used for afforestation, especially if different functions like timber production, recreation and nature can be combined. Other plausible possibilities are nature reserves or mixed designation, like dispersed housing, hobby farming, etc. The production of industrial crops does not require large amounts of land and biofuel production is economically not viable. These are therefore less realistic options for the vacated land. 3.4 I m p l i c a t i o n s A doubling of the CO2-content and an increase in t e m p e r a t u r e seem to have a positive influence on crop production. The implications of a climate change as assumed in this study are however small for land use, compared to the influence of autonomous changes. In general, also without climate change, it may be expected that the area built-up land will increase but the agricultural area will decrease at a faster rate. This may offer possibilities for nature development and afforestation. Possible implications for morphological processes in the Rhine basin are briefly discussed in Sections 7 and 8 and for hydrological processes in 6.
It should be noted t h a t in this study only average changes in climate were considered. Changes in for example frost risk or extreme events such as hail storms have probably a larger influence on average yields and yield variability and consequently on land use. However, due to lack of information on changes in these phenomena, they were not taken into account. 4.
FORESTS
H.J.M. Lankreijer Department of Physical Geography, University of Groningen Kerklaan 30, 9651 NN Haren, The Netherlands Abstract The possible impact of an increase in CO2 on the hydrology of forests is evaluated using sensitivity analysis and a climate scenario on an one-dimensional model of forest hydrology. Water use of forests is affected by plant physiological and meteorological variables. Doubling of CO2 leads to a decrease of s t o m a t a l conductance, resulting in a decrease in transpiration of 10 to 30%. The evaporation of rainfall interception by the canopy is increased due to a higher leaf area index and higher temperatures. Total interception increases, but the ratio between interception and precipitation decreases. Simulating a small increase in forest canopy increases the evapotranspiration only weakly and the higher precipitation in the scenario is mainly passed on to drainage. Drought damage in summer should reduce, but winter discharge may strongly increase. 4.1 I n t r o d u c t i o n A change in the concentration of C 0 2 as well as a possible climate change will have direct and indirect effects on the water use of plants, including trees. The changing
853 concentration of ambient CO2 directly effects physiological processes in the plant. The indirect effect results from the change in meteorological variables. Forests are aerodynamically rough and are therefore strongly coupled to atmospheric conditions. As a result, changes in the atmosphere might affect forests stronger than other vegetation types. The aim of this study is to estimate the consequences of a climatic change associated with a doubling of the atmospheric CO2 concentration, for the water balance of forests. The results may be used in other studies in the subtheme Regional Hydrology. Given the direct effect of CO2 on plant physiology, the project is also part of subtheme Terrestrial Ecosystems. The water flow in forests can be divided into interception of rainfall, transpiration by the canopy and drainage to groundwater. Interception and transpiration depend on meteorological variables and characteristics of the canopy. Transpiration is regulated by the stomatal conductance. Because of the turbulent flow of air in the canopy the dependency of interception and transpiration on meteorological conditions is much stronger for forests than for low vegetation. Soil characteristics determine in general the availability of water and the rate of drainage. A simultaneous change in atmospheric CO2 concentration and in climate influences the forest ecosystem in a complex way. Photosynthesis, water use efficiency, growth, canopy structure, nutrient circulation, species composition and phenology are all affected by a climate change. The interrelated and partly unknown processes involved make an analysis of the effects difficult and the results uncertain. Also the different reactions per species makes it difficult to generalize results. Some species, like several coniferous trees, show no reaction of s t o m a t a l conductance to changed CO2 concentrations. In general, pl ant physiological studies show that an increase in CO2 results in higher growth rates, lower stomatal conductance and increased water use efficiency. To simulate the water use of a forest, a realistic model of stomatal conductance (Gs) is needed. However, the exact relation between stomatal regulation, plant physiological processes and environmental variables is not fully known. This has resulted in a variety of empirical models simulating Gs. In this study a well known empirical parameterization of Gs is applied. Given the available data and the existing uncertainties in stomatal behaviour this parameterization is believed to be adequate. However, it is expected that in the near future the stomatal regulation will be simulated more realistically. 4.2 M e t h o d
Model A one-dimensional model is developed to simulate the water balance of a forest on an one hour time scale. The model is based on the model used by Dolman (1988) to simulate the water balance of a coniferous forest and is devided into three main submodels. Transpiration is simulated using the Penman-Monteith equation. The Gash-Rutter (Gash,1979) approach is applied to simulate the interception of rainfall. The soil water balance is simulated on a daily time scale by a simple
854 bucket type model. The amount of water exceeding field capacity is considered as precipitation excess and drained. Actual s t o m a t a l conductance is calculated from solar radiation, a t m o s p h e r i c humidity, air t e m p e r a t u r e and soil w a t e r deficit using the regression equation according to Jarvis (1976) and Stewart (1988). Data Five data sets of different forests in Europe were available to calibrate the model. These have been analyzed for their potential use in this study. The data sets of the Thetford forest (1976) in England and Ede (1988/1989) in The N e t h e r l a n d s are used. The calibration of the coniferous forest in Thetford is described by S t e w a r t (1988), and the calibration of the deciduous forest in Ede is derived from Hendriks et al. (1990) and Ogink-Hendriks (1994). The datasets of Ede did not include winter measurements. As water use during winter is limited due to low t e m p e r a t u r e s and low irradiation, this restriction of data is permissible to calibrate the model. The w a t e r balance is simulated over 5 years using the KNMI-data set of'De Bilt'. This d a t a set covering 1974 - 1978, consists of hourly values of air t e m p e r a t u r e , air h u m i d i t y , global radiation, windspeed and precipitation. The totals of precipitation of these years were 992, 635, 536, 813 and 643 mm respectively; on average 724 mm. The average over 1961- 1990 is 802 mm. The period of 5 y e a r was r e l a t i v e l y dry, with 1974 a wet y e a r and 1976 a very dry one. The meteorological variables of the KNMI data set were measured above grass and are transformed to above forest conditions according to Nonhebel (1987). The forest c h a r a c t e r i s t i c s are described by the calibrated p a r a m e t e r s of the 'Ede' and Tnetford' forests.
4.3 Sensitivity and climatic scenario analysis The influence of the main model p a r a m e t e r s on interception and t r a n s p i r a t i o n were analyzed by sensitivity analysis. To integrate the results with the results of other impact research groups within the National Research Program, the scenario KNMI-2 as described in Section 2 is applied to the 'De Bilt' data. In this study this scenario is n a m e d scenario-2. The changes in t e m p e r a t u r e and precipitation are given in table 4.1. In the scenario the relative humidity is held identical to the relative humidity of the unchanged climate. Other meteorological variables are not changed. The amount of precipitation in the scenario increases strongly compared to climate scenarios described by IPCC or Kwadijk (1993). The increase in precipitation in the scenario is regarded as an increase in precipitation intensity, and not as an increase in duration. In order to apply the scenario an e s t i m a t i o n m u s t be m a d e of the forest p a r a m e t e r s in a changed climate. In particular the leaf are index (LAI) is an i m p o r t a n t parameter. According to the review by Idso and Idso (1994), doubling CO2 increases dry weight by 24% when water is not limiting, and by 58% w h e n w a t e r is limiting. Trees may even be more responsive to a CO2 increase t h a n herbaceous plants, although most experiments are done on seedlings, leafs or small trees. According to the same authors, average increase in dry weight w h e n nutrients are limiting, still amounts to 48%. With limited nutrients and high CO2 the increase will be concentrated in the roots. It is unclear whether the increase in growth is sustainable. Due to the use of unacclimated plants and leaves in most experiments, and the short periods over which m e a s u r e m e n t s are made, it is h a z a r d o u s to transfer the results of these studies to forest (Eamus and Jarvis,
855 1989). For instance, the response of the assimilation rate of acclimated plants seems to be 50% lower than of unacclimated plants due to the lack of active sinks for the assimilation products (Cure and Acock, 1986). In the scenario a modest increase of 5% in LAI and storage capacity is applied. It is expected that growth of the canopy will be limited by low nutrient availability and the maximum LAI possible considering the radiation in the canopy. The direct effect of increased CO2 is simulated by a decrease in stomatal conductance of 30%. Based on present knowledge, these changes are regarded as realistic, though variations due to varying species composition and forest site may be large. Table 4.1 Scenario use as input in model simulations. Change of actual temperature per hour in ~ and hourly precipitation in %. Number of precipitation days is unchanged
Temperature change Precipitation change scenario 2
Winter
Spring
Summer Autumn Year
3.0 19.2
2.3 9.5
3.7 16.2
3.4 10.4
3.1 13.8
4.4 R e s u l t s a n d c o n c l u s i o n s
Interception The interception of rainfall is especially sensitive to changes in evaporation rate, leaf area and related storage capacity of the canopy. Changes in temperature, air humidity and windspeed strongly affect interception. A change in air humidity of 20% results in a change in average air humidity deficit from 0.6 g/kg to 1.8 g/kg. Interception changes by about 50% for coniferous forest and by 60% for deciduous forests. An increase in storage capacity of 20% results in an increase in interception of 10% for coniferous forest and 7% for deciduous forests. In applying the scenario, the small increase in interception for both forest types (Figure 4.1) is mainly caused by the increase in storage capacity. Relative humidity is unchanged and evaporative demand of the air is hardly increasing. Although precipitation increases strongly, it hardly affects interception because the increase is concentrated in winter, when evaporation is low. As a result both forest types show a small decrease in the ratio of interception and total precipitation.
856 Precipitation
4 ,
1974
I
I
1975
1976
el
;.
1
Deciduous forest
i
i
1978
Avg
Coniferous forest
Interception
E 50
I
I
I
I
I
i
I
l
1974
197s
197e
1977
197s
--
Avg
1974
I~7s
197e
197r
lgTe
Aw
I
1975
1976
1977
1978
Avg
1974
1975
1976
1977
1978
Avg
Transpiration
1974
Precipitation excess E~ 35o
Normal
~
Scenario2
Figure 4.1 Yearly totals of precipitation, interception transpiration and excess simulated with normal climate and scenario 2
Transpiration Using the P e n m a n - M o n t e i t h equation, t r a n s p i r a t i o n of forests is sensitive to changes in m a x i m u m stomatal conductance, temperature, air humidity and soil water availability. The sensitivity of transpiration is caused by soil water. Due to limited soil water availability, transpiration is reduced after some time. So in most years, when transpiration is enhanced during winter and spring, water shortage occurs and reduces the t r a n s p i r a t i o n in summer. On the other hand, w h e n t r a n s p i r a t i o n is decreased, more water is available and t r a n s p i r a t i o n during s u m m e r is not so often limited. For some years, this results in a higher total transpiration when transpiration enhancing parameters have lower values. This includes interception. When interception is increased, less water is available for transpiration. Application of the scenario shows a small decrease in transpiration over the five year period. In dry years transpiration is limited due to low soil water content during summer. The strong increase in precipitation with a climate change leads to a higher availability of soil water, so transpiration in those years increases. But
857
due to the lower stomatal conductance, transpiration decreases by 10-30% most of the time when water availability is not limiting (Figure 4.1). Forest water balance
The annual water use of forest is simulated to change between -20% and +10 % depending on water availability. The average change over the 5 years is close to zero. Water use is increased when the forest stands on soil with low water availability. The increase in precipitation results in large precipitation excesses and a reduction in the number of days with water shortage (Table 4.2 and 4.3). In winter, when the evapotranspiration of the forest is low, the large increase in precipitation will drain almost completely to the groundwater. Large discharges can be expected, especially in deciduous forests. Figure 4.2 shows a total increase in precipitation excess of 60%, with high peaks during winter. Thetford/coniferous
5oo--
"~2oo o_ loo
... 9
....
9
o
Uljllllll,ll~'l 1
5
.
...
.
I r I I ' l I' I ~'11 I I ' l ' l ' l " l I"1 ~ 9
13
17
21
25
29
33
lll]r'll 37
41
45
It 49
53
Week
Ede/deciduous 4oo
t 350
9. . .
..
.
~ loo1 5O o -T-~ I i l l 1
5
I I 1 ] I 1 1 1 1 F7 i ~ |--I I l-I I ~ t I t .I..I.(1 9
13
17
21
25
29
33
I I vil 37
it 41
I"l
I1,
I 45
I ~.1.
II
49
53
Week
9
Scenario
2
Normal
Figure 4.2 Cumulative precipitation excess per week for Thetford and Ede forests Table 4.2 Number of days with soil water deficit above maximum for Ede deciduous forest
Normal Scenario 2
1974
1975
1976
1977
1978
Avg
0 0
39 18
46 31
13 0
22 0
24 9.8
858 Table 4.3 N u m b e r of days with soil w a t e r deficit above m a x i m u m for Thetford coniferous forest
Normal Scenario 2
1974
1975
1976
1977
1978
Avg
7 0
42 18
91 59
15 10
46 17
40.2 20.8
Implications According to the simulation study winter discharge will increase strongly and s u m m e r droughts will decrease. The increase in winter discharge results from the strong increase in precipitation and not from the decrease in transpiration, which is low in winter. Compared to deciduous forests, coniferous forests diminish winter discharge. It should be noted t h a t the results of the study are strongly dependent on the expected increase of precipitation, the decrease in stomatal conductance, the small increase in leaf area and the assumed constant relative humidity. On the short t e r m the s t o m a t a l conductance of most C3 species at elevated CO2 levels decreases, but long t e r m effects are h a r d l y known at present. Given the uncertainties in these parameters, the limits of confidence of the present study are very wide. The p r e s e n t policy to replace coniferous forest by deciduous forest to limit evaporation, will further increase drainage in a greenhouse climate. This m e a n s that frequent flooding can be expected in winter when the soil is already saturated. Like the prediction of future climate, the prediction of the impacts of climate change on w a t e r use of forest systems is hazardous. An important reason is t h a t data to validate the model are scarce. The response of trees to elevated CO2 levels might mirror t h a t of other C3-plants, but may also differ because trees are woody and perennial. Experiments with increased CO2 concentration on fully grown forest trees are recommended to improve the confidence level of the present studies. 5. L O W L A N D H Y D R O L O G Y J. Postma, L.C.P.M. Stuyt and P. Kabat Research Institute: DLO-The Winand Staring Center (SC-DLO) P.O.Box 125, 6700 AC Wageningen, The Netherlands
Abstract Dynamic computer simulation models were used to carry out scenario studies, forecasting the possible effects of sea level rise and climate change on physical processes which are crucial in regional- and agro-hydrology. These effects call for w a t e r m a n a g e m e n t m e a s u r e s on a regional scale. Attention was focused on changes in hydrology in the upper soil layers where these effects interfere with soil
859 water dynamics. A modified version of the two-dimensional groundwater flow model MOC of Konikow & Bredehoeft was used to simulate density-dependent deep groundwater flow and salt transport. Soil water dynamics and salt transport in the u n s a t u r a t e d zone were simulated with the one-dimensional model SWAP. A sea level rise of 1.2 m (worst-case scenario of IPCC, 1990), gradually imposed during a century, affects the seepage rate into polders in the studied area almost instantaneously but at a negligible rate. During the simulated period, the salinity of the seepage w a t e r r e m a i n s unaffected due to the low flow velocities of the g r o u n d w a t e r and the great path lengths to be travelled by the g r o u n d w a t e r between the coastal area and the polders. In contrast, climate change significantly affects crop production, viz. potential and actual transpiration. 5.1 I n t r o d u c t i o n
Climate change will interfere with low-coastal hydrology int two different ways, namely sea level rise and altered meteorological conditions near the land surface. Sea level rise will probably cause increased seepage rates in low-coastal regions, leading to s a l i n i z a t i o n of (shallow) ground- and surface waters. Altered meteorological conditions will affect the exchange of water and energy at the soil surface, and t h u s soil w a t e r dynamics in the u n s a t u r a t e d zone and crop production. In low-coastal regions of The Netherlands, integral water management influences the open water and the shallow groundwater systems. Agriculture, horticulture, n a t u r e conservation, domestic and industrial w a t e r supply are involved on a regional scale. As climatic change is likely to interfere with integral water supply and demand, an investigation of its possible consequences is called for. The climatic change was simulated using meteorological relationships from the KNMI (see Section 2), based upon a temperature rise of 1 ~ Simulations of the proposed t e m p e r a t u r e rise of 3 ~ (IPCC, 1990) was abandoned, because of sensitivity of the available crop varieties to the changes in t e m p e r a t u r e sums. Adapting these and other physiological plant parameters to such comparatively extreme conditions was considered to be unreliable at this time. It is to be expected t h a t varieties suited to changed conditions will be available when they become necessary. The consequences were assessed through a series of scenario studies, made with dynamic computer simulation models which were modified for this study. These studies were made in a vertical cross-section through the island of Voorne-Putten in the SW-Netherlands. This island was selected because it lies below sea level, and there is a certain amount of saline seepage there already. Also, investigations were made here earlier, providing essential data. The effect of sea level rise on saline seepage was simulated with the 2-D groundwater flow model MOC (Konikow & Bredehoeft, 1978) in cooperation with G. Oude Essink of the Technical University of Delft. Soil water dynamics and crop production were s i m u l a t e d using the SWAP model. SWAP is an i n t e g r a t e d s i m u l a t i o n tool consisting of SWACROP, a quasi 2-D model of the water (plus soluble salt) balance of a cropped soil including drainage and irrigation (Feddes et al., 1994), and WOFOST: a water-limited crop production model (van Diepen et al., 1988) made at the DLO-Centre for Agrobiological Research (CABO-DLO).
860 5.2
Methods
C l i m a t e scenarios To create a climate scenario, the methods were used t h a t are discussed in Section 2. Radiation, humidity and wind and the pattern of rainfall are assumed to r e m a i n unchanged. The increase in t e m p e r a t u r e used in the scenarios is 1 ~ resulting in a change to a n n u a l precipitation of-2% to +9%, depending on the temperature. The meteorological files of the years 1966, 1976, 1979, 1985 and 1986, r a n g i n g from very dry to very wet, were selected as input for the changed climate. Crop production and water use were calculated for these years, first without, then with the climate scenarios. The differences in production show the effect of climate change.
C a l c u l a t i n g crop p r o d u c t i o n with SWAP The island of Voorne-Putten, surface area 19025 ha., was divided into 761 subareas. Soil physical properties, open water levels, drainage properties, salinity and seepage rates, and land-use were collected for each subarea, 461 of which are cultivated. Production of the most frequently grown crops, potatoes, sugarbeets, w i n t e r w h e a t and grass, was calculated of the 461 subareas, for the five selected years, and calibrated with estimated actual harvests. Production and water use for the changed climate was then calculated by using the same years, changed by the climatic change. Higher temperatures will cause: maintenance respiration to increase; - plant organs to age faster, inhibiting daily increase and harvest total of dry matter; - t e m p e r a t u r e sums to increase faster, causing the crop to flower, m a t u r e and/or ripen (too) early. Higher atmospheric CO2 affects the crops (of the C3 plant type) by 4 i m p o r t a n t mechanisms (Wolf & van Diepen, 1993): - Leaf thickness increases, meaning specific leaf area decreases, - Light-use efficiency (crop production per unit radiation) increases, - Maximal assimilation rate increases, - The crop can absorb sufficient quantities of CO2 in a shorter time, keeping the s t o m a t a open for a shorter time, and so reducing transpiration. W a t e r use efficiency is increased this way. The simulations were done with the same crops, but with different crop-varieties assessed to give a realistic yield under the associated climatic conditions, by changing the physiological p a r a m e t e r s of the crop models, cf. Table 5.1 (BoonsPrins et al., 1993).
861 Table 5.1 C h a n g e s in p l a n t physiological p a r a m e t e r s to a d a p t to h i g h e r t e m p e r a t u r e s a n d raised CO2-1evels (from Wolf & van Diepen, 1993) specific leaf
light-use maximal temp. sum area efficiency assimilation (m 2.kg-1) (kg.ha-l.h-1 rate /J.m-2.s-1) (kg.ha-l.h-1)
temp.sum before flowering (~
surface until maturity (~
resistance (s m-l)
Winter wheat l'CO 2 2"CO 2
18.0 14.4
0.45 0.55
40 80
1048 1290
1258 1171
40
Potatoes 1"CO2 2"CO2
18.0 14.4
0.45 0.55
40 80
150 150
1550 1800
30
Grass l'CO 2 2"CO2
25.0 20.0
0.45 0.55
40 80
-
-
65
Sugarbeets 1"CO2 2"CO2
18.0 14.4
0.45 0.55
40 80
573 483
1909 2194
30
Groundwater flow modelling with MOC After several experiments, 3-D simulation of g r o u n d w a t e r flow was discontinued due to severe limitations of the available models. Instead, g r o u n d w a t e r flow was s i m u l a t e d in a v e r t i c a l l y oriented, 2-D cross-section t h r o u g h V o o r n e - P u t t e n , r u n n i n g w e s t to east, w i t h dimensions 200 m (depth) by 25 k m (length). The g r o u n d w a t e r flow model used was MOC (='Method Of Characteristics'), version 3.0 of 1989, w h i c h w a s developed by the US Geological S u r v e y (Konikow a n d Bredehoeft, 1978) as a t r a n s i e n t solute t r a n s p o r t model, including h y d r o d y n a m i c dispersion, t h r o u g h the horizontal plane. In order to suit the model for application in v e r t i c a l l y oriented cross-sections, it was a d a p t e d for d e n s i t y differences of g r o u n d w a t e r (Oude E s s i n k , 1993). In the model, the chloride c o n c e n t r a t i o n d e t e r m i n e s g r o u n d w a t e r density. The n u m b e r of grid cells is 100 (horizontal direction) by 20 (vertical direction); all cells are 250 m long by 10 m high. The g e o m e t r y of the geohydrological s y s t e m at the cross-section t h r o u g h VoorneP u t t e n is depicted in Figure 5.1. Geohydrological p a r a m e t e r s of the subsoil, initial salinities and b o u n d a r y conditions for g r o u n d w a t e r flow were derived from Wit (1987), DGV-TNO (1984), Oude E s s i n k (1993) and P o m p e r (1983). MOC requires the ratios t r a n s v e r s a l to longitudinal conductivity and dispersivity to be constant in the entire modelling domain; these were set to 0.1. Initial salinities following are shown in Figure 5.2. The following boundary conditions were imposed. The base is a no flow b o u n d a r y . Along the seaside and inland b o u n d a r i e s w h e r e h y d r o s t a t i c conditions are assumed, constant piezometric levels and salinities are m a i n t a i n e d , determined by m e a n sea level, w a t e r levels in bordering channels and the density of the water. Along the upper boundary, constant phreatic levels are m a i n t a i n e d in polder areas. These levels are determined by the w a t e r levels in open channels and
862 collector drains. At the sand dune areas a constant rate of groundwater recharge is maintained ( 180 mm.yr- 1).
C a l i b r a t i o n o f MOC The geometry of the island of Voorne-Putten imposes restrictions to the modelling of groundwater flow. Simulation accuracy in a 2-D vertical cross-sectional area is hampered by the fact that important boundary conditions to groundwater flow, i.e. pressure heads at the nearby n o r t h e r n and southern shorelines of the island cannot be incorporated in the model. Hence, calculated seepage rates will be lower t h a n observed ones, particularly in the central area of the domain where the effect of the inland and seaside boundary conditions of the groundwater velocity field are c o m p a r a t i v e l y insignificant. It was therefore decided to concentrate model calibration in the area bordering the seaside boundary where the effect of sea level rise was to be simulated. In addition, the area used for calibration was confined to the bottom layer of the upper aquitard and the first aquifer because of the high r e s i s t a n c e to flow of the lower a q u i t a r d (10000 d.; P o m p e r , p e r s o n a l communication). The calibration was made for seepage rates through the upper a q u i t a r d for the reference case, using the rates established by Wit (1987), by varying the kSAT of (groups of) model cells within ranges, derived from existing information (Figure 5.3). All seepage rates are averaged for specific subareas, mainly polders, with uniform open water levels. j -." / / ,~: ~ / y .
.......... Northsea ..... i"
dikes
dunes / '\ "~ / \ \ polders 9 t ~ ~ - . . . , - . - - , - ~ ~J//////////J//'///////J
(0.0 to -2.0 m) b e l o w s e a level ,___ \ 9 .... - .... ~,.......-;~,---,,---.................. = , , - ~ ~ ~ - ~ - ~ Spul uu_inK er_K e__ P_:_ ! _m_ult i p l e k's!j ~/~/z'/7-/7-/z~'/z/~/7"/7~,/7~/z/~, .........
~ X X X X X X X X X X X X X X X X X X ~ - : I i f f I f l I ! [ I I ! : : , C a l agl s ( t ~])
Kreftenheye
>.
,
C :D
;
4--'
....:i: ...'.:"
m/d)!fi!:~///X/////y//J/J///'///'///'///'SJJ~J~
F. ( 2 0
'-
m/d)
~J 4_.,
~/////////////////////////////~/, Kedichem
F. ( 0 . 0 0 2
m/d)
m
o
,5 ""-7- c
d)
,. 9
.
.
..~ MOC cells (200~ 10
m)
-,
%
~,
%.
"
.. "-,.,.,~ " " .'"i-4,
. . . . . .
.
..
.
9
. .
"
'
...9
.
. . .
9
" 9 9
~ ' %...k~.
"
9 .
:.. .
9
.:
;
i".
9
" . ..
:
~,,,,,
...
"
"m_..~L... " '
9 Maa,,
sluis
'
.
9
"',%o.._ .
~"
"
"
'
.. '
..
'
"~. "'"%,
"
"
"--~oo_ .
.~,~
. ~ -
..:
.
" " 9
. .
"
::.
.
' '
.
.
"b%.
..
9
"
" "
. i...
"
9 "
.
: " "
\%,
X.~
..:
.
.
\"~. ..
_~.~
[..
9 .
.
-~\ 9
. . . . " i......
o
9
"
" ..".".i '.".
\
--%
..
.
.
.
1~o.o
leO
~:~:: . . . . . . : ~i ,=
.
. 9
9
and
.
..
......
o
"
..
.
with
. 9
140.0
o
i
i.
9
\
.
.
.
9
.
26.0
100 years,
~
~0 -::---: 9 ~.-: .- .~.~: ,- ''~- .: ~
'
. .
o
~_~ i::
~i~...... :...."?.'~
.~ ::
.
~
i 9 e o . o
lOO
22.,5
.
.
.
20.0
(Km)
: ': eo.o
17.6
MOC
~
....................................................................................
16.0
Coastline
ll~~~':iiI"~::::, :~:.~...:., ~:::~".i"":~:.... i: ~ ~ . ~.~...'... ~ ' "
~
..~
......
from
~ ~: :.... , . ~ - ~ - ~ _ ~ ~ ~ , 9
..
" "
10.0 Distance
Figure 5.4 Differences in groundwater without sea-level rise, with
L
..
.
.
9. . . . . . . . . . . . . . . . . . . . . .
: 9
9
"
"
. " % o 4 ~ . ~
:.
~ooo
0.8 kg/m2 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.00
,.::::~
Study
m~t
'Keenf
a. Keent.
iiiiiiiiiiii!iiiiiiiiiiiiiiiiiiimiiiiiiiiii :: ii!i!i?!!iii!iii!i;iii)iiiiiiiiiiiiiiiii?iiii.;ii:: :
2.26 kg/m2 1.99 1.73 1.46 1.20 0.93 0.67 0.40 0.14
. . . . . . .
Study
area
~ili~: i:i=:i;:ii
'V'triksche Plaat'
b. Variksche Plaat. floodplain
~
......
minor river dike
former floodplain protected by river dike
{...........'".. escarpment
major river dike
~
.
.--
woe
groyne
_
_ V
a
r
i
~
Figure 8.6 E s t i m a t e d s e d i m e n t a c c u m u l a t i o n after block kriging. Grid cell size = 10 x 10 m2
b''h
894 gradient is mainly caused by the amounts of deposited sand; no gradient was found in the amounts of silt and clay. The amounts of sand deposited j u s t behind the levee show a high spatial variability. If a floodplain has been completely s u b m e r g e d for several days, local depressions have almost no effect on the sediment deposition. The amounts of sediment deposited on different floodplains can vary considerably. Observed average values after the flood of december 1993 range between 1.0 and 6.6 kg/m2, equivalent to 0.8 respectively 5.4 mm. These differences between floodplains seem mainly related to the flow pattern of the water during inundation. During minor floods, when a floodplain is inundated only during a few days, the p a t t e r n of sediment accumulation is much more related to local differences in elevation and inundation times than during large floods. The amount of deposited sediment is far less than proportional to the magnitude of the flood. This implies t h a t minor floods contribute considerably to the yearly floodplain sedimentation rate. From the results of the individual floods, the present average yearly sedimentation rates will be estimated for different floodplain sections. This part of the study is still in progress.
B. Sedimentation rates during the past decennia. The heavy metal profiles obtained from the floodplain soils generally have the same shape as the pollution curve reconstructed from the dike-breach pond sediments. Profiles from floodplains with high sedimentation rates have a higher pollution in the upper 10 cm and have a higher total pollution. Also the maximum concentration is greater and is found at greater depth. The results show that the total soil pollution can be very different from the pollution in the upper 10 cm. A one-dimensional sedimentation model is being developed that simulates the floodplain sedimentation with contaminated sediments. Using this model the sedimentation rates will be reconstructed. The average sedimentation rates during the past decades obtained from the Pb-210 samples range between 1.6 and 0.1 cm/yr. C. Sedimentation rates since the floodplain formation. This method applies to lateral bars and can be used in a limited number of undisturbed floodplains. At the beginning of the floodplain formation the sedimentation rates varied between 1 and 3 cm/year, which is 3 to 4 times as great as at present. Two examples are shown in table 8.2.
895 A. Klompenwaard: 1800 - 1990
B. Variksche Plaat: 1860 - 1990
year
elev. (m a.s'l.)
sed. rate (mm/yr)
year
1800 1850 1900 1950 1990
10.6 11.2 11.7 12.0 12.3
16.0 11.0 7.6 6.2 4.8
1860 1900 1950 1990
N
i
elev. (m a.s.I.)
sed. rate (mm/yr)
4.5 5.4 6.1 6.5
30.0 18.0 11.0 8.5
Ame.m 2 KL'*I~ ~
~
~iTme~e'/ .....~::
.::~-".....
Lobith ,,........
[iiiiiiiiii
i] floodplains
KL = Klompenwaard VP = Variksche Plaat
Table 8.2 Reconstruction of average yearly sedimentation rates using historic data; two examples
A s s e s s m e n t o f the i m p a c t o f c l i m a t e change on f l o o d p l a i n s e d i m e n t a t i o n rates A. Assessment of impact on floodplain inundation times. The effects of climate changes on the Rhine peak discharge frequencies and exceedance times according to the AP and BaU scenarios are given in Section 6. The BaU scenario causes a significant increase in inundation times of floodplains. B. Assessment of local sedimentation rates using a sedimentation model. A proper calibration of the WAQUA/DELWAQ model requires more d a t a on actual sedimentation than those obtained of the flood of J a n u a r y 1993, and preferably for l a r g e r a r e a s t h a n the V a r i k s c h e Plaat. Therefore, also the s e d i m e n t measurements of the December 1993 flood will be used for calibration. Preliminary results of the model indicate that the sedimentation rates on the Variksche Plaat under the BaU climate scenario increase by only 1% if the sediment concentration of the Rhine only depends on hydraulic conditions. Sedimentation rates increase by almost 20% when increased sediment production is taken into account. The effect of climate change for this particular floodplain seems to be insignificant. The main reason for this is that in floodplains directly bordering the main channel only minor amounts of sediment are deposited during high discharges since the flow velocities are then too high for sedimentation. The increase in sedimentation rates will be
896 larger on floodplains t h a t are separated from the main channel by a s u m m e r dike and w h e r e relatively large a m o u n t s of sediment are deposited d u r i n g high discharges.
C. Sensitivity of large scale potential sedimentation rates. The BaU scenario has a strong impact on the potential floodplain sedimentation (figure 8.7 and 8.8). According to the estimator using sedimentation areas, potential sedimentation rates will increase by a factor 1.5 to 2, depending on the sediment rating scenario. The effect on the estimator using sediment volumes is even larger: potential sedimentation rates will increase by a factor 1.7 to 3. When scenario 4 (only land use change) is t a k e n as a reference, potential sedimentation m a y increase by a factor 2.5 respectively 4.5. As demonstrated by the WAQUA/DELWAQ model study of the Variksche Plaat, the increase in the amounts of sediment t h a t really will be deposited may be less large. This strongly depends on the morphological properties of the floodplain sections. For a proper estimate a further analysis of t h e s e d i m e n t a t i o n process u s i n g s e d i m e n t t r a p s and models such as WAQUA/DELWAQ is required in different types of floodplains. 25
:i~~... ~:-i ::-!--!-i-::,/], ~. ~-~;i~i
20
:
:
:
:
! .......... '~ ' ~' ~ - ~ ~ -~
=--~176176 " =~
:',:ILII~,:"~-~L!"::"',
. .:_ .:- - :. - -:- - :- - -:- - : - i ~ )
\-:
- ~ ~ , ~ \ -
.~
9
,. - -:- - : - -:- - :- -
=
=~
--!--:--;--:--:---:-';A;~~-i--~--:-:--6000 m3/s); this will be over 15% when climate changes. If no measures against soil erosion are taken by reducing the area of arable land or by soil conservation programmes, the production of sediment will increase. Increased, sediment production will lead to about 20% higher concentrations of suspended sediment at low discharges. When the area of arable land decreases, background suspended sediment concentrations will decrease by some 40%. The amounts of suspended sediment transported at discharges when all floodplains are inundated is hardly affected by the expected changes in land use. The increase in sediment load, transported at high discharge is mainly the result of changes in the discharge regime. Floodplain s ed i me nt a t i on is a complex process. It depends on floodplain characteristics, discharge frequency distributions and sediment concentrations, and therefore shows a high variability in time and space. Depending on site characteristics, present sedimentation rates range between 0.5 and 15 mm per year. Present sedimentation rates are 3 to 4 times as low as they were in the past when the floodplain surface was 1 to 2 m lower and flooding occurred more frequently. This implies that removal of summer dikes and lowering of floodplain surfaces as proposed in nature rehabilitation plans will strongly increase sedimentation rates. The effects of climate change on floodplain sedimentation, as evaluated using the BaU scenario can be considerable. This is mainly caused by an increased frequency of high discharges. As the relationship between effective discharge and the area of inundated floodplains is non-linear, even minor changes in discharge and sediment concentrations will have a strong effect. As a result, the potential floodplain sedimentation is very sensitive for the BaU discharge and sediment rating scenario conditions. The effective sedimentation rates may be different, depending on the morphology and type of floodplain. On floodplain sections directly bordering the main channel and without a summer dike, changes in sedimentation rates are relatively unimportant. However, the effect on floodplains t h a t are situated behind a summer dike, is expected to be relatively high. As it was found that floodplain sedimentation is highly variable in space and during individual floods, predictions on future sedimentation rates cannot yet be done with high accuracy. Further analysis on the effective sedimentation on different types of floodplain sections is therefore required. A p a r t from the sediment quantities, sediment quality is a matter of concern. A reconstruction of the heavy metal pollution of the river Rhine during the past century shows that maximum pollution occurred between 1960 and 1970. The present distribution of heavy metals in floodplain soils is related to this pollution history and the floodplain sedimentation in the past. At many locations the soil
898 quality of the upper 10 cm gives an underestimation of the total pollution of the whole profile. Although the heavy metal content has considerably decreased since 1970, an accelerated sedimentation will still accumulate large amounts of heavy metals on the floodplains in The Netherlands. As average sedimentation rates on most floodplains are in the order of 1 mm/yr, the quality of the upper 50 cm of the floodplain soils will only little improve by the accumulation of sediment of improving quality during the forthcoming century. 9. ECONOMY, SAFETY, ENVIRONMENT F.R. Rijsberman Resource Analysis Zuiderstraat 110, 2624 SJ Delft, The Netherlands 9.1 I n t r o d u c t i o n
In the following pages the results of the hydrology-related projects in the NRP, and their implications for economy, safety and environment, are compared with the results and conclusions of earlier studies. The objective of this comparison is to evaluate whether the NRP-sponsored research has led to changed insights into the impacts of hydrology-related impacts of climate change on the economy, safety and environment in The Netherlands. During the period when IPCC produced its First Assessment Report a comprehensive study was undertaken in The Netherlands to assess the Impacts of Sea Level Rise on Society (ISOS). The ISOS study (Peerbolte et al., 1991) focused largely on sea level rise, but attempted to assess the impacts of changes in hydrology on various aspects of the economy, safety against flooding and the environment as well. The study attempted to integrate data, knowledge and modelling results available in 1988-1991 to produce an assessment of the impacts of climate change for The Netherlands that can be considered as a state-of-the-art review at that point in time; as such it is used in the following pages as a baseline against which the progress made through the NRP programme can be evaluated. It has to be noted at the outset that the approach adopted in the ISOS and NRP studies is quite fundamentally different. The former study made relatively quick and dirty scenario-assumptions for local changes in climate and for a series of physical (intermediate) effects of such changes, e.g. on discharge of the River Rhine, and subsequently attempted to estimate socio-economic impacts resulting from these changes. Impacts were expressed, where possible, in monetary terms for damages such as production losses in agriculture, flooding damages for industry, additional costs of provision of cooling for power plants, and increased costs of water management measures such as pumping drainage water. The hydrology related NRP studies focus largely on the development of regional climate change scenarios (Section 2) and the direct physical effects of these changes in terms of soil erosion, river discharge, sediment transport, sediment deposition, saline seepage and the hydrology of forests. The hydrology-related NRP studies do not assess the socio-economic impacts of these physical effects directly,
899 w i t h the exception of the analysis of changes in agricultural production caused by changes in t e m p e r a t u r e , evaporation, precipitation and CO2-concentrations. 9.2 C l i m a t e s c e n a r i o s One of the more i m p o r t a n t steps forward in the last few y e a r s is the improved regionalization of the climate change scenarios. Both earlier studies as well as the N R P studies are based on the IPCC Business as U s u a l (BaU) scenario, b u t the regionalization of the hydrology-related climate variables is quite different. For c o m p a r i s o n t h e scenarios u s e d in the I S O S - s t u d y are p r e s e n t e d h e r e a f t e r , t o g e t h e r w i t h the scenarios on which the N R P studies are based. One ISOSscenario is the so-called Average (AV) scenario in which all variables, including hydrology-related variables, have their expected values. Another scenario, in which the e s t i m a t e d s t a n d a r d deviations were deducted from the expected values for all variables, is referred to as the Unfavourable (UNF.ALL) scenario. The hydrology related NRP-studies refer to either the regional climate scenarios for The N e t h e r l a n d s developed in NRP (Section 2) or the scenarios for the Rhine Basin developed by Kwadijk (1993). These are also expected values and should therefore be compared with the AV scenario in ISOS. Both scenarios are presented in Table 9.1, together with the two earlier scenarios. Table 9.1 Comparison of regionalized climate change scenarios used in the hydrology related NRP studies as well as in earlier studies Variables Ta Tw Ts Pa Pw Ps Qw Qs Ea=Ew=Es SLR
ISOS AV
ISOS UNF.ALL
NRP
Kwadijk
+3 +3 +3 + 10% + 10% + 10% + 5% -5% + 10% 0.6
+3 +3 +3 +15% + 5% + 10% - 10% + 15% 0.85
+3 * 2.3-3.7 * 2.3-3.7 + 13% * 3%-21% * 3%-21% ** + 15% * * - 10%
+3.5 + 4.3 + 2.9 + 11% + 19% +4%
* m o n t h l y values ** computed values from N R P studies ( P a r m e t et al., 1994) Where: Ta = average a n n u a l T e m p e r a t u r e increase in degrees C Ts = s u m m e r T increase in degrees C Tw = winter T increase in degrees C Pa = average precipitation change in % Ps = s u m m e r precipitation change in % Pw = winter precipitation change in % Qs = s u m m e r river discharge change in % Qw = winter river discharge change SLR = sea level rise in m Ea = average a n n u a l evaporation change in %
900 As can be seen from Table 9.1, the changes in precipitation assumed in the recent scenarios (Section 2) are considerably greater than in the earlier (ISOS-AV) scenario; even greater than in the scenario that was considered "extreme" a few years ago (UNF.ALL). Similarly, the expected changes in river Rhine discharge resulting from the recent NRP RHINEFLOW study considerably exceed the average changes expected earlier (AV scenario). In addition, the NRP studies (Section 6) conclude that the frequency of low flow months (Q< 1000 m3/s) increases with 60%. The conclusion can be drawn from the scenarios presented above that the regional climate change studies undertaken in recent years have shown that the hydrological changes associated with the IPCC-BAU scenario are considerably greater than expected before. Consequently, the hydrology-related impacts of climate change on the economy, safety and the environment can also be expected to be relatively more important than estimated before.
9.3
Impacts on safety against flooding
In the ISOS study it was assumed that the design discharge used for flood protection infrastructure (a discharge of 16,500 m3/s at Lobith, with a recurrence interval of 1250 years) would increase 10%. This resulted in increased design dike crest levels for the river dikes that varied between 0.1 and 0.5 m for different dike sections to maintain the same safety against flooding. The cost of raising river dikes in the non-tidal part of the country (necessitated both by sea level rise and increased river discharges) was estimated to be in the order of 2,500 million guilders over a 100 year period. The cost of raising sea and river dikes was the largest direct economic impact of sea level rise determined in the ISOS-study. The NRP RHINEFLOW study concludes that the average winter discharge increases 15% (Qw= +15%), but u since only a 25 years period was simulated does not allow conclusions to be drawn concerning changes in discharges at frequencies relevant for flood analyses, e.g. discharges with a recurrence interval of 1250 years (Section 6). In a related study, Kwadijk and Middelkoop (1994) used the RHINEFLOW model to assess the probability of exceedance for discharge peaks under possible future climate conditions. For peak discharges relevant for flooding of floodplains (up to Q= 6.500 m3/s) they found that a precipitation increase of 20% leads to a 30% higher two-year peak discharge. For larger design discharges, used for the design of flood protection structures, reliable results could not be obtained because the length of the discharge record was too short. This implies that the hydrology-related NRP studies have not, so far, improved the assessment of floodsafety related impacts. Further analyses of the changes in the peak discharges of the River Rhine are therefore recommended.
9.4
Economic impacts
I m p a c t s on a g r i c u l t u r a l p r o d u c t i o n a n d l a n d use The NRP study (Section 3) analyzed changes in crop yield potential for a time horizon of 2040-2050 with a biophysical simulation model (WOFOST) for Ps=0% en PW=+10%, Ts=l.5, Tw=2, double CO2 concentrations, and computed (Penman) evaporation rates (Ea= +15%). It concludes that the changes would result in average increases in yields of wheat, grass, sugarbeets and corn on sandy soils in the Rhine river basin of 40, 33, 25 and 12 %, respectively. Based on analysis of autonomous developments such as population growth, agricultural production
901 levels and developments in the world and EU market, land use patterns have been simulated, with and without climate change. Without climate change, the range of land use scenarios for the Rhine Basin in the decade 2040-2050 shows that the land used for agriculture will decrease by 0 to 3 million hectares. With climate change, land used for agriculture decreases an additional 0.2 million hectares. Both earlier studies and analysis carried out in the NRP (Section 5) conclude that possible increases in saline seepage through the groundwater due to sea level rise are negligible. Parmet et al. (Section 6) conclude that the summer discharge of the River Rhine would decrease by about 10% and the number of low flow months (Q< 1000 m3/s) increases with about 60%. It has not been analyzed in the NRP what the surface water related implications of these drier summer conditions are, but these could be considerable. The analyses of earlier studies concerning climate change impacts on agricultural production focused on surface water management related issues, including agriculture drought damage, agriculture salinity damage, sprinkler irrigation cost; (discharge) pumping capacity and cost. Conclusions were, for instance, that hydrological dryness influences drought damage substantially, e.g., Ps=-7% (and assumed Es=+15%) resulted in 38% increase of drought damage cost from about 350 to about 500 million guilders per year (Peerbolte et al., 1991). It was also concluded that reductions in average summer discharge of the River Rhine, e.g., Qs=-7%, had only minor impacts on agricultural production. Summarizing, the NRP studies have concluded that there is a potential for significant increases in agricultural production as well as a potential of significant increases in drought damage. From a perspective of the socio-economic impacts on agriculture, the forecasted increase in productivity could increase the comparative advantage of agriculture in The Netherlands, but this will depend on a host of other factors as well (e.g., surface water related drought damages, and w a t e r management costs). The surface water related impacts of the considerably lower summer discharges of the Rhine have not been investigated in the NRP studies, but based on earlier studies it is expected that these could be quite significant for The Netherlands.
Increased costs o f electricity p r o d u c t i o n The ISOS study concluded that the additional costs of electricity production of a decrease in summer discharge of the River Rhine (Qs=-7%) would be in the order of 6 million guilders per year. In Section 6 it is concluded that the average reduction in summer discharges could be higher (Qs=-10%) and that the frequency of low flow months, which can be critical for design of the cooling system, increases considerably. The additional costs of electricity production are therefore likely to be significantly higher than estimated in the earlier studies. Water m a n a g e m e n t costs The larger decrease in average summer Rhine discharges, and increased frequency of low flow months, can also be expected to impact several surface water management factors, particularly the effort required to control the salinity intrusion through the Nieuwe Waterweg (harbour of Rotterdam), navigation, and flushing of polder areas. Increased costs for shipping have not been analyzed in the
902 climate change impact studies to date. Through other studies it is known, however, that the costs for navigation increase rapidly when discharges decrease. Shipping costs in The Netherlands increase from 85 million guilders per week to 130 million guilders per week as the discharge decreases from 1600 to 1000 m3/s (Anonymous, 1990).
F l o o d i n g in f l o o d p l a i n s Earlier studies by Licht (1990) estimated t hat for SLR= 0.85, Qs=-10%, en Qw=+10%, the expected annual economic losses for the brick industry are in the order of 2 million guilders per year due to increased flooding, whereas for dairy farming the increased flooding in winter is more or less compensated by decreased flooding in summer. The NRP study results (Section 6) of Qs=-10% and Qw=+15% would probably result in somewhat higher losses for both the brick industry and dairy farming, but these impacts are still quite small. I m p a c t s on the environment The NRP studies evaluated changes in the following variables that may, in turn, cause environmental impacts" sediment production (Section 7) and suspended sediment t r a n s p o r t and suspended sediment deposition on embanked river floodplains (Section 8); - hydrology of forests (Section 4). -
Earlier studies for The Netherlands evaluated a number of climate change impacts on the environment, but most of these related to coastal environmental aspects. Where, for instance, the ISOS study did look at environmental impacts (impact of flooding on natural areas in river floodplains; decreased biomass phytoplankton and chlorophyll in fresh water ecosystems), the assumed changes in hydrology were too small to determine significant impacts on the environment. The NRP studies showed that sediment production in the River Rhine catchment area may increase by about 20% due to climate change (Section 8), but this increase will be more than balanced by autonomous land use changes that cause a decrease in sediment production. The NRP studies also show that a much higher part of the suspended sediment will be transported at higher discharges, which will lead to a considerably higher deposition of suspended sediment on the embanked floodplains (Section 8). If this sediment remains as polluted as it has been during the last decades than this would lead to a buildup of pollutants on the floodplains which will have serious consequences for the environment. Fortunately, there is a trend towards improving sediment quality in the River Rhine that may mitigate this potential impact. The studies on the hydrology of forests shows t h a t increasing CO2 concentrations may lead to lower evapotranspiration and hence reduced drought damages in summer. Generally speaking, the value of the conclusions of these studies would increase if they were integrated into a common framework for analysis. The changes in the hydrology-related climate variables that follow from the scenarios discussed above, as well as the computed changes in river discharge, are likely to have significant consequences for natural ecosystems, such as desiccation of floodplains in summer. Such ecosystem related consequences have not been looked at in the hydrology-related NRP studies, however.
903 9.5 C o n c l u s i o n s
The changes in hydrology-related climate variables t h a t follow from NRP regionalized climate scenarios are significantly greater those assumed in earlier studies. NRP studies have analyzed the changes in average summer and winter discharges of the River Rhine, as well as the changes in the 2-year peak discharges and frequency of low flow months. The results of these studies show that the discharges of the River Rhine change considerably more than assumed earlier. Peak discharges at recurrence intervals relevant for analysis of impacts on flood safety have not yet been analyzed. Further research in this area is recommended. NRP studies have shown that the expected changes in temperature, precipitation and CO2-concentration have the potential for increased agricultural productivity in The Netherlands. The economic consequences of such changes also depend on changes in drought damages and water management costs. The hydrology related NRP studies have forecasted relatively greater decreases in average summer discharges of the River Rhine than assumed earlier, and significantly increased frequencies of low flow months. This is likely to have considerable impacts on drought damages in agriculture, water management costs, costs of navigation, and costs of electricity production. These surface-water related drought effects may well be the most important economic impacts caused by changes in hydrological climate variables. Several NRP studies have assessed physical effects due to changes in hydrologic variables, such as sediment production, sediment transport and deposition, and hydrology of forests. The expected change in hydrology is likely to lead to increased sediment production, a larger deposition of sediment on the embanked floodplains, and a decrease in the evapotranspiration of forests. These physical effects are likely to have impacts on the environment, but these have not been estimated directly. It is recommended that the analysis of impacts of the various physical effects studied in the future hydrology-related NRP projects is conducted in an integrated framework. 10. R E F E R E N C E S
Anonymous. 1990. Beleidsanalyse waterhuishouding scheepvaart, bevindingen van de werkgroep scheepvaart (in Dutch). RIZA-report 90.005. Asselman, N.E.M. and H. Middelkoop, 1993. Floodplain sedimentation; quantities, patterns and processes. Geopro-93.09. Dept. of Physical Geography, Utrecht University. Accepted for publication in E a r t h Surface Processes and Landforms. Boer, G.J., N.A. McFarlane and M. Lazare, 1992. Greenhouse gas-induced climate change simulated with the CCC second-generation general circulation model. J. Climate, 5: 1045-1077. Boons-Prins, E.R., G.H.J. de Koning, C.A. van Diepen and F.W.T. Penning de Vries, 1993. Crop specific simulation parameters for yield forecasting across the European Community. Simulation reports CABO-TT, no. 32, CABO-DLO. Wageningen, The Netherlands.
904 Buishand, T.A. and A.M.G. Klein Tank, 1994. Regression model for generating time series of daily precipitation amounts for climate change impact studies (submitted to Stochastic Hydrology and Hydraulics). Bultot, F., Coppens, A., Dupriez, G.L., Gellens, D. and F. Meulenberghs, 1988. Repercussions of a CO2 doubling on the water cycle and on the water balance, A case study for Belgium. Journal of Hydrology 99: 219-347. Bultot, F., Gellens, D., Spreafico, M. and B. Sch~idler, 1992. Repercussions of a CO2 doubling on the water balance - a case study in Switzerland. Journal of Hydrology, 137: 199-208. Cure, J.D. and B. Acock, 1986. Crob responces to CO2 doubling: a literature survey. Agri. For. Meteor., 38" 127-145. DGV-TNO, 1984. Grondwaterkaart van Nederland, (Groundwater map of The Netherlands), Rotterdam 37 oost, 37 west). (In Dutch). Rapport GWK 35, Dienst Grondwater-verkenningen TNO, Delft-Oosterwolde. Dolman, A.J., Stewart, J.B. and Cooper, J.D., 1988. Predicting forest transpiration from climatological data. Agri. For. Meteor., 42: 339-353. Drift, J.M.W., H. Middelkoop & N.E.M. Asselman, 1994. Estimation of the effects of climate and land use change on the production of fine sediment by soil erosion in the catchment area of the river Rhine. Internal report, Laboratory of Physical Geography and Soil Sciences, University of Amsterdam; Department of Physical Geography, University of Utrecht, 12 pp. Eamus, D. and Jarvis, P.G., 1989. The direct effects of increase in global atmospheric CO2 concentration on natural and commercial temperate trees and forests. Adv. Ecol. Res., 19: 1-55. Feddes, R.A. and P. Kabat, (eds.) In prep. 1994. SWAP. A model to simulate the Soil Water Atmosphere Plant interactions. Part I. Theory and model description. Simulation Monograph, Pudoc, Wageningen. Hendriks, C.M.A., 1994. Biophysically-based, spatial analysis of possible climate change impacts on forest yield potentials and water use in the Rhine basin, volume 3. (In press). Hendriks, C.M.A., 1994. Biophysically-based analysis of possible climate change impacts on forest yield potentials and water use in the Rhine basin. Volume 3 of Land use projections for the Rhine basin based on biophysical and socioeconomic analysis. Winand Staring Centre-RIZA report 85.3. Hendriks, M.J., Kabat, P., Homma, F. and Postma, J. 1990. Research into the evaporation of a deciduous forest. Measurements and simulations. Report 90, Staring Centrum, Wageningen, The Netherlands, 95 pp, (in Dutch). Houghton, J.T., Callender, B.A. and Varney, S.K. (eds.), 1992. Climate Change 1992, The Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press, Cambridge. Idso, K.E. and Idso, S.B., 1994. Plant responces to atmospheric CO2 enrichment in the face of environmental constraints: a review of the past 10 years' research. Agri. For. Meteor., 69: 153-203. IPCC, 1990. 'First Assessment Report'. In: Land and Water International no. 69. Isarin, R.F.B. & H.H.A. Berendsen, 1992. Morfodynamiek van de rivierduinen langs de Waal en de Lek. Rapport GEOPRO-92.08. Vakgroep Fysische Geografie Universiteit Utrecht. Jarvis, P.G., 1976. The interpretation of variations in leaf water potential and stomatal conductance found in canopies in the field. Phil. Trans. R. Soc. Lon. B., 273: 593-610.
905 Klein Tank, A.M.G. and T.A. Buishand, 1993. Modelling daily precipitation as a function of temperature for climate change impact studies. Scientific Reports WR 93-02, KNMI. De Bilt, The Netherlands. Klein Tank, A.M.G. and T.A. Buishand, 1993b. The occurrence of rain in a changing climate. KNMI Memorandum 93-04. Klein Tank, A.M.G. and T.A. Buishand, 1994. Daily precipitation amounts in a future climate derived from temperature and surface air pressure. KNMI Memorandum 94-03. Klein Tank, A.M.G. & G.P. KSnnen, 1993. The dependence of daily precipitation on t e m p e r a t u r e . In: Proceedings of the 18th annual climate diagnostics workshop, Bolder, Colorado. US Dept. of Commerce. KNMI, De Bilt, The Netherlands. Konikow, L.F. and J.D. Bredehoeft, 1978. Computer model of two-dimensional solute transport and dispersion in ground water. U.S. Geol. Surv. Techn. of Water Resour. Investigat., Book 7, Ch C2, 90. Kwaad, F.J.P.M., 1991. Summer and winter regimes of runoff generation and soil erosion on cultivated loess soils (The Netherlands). Earth Surface Processes and Landforms, Vol. 16, 653-662. Kwadijk, J.C.J., 1993. The impact of climate change on the discharge of the river Rhine. Thesis, University of Utrecht. Kwadijk, J. and H. Middelkoop. 1994. Estimation of the impact of climate change on the peak discharge probability of the River Rhine. Climatic Change. 27: 199-224. Licht, P.M., 1990. Beleidseffecten evaluatie binnen rivierbeheer (in Dutch). Thesis, University of Twente. Delft Hydraulics report Q1065/H472. Manabe, S. and Stouffer, R.J., 1980. Sensitivity of a global climate model to an increase of CO2 concentration in the atmosphere. J. Geophys. Res., 85: (C10), 5529-5554. Manabe, S., Wetherald, R.T. and Stouffer, R.J., 1981. Summer dryness due to an increase of atmospheric CO2 concentration. Climatic Change, 3: 347-385. McFarlane, N.A., G.J. Boer, J.P. Blanchet and M. Lazare, 1992. The Canadian Climate Centre second-generation general circulation model and its equilibrium climate. J. Climate, 5: 1013-1044. Middelkoop. H, 1991. Impact of climatic change on sedimentation on the bottomlands ("uiterwaarden") in The Netherlands. Progress-report GEOPRO91.023. Vakgroep Fysische Geografie Universiteit Utrecht. Middelkoop, H. & N.E.M. Asselman, 1993. Assessment of sedimentation rates on floodplains, a case study in The Netherlands. Poster presentation at the 5th International Conference on Fluvial Sedimentology, July 1993, Brisbane, Australia. Middelkoop, H., E.L.J.H. Faessen & H.J. Huizinga, 1992a. Historische morfologie, hydrologie en ecologie van de Waal tussen Pannerden en Nijmegen. Inundatie en landgebruik van de uiterwaarden en morfologie van het zomerbed tussen 1770 en 1830. Rapport GEOPRO-92.06. Vakgroep Fysische Geografie Universiteit Utrecht, (in Dutch). Middelkoop, H., N.J. van den Berg, E.L.J.H. Faessen & H.J.A. Berendsen, 1992b, Morfodynamiek van nevengeulen van de Waal: een historisch overzicht. Rapport GEOPRO-92.07. Vakgroep Fysische Geografie Universiteit Utrecht.
906 Middelkoop, H. & M. Deurloo, 1993. Geomorfologische en Historisch geografische waardering van het uiterwaardengebied rond Sint Andries. Toetsing van een inrichtingsschets. Rapport GEOPRO-93.13. Vakgroep Fysische Geografie Universiteit Utrecht, (in Dutch). Middelkoop, H., & W.P.A. van Deursen, 1993. Modellering inundaties uiterwaarden Case study Gelderse Poort. Rapport GEOPRO-93.01. Vakgroep Fysische Geografie Universiteit Utrecht, (in Dutch). Middelkoop, H. & H.J. Huizinga, 1992. Assessment of suspended sediment concentrations in the rivers Rhine and Waal during the high water period on April 2nd, 1988 using LANDSAT TM data. Rapport GEOPRO-92.01, Vakgroep Fysische Geografie Universiteit Utrecht. Middelkoop, H. & M. van der Perk, 1991. Een reconstructie van de opslibbing van uiterwaarden. Rapport GEOPRO-91.06. Vakgroep Fysische Geografie Universiteit Utrecht. Middelkoop, H., E.L.J.H. Faessen & H.J. Huizinga, 1992a. Historische morfologie, hydrologie en ecologie van de Waal tussen Pannerden en Nijmegen. Inundatie en landgebruik van de uiterwaarden en morfologie van het zomerbed tussen 1770 en 1830. Rapport GEOPRO-92.06. Vakgroep Fysische Geografie Universiteit Utrecht, (in Dutch). Nonhebel, S., 1987. Water use of Dutch forests: a simulation study. Report 7G, Studiecommissie Waterbeheer Natuur, Bos en Landschap, (in Dutch). Ogink-Hendriks, M.J., 1994. Modelling surface conductance and transpiration of a oak forest in The Netherlands. Agri. For. Meteor., (Submitted). Oude Essink, G.H.P., 1993. Effect of Sea Level Rise on the Groundwater Flow System through Amsterdam Waterworks and Haarlemmermeer polder, The Netherlands. Proc. UNESCO Conf. on Sea Level Changes and their Consequences for Hydrology and Water Management, Noordwijkerhout, The Netherlands. Parmet, B., 1993a. Impact of climate change on the discharge of the Rhine. Change 15: 1-3. Parmet, B. and M. Mann, 1993b. Influence of climate change on the discharge of the River Rhine - a model for the lowland area. IAHS publication 212: 469477. Peerbolte, E.B., J.G. de Ronde, L.P.M. de Vrees, M. Mann, G. Baarse, 1991. Impact of Sea Level Rise on Society. A case study of The Netherlands. Report GWAO 90-016. Rijkswaterstaat, Den Haag. Pomper, A.B., 1983. Geohydrological situation and observations on the hydrochemical groundwater situation of the western Netherlands. Geologie en Mijnbouw 62: 3/4. Roetter, R., 1994. Biophysical classification of the Rhine basin as a frame for land use projections. Volume 1 of Land use projections for the Rhine basin based on biophysical and socio-economic analysis. Winand Staring Centre-RIZA report 85.1. Roetter, R. and C.A. van Diepen, 1994. Biophysically-based analysis of possible climate change impacts on crop yield potentials and water use in the Rhine basin. Volume 2 of Land use projections for the Rhine basin based on biophysical and socio-economic analysis. Winand Staring Centre-RIZA report 85.2.
907 Roetter, R. and C.A. van Diepen, 1994. Biophysically-based, spatial analysis of possible climate change impacts on crop yield potentials and water use in the Rhine basin, volume 2. In press. Rotmans, J., 1990. IMAGE: An integrated model to assess the greenhouse effect. PhD Thesis, Kluwer, Dordrecht. Sch~idler, B., Spreafico, M., Bultot, F. and D. Gellens, 1992. Evaluation Wasserhaushaltmodelle. Vorstudie Nationales Forschungsprogramm 31: "Klima~inderungen und Naturkatastrophen". Stewart, J.B., 1988. Modelling dependence of surface conductance on environmental conditions. Agri. For. Meteor., 43" 19-35. Van Diepen, C.A., C. Rappoldt, J. Wolf and H. van Keulen, 1988. Crop growth simulation model WOFOST version 4.1, documentation. SOW-88-01. Center for World Food Studies, Wageningen, The Netherlands. Van Dijck, S.J.E., 1993. Palaeomagnetic research of the sediments from the dike burst pond in the river foreland of the Waal at Wamel (The Netherlands). Student report, Vakgroep Fysische Geografie Universiteit Utrecht. Van Dinter, M., 1993. Palynologisch onderzoek naar wielopvullingen in het Nederlandse rivierengebied. Student report, Vakgroep Fysische Geografie Universiteit Utrecht, (in Dutch). Van Dinter, M., H. Middelkoop, & B. Derks, 1992. Pollen analysis of dike burst ponds near Nijmegen, The Netherlands. Poster presentation at the 8th International Palynological Congress, Aix-en-Provence, 1992. Van der Drift, J.W.M., 1994. The impact of temperature and rainfall changes (climate change) on land degradation in source areas of the suspended sediment load of the Rhine. NRP Project no. 852089, Interim report no.l, Laboratory of Physical Geography and Soil Science, University of Amsterdam, 33 pp. Van der Drift, J.W.M., Middelkoop, H. and N.E.M. Asselman, 1994. Estimation of the effects of climate and land use change on the production of fine sediment by soil erosion in the catchment area of the river Rhine. Internal report, Laboratory of Physical Geography and Soil Science, University of Amsterdam, Department of Physical Geography, University of Utrecht, 12 pp. Veen, A.W.L. and Dolman, A.J., 1989. Water dynamics of forests: one-dimensional modelling. Progress in Physical Geography, 13: 19-35. Veeneklaas, F.R., L.M. van de Berg, D. Slothouwer and G.F.P. IJkelenstam, 1994. Rhine Basin study: Land use projections based on biophysical and socioeconomic analyses. Volume 4, Land use: past, present and future. Report 85.4, Winand Staring Centre (SC-DLO), Wageningen, The Netherlands. Viner, D. and M. Hulme, 1993. Climate change scenarios for impact studies in the UK: General circulation methods and applications for impact assessment. Climatic Research Unit, University of East Anglia, Norwich. Washington, W.M. and Meehl, G.A., 1983. General circulation model experiments on the climatic effects due to a doubling and quadrupling of carbon dioxide concentration. J. Geophys. Res., 88 (Cll): 6600-6610. Wateren-de Hoog, B. van der, & H. Middelkoop, 1992. Floods in the River Rhine and atmospheric circulation patterns. Rapport GEOPRO-92.02. Vakgroep Fysische Geografie Universiteit Utrecht. Wigley, T.M.L., T. Holt and S.C.B. Raper, 1991. STUGE, an interactive greenhouse model: Users manual, Climate Research Unit, Norwich.
908 Wischmeijer, W.H. and D.D. Smith, 1978. Predicting rainfall erosion losses - a guide to conservation planning. US Dept. of Agriculture, Agriculture Handbook No. 537. Wit, K, 1987. Wateraanvoerbehoefte Zuidhollandse Eilanden en Waarden (Fresh water requirements of the south-western island and polders of Zuid Holland), I.C.W. Nota nr. 1801, Winand Staring Centre, RIZA., Wageningen (in Dutch). Wolf, J, and Diepen, van, C.A., 1993. Effects of climate change on crop production and land use in the Rhine basin. In: Geijn, S.C., Goudriaan, and Berendse, F., editors. Climate change; crops and terrestrial ecosystems. Agrobiologische Thema's 9. 1993, CABO-DLO, Wageningen, The Netherlands. Wolf, J. and van Diepen, C.A., 1991. Effects of climate change on crop production in the Rhine basin. Report 52, Winand Staring Centre, RIZA, Wageningen.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
911
IMPACT OF CLIMATE CHANGE ON THE DISCHARGE OF THE RIVER RHINE
B. Parmet "), J. Kwadijkb) and M. Raak a) a)
RIZA, Institute of Inland Water Management and Waste Water Treatment Section Rivers, P.O. Box 9072, 6800 ED Arnhem, the Netherlands
b)
University of Utrecht, Department of Physical Geography P.O. Box 80115, 3508 TC Utrecht, the Netherlands
Abstract Climate change influences the water balance of drainage basins in several ways. In a project of the International Commission for the Hydrology of the Rhine basin the possible consequences for the discharge regime of the Rhine are investigated. In the first phase of this project detailed models have been developed and applied for selected sub-basins and a simple water balance model has been developed for the whole Rhine basin. Results are presented for climate scenarios assuming an increase in temperature of about 3~ and an increase in annual precipitation. The consequences of such a climate change are largest in the Alpine part of the Rhine basin, and are also considerable in other parts of the basin and for the basin as a whole. In general the Rhine changes towards a rain-fed river. The winter discharge increases, which can have consequences for safety, and summer discharge decreases with consequences for shipping, industry, agriculture and ecology.
1. Introduction
Climate change influences the components of the water balance of drainage basins in several ways. Precipitation patterns may change and because of a higher temperature also the timing and amount of snow fall and snow melt. Evapotranspiration is directly influenced by an increase in temperature. An increased CO2-concentration affects plant physiology and may lead to changes in water use too. Furthermore climate change may induce changes in land use, which is an important factor in evapotranspiration and runoff processes. Changes in these water balance components will of course affect the discharge. The River Rhine is economically the most important river of Western-Europe. Its drainage basin, see figure 1, stretches from the source in Switzerland to the mouth in the North sea, covering an area of 185.000 k n l 2 and is habitat to about 55 million people. The river is one of the most intensively navigated inland waterways in the world and is of major importance for the supply of water to large economically important areas. Changes in its discharge regime can have consequences for safety and for the water availability for shipping, industry, domestic use, agriculture, the natural environment and recreational purposes.
912 Against this background, the International Commission for the Hydrology of the Rhine basin (CHR) initiated a project to assess the consequences of climate and land use changes for the discharge regime of the River Rhine. Since a proper tool for this was lacking, the main purpose of the project was to develop a water management model for the whole Rhine basin. This model should be used to analyze the changes in average and extreme discharges and the effectiveness of mitigating measures. Several institutes from the Rhine riparian states cooperate in the project (Parmet, 1993). The Netherlands contribution to this project is incorporated in the Dutch National Research Program on Global Change (NRP).
0# \
120 km
o
\
........--"
,..,
,,) ,J
~~176 \
.J
-\(y :_ )
( "",3
Figure 1. Drainage basin of the River Rhine
2. Method
The model will be developed in two phases. In the first phase model development takes place along a bottom-up and a top-down approach. In the second phase both approaches will be combined. Along the bottom-up approach hydrological models are developed for selected sub-basins. These models are detailed in time and space and can be used to simulate the effects of climate and land use changes on average and peak flows. Sub-basins were selected in three distinct areas; the Alpine area, the Middle Mountains and the Lowland area. Different model concepts are used because the relevant hydrological
913 processes differ. Snow accumulation and melt, for example, is very important in the Alpine area and groundwater flow for the Lowland. Along the top-down approach a GIS-based water balance model is developed with a course resolution in time and space. This model can be used to study the sensitivity of the average discharge regime of the river Rhine and its main tributaries for climate change. The first phase of the CHR project is almost finalized. Along the bottom-up line, several detailed models are developed and also existing models are applied for representative subbasins. For the Alpine area an existing hydrological model, the IRMB model, was applied for several small drainage basins (Bultot, 1992, Sch/idler 1992). In the Middle Mountains area, a model will be developed for the Saar basin, a sub-basin of the Mosel. Results are not yet available. The Lowland model is developed for the drainage basin of the Overijsselsche Vecht (Parmet and Mann, 1993, Parmet and Raak, in prep). This model consists of a hydrological component, that is used to compute the daily evapotranspiration and discharge for sub-basins, and a flow-routing component, that combines the sub-basins and routes their discharges towards the mouth of the Overijsselsche Vecht. For the Rhine basin as a whole the water balance model RHINEFLOW was developed (Kwadijk, 1993). It is a simple water balance model based on a Geographical Information System.
3. Results 3.1. Effects of climate change in representative basins in the Alpine area With the IRMB model the effects of a climate change for several components of the water balance were simulated for three drainage basins, Murg, Ergolz and Broye. A climate scenario as defined by Bultot was applied (Bultot, 1988). The monthly temperature and precipitation changes are given in table 1. Changes in physiological behavior of plants were not taken into account. Simulations with changed climate were carried out for the period 1981 to 1988. The simulations showed an increase of annual potential evapotranspiration of 10%. Actual evapotranspiration increased somewhat less because during the summer period there is a slight decrease in soil moisture. Discharge increased during the winter period with 10%. This is due to the fact that the amount of winter precipitation increases and less precipitation is stored as snow. Furthermore the accumulated snow melts faster. The duration of the snow cover decreases considerably, especially below an altitude of 1500 m. Discharge in summer decreased with about 15%. This follows from a reduced contribution of melt water, a larger evapotranspiration and a slight decrease in precipitation. The total annual discharge hardly changes. The daily maximum discharge increases and the daily minimum discharge decreases. 3.2 Effects of climate change in representative basins in the Lowland area The climate scenario that was used for the Lowland area was based on a method developed in the framework of the NRP (Klein Tank and Buishand, 1995). The monthly changes in temperature and precipitation are given in table 1. Compared to the other scenarios used in this study, this scenario is rather wet. Computations were carried out for the basin of the Overijsselsche Vecht for the period 1965-1990. Changes in plant physiological characteristics were taken into account.
914 An increased CO2-concentration influences plant physiology. For most plants the water use efficiency increases and the biomass production increases. An increase in temperature for the temperate zones generally leads to an increase in production too. Present knowledge indicates, for doubled CO2-concentrations and an increase in temperature of about 1.5~ a small decrease in evapotranspiration for most crops and forests (Roetter en van Diepen, 1994; Hendriks, 1994). Based on this knowledge, plant physiological parameters were provisionally adapted in the Lowland model (Parmet and Raak, in prep).
Table 1 Monthly temperature (T) and precipitation (P), used for representative Alpine basins and for the representative Lowland basin Month
J
F
M
A
M
J
J
A
S
O
N
D
Talpine, ~ Palpine, % 1)
3.1 10
3.4 14
3.4 11
3.1 10
2.8 -1
2.7 -2
2.5 -2
2.3 -2
2.3 0
2.7 6
2.8 10
3.2 10
Tlowland, ~ Plowland, %
3.0 21
3.0 20
2.3 15
2.3 13
2.3 5
3.7 12
3.7 11
3.7 9
3.4 3
3.4 8
3.4 19
3.0 18
1) The percentual change is an average for the three basins Murg, Ergolz and Broye (Sch~idler, 1992)
700
E d L.. ~m
6oo 500
t-
.-
400
E m E
300
E
2oo
r-
100
U
f
f
f
~
Reference --
- Scenario
r
< 0
2
i
I
i
I
I
I
5
1o
.50
1 oo
200
500
Reccurence
1250
period, [years]
Figure 2. Discharges for different recurrence periods for reference and scenario conditions, as computed with the lowland model for the drainage basin of the Overijsselsche Vecht.
915
Simulations for the lowland showed an increase in actual evapotranspiration of about 7%. The CO2 effect on water use does not compensate for the increase in temperature of 3 ~ The annual discharge increased with 22%. Winter discharge increased with 25%. The increase in evapotranspiration in summer does not exceed the increase in precipitation. Discharge in summer increased with 19%. The maximum discharge increased considerably. The distribution of annual maxima can be described using a Gumbel distribution (Mendel, 1993). According to fitted Gumbel functions for the reference (r 2 = 0.98) and scenario (r 2 = 0.97) simulations, peak flows with different recurrence periods change as indicated in figure 2. River dikes in the Netherlands are designed for a discharge with a recurrence period of 1250 years. The figure shows an strong increase of 26% in this design discharge for the river Vecht. 3.3 Effects of climate change, Rhine basin Consequences for the whole Rhine basin have been computed with the RHINEFLOW model. The sensitivity of the discharge regime was examined with a wide range of climate scenarios for the period 1956 to 1980 (Kwadijk, 1993). Here the results of computations with one scenario, the so-called BAU-best scenario, are presented. The scenario is based on the IPCC Business as Usual scenario (IPCC, 1991). It is given in table 2. Changes in land use and physiological characteristics of plants were not taken into account.
Table 2 BAU-BEST scenario for temperature (T) and precipitation (P), for different parts of the Rhine basin (Kwadijk, 1993) Year
Summer
Winter
Part of Rhine basin
T, ~
P, %
T, ~
P, %
T, ~
P, %
North Middle South
3.5 3.5 3.5
11 8 7
2.9 2.9 2.9
4 -1 -4
4.3 4.2 4.1
19 19 19
For the Alpine part of the Rhine basin, the changes as computed with RHINFLOW have the same direction as the results for the representative Alpine basins. As can be seen from figure 3, winter discharge increases. This is caused by increased precipitation and snow melt. During summer the discharge decreases due to a smaller contribution of melt water, increased evapotranspration and a slight decrease in precipitation. The increase in winter discharge is much larger than for the representative basins, upto 100% with an average of 60%. This can be explained partly from the used scenarios. Both the increase in temperature and in precipitation is smaller for the Bultot scenario compared with the BAUbest scenario. Furthermore it can be explained by differences in model components, especially the snow component, and off course the considered area is not the same. The changes during summer are comparable, both for the alpine area as a whole and for the
916 representative basins in the alpine area, a decrease of about 15% was computed. The changes for the area downstream, the Middle and Lowland part, are much less pronounced. The discharge increases during winter and spring and decreases during summer and autumn, as can be derived from figure 3. The increase in evapotranspiration causes the soil water deficit to increase. As a result, summer discharge decreases, but because part of the winter surplus is stored as groundwater, not untill July. The water surplus during autumn is partly used to replenish soil water, which explains the decrease of discharge during autumn. Because the scenario used for the representative basin for the lowland is wetter, especially for the summer period, than the BAU-best scenario, the changes in discharge are not directly comparable. 120
100
o~~ 80 ,.._,
Alpine part -- - Middle & lowland part
40
- - - Outlet basin 20 l-9
~
R li"~
~
~
~
.c
0
0 -20 -40
o>
Z
~
O
~
~
~LL
~
~
~_
,
.E "1o
X
X
X
X
X
:
::::i~:::::: ........ ::::::::::::::::::::::
::iii.i:...
.=
' ~.-
:i:~i:iJ:i::~
~..
X Arnhem
Variksche Plaat
9
.
J~::.-|~
o_
Lobith
....... ' ~ I ~ ~ r
~ ~
~
~ ~
-
f "
~se .... K e e n t /
o .....!..
Nijmdt]en
..~_~ --
X
.:::, o
~o ~m
~''";~':=
x = method applied at floodplain section [ ij~iii~i~i~i~!~i] floodplains
Figure 1. Study area and location of investigated floodplain sections. 2. FLOODPLAIN
GEOMORPHOLOGY
AND
GENESIS
The genesis and geomorphology of the embanked floodplains were investigated by geomorphologic mapping and corings, and by analysis of old river maps and historic records of water levels. Old river maps provide a rough indication of the beginning of sedimentation on the enclosed floodplains. The present embanked floodplains along the rivers Rhine and Meuse consist of point bars of meandering rivers as well as lateral bars separated by side channels. Many lateral bars were formed only during the last two to three centuries. Lateral bars are mainly found along the river Waal. The development of lateral bars is related to land reclamation from the river bed. Using old maps, a succession scheme was developed showing the stages of development of lateral bars, side channels and vegetation (Middelkoop et al., 1992). Assuming that the yearly sedimentation rate is proportional to the product of inundation time and suspended sediment concentration it was found that at the beginning of the floodplain formation the sedimentation rates were 3 to 4 times as great as at present, and varied between 1 and 3 cngyear.
933 3. R E C O N S T R U C T I O N OF F L O O D P L A I N S E D I M E N T A T I O N RATES USING HEAVY METALS AS A T R A C E R The river Rhine sediments that are deposited on the embanked floodplains in the Netherlands are contaminated with pollutants, including heavy metals. The heavy metals in the floodplain soil profiles were used as a tracer to calculate floodplain sedimentation rates. Sediment accumulated in dike-breach ponds was analysed to reconstruct changes in heavy metal pollution of the river Rhine sediment during the past 200 - 300 years. From historic dates, Pb-210 dates, palynological information and variations in sediment compaction, a timedepth control of the sediment fill of the dike-breach ponds was obtained. The reconstructed changes in the heavy metal pollution of the Rhine sediment are shown in figure 2. The heavy metal pollution increased during the first half of this century; maximum pollution occurred around 1960; since 1970 the heavy metal pollution has strongly decreased. From various sections of the embanked floodplains, samples from vertical soil profiles were collected and their heavy metal content was measured. The samples were taken from sites with different inundation frequencies, local elevations, and distances to the main channel. The heavy metal content of floodplain soils was related to these factors. Average sedimentation rates during the past decades were reconstructed by comparing the heavy metal profiles in the floodplain soils with the pollution history of the Rhine. The heavy metal profiles obtained from the floodplain soils generally have the same shape as the pollution curve reconstructed from the dike-breach pond sediment. Depending on the sedimentation rate, the vertical profile is more or less stretched, and the total heavy metal content varies (figure 3). The total soil pollution can be very different from the pollution in the upper 10 cm. The sedimentation rates for the past decennia that were estimated from the heavy metal profiles range between 2 and > 15 mm/yr.
(Jrl 0 0
0 0 0
._L 01 0 0
I~ 0 0 0
tO 0 0
0 1980 1
1960 I
1940 2
...... ,~-;~- ....... ..i-:........... J,........... 4............ 1920 1900
3 average sedimentation rates (mm/yr):
,50 4 o "13 :3"
2-4 -
15-18
s
~s
v3
o 0
o 0
o 0
200
9
i
Cd, Cu, Pb (mg /kg )
Figure 2. Heavy metal pollution history of the river Rhine reconstructed from a dike-breach pond along the river Waal.
Figure 3. Zinc profiles of floodplain soils as a result of different sedimentation rates.
934
4. PRESENT FLOODPLAIN SEDIMENTATION RATES Present floodplain sedimentation rates and their spatial variability were measured after the two floods of 1993 and one flood in 1994 using a total of about 800 sediment traps of artificial grass. Measurements from the individual traps were interpolated using block-kriging to create raster maps of sediment accumulation (figure 4). The patterns shown on the maps were correlated with floodplain morphology and sedimentation mechanisms (Asselman & Middelkoop, in press.).
0,
....
>1.0 kg/m 2 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 500 m
....~:~;!!!!!!i!i;ii~.:.. ===================================== 9 . ~ .-.
>9.0 k g / m 2 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0
sand sheet
B. K e e n t - D e c e m b e r 1993.
A. K e e n t - J a n u a r y 1 9 9 3 .
5.5 k g / m 2
2.4 kg/m 2 1.8 9~ i
'iiii!ili%~.iiiiiiiiii!iiii!:' ....
4.5 5.0 4.0 3.5 3.0
i .......-:i:i:i" "
1.5
1.2 0.9 0.6 0.3 0.0
i
C. VarikschePlaat- January1993.
2.5
:.:.!
, iiiiiil}iiP"i iiiiiiiIiiii} 500 m
:.
..-.-.:,-.:.:.:.-.'..-,:.-. D. V a r i k s c h e P l a a t - D e c e m b e r
1993.
Figure 4. Sediment accumulation after a single flood, measured by sediment traps. January 1993: minor flood; December 1993: extreme flood.
During the flood, sand is locally deposited directly behind a levee. The amount of deposited (fine) sand decreases exponentially with distance from the levee. This pattern agrees well with existing diffusion models. At distances larger than 50 to 100 m from the river channel the accumulated sediment consists mainly of material < 53~. Here, differences in sediment accumulation are determined by the local topography, which causes differences in inundation times and ponding (figure 4c). The total amounts of suspended sediment < 531.t deposited in the investigated floodplain sections during the high flood of December 1993 ranged between 1.20 and 3.98 kg/m 2 along the river Waal, and between 1.0 and 2.0 kg/m 2 in the study areas along the river Meuse. Equivalent thicknesses of sediment accumulation during the flood ranged between 0.8 mm and 3.2 mm in the central parts of the floodplain sections. The estimated total amount of suspended sediment deposited during the flood on the entire river Waal floodplain was 0.24 Mton. This is about 19% of the total suspended sediment load transported through the river Waal in the same period. It is 7.7% of the average yearly load of suspended sediment transported by the river Rhine into the Netherlands. Comparison of sediment accumulation of different floods shows that the amount of sediment deposited on a floodplain increases less than proportionally with the flood magnitude.
935 5. I M P A C T OF C L I M A T E CHANGE ON FLOODPLAIN SEDIMENTATION RATES 5.1 L o c a l s e d i m e n t a t i o n rates
Sedimentation on a small floodplain section was simulated using the 2-dimensional WAQUA-DEWAQ model (Ubels, 1986). For a series of stationary discharge stages, the average yearly sediment accumulation was calculated by multiplying the sedimentation rates by the associated frequency of occurrence and suspended sediment concentration. The total yearly sedimentation is the summed total of the products. Sedimentation rates under changed climate conditions were calculated by using the BaU flow duration curve (Kwadijk & Middelkoop, 1994) and using the sediment rating curves provided by Asselman (1994). The WAQUA-DELWAQ model demonstrates that sedimentation becomes very ineffective at high discharges (figure 5). Preliminary results indicate that under the BaU climate scenario the sedimentation rate in the test area increases by only 3% if both climate and land use change. A climate change alone, however, increases the sedimentation rates by about 50%. Sedimentation rates will increase much more on floodplain sections behind a minor dike and where sedimentation only occurs during relatively high discharges.
tonh
kg/m2/yr
reference scenario 8.00
i i [ [ i ........i - - - - ~ ..................................
present auton, land use change only
6.00
4.00
climate scenario =
BaU-climate
.t
BaU-climate + land use
2.00
0.00 3500
4500
5500
6500
7500
.
o o
8500
discharge (m3/s)
Figure 5. Effective sedimentation on a low floodplain section for different discharges.
.
o o
.
o o
.
.o o
.
o o
o o
~
~
o o o
o o o
Rhine discharge (m3/s)
Figure 6. Sediment load transported over the embanked floodplains for different scenarios.
5.2 Sensitivity of large scale potential s e d i m e n t a t i o n rates
The potential sensitivity of the sedimentation rate on the entire embanked Waal floodplain for a climate change was investigated from changes in the annual amount of river sediment load that is transported over the embanked floodplain. The 1-dimensional SOBEK model (Huyskens & Barneveld, 1994) was used to calculate for a series of discharge stages the percentage of the river discharge transported over the embanked Waal floodplain. At each stage, the sediment load transported over the floodplain was calculated from the product of (1) the discharge, (2) the suspended sediment concentration and (3) the relative frequency of occurrence. The summed totals of the product for all discharges gives the average yearly load,
936 expressed in Mton/yr. This is the amount of sediment is potentially available for deposition. The effect of the BaU discharge scenario and four sediment rating scenarios on this sediment load was calculated to investigate the possible impact on floodplain sedimentation. The BaU climate scenario will increase the average sediment load transported across the floodplains by a factor 3.5 compared to the present situation and by a factor 3.2 compared to the situation with autonomous land use changes (figure 6). The changes in effective sediment deposition will be smaller than the changes in the sediment load. 6. CONCLUSIONS The present floodplain geomorphology comprises both pointbars of meandering rivers and lateral bars that have been formed during the past centuries. Floodplain sedimentation depends on floodplain characteristics, discharge frequency distributions and sediment concentrations, and therefore shows a high variability in time and space. Present floodplain sedimentation rates range between 0.5 and 15 mm per year. The effect of the BaU climate change on floodplain sedimentation are considerable. High discharges will occur more frequently and sediment concentrations are expected to increase as a result of climate change. Under the BaU climate scenario, the yearly amount of sediment transported over the embanked Waal floodplain is more than three times as large as under present climate conditions. The increase of the effective sedimentation rates will be smaller, depending on the type of floodplain section. Sedimentation rates are expected to increase at least by about 50% on floodplain sections directly bordering the main channel and without a summer dike. The sedimentation rate on floodplains that are situated behind a summer dike is expected to increase more than 50%. If also the effect of changes in land use are taking into account, changes for low lying floodplains are insignificant. Nevertheless, the average sediment load transported over the entire floodplain increases still by a factor 2.8.
REFERENCES
Asselman, N.E.M., 1994: The impact of climate change on suspended sediment transport in the river Rhine. Department of Physical Geography, Universiteit Utrecht. Asselman, N.E.M. & H. Middelkoop, in press.: Floodplain sedimentation: Quantities, patterns and processes. Accepted for publication in: Earth Surface Processes and Landforms. Huyskens, R.B.H. & H.J. Barneveld, 1994: Sobek-model voor de Nederlandse Rijntakken. Rapport Q 1895, WL. Kwadijk, J.C.J., 1993: The impact of climate change on the discharge of the River Rhine, Thesis. Universiteit Utrecht. Kwadijk, J.C.J. & H. Middelkoop, 1994: Estimation of the impact of climate change on the peak discharge probability of the River Rhine. Climatic Change, Vol.27, issue 2, pp. 199-224. Middelkoop, H., N.J. van den Berg, E.L.J.H. Faessen & H.J.A. Berendsen, 1992: Morfodynamiek van nevengeulen van de Waal: een historisch overzicht. Vakgroep Fysische Geografie, Universiteit Utrecht, Rapport GEOPRO 1992.07. Middelkoop, H., 1994: The impact of climate change on the sedimentation rates on the embanked floodplains in the Netherlands. Report of the NRP-project NOLK/002/90. Department of Physical Geography, Universiteit Utrecht. Ubels, J.W., 1986: Verantwoording van het hoogwateronderzoek op de Boven-Rijn, de Waal, het Pannerdensch Kanaal, de Neder-Rijn, de Lek en de IJssel. Nota nr. 86.035. DBW/RIZA, Arnhem.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
937
The impact of climate change on suspended sediment transport in the river Rhine Nathalie E.M. Asselman Department of Physical Geography, Utrecht University, PO Box 80.115, 3508 TC Utrecht, The Netherlands
Abstract Erosion, transport and deposition of fine suspended sediments are both directly and indirectly influenced by climate conditions. In this study, the suspended sediment transport regime of the river Rhine under present and future climate conditions was assessed. The impact of climate change on the sediment transport regime was investigated using sediment rating curves in combination with flow duration curves, developed using the BaU-climate scenario, and three sediment production scenarios. The results indicate that a climate change under changed land use conditions will result in a 14% increase in total annual suspended sediment load. A larger part of the yearly sediment load will be transported at discharges over 4000 m3/s. This probably results in increased floodplain sedimentation rates.
1. INTRODUCTION The increased emission of C O 2 and other greenhouse gases during the 20th century, is expected to enhance the greenhouse warming of the lower atmosphere. This may cause a world-wide climate change in the forthcoming decades, resulting in changes in temperature and precipitation. The climate induced changes in vegetation cover and water discharge in turn will affect erosion, transport and deposition of suspended sediments in the river Rhine. Within the scope of the National Research Program (NRP 1) the impact of climate change on discharge, and the suspended sediment transport regime of the river Rhine was studied. The IPCC "Business as Usual" (BaU) scenario projected on the Rhine catchment, was used as climate change scenario (Kwadijk, 1993). The aim of this study is to investigate the processes of sediment transport through the river Rhine under actual climate conditions, and to assess the effect of climate change on the suspended sediment transport regime, depending on changes in discharge and sediment supply to the rivers.
938 2. M E T H O D S
Since wash load is a non capacity load, the amount of fine suspended sediment transported by the river Rhine depends on the availability of loose material and to a lesser extent on the capability of the river to transport this material. Therefore, sediment transport rates cannot be calculated using stream power related transport formulas. Instead, the so-called rating curve technique can be used. A sediment rating curve describes the average relation between discharge and suspended sediment concentration. The most commonly used relationship between discharge and suspended sediment concentration is a rating curve in the form of a power function (a.o. Walling, 1974). In this study a power function with additive constant term was used: c = /9 + a . Q b where c is suspended sediment concentration (mg/1), Q is river discharge (m3/s) and a, b and p are regression coefficients. Sediment rating curves were developed for five gauging stations along the river Rhine (figure 1), using daily discharge and suspended sediment concentration data, measured by the Bundesanstalt fir Gewisserkunde (BfG), Germany.
iiiiiii iiii!iiii \
o
,00k=
ERMANY ~ / /
t...i,
o
..
N .
ANDERNACH-WEISSEI~THURM~ ~ k~.,/..~-"'--" ,J 2'/"
KAuB
~
....I,..'
"t. 50 >50 10-25 5-10
++ ++ n.a. ++ +
+ + ++ + -
Transportation Hydrogen Methanol Ethanol Electric vehicles RME
10-25 10-25 5-10 10-25 5-10
+ + -
+ + -
Residential & commercial Insulation Hydrogen Heatpumps Efficient appliances
10-25 10-25 10-25 < 5
++ + ++ ++
++ + ++ ++
5-10 5-10 5-10 5-10 5-10 5-10
++ ++ ++ ++
+ ++ ++
Industry More CHP More natural gas Heatpumps Hydrogen CO 2 removal Savings ++ + n.a.
= = = =
achieves maximum potential achieves limited potential not applied o p t i o n is n o t a v a i l a b l e in t h i s s c e n a r i o
+ ++
++
++
+
+
+
+
++
++
++
++
1066
Table 1 shows an overview of reduction options in different scenarios. A general cost figure in DFL per tonne CO e for each option is not available, as the costs and the CO 2 reducing potential depend on the scenario conditions. For example the reducing potential for electricity generating or consuming technologies and for combined heat and power generation (CHP) are largely determined by reference technologies and load patterns. As these conditions vary between scenarios, the attractiveness varies accordingly. Cost figures per tonne CO 2 in the literature should thus be considered with care, as scenario conditions determine their validity. The general picture from table 1 shows the greatest cost-effective potential in electricity generation and in the residential and commercial sector. Shifts in the transportation sector prove to be very costly, while the potential for shifts in the industry is limited (at least concerning energy related options in the industry, integrated chain management shows a very different picture, see section 4). Conversion savings like e.g. COe-free hydrogen and methanol production are in table 1 allocated to final consumption. The potential in table 1 is only an indication; these figures cannot be added straightforward as reduction options show interaction (e.g. through limited CO 2 storage potential, see figure 3).
3. I N T E G R A T E D R E D U C T I O N OF G R E E N H O U S E G A S E S
The sensitivity of emission reduction results from consideration of non-CO e GHGs (CH 4, N20, CO and halocarbons), was studied with an extended MARKAL database. Emissions of GHGs which occur outside the Netherlands, but which are related to the Dutch final energy use were also included. The upstream GHG emissions include emissions from mining, processing and transport of energy carriers. Such system boundaries differ from the ones commonly used for national emission accounting, but they coincide with emission definitions in full fuel cycle analysis and life cycle analysis. The warming impacts of emissions of different GHGs were compared using the Global Warming Potential (GWP) concept. Incorporation of non-CO e GHGs and upstream GHG emissions in the analysis appears to affect the effectiveness of reduction options. Total upstream CO 2 emissions and non-CO 2 GHG emissions account for 10-15 % of total energyrelated GHG emissions. Upstream COe emissions and CH 4 emissions are dominant. The impact of other greenhouse gases on the optimisation was analysed, using a CO 2 "penalty". In the penalty concept, CO 2 emissions are valued externally with a fixed sum per tonne CO e. COe emissions are minimised again in a cost-effective way. Table 2 shows the contribution of groups of options to emission reduction in two approaches. In the 'only direct CO 2' approach the non-CO 2 GHG emissions and the upstream emissions have been neglected, while in the 'all GHG' approach these emissions were included. At two emission penalties (100 and 200 DFL/tCO2), CO 2 removal at coal-fired facilities appears to reduce less direct COe emissions than in the 'all GHG' approach. On the other hand, renewables play a more important role in the 'all GHG' approach. For most other options, such as end-use savings and efficiency
1067 improvements the results are less sensitive to the inclusion of non-CO 2 GHG and upstream GHG emissions.
Table 2 Contribution of options to reduction of direct CO 2 emissions in cost-optimal emission reduction strategies in 'all GHG' approach and in 'only direct CO 2' approach (DZ scenario, 2030).
Savings on end-use Savings in conversion Fossil fuel substitution CO 2 removal, coal-fired CO 2 removal natural gas-fired
100 DFL/tCO 2 penalty all GHGs only direct CO2 16.0 15.3 21.5 22.5
Renewables
Total Reduction
200 DFL/tCO 2 penalty all CHGs only direct CO2 20.7 20.1 22.1 22.9
10.7 27.5 0.0
9.1 33.9 0.0
0.0 30.9 39.9
0.0 35.1 29.7
8.4
5.8
18.7
14.9
84.2
86.7
132.4
122.7
140 ..J
uJ
>uJ J o
upstream
CO2
120 s~
~ "~ "~ ~" "~ " " ' " ~ ~ " " ' "
~ ~ ~ ,_ --',,Z,..X. ... " " ' " " ' " - . .
o
r 100 """ ".~.
W 0
" ~.d ~ ~,.,~,,,,
wZ rr
w
"~"...... "",,
.....'"
..'"" 9......
'~'.. \".....
...,"" ",
\
,,,,
80
".....
\\
\\
"'""-..
halocarbons .........................
w IT IT
oLL
60
"~
-
_oz'Z2040 ~ _
\
x\
'"",...
Q
O9 CO __ W
0
0
i 20
i 50 EMISSION
i 100 PENALTY
I 200
i 500
I 1000
[DFLII'CO2 equivalent]
Figure 4: Indexed emissions for various greenhouse gases at different emission penalties (DZ 2030).
The emission levels which resulted from the enforcement of penalties are shown in figure 4 for the year 2030, indexed to the emission level in the reference case. Note that the horizontal axis has a logarithmic scale. As expected the levels of direct CO 2 emissions decrease with rising emission
1068 penalties. The gradual reduction is achieved by a mix of options, with prominent roles for energy saving, savings in conversion, CO 2 removal and renewables. Upstream CO e emissions show an initial increase, but decrease at emission penalties above 200 DFL/t CO e. The increase is caused by shifts towards more coal with CO2 removal for power generation at emission penaties between 100 and 200 DFL/t CO 2. Coal production shows relatively high u p s t r e a m CO 2 emissions. The path of the CH 4 emissions is partly a result of specific CH 4 a b a t e m e n t measures, such as technical measures at offshore gas production, and the path is partly a result of changes in the fuel mix. At the lowest penalties (20 and 50 DFL/tCO 2) CH 4 emissions will be reduced by measures at gas production facilities and by a reduced coal consumption. The strong emission decrease at 100 DFL/tCO 2 is a result of a move away from certain coal types and n a t u r a l gas imports which are linked with high production emission levels. The alternatives, surface-mined coal and natural gas transported through high technical standard pipelines, have lower CH 4 emission levels. Replacement of cast-iron natural gas distribution networks is attractive at 200 DFL/tCO e. The increase of CH 4 emissions at 500 DFL/tCO 2 results from the increased consumption of natural gas which is mainly used for hydrogen production. The emissions of halocarbons show a peak at 175 DFL/tCO2, caused by an increased use of heatpumps. This is offset at higher penalty levels by improvements in cooling devices that reduce halocarbon emissions.
4. CO 2 E M I S S I O N R E D U C T I O N IN T H E I N T E G R A T E D E N E R G Y A N D MATERIALS SYSTEM
While CO 2 is generally considered as an energy related problem, this depends on the point of view. For the Netherlands, industrial materials production is responsible for approximately one third of the national CO 2 emissions (50-60 vs. 160 Mt). This part of the CO 2 emissions can be influenced by changes in the materials system. The environmental impacts of energy systems (energy production and consumption) and materials systems (materials, products and waste materials) are closely related. Oil is used as feedstock for plastics, waste is incinerated for energy recovery. Wood can either be used as construction material or energy carrier or in a sequence of both applications. An integrated approach for both systems should enable the identification of ways to reduce CO 2 emissions with lower costs. The existing energy system model was extended to represent the materials system. The model describes the whole Dutch materials system, and it includes all processes "from cradle to grave"; figure 5 shows the materials system model structure. All material flows are modeled that are related to end-use of materials in products in the Netherlands.
1069
1 Primary production
2 Recycling [
Material
3 Product Assembly
Product
4 Product Use
5 Removal & separation
6 Energy recovery
Waste material 7 Disposal
Figure 5: Materials system model structure.
A large effort was put into the characterisation of 29 materials, 20 product groups and 30 waste materials and some 200 processes which link the material flows. Appendix 1 shows the relation between materials and products. CO 2 reduction options in the materials system include: industrial energy savings; - CO e removal from industrial plants and storage; reduction of materials consumption (e.g. re-usable packaging); materials substitution; biogenous fibre materials; improved waste collection and separation systems; waste recycling, cascading and energy recovery. Figure 6 shows the model results for CO e emissions in the base-case (no CO 2 reduction). The materials system that is defined on the end-use principle is again responsible for approx, one third of the CO e emissions from the energy system (with national boundaries). This is important, as large Dutch industrial CO 2 emissions are generally dismissed as being related to exports. These results prove however t h a t these export-related CO 2 emissions are offset by importrelated CO 2 emissions. The emissions from the materials system (M) are stabilised in time, while the total emissions from the energy system (E) increase. On one hand, this stabilisation is caused by improved efficiency and recycling; on the other hand dematerialisation plays an important role. -
-
-
-
-
-
1070 [Mt CO2/YEAR] 250
ENERGY SYSTEM (E)
200
MATERIALS SYSTEM (M) ......
150
100
50
0
I
2000
I
I
2010
I
I
2020 [YEAR]
I
I
I
2030
I
2040
Figure 6: Base case CO 2 emissions for the energy system (E) and the materials system (M) (DZ).
[Mt CO2/YEAR] 200
RENEWABLES
150
~
WITH CO2-REMOVAL
~
OTHERCO2-REMOVAL
H2 FROMNAT.GAS
MAINLY COAL FOSSIL FUEL SUBSTITUTION
iiii!!ii:i!ii:ii:i:i!ii:i:i}ii!i!i:i:i:i!i::!ii!
i! m
100
i[
~TE~,~,s O~T,O~S
I ~176176
s~'~s
50
constant 20 40 60 CO2 EMISSION REDUCTION [%]
80
Figure 7: Emission reduction allocation for the integrated energy and materials system (DZ 2030).
1071
Considering emission reduction in the materials system results in significant cost reduction. The long term marginal CO 2 reduction costs decrease by NLG 50-100 as costly reduction options in the energy system can be avoided. Figure 7 shows the structure of emission reduction options in the integrated energy and materials system. Comparing figures 3 and 7 shows what type of CO 2 emission reduction options in the energy system can be avoided at certain reduction targets. Generally speaking, savings in conversion and end use are reduced. The largest shift is however related to CO 2 storage. The storage capacity is limited. As more CO 2 reduction can be achieved in the integrated energy and materials system at certain costs without storage, less storage per P J is required. The consequence is that the limited storage capacity can be used less effectively, but at lower costs. The 'hydrogen economy' (hydrogen from natural gas with CO 2 removal) is introduced later, while CO 2 removal at coal fired power plants is still used at higher reduction targets compared to the results for the stand-alone energy system.
INCREASE COMPARED TO BASE-CASE [%] CEMENT E'~ STEEL PLASTIC ~/~ WOOD ALUMINIUM
60 40 20
~20 -40
"
i
2000
i
i
2010
i
i
2020
I
I
i
~
20:30
[YEAR] Figure 8: Shifts in materials consumption due to CO 2 emission reduction (DZ -6O%).
As a result of CO 2 reduction, materials in products are substituted. Some options for reduced materials consumption (e.g. in packaging) are already included in the baseline; other environmental policies may cause such a shift.
1072 Figure 8 quantifies material substitution effects due to CO 2 reduction. The main shifts are in the construction and transportation areas. Traditional brick/concrete buildings are replaced by wooden skeleton buildings. The energy consumption per tonne for brick and cement production is relatively low, compared to other materials. The relatively high CO 2 emission for traditional buildings is caused by the large amount of materials that is required per house and because of inorganic CO 2 emissions from cement production. In the transportation sector, cars and trucks shift towards more aluminium and plastic is used instead of wooden pallets and crates. The fuel savings due to light weight constructions are in this area the main drive. The net result of materials substitution is a decrease in the use of cement, while the use of wood and aluminium increase after 2015. The use of steel and plastics remains constant. These results prove to be very sensitive to assumptions concerning assembly costs for different product options. The impact of 60% CO 2 reduction on product life cycle costs is generally below 10 %. If e.g. assembly costs for an aluminium car are 15% higher as for a steel car, the shift from steel to aluminium is not cost effective. The uncertainty range in future production costs is however in this order of magnitude. This problem occurs for most products. Other shifts in the materials system occur in materials production due to shifts from one production technology to another and occur also in waste management. For some materials recycling is favoured (e.g. plastics), while for others (elastomeres, biogenous fibre materials) incineration seems the best solution. These shifts are not discussed in further detail in this paper.
5. C O N C L U S I O N S
The model calculations for the stand alone energy system indicate that significant CO 2 emission reduction is possible. Changes in the residential and commercial sectors (conservation, high efficiency equipment such as heatpumps, etc.) and in electricity generation (fuel switching, cogeneration, etc.) appear more cost effective than those in industry and transport. Significant savings, mainly in the residential and commercial sectors, are still possible but will need to be supplemented by measures in the supply sectors to reach more ambitious targets. CO 2 removal and storage options are relatively cost effective, but are to be considered as transient towards more sustainable configurations only. Biomass and wind provide relatively cheap renewable energy, but have limited potential. Photovoltaic solar energy could serve as backstop technology only: large potential but high costs. The preferred measures are not significantly influenced by taking upstream CO2 and other greenhouse gases into account. Including upstream CO2 and m e t h a n e emissions makes a noticeable difference, but nitrous oxide, carbon monoxide and halocarbons are less important when assessing future Netherlands energy systems. Production of materials is associated with considerable energy use. Today it constitutes approximately one third of the Netherlands CO e emissions. Changes in material flows and material systems technologies appear important
1073 contributors to cost effective C02 reduction strategies. Model calculations indicate an increased use of aluminium and wood, while the use of cement decreases. This is caused by changes in the construction and transportation sectors. Interaction between the materials system and the energy system at large are shown to be of importance and require further attention. The integrated assessment of energy and materials systems reveals more cost effective options than were found in the energy system alone. Moreover, options within the materials system are often truly sustainable, an important feature supporting current long term policy strategies. International trade issues complicate the development and implementation of integrated chain management policies, especially for open, trade oriented economies like the Netherlands. Therefore extending the coverage of this type of studies to the European level is required. This would also provide valuable extra insights into the interactions between a broader array of energy system configurations and materials systems.
6. R E F E R E N C E S
1 P.A. Okken, T. Kram, P. Lako, J.R. Ybema, J. van Doorn, D. Gerbers: Drastische CO 2 reduktie - Hoe is het mogelijk. (Drastic CO 2 reduction) ECN-C--92-066. Petten, J a n u a r y 1993. 2 P.A. Okken, J.R. Ybema, T. Kram, P. Lako, D. Gerbers: Energy systems and CO 2 constraints. ECN-C-93-014. Petten, March 1994. 3 J.R. Ybema, P.A. Okken: Full fuel chains and the basket of greenhouse gases. ECN-C-93-050. Petten, December 1993. 4 D.J. Gielen, P.A. Okken: Optimisation of integrated energy and materials systems. ECN-C--94-010/011/012. Petten, June 1994. 5 T. Kram: National energy options for reducing CO 2 emissions, Volume 1: the international connection. A report of ETSAP/Annex IV. ECN-C--93-101. Petten, December 1993. 6 T. Kram et. al.: Koleninzetstudie (KIS). (Coal use study). ECN-C--91-072. Petten, November 1991.
1074
A P P E N D I X 1: M A T E R I A L REQUIREMENTS ALTERNATIVES
FOR
PRODUCT
3 O
O
roads Brick roads Concrete roads Tr..~.pic wood waterworks Steel waterworks Concrete waterworks Plastic waterworks Other build./infrastr. ....................................................................... .................................... ....................................................................... ....................................................................... Steel cars Aluminum cars Plastic cars .................................... ................................... ................................................................... .................................................................. tee tru s Aluminum trucks Plastic trucks Reference machinery
Max. plastic appliances Other objects at home Steel furniture Wooden furniture Other int. decoration .:.:.:.:.:.:+:.:.:.:.:.:.:.:.:.:.:.:.:.:.:+:.:.:.:.:.:.:.:.:.:.:.:.: :.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:. .................................... Steel cans Aluminum cans Glass multiple use bottle Paper/board packaging Plastic disp. packaging PET multi le use ack. One-way glass bottle Degradable plastic pack. Sanitary paper Wooden pallets Plastic pallets Other ind. packaging Plastic clothing Nat. org. clothing ........................................................... ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: Agents Co_~.m..post N-fertilizer Paint Lubricating oil Detergentia Chlorine Na(OH)
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1075
Energy conservation and investment behaviour of firms Merlijn Gillissen and Hans Opschoor Ecological Economics Group, dept. of Spatial Economics, Vrije Universiteit Amsterdam, De Boelelaan 1105, 1081 HV Amsterdam, The Netherlands
Abstract This paper analyzes the determinants and barriers of energy conservation investment behaviour. A number of barriers were found in a literature survey. A three-phase investment model on the micro level was constructed. Hypotheses derived from the model were empirically tested by analyzing a survey of more than 300 Dutch Firms. Economic variables seem to determine investment behaviour to a large extent.
1. Background and problem description To reduce the emissions of Greenhouse gases (GHG), a reduction in CO2 emissions is necessary. Energy conservation (EC) is considered as one of the major strategies to achieve this. Industry (in its broad sense: agriculture, manufacturing, services) is one of the main users of energy and potentially an important energy conserver. A main objective of Dutch policy is to speed up the energy efficiency improvement from approx. 1% p.a. to 2.2% p.a. in the year 2000 (Nota Energiebesparing, 1989) to reduce CO2 emissions by 3-5% in 2000. It has been recognized (Blok 1991) that there exists large potentials for energy conservation in industry. Calculations with the data base ICARUS (see De Beer et aL 1993) show that the technical potential for energy conservation can be as much as 30% on average. Not all technologies are profitable. However, if one applies economic evaluation criteria, there still remains a profitable potential for energy conservation of about 20% (V.d.Werff and Opschoor 1992; Ayres 1994). Problems arise when one tries to apply the results of ICARUS to industry: large differences exist between what ICARUS indicates as profitable and what firms think is profitable. This study analyzes the differences between ICARUS's results and observed implementation behaviour of firms, in terms of determinants and barriers to the adoption of EC-technologies. The result is a theoretical implementation model which is empirically validated. In a second part of the study a set of realistic energy policy scenarios are constructed and these scenarios are applied to the implementation model of the first stage. The result here will be a simulation model that assesses the impacts of energy policy instruments on implementation behaviour and estimates how much the adoption process of EC-technologies can be accelerated and what the results are in terms of additional energy conservation.
1076 2. Methodology
A literature survey looked into investment decision in general and an application of investment theory to energy conservation and identified theoretical barriers that might arise. In this framework important theoretical determinants and barriers to energy conservation adoption have been derived and a conceptual model was constructed (section 3). Next, a survey among more than 300 Dutch firms was held. Its results were used to empirically validate or reject the hypotheses derived from the theoretical framework about the determinants of and barriers to the investment decision (section 4). One part of the survey focused on the information on and implementation of the six most applicable EC-technologies in a sector. Another part of the survey focused on variables related to the theoretical determinants and barriers. Thus, it is possible to estimate the impacts of the variables on investment behaviour (section 5). The second part of this study entails the estimation of the effects of additional energy policy on investment behaviour of firms. A set of plausible energy policy scenarios for the future (1994-2015) was constructed. These scenarios are used as an input for a simulation model that is currently built. For a schematic synthesis, see figure 1.
Technical potential o n sectoral level
l
Profitable potential o n sectoral level
Policy
l
Policy scenarios on: - energy taxes - financial incentives - direct regulation
- information c growth
S e c t o r a l model
of investment behaviour of farms for e n e r g y
I demand
t
Energy s a v e d by sector F i g u r e 1" Schematic representation
of project structure
3. Potential determinants and barriers
In a perfect world (e.g certain cash flows, free and full information, independence between technologies and unlimited access to capital markets), a profit maximizing firm would implement all available technologies that have a positive net present value. However, introducing imperfections lead to the existence of barriers that prevent firms from implementing EC-technologies. The potential barriers can be categorized in the follov'ing groups (see Gillissen, 1994a): a. e c o n o m i c barriers: i) low expected energy prices; ii) uncertainty due to
1077 expected fluctuations in energy prices; iii) low expected revenues due to low energy bill; iv) budgetary problems; v) too high required return on investment; b. physical/technology barriers: i) reduction in production quality; ii) bounded rationality; iii) "technology-lock"; iv) information gap; c. management barriers: i) no specialized personnel; ii) no interest in energy conservation by management; iii) no priority to conservation (high opportunity costs); iv) present technologies are not fully depreciated; v) lack of pressure. Potential determinants of energy conservation are, for example, firm size, the presence of an energy coordinator and R&D department. External pressure and bilateral agreements may also speed up the implementation process.
4. Modelling energy conservation implementation behaviour As a complement to ICARUS (where only EC-technologies are listed), our model provided detailed information to which extent the EC-technologies in are actually being implemented by firms. The model consists of three "modules". The first module analyses the information process of firms. Variables that describe the information capacity for energy conservation are: the number of information channels, the presence of an energy coordinator, R&D department or environmental care system. Other important variables that represent the importance of energy conservation technology information are: firm size, the energy bill, the complexity and costs of a EC-technology. Together, these variables serve as explanations for the level of information of a firm. Lack of information might lead to an information gap, which is a barrier to the adoption process. The second module analyses the economic evaluation process by firms. Technologies are judged on their expected profitability. The profitability as perceived by the firm might differ from the profitability as calculated in ICARUS, because of uncertainty and firm specific expectations about for instance energy prices. Other variables include possible biases in perceptions through a low priority for energy conservation in comparison with "core business activities". The implementation stage is analyzed in the third module. Rational behaviour theories predict that a firm will only implement technologies it considers to be profitable. However, there may be physical barriers that prevent a profitable technology from being implemented, whereas non-economic influences cause an unprofitable technology to be implemented. Possible barriers and positive influences were named above. Figure 2 shows the conceptual framework of the implementation process in firms.
5. Results of modelling energy conservation implementation The empirical modelling stage consisted of two steps. The first step empirically identified the most important determinants and barriers from a set of more that 100 possible influential factors (see Gillissen and Opschoor, 1994). Indicators of the degree of information and implementation were constructed and the influ-
1078
Conceptual model energyconservation
external:
=known V-] -unknown
Energy
teclmical
technical filter of a fmn
conservation
options based on ICTUS
l
potential ] of energy- I conservation measures of a firm
Influential factors on profitability-criteria: 1) size of project 2) marke~ition of a fLrm 3) non-core business ~ [non-profitable energy (~ profitability projects r - | criteria of ~.a firm T profitable energy projects
I i t i
information index
(;nergYprio~I/uctuati and ~. I I Qgovemmental~ policy S profitable I~~Pr~ /~. liquidity& solvaenergy investment" bility constraints options op~ons / marketexpectations and marketprospects actually implementeAprojects among which: energyprojects
noriti~j ~
' k
environ~, mental
awareness
xtemal ! pression ~
/ i'
governmental policy
I internal f a R&D c t O rfirm s:l) 2) planninghorizon 3) depreciationmethod
Figure 2: Conceptual framework of a 3-phase investment model ence of firm specific variables was assessed. The results suggest that energy in considered as one of the production factors, and that investments to reduce the use of energy i.e. by EC-technologies are made largely on a economic evaluation, taking into account the physical and financial constraints. Determinants are: firm size, return on investment, the availability of capital, the possibility of early depreciation. Barriers that prevail are: uncertainty due to fluctuations in energy prices, budgetary problems, poor financial market expectations, a lack of knowledge of EC-technologies and the complexity of those technologies. Variables that do not seem to influence the implementation decision are the "core business" argument, the size of energy bill and the presence of an energy coordinator or R&D department. Decisions on EC-investments do not basically differ from the decisions on "core business" investments (see table 1). The three phase investment model was estimated in the second stage (see Gillissen, 1994b). Preliminary results seem to confirm the results of the first step, with a few changes: the role of covenants stimulates the information and knowledge about EC-technologies. Also the expected positive role of the energy coordinator could sometimes be proven. Again, complex technologies were less known that simple measures.
1079 6. Policy simulation
The policy simulation part evaluates constructed energy policy scenarios on their contribution to energy savings. Scenarios consist a set of economic and regulatory instruments, combined with expectations regarding economic growth and energy prices. The instruments are constructed on the basis of actual and intended energy policy (VNEB, 1993); other scenarios line with "Scanning the future" and "Milieuverkenning 3". The advances in implementation, as a consequence of such a policy, will be calculated on a yearly basis up to the year 2015. Energy policies are then evaluated on their estimated contribution of additional energy savings. Instruments (control variables) that are analyzed include energy taxes, energy subsidies, the effectiveness of covenants, and information policy to reduce the information gap.
Table I: Determinants of EC-investment decision process
Important
variables
- Firm size -Information - Availability - depreciation
sources of capital moment
Less i m p o r t a n t
variables
- Size o f e n e r g y bill - Distance to core business - Low expected energy prices - competition
References
Ayres, R.U. (1994), "On economic disequilibruim and free lunch", Environmental and Resource Economics vol 4 (5) pp 435-454 Beer, J.G. de, Wees, M.T. van, Worrell, E, Blok, K, "ICARUS, the potential of energy efficiency improvement in the Netherlands up to 2000 and 2015", dept of Science Technonlogy and Society, report nr 94013, Utrecht, The Netherlands Blok, K (1991) "On the reduction of Carbon Dioxide emissions", Ph.D. thesis, University of Utrecht, Dept of Science Technology and Society, Utrecht, The Netherlands Gillissen, M. (1994a), "Energy conservation investments, a rational decision?", series research memorandum no 33, Vrije Universiteit Amsterdam, Faculty of Economics, The Netherlands Gillissen, M. (1994b), "Empirical modeling of energy conservation investment processes of firms", paper to be presented at the Summer Study of the ECEEE, june 6-10 1995, Cannes
1080 Gillissen M. and Opschoor, J.B. (1994), "Energy conservation investment behaviour, an empirical analysis of influential factors and attitudes", paper presented at the Regional Science Association, august 22-24 1994, Groningen, The Netherlands Nota Energiebesparing (1989): Energy Conservation Policy, Report of the Dutch Ministry of Economic Affairs, The Hague, The Netherlands VNEB (1993): Vervolg Nota Energiebesparing (Second Energy Conservation Policy), Report of the Ducth Ministry of Economic Affairs, The Hague, The Netherlands Werff, R.L. van der, Opschoor, J.B (1992). "De potentiele energiebesparing van Nederlandse Bedrijfstakken" in" Economische Statistische Berichten, vol 77 (3884), pp 1069-1072
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1081
Long T e r m E n e r g y Efficiency I m p r o v e m e n t Jeroen de Beer, E r n s t Worrell, Kornelis Blok
D e p a r t m e n t of Science, Technology and Society, Utrecht University, P a d u a l a a n 14, 3584 CH Utrecht, The Netherlands Phone: +31-30-537638; e-mail:
[email protected] Abstract The opportunities for long term energy efficiency improvement in industry have been studied. Three studies are described. The first study was directed at making a preliminary survey of technologies that might reduce the end-use demand of industrial processes on the long term. The second study focused on the development of a methodology to make a more profound analysis of the long term potential. The third study describes a database for energy efficient technologies. It is concluded that, after technologies t h a t are currently technically feasible have been implemented, there still exists a considerable (new) potential for improvement.
1. INTRODUCTION A major option to reduce C O 2 emissions is efficiency improvement of energy end-use and energy conversion processes. Often it is stated that the ultimate potential of energy efficiency improvement in the western world is several ten percents. A large n u m b e r of studies showed that this potential is technically feasible within 10 years [see e.g. Lovins and Lovins, 1990; Pilavachi, 1993; ETSU, 1984; Giovannni and Pain, 1990]. However, it is also claimed t h a t in the longer r u n this potential is much larger, i.c. more t h a n 80% or even 90% [see e.g. Ayres, 1988; Jochem, 1991]. In this paper an overview is given of work performed to assess the potential of efficiency improvement on the long term. The results can be of use in R&D priority setting, developing CO 2 mitigation strategies, and long term energy infrastructure planning. Three studies addressed the opportunities of long term energy efficiency improvements. The approach and results of these studies will be described in this paper.
1082 2. S U R V E Y OF E N E R G Y E F F I C I E N T T E C H N O L O G I E S I N I N D U S T R Y
The first study is directed at making a preliminary survey of technologies t h a t might reduce the end-use demand of industrial process on the long term [Smit et al, 1994]. The technology descriptions are based on readily accessible literature, where necessary supplemented with data provided by experts on the specific sector or technology. The descriptions are divided into two sections. The first section gives a description of the reference technology and the new, energy efficient technology. Also information on state-of-the-art, ongoing R&D, and applicability of the technology is presented. The second section gives preliminary economic and energetic parameters. In table I some results of this research are presented. It must be emphasized that the results are based on a limited literature research, in some cases supplemented with consultation of experts. A more thorough analysis of the energy efficiency improvement potential is topic of the second study.
3. D E V E L O P M E N T A N D T E S T I N G OF A M E T H O D O L O G Y
The second study focuses on the development and testing of a methodology to m a k e an accurate analysis of the long term energy efficiency improvement potential. The methodology developed so far starts with the determination of the m i n i m u m energy requirement to perform a certain energy function and of the energy losses associated with performing the energy function with the current technology. The question posed is, can these losses be reduced without changing the current technology? And, if such is not the case, are technologies perceivable t h a t can reduce the energy losses? After having compiled a list of efficiency improvement technologies, an assessment of the possible technological development is made. A list of determinants of technological development is filled out on the basis of a literature review and consultation of experts. A study following this line of research has been conducted for the sector 'paper and board' industry. Furthermore, two studies are underway for the sectors 'iron and steel industry' and 'cement industry'. Here we will present some results of the studies for the paper and board industry and the iron and steel industry. Theoretically, the minimum energy demand for making paper out of wood pulp is very small (compared to a present average primary energy demand of about 10 GJ/ton paper). The operation with the largest energy losses is the steam generation (in a CHP-unit or boiler). Steam is mainly required for drying of the paper against steam heated driers. Elimination of these losses is only possible by making paper without the addition of water. However, this has a large negative effect on the product characteristics. Five technologies were selected t h a t have the opportunity to reduce the energy losses.
o m~
b~
b~
.,-i
o~ ~J
=.~ .~~
~
~-
m
L~
Z
L~
~
~
~ m
~
,
~--~
~
m
,--I
~
~
CO,
N
~ ~ ~
N
~~.~ ~ ~,~o~ ~
~I~ ,
~
~
|
o
o
~
~
~
o
r~
94 ~
~,'~1
0
9.~ ~ , ~
119
s~C
0
~ ~
,~
~1
~D ~
~.~ ~'~
II
.~
1083
..~
.1084 The third step of the methodology, assessment of the technological development, resulted in the selection of two technologies t h a t are most promising for reducing the energy consumption of a paper mill: condensing belt drying and impulse drying. R&D to these technologies is concentrated in Finland and the USA. Also in Italy, Sweden and Germany research is being conducted. These technologies have the ability to reduce the specific steam d e m a n d by 50-75% [Beer et al. 1993]. For the iron and steel industry only the first two steps and part of the third step of the methodology have been performed so far. The specific energy r e q u i r e m e n t of an efficient integrated steel plant (Hoogovens, the Netherlands) is 19.7 GJ/ton crude steel. On the short term a reduction to 16.8 GJ/ton is technically feasible. However, the thermodynamical minimum energy requirement for the reduction of iron oxide is only 6.2 GJ/ton. An exergy analysis of an integrated steel mill revealed that the room for improvement of the current process is limited. Larger improvements seem only feasible when another production route is chosen. Several technologies are outlined: an increased share of secondary steel making, advanced iron making processes (e.g. plasma processes), direct steel making (in-bath melting of iron, ore-to-powder steelmaking), near shape casting, and using hydrogen as reductant. The largest efficiency improvement is achievable with more secondary steel making. Taking into account an improved efficiency of electric arc furnaces and a higher energy demand for scrap benification a specific primary energy demand of about 7 GJ/ton steel seems possible in the long term. A combination of efficient technologies for primary steel making might reduce the specific p r i m a r y energy demand to about 12 GJ/ton steel [Beer, 1994].
4. A D A T A B A S E O N E N E R G Y E F F I C I E N T T E C H N O L O G I E S
A database (called ICARUS) on the potential and costs of energy efficiency improvement measures that can be applied in different sectors of the Dutch economy between 1990 and 2000 has been constructed [Beer et al, 1994]. D a t a were acquired using a sectoral, bottom-up approach. In figure 1 the results are summarized in a supply curve. The figure indicates a technical potential for efficiency improvement for the period 1990-2000 of 36%. If only those measures with a positive net present value are taken into consideration, the potential is 29% (we call this the economic potential). It is also possible to collect data on long term energy efficiency improvement technologies in a database like ICARUS. We have done this for technologies t h a t probably will be commercially available before the year 2015 [Beer et al, 1994]. The results are also summarized as a supply curve, see figure 1. Of course, on this longer term the uncertainty in the data is larger t h a n for the period 1990-2000.
1085
50t
2000
2015
40
30
Projected primary energy demand in the year 2000:3510 PJ in the year 2015:4915 PJ
~ 2o 100 ......................................................
,.........................................................
lO 20 3040 50 I" 0%
29% ,
10%
,
,
43% ~
20% 30% 40% Cumulative saving (%)
,
50%
60%
F i g u r e 1: Supply curves of energy efficiency improvement measures for the periods 1990-2000 and 1990-2015. On the horizontal axis the cumulative improvement potential is given as percentage of the projected energy demand without efficiency improvements. The European Renaissance scenario is used with growth figures based on physical growth. Vertically the specific energy efficiency improvement costs are depicted. Energy prices are taken from [EZ, 1994]. The low energy price scenario is used. A discount rate of 5% is used.
From figure 1 it can be seen t h a t for the period 1990-2015 the technical potential is 56%. This would m e a n a decrease in specific energy consumption in 2015 of 31% compared to 2000. The economic potential for the period 19902015 is 43%.
5. CONCLUSIONS It can be concluded t h a t the potential of energy efficiency i m p r o v e m e n t is not limited to the currently technically feasible technologies. In the steel m a k i n g process, for instance, the long t e r m potential is about three times as large as the short t e r m potential. An a s s e s s m e n t of the technological development by filling out a list of d e t e r m i n a n t s of this development, gives insight in the probability t h a t a technology will be commercialized and the time scale for this to happen. This information can be of aid in R&D-priority
1086 setting. For instance, all R&D to innovative drying techniques in paper making is currently concentrated in four or five countries, but not in the Netherlands. Therefore, national R&D-funds can better be applied for the development of other energy efficient technologies.
REFERENCES
Ayres, R.U. (1988) Energy Inefficiency in the U.S. Economy: A New Case for Conservation, Dept. of Engineering and Public Society, Carnegie Mellon University, Pittsburgh, 1988. Beer, J. de, E. Worrell and K. Blok (1993) Energy Conservation in the Paper and Board Industry on the Long Term, Department of Science, Technology and Society, Utrecht University, report 93006, Utrecht. Beer, J. de (1994) Long Term Energy Efficiency Improvements in the Iron and Steel Industry, Working paper prepared during a three months stay at IIASA, Laxenburg, Austria (draft). Beer, J.G. de,. M.T. van Wees, E. Worrell and K. Blok (1994) ICARUS-3; The Potential of Energy Efficiency Improvement in the Netherlands up to 2000 and 2015, Department of Science, Technology and Society, Utrecht University, report 94013, Utrecht. ETSU (1984) Energy Use and Energy Efficiency in UK Manufacturing Industry up to the year 2000, Energy Technology Support Unit, Energy Efficiency Office, Harwell, UK. EZ (1994) Ministry of Economic Affairs, Van wereldmarkt tot eindgebruiker, Energieprijzen voor de periode tot 2015 (From world market to end user, Energy prices for the period until 2015), beleidsstudies energie nr7. 1994. Giovanni, B. and D. Pain (1990) Scientific and Technical Arguments for the Optimal Use of Energy, Centre Universitaire d'Etude des Probl~mes de l'Energie, Universit~ de Gen~ve. Jochem, E. (1991) Long-term potentials of rational energy-use - the unknown possibilities reducing greenhouse gas emissions, Energy & Environment, 2, 1, pp. 31-44. Lovins (1990) Various report by A. Lovins and Lovins, for instance: The State of the Art: Lighting, Rocky Mountains Institute, Snowmass CO. Pilavachi, P.A. (ed.) (1993) Energy Efficiency in Process Technology, Proceedings of the International Conference, held in Athens, Greece 19-22 October 1992, Elsevier Science Publishers, London. Smit, R., J. de Beer, E. Worrell and K. Blok (1994) Long Term Industrial Energy Efficiency Improvement: Technology Descriptions, Department of Science, Technology and Society, Utrecht University, report 94076, Utrecht.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1087
Energy efficiency improvement in industrial sectors: international comparisons G.J.M. Phylipsen, E. Worrell and K. Blok
D e p a r t m e n t of Science, Technology and Society, Utrecht University, P a d u a l a a n 14, NL-3584 CH Utrecht, The Netherlands
Abstract. Eight major industrial processes areresponsible for over 50% of industrial energy consumption in most countries. The energy efficiency of these processes was determined in a number of countries, with appropriate corrections for structural differences between countries. It is shown t h a t considerable differences occur between countries, but t h a t m a n u f a c t u r i n g industry in Eastern Europe in general is less efficient t h a n in EU countries. In all cases efficiency is worse t h a n w h a t is technically and economically feasible. International comparisons provide information on energy efficiency differences, insight into technological differences between countries and into costs requirements for efficiency improvements. The comparisons can be used in international climate change negotiations and in the field of bilateral or multilateral cooperation.
1. I N T R O D U C T I O N
It is well-known that there are large differences in energy efficiency between countries. Up to now comparisons between countries are done on a national level, using aggregate measures, like the energy consumption per capita or energy consumption per unit of GDP [1]. Such comparisons can give a first impression of the differences between countries, they do not give much insight into the causes of differences or ways to reduce them. In this paper we will make a more detailed comparison between countries to show it can be used for international climate change policy making. First, we will discuss the results of a comparisons between countries for two main sectors. Subsequently we will evaluate the use of such comparisons for policy formulation.
1088 2. I N T E R N A T I O N A L E N E R G Y E F F I C I E N C Y C O M P A R I S O N S
The level of energy consumption in an economic sector is determined by three factors: the level of h u m a n activity, the mix of activities (the structure) and the energy efficiency within the sector (the energy consumption per unit of activity) [2]. All of these can be a subject of policy to reduce energy-related CO 2 emissions. Of the three, improving energy efficiency may be considered to be the most important option on the short term. For energy end-use activities two measures for (the inverse of) energy efficiency are used: 9 energy intensity: energy consumption per unit of value added; 9 specific energy consumption: energy consumption per physical unit of h u m a n activity (e.g. person-kin of transportation, tonnes of steel produced). In general the second measure gives a better insight in the technological characteristics of the use of energy. The first measure is also influenced by other factors, like feedstock and product prices. The second measure is not applicable to all sectors as not for all sectors a good physical indicator o f h u m a n activity can be defined.
2.1 Energy efficiency in heavy industry In the heavy industry the activity level can generally be measured i n tonnes of product, so energy efficiency is measured as Specific Energy Consumption (SEC). In an earlier analysis [3] we have identified t h e mix of feedstocks (e.g. primary or secondary feedstocks) and product mix as structural factorsl The potential for energy efficiency improvement is established by comparing the present SEC of a country with a 'best practice' SEC. Best practice SEC is here defined as the lowest SEC observed in a sector or plant in Europe in the reference y e a r (1988). In calculating the best practice SEC we take into account the structural effects mentioned before. It should be noted t h a t the energy efficiency improvement potential is time-dependent. On the basis of additional information [4..11] we extend the analysis to countries outside the European Union. The sectors we have studied are fossil fuel-based electricity production, refineries, iron & steel production, ammonia production, the paper & board industry, the cement industry and the chemical industry. Here we present the r e s u l t s for ammonia and steel production. 9
A m m o n i a production Ammonia can be produced by partial oxidation of oil residues and by steam reforming of natural gas. Steam reforming of natural gas is the more
1089 energy efficient process of the two. Eighty percent of the worlds ammonia production is produced by steam reforming of natural gas. The best practice S E C of 28 GJ/tonne ammonia is derived from the ICI-AMV steam reforming process (1988) [3]. Information for non-EU countries is retrieved from [4,6,8,-
11]. 9
Steel production
Steel production can be based on the Basic Oxygen Furnace (BOF) route or on the Electric Arc Furnace (EAF) route. The BOF route uses iron ore and scrap to produce primary steel, resulting in a higher quality of steel, but consuming more energy. The EAF route uses scrap only as feedstock for secondary steel production. Product type also influences the SEC. We distinguish slabs, hot rolled products and cold rolled products in the BOF route. The best practice values are based on the Hoogovens plant in the Netherlands (BOF route) and the Badische Stahlwerke plant in Germany (EAF route) [3]. Information for non-EU countries is also retrieved from [4,6,8,9,10,11].
The results are depicted in figures 1 and 2 for ammonia production and steel production respectively. From these pictures we see t h a t there are cases of developing countries and countries in transition t h a t are less efficient t h a n the European countries, but t h a t also in some cases there are no such differences.
sR1
GR
oll
CR POL
E3O E o20 (5
B
DK
F
O GR IR
I
L NL Country
P
E
UK
POL CR SR
Figure 1. Comparison of the specific energy consumption in ammonia production for various countries. The bars represent the present SEC, whereas the solid line represents the best practice (steam reforming of natural gas).
1090
35
30 -
I'
25"
Figure 2. Comparison of specific energy consumption in steel making. i represents present SEC and [::] represents best practice (with current feedstock and product mix). The solid lines indicates improvement potentials.
China
I
SR
POL CR
t5
++
-I
10
0 0%
i
i
:
i
20%
40%
60%
80%
100%
Share EAF
2.2 Discussion The methodology used heavily depends on data retrieved from international statistics. Production and energy statistics are main sources of information for studies like ours. Errors and deviations in these statistics will affect the reliability of the results. For non-OECD countries the accuracy of the statistics is in general less reliable than that of OECD countries. Improvement of available statistics, in an internationally harmonized way, is needed to improve the results of this type of analyses. There is also a strong need for the design of common methodologies to calculate energy efficiencies. The developed methodology needs to be tested for applicability in other sectors, t h a n the ones described in this paper.
4. A P P L I C A T I O N O F I N T E R N A T I O N A L E N E R G Y E F F I C I E N C Y COMPARISONS
In this section the applicability of international energy efficiency comparisons for the development of international climate policy will be discussed.
9Improvement of the knowledge on potentials and costs of energy efficiency improvement. For m a n y studies on potentials and costs of energy efficiency improvement carried out before, results are difficult to compare. International comparisons
1091 of energy efficiency, followed by a sector-by-sector comparison of costs can form the basis for a better understanding of the real differences in potentials and costs for energy efficiency improvement. 9I n t e r n a t i o n a l agreements on energy efficiency levels.
In international negotiations on CO 2 emission reduction several approaches (e.g. equal relative emission reduction) lead to objections from part of the countries involved. An alternative approach is to close agreements on energy efficiency levels by sector (taking into account structural differences) t h a t should be obtained by participating countries. International comparisons of energy efficiency are the first step of evaluating possibilities and effects of such agreements. 9I n t e r n a t i o n a l technological cooperation.
In international cooperation regarding C O 2 emission reduction (for instance in the field of 'joint implementation') international comparisons of energy efficiency can give an important contribution by steering the cooperation activities by indicating which sectors in which countries should have the highest priorities; furthermore, they can give an indication which type of transfer is most needed: investment capital, knowledge and education, licences, etc.
5. C O N C L U S I O N S
International comparisons of energy efficiency can be done for m a n y sectors, covering a considerable part of world energy demand. Development of common methodologies of measuring energy efficiency and improving the availability and quality of data is necessary. Taking into account the possible applications of international energy efficiency comparisons in international climate policy, such a task seems worthwhile undertaking.
6. R E F E R E N C E S
1. 2. 3.
See for example: The State of the Environment, OECD, Paris, 1991. L.J. Nilsson: "Energy Intensity Trends in 31 Industrial and Developing Countries 19501988" Energy, 18 (1993) pp. 309-322. L. Schipper, R.B. Meyers, R. Howarth, R. Steiner: Ener~zv Efficiency and Human Activity: past Trends, Future Prospects, Cambridge University Press: Cambridge USA; 1992. E. Worrell, R.F.A. Cuelenaere, K. Blok, W.C. Turkenburg: "Energy Consumption by
1092
4. 5. 6.
7. 8. 9. 10. 11.
Industrial Processes in the European Union", Energy (forthcoming). E. Bossenbroek: Energiegebruiken in Polen; trends en vergelijkingen met de EG/OECD, Dept. of Science, Technology & Society, Utrecht University, November 1993 J. Garcia del Valle, A. Torres: Outlook of Latin American Cement Industry, in: J. Sichis: Energy Efficiency in the Cement Industry, Elsevier Applied Science, London, 1990 S. Meyers, L. Schipper, J. salay, A. Gromadzinski, E. Hillie, P. Kaleta, M. Kumanwoski, J. Maron, J. Norwisz and S. Pasierb, Energy Use in Poland, 1970-1991: Sectoral Analysis and International Comparison, LBL, Berkeley, July 1993 M. Ross, L. Feng, "The Energy Efficiency of the Steel Industry in China", Energy 5 16, (1991), pp.833-848. Statistical information on the Czech Republic, Slovak Republic and Poland, 1990. Statistics on Energy in the Steel Industry (1990 Update), International Iron and Steel Institute, Brussels, 1990 Steel Statistical Yearbook 1992, International Iron and Steel Institute, Brussels, 1992 B. Vallance, East-West Comparisons of Energy Efficiency in Energy Intensive Industries, Symposium on Energy Efficiency and Economic Transition in Central and Eastern Europe, Paris, 25-28 May 1993
Abbreviations
B Br EU L CR D DK F E GR I IR NL POL SR UK
Belgium Brazil European Union Luxembourg Czech Republic Germany Denmark France Spain Greece Italy Ireland Netherlands Poland Slovak Republic United Kingdom
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1093
Carbon dioxide removal studies in the Netherlands K. Blok D e p a r t m e n t of Science, Technology and Society, Utrecht University, P a d u a l a a n 14, 3584 CH Utrecht, The Netherlands
Abstract
An explorative research programme o n C O 2 removal has been carried out in the Netherlands. The goal of this programme was to obtain a better understanding of the technical and economic feasibility of the recovery of CO 2 from flue gases and synthesis gases and the sustainable storage of CO 2 outside the atmosphere. As far as CO 2 recovery from power plants is concerned, options based on coal gasification with CO 2 recovery turn out to be most energyefficient. Of the remaining recovery options chemical absorption from flue gases, using amines, seems most promising. A number of recovery options based on membrane technologies have been identified, but most of them still require considerable development. In manufacturing industry, attractive options for CO 2 recovery are available in refineries equipped with residue gasification, and in the ammonia fertilizer industry. More costly options were identified in the iron and steel industry and in the petrochemical industry. CO 2 storage in aquifers is technically feasible. When injecting CO 2 in aquifers part of the water already present will be displaced. The main mechanisms for this displacement will be gravity segregation and viscous fingering, as was shown by simulation calculations. The Dutch subsurface contains a large number of aquifers that are potentially suitable for CO 2 storage.
1. I N T R O D U C T I O N
In the period of 1991 to 1993 an explorative research programme on CO 2 removal has been carried out by a number of companies and research institutes in the Netherlands. The goal of this programme was to obtain a better understanding of the technical and economic feasibility of the recovery of CO 2 from flue gases and synthesis gases and the sustainable storage of CO 2 outside the atmosphere. Before this research programme started a number of publications on
1094 carbon dioxide recovery and storage had already been issued in the Netherlands with special emphasis on carbon dioxide removal from IGCC power plants and on storage of carbon dioxide in depleted natural gas fields. In the new research programme the main emphasis was on studying a range of techniques to recover carbon dioxide from gas streams in detail. Some studies were devoted to the recovery of carbon dioxide from industrial processes. One study explored the possibilities of CO 2 storage in aquifers. A separate study was directed at the calculation of the efficiencies and costs of complete power plants in a comprehensive way. A more extended overview of the results is given in
[1]. The research programme was sponsored from various sources, the main ones being the Ministry of Housing, Physical Planning and Environment and the National Research Programme on Global Air Pollution and Climate Change. The total budget amounted to about 1.5 million Dutch guilders (1 Dfl = $0.6 - ECU 0.45).
2. C O 2 R E C O V E R Y B A S E D ON COAL G A S I F I C A T I O N
It was confirmed that carbon dioxide recovery based on coal gasification shows the smallest decrease in efficiency of electricity production. The detailed analysis carried out by the research institute of the electric utilities, KEMA, showed that efficiencies of over 36% can indeed be attained in combination with carbon dioxide recovery (table 1). Two widely differing CO 2 recovery technologies were studied. In the first case a shift reaction is applied after gasification resulting in a fuel gas mainly consisting of hydrogen and carbon dioxide. For separation of hydrogen and carbon dioxide a number of options were studied: freezing out the CO2, membrane separation, hydrogen recovery, physical absorption and chemical absorption. The best option is physical absorption (using selexol). By freezing out the CO 2 the required high degree of CO 2 recovery (to less than about 120 g/kWh) can not be attained. When this limitation would not be set, freezing out the carbon dioxide should certainly be taken into consideration. Using membranes at low temperatures is not attractive, mainly due to the high hydrogen loss. Combined with high-temperature gas clean up membrane applications may become of interest, although the membranes with the required high H2/CO 2 selectivities are still under development. Chemical absorption systems have a too high energy demand. Hydrogen recovery techniques showed considerable hydrogen losses. The components for the favoured shift/selexol concept are commercially available but were never applied in this combination. It is concluded t h a t the technology is ready for demonstration. The second IGCC approach makes use of a gas turbine in which the fuel is combusted in a mixture of oxygen and recycled CO 2. The
1095 Table 1. Some key results for complete power pla nt concepts as calculated by KEMA. P r e l i m i n a r y cost es tima te s ra nge from 60 - 80 Dfl per tonne of CO 2 avoided, being the lowest for the IGCC options. Electricity production costs are e s t i m a t e d to be 0.14 - 0.15 Dfl/kWh for the coal cases and 0.095 Dfl/kWh for the n a t u r a l gas case. Type of p l an t m e t h o d of recovery
Net conversion efficiency (%)
Specific CO 2 emission (g/kWh)
IGCC - CO2/O2-combustion (Texaco)
34.8
5
IGCC - CO2/O2-combustion (Shell)
36.0
30
IGCC - shift & physical absorption (Texaco)
36.4
139
Pulverized coal - chemical absorption (retrofit)
29.7
105
N a t u r a l gas fired combined cycle - chemical absorption
44.9
86
42 - 43
800 - 820
52
390
Reference coal fired power plant Reference n a t u r a l gas fired power p l a n t
combustion products (mainly CO 2 and water) are expanded t h r o u g h the t u r b i n e section. After cooling in a h e a t recovery s t e a m g e n e r a t o r and removal of the water, the CO 2 is recycled to the gas t u r b i n e compressor. P a r t of the compressed CO 2 is used in the gas turbine combustion chamber, the r e m a i n d e r is exported. As CO 2 is the m a i n working fluid in the gas turbine, the properties differ strongly from a conventional gas turbine. The results of i n t e g r a t i n g such a gas turbine in an IGCC p l a n t are given in table 1. The m a i n bottleneck for the application of this scheme is the fact t h a t such a C02-gas-turbine is not available at present. S t a r t i n g such a costly development process is only justified if it gives clear (cost or efficiency) a d v a n t a g e s above the IGCC/shift/selexolprocess m e n t i o n e d before.
3. CHEMICAL A B S O R P T I O N / O T H E R R E C O V E R Y T E C H N I Q U E S As the f u t u r e of coal gasification is still u n c e r t a i n it is advisable to develop other CO 2 recovery techniques as well. A n u m b e r of other options for CO 2 recovery was evaluated. In most cases chemical absorption, using amines, is the most attractive alternative.
1096 For the recovery of C O 2 from flue gas of conventional coal-fired power plants the use of gas separation membranes is more expensive t h a n chemical absorption. This is mainly due to the high power requirements for the compression of the flue gases. When chemical absorption is applied, the use of gas absorption membranes, is of interest. Gas absorption membranes are used in conjunction with chemical absorption liquids where the conventional absorption column is replaced by a m e m b r a n e contactor. This modification could increase the conversion efficiency of the power plant by approx. 0.5%. This improvement is mainly due to a reduction of the pressure drop over the absorber. Gas absorption membrane systems, however, are still under development. For natural-gas-fired combined cycle power plants the most costeffective option also is chemical absorption, with an overall conversion efficiency of approx. 45%. An alternative is a power plant based on a gas turbine using combustion in a CO2/O2-mixture. Also a system based on m e t h a n e reforming of natural gas (to a large extent similar to an IGCC plant) was investigated. However, it showed a low efficiency: about 37%.
4. C O 2 R E C O V E R Y IN M A N U F A C T U R I N G I N D U S T R Y
Twenty plants in manufacturing industry with the largest C O 2 emissions in the Netherlands together are responsible for about 20% of the total Dutch CO 2 emissions. Main sectors are refineries, the iron and steel, petrochemical and fertilizer industries. Carbon dioxide recovery can be accomplished in refineries equipped with a residue gasification unit. Residue gasification is expected to be a good solution in the development towards low sulfur oil products and deeper conversion. The gasification product is fed to a shift reactor in order to produce hydrogen for other refinery processes. The carbon dioxide that is co-produced can be recovered easily. In this way about one quarter of the CO 2 emissions in future refineries can be avoided. Another attractive option is available in the fertilizer industry. In producing ammonia, which is one of the main feedstocks for fertilizer production approx. 50% of the CO 2 output of the fertilizer industry is already recovered. At present part of this amount is utilized, the remainder is vented to the atmosphere. CO 2 recovery can be applied on this stream by just compressing it to transportation pressures. Both for the refineries and the fertilizer industry estimated mitigation costs are in the order of 20 Dfl per tonne of CO 2 avoided. More costly options were identified in the iron and steel industry: recovery of CO 2 from blast furnace gas; and in the petrochemical industry: the use of low-temperature waste heat (100 - 150 ~ for supplying the reboiler duty of a chemical absorption process.
1097 5. S T O R A G E O F C A R B O N D I O X I D E
According to one of the studies, CO 2 storage in aquifers is technically feasible. When injecting CO 2 in aquifers part of the water already present will be displaced. The main mechanisms for this displacement will be gravity segregation and viscous fingering. Extended simulations of the behaviour of CO 2 have been carried out for sample reservoirs; in one of these aquifers 15,000 tonnes of carbon dioxide per day can be injected during 8 years. After this period CO 2 breakthrough is observed at the spillpoint. The Dutch subsurface contains a large n u m b e r of aquifers t h a t are potentially suitable. Taking a number of constraints into account the total aquifer storage capacity for CO 2 a prudent estimate of the storage capacity of 1.2 Gtonne CO 2 is made. The main chemical effect of carbon dioxide in aquifers is its effect on carbonate chemistry. The decrease of the pH due to the dissolution of carbon dioxide will cause solution of carbonates. This effect is so small t h a t weakening of the porous structure is not expected. However significant changes in permeability may occur. If the seal of the structural trap is a clay layer, drying of this layer could reduce its tightness. The costs of injection (departing from a delivery pressure of 110 bar) are estimated to be 0.7 and 1.2 Dfl per tonne of CO 2 injected for aquifers above and below 1000 m depth respectively.
6. C O N C L U S I O N S As far a s CO 2 recovery from power plants is concerned, options based on coal gasification with CO 2 recovery turn out to be most energyefficient. Of the remaining recovery options chemical absorption from flue gases, using amines seem most promising. A number of recovery options based on membrane technologies have been identified, but most of them still require considerable development. More t h a n at present, attention should be paid to CO 2 recovery options outside the electricity production sectors, e.g. in manufacturing industry. Storage of CO 2 in aquifers seems to be technically feasible, but the total storage capacity, taking into account strict conditions, is limited. It is felt that with respect to underground storage of carbon dioxide, especially in aquifers, the largest uncertainties exist. F u r t h e r investigation of this option by reservoir simulations, laboratory experiments and field tests should have a high priority in further R&D planning.
1098 7. R E F E R E N C E
1 K. Blok: Final report of the Integrated Research Programme on Carbon Dioxide Removal and Storage", Department of Science, Technology and Society, Utrecht University, The Netherlands.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1099
STORAGE OF CARBON DIOXIDE IN AQUIFERS IN THE NETHERLANDS L.G.H. van der Meer, R. van der Straaten and J. Griffioen TNO Institute of Applied Geoscience EO. Box 6012, 2600 JA Delft, The Netherlands
Abstract This paper presents the results of a study about the technical feasibility of the underground storage of carbon dioxide (CO2) in aquifers. Special attention was paid to physical processes, limiting geological conditions and geochemical and environmental aspects. The CO 2 storage capacity of aquifers below the Dutch onshore is estimated based on these results. In addition, the long-term CO 2 storage potential of a hypothetical CO 2 storage reservoir is estimated. 1. INTRODUCTION The investigations were commissioned by the Dutch Ministry of Housing, Physical Planning and Environment and the Dutch National Research Programme on global air pollution and climatic change. This study of the technical feasibility, limiting geological conditions and consequences of carbon dioxide storage in aquifers was carried out as part of a programme entitled: A preliminary research programme for CO 2 removal and storage.
2. DISPLACEMENT BEHAVIOUR In order to elucidate the dispersive character and the fluid flow mechanism of C O 2 in an aquifer system we have unravelled the individual mechanisms affecting the displacement process. In general, the dispersion or spreading out of CO 2 in an aquifer can be described at three different scales: pore-scale, stratum-scale, and reservoir scale. Each scale is characterized by a particular process. Although smaller scale processes are active in the dispersion process at reservoir-scale, they will play only a minor role. All the processes and/or effects are understood and well described in the literature. For further information, the reader is referred to publication of van der Meer 1. If CO 2 is injected into an aquifer, it will be able to displace the pore water in the aquifer to a large extent. The displacement process is determined by many individual mechanisms related to fluid properties and the specific conditions of the rock matrix. One of the most important parameters in this displacement process is the relative ability of the two fluids to flow in the porous medium. This property is referred to as the relative mobility of the fluid. When one fluid displaces another, the mobility ratio (M) of the displacement is defined as the mobility of the displacing fluid divided by the mobility of the displaced fluid. For average reservoir parameters, we calculated M=40 for an aquifer at 800 metres depth and M=13.2 for an aquifer at 1800 m depth. This means that CO 2 is 13 to 40 times as mobile as the formation water and because the CO 2 is pushing the water, it tends to by-pass the water.
1100
The effect of one fluid being displaced by another can be considered as a complex process. The large differences in the physical properties of the three main items (two fluids and the reservoir rock) for the adopted range of depths make it difficult to predict the results of their interaction in a displacement process. A CO2/water displacement process will be dominated by a gravity segregation effect. A layered permeability distribution i.e. a large kv/kh ratio will have a negative influence on the upwards migration of CO 2. The calculated mobility ratios for a process in which CO 2 displaces water enable us to predict substantial viscous fingering effects. The resulting areal sweep efficiency will be in the order of 25 to 60 %, whereas the vertical sweep efficiency will be very small (in the order of 2-25 %), due to the combined effects of gravity segregation and viscous fingering. With the exception of the permeability distribution, all other small and medium scale effects will have an insignificant influence on the displacement process.
a.
.038PVI
b.
.076PVI
c.
.114PVI
d.
.150PVI
e.
.190PVI
f.
.228PVI
Fig. 1. Results of numerical displacement simulations. Concentration distribution maps for increasing time slices. (PVI = Pore Volume Injected)
3. G E O C H E M I C A L A S P E C T S Two types of geochemical processes are associated with the injection of CO 2 in deep-seated aquifers. The first is enhanced dissolution of carbonate minerals due to an increase in the dissolved CO 2 in formation water. The amount of dissolution is almost independent of depth (and temperature) for depth below 750 m. The total groundwater composition is not greatly affected by this process. Effects on aquifer properties (permeability and porosity) are also small. The second process relates to the characteristics of electric double layers of clay minerals. The double layer thickness of (swelling) clay minerals depends on the di-electric constant of the fluid present. The change from water to CO 2 as pore fluid may lead to a decrease in double layer thickness for swelling clay minerals such as smectite. This may effect the aggregate structure of clay minerals. Unfortunately, no applicable information was available on this topic. Clay minerals with a swelling interlayer may shrink. The associated consequences for the permeability of the aquifer and the sealing characteristics of cap rock need to be investigated.
1101
4. L I M I T I N G A S P E C T S OF CO 2 I N J E C T I O N IN AN A Q U I F E R Much information about aspects limiting fluid injection in the subsurface was obtained from the practice of flooding with water when extracting oil. Flooding with water is the main fluid injection method. This information yielded two possible limiting aspects in respect to CO 2 storage in aquifers: well/formation damage and injection pressure. Laboratory and field studies indicate that almost every operation that has to do with drilling, completion, workover, production, particle induction and stimulation are a potential sources of damage to well injectivity. After evaluating all possible causes of well damage, we have concluded that well damage can have no direct limiting effect on CO 2 injection. All problems associated with well clogging or well damage are understood and technically solvable. The injection of fluids into an aquifer will result in an increase of the fluid pressure of the aquifer: this causes the grain pressure to decline. This shift in pressure regime can cause fracturing of the rock matrix, opening up existing faults and/or induction of microseismicity. These effects depend largely on the mechanical properties of the reservoir rock. If the average aquifer pressure exceeds the overburden pressure, there is a risk of absidence. 5. E N V I R O N M E N T A L A S P E C T S The major risks of the underground storage of C O 2 are suffocation, groundwater acidification and pollution, and damage by CO 2 blow outs or absidence of the earth's surface. If large amounts of CO 2 leak to the surface they will create blanket-like cloud of CO 2 that fills topographic depressions. Since this CO 2 will drive away all oxygen, any people or animals that enter theses areas may suffocate. Malfunctioning of the CO 2 injection system can be reduced by the use of appropriate materials and by intensive maintenance. A simple additional device, integrated in the pressure monitoring system, could shut off the failing subsystem from the rest of the system and limit the emission of CO 2 to minimal quantities. If large amounts of CO 2 escape the reservoir rock and invade the subsurface, the groundwater may be affected. Groundwater naturally contains CO 2. Escaped CO 2 could increase the natural CO 2 concentration of the groundwater. A tenfold increase of CO 2 concentration in the groundwater will decrease the pH number by 1. The risk of CO 2 escaping from a storage location can be reduced by introducing peripheral observation wells. As a result of manmade pressure changes in the subsurface the earth's surface may gradually sink or rise. A symptom of these changes is the occurrence of microseismicity. Several cases of sinking or subsidence are well known and have been extensively documented. The data on the occurrence of absidence is limited but it is understood that the same theories as for subsidence can be applied. Regular monitoring of the possible rise of the earth surface is recommended. 6. S U B S U R F A C E A S P E C T S The similarities between natural gas storage in aquifers and C O 2 storage in aquifers are obvious. The technical reservoir engineering knowledge gained in underground gas storage can be directly applied. In the following sections the subsurface aspects of CO 2 storage are discussed, using a hypothetical aquifer. We deal with subsurface aspects from the surface downwards. From the results of calculations it can be concluded that all the pipeline diameters we investigated (4.0-7.0 inch) are capable of delivering the CO 2 at the aquifer injection location. A smaller pipeline diameter or an increased injection flow rate will reduce the CO 2 delivery pressure at this location.
1103
lopment during these two time periods. A simulation model was constructed, representing a 30x30 km part of the subsurface. An injection period of 50 years followed by a shut-in period of 100 years was simulated. Figure 3. shows the results of this simulation run. The delta CO 2 distribution map shows only the upper part of the subsurface model. The observed CO 2 bubble diameter at the top of the storage location can be estimated as 16 km at the end of injection period and grows to 18 km during the shut-in period. CO2 movements are only active if there are large differences in pressure between the injected CO 2 bubble and the constant pressure boundary of the model. From the simulation results it can be concluded that CO 2 storage in a quasi-infinite aquifer is possible. It is however impossible to define a storage efficiency factor due the infinite nature of the storage location. From all simulation work performed it can be concluded that the suitability of aquifers depends entirely on their size, within the boundary conditions stipulated. Displacement process will be dominated by channelling, viscous fingering and gravity segregation.
Iil~~~f~ifBi[
lill!i[!l l i l | | | | B ! i | | | R l | | | | l l i | l l U [ i'1==Ill =lII [[ t',litIll I Ill I!*,IH~ I l l l / l l l l l l l l l l i l ~ l i l l l t M M l I R l l J i ~ l l l
llillWlililUlllli
g / ~ D i
II
i~!i~ii!iii!i
I'II
|gltlll: "....... I I B D I I ,
Ii!ilI|~
iii!iiiii!
'
i~!ili::ii!: i
'ii'i [ i;i~:~; .....I ,,,WMItI~,,,~ ,..
~_-.L~.::'..':'~-'--
.......
I
Ini~ Inlnaq ~
IqnE]nt~
~
[ i q l ~ I~lIIOl I l l . i l l [ l ~ I ~ l ~ I I
illIliI! [
-.02
-.06
-.10
~
............. :::::";'--
I31qnE3Iqai~qnImad ~lllll'.lPIll',l q iI:llmJnli~ IIqnE ~
-.~14
~ I l l l I h q '.iII I I: Ih q [i IIi Ie IlII [ l i I l l I i ~ I i q [
PlamlqnElninEIIqnE2qn~
[iI~Iiq [
qnE]qnIDmqnDqaiTb ~
i~l~llII ( l l l ~ l I~l~lll I l i q g I lilIleq [
qdcsqm~
qln~qmIE]qn~lqnl]nb ~
qkn n
~
~
]
[
[
~
~
-."18
Fig. 3. Map of the difference in CO 2 concentration between
150-year map and the 50-year map
7. CO 2 STORAGE CAPACITY IN THE NETHERLANDS The underground CO 2 storage capacity of the Permian to Quaternary aquifers of the Dutch onshore has been estimated from published data about the subsurface. First, an inventory was made of potentially suitable aquifers for CO 2 disposal (permeability > 50 mD, a depth below 800 m and covered by cap rock) and information was gathered on net reservoir thicknesses
1102
We investigated the sensitivity of aquifer parameters and the scale of the CO 2 injectivity in an aquifer. A computer program was written to compute the pressure at increasing drainage radius as function of the permeability and the skin factor. Analysis of the results clearly shows that the aquifer permeability and the well skin factor are the controlling parameters of a CO 2 aquifer storage operation. It was observed that in nearly all cases when the permeability is 0.025 bll-n2 there are large pressure gradients near the well bore. Clearly, the overall aquifer permeability will play a decisive role when selecting potential aquifers for CO 2 storage. An aquifer in the Netherlands was selected to investigate and estimate the technical reservoir aspects of CO 2 storage in aquifers. (Aquifer data: porosity Brussel sand 30 - 36 %, permeability .05 - .6 gm 2, thickness 50 m). From the outset it was assumed that 6 wells would inject 15 000 ton a day of CO 2. This, in combination with the domed shape of the aquifer under study, makes it possible to reduce the simulation model to one-sixth of its original aquifer size. A pie-slice segment, with an angle of 60 degrees, was selected. The results of the CO 2 storage simulation runs reveal that CO 2 will breakthrough at the spillpoint after a cumulative CO 2 injection of 5.921 x 109 Nm 3. The results clearly indicate that the CO 2 distribution is dominated by gravity segregation. If we compare the results of the theoretical storage volume calculation with the results of simulation, than only 4.3 % of the volume is used. Figure 2 is a graphical representation of the model selection procedure and shows in cross section the CO 2 distribution at breakthrough. A further parameter sensitivity study 2 has shown that the CO 2 storage efficiency of a predefined part of an aquifer is small. For practical purposes a CO 2 storage efficiency of 1 to 6 % can be used, depending on the vertical transmissibility of the potential reservoir. I..
4500 m Selected reservoir A
A'
750 mSS 800 ~ss--~.
o~ o
,,
[
.:. :
". "-..
A'
-. 800 mSS 775 mS.S \ 750mSS \'-.\
ii::2a ij
S . . . . . . .
.... :.: "".. !: :.. ' " : . ~77T: ........ ~,...... 9............ ~,.,..:.:.: .................................................. :"~......:,:::,:ii
E
o
0.84 0.0 COa concentration
Fig. 2. Selection of simulation model, and the simulation results.
All the above work and reported efficiency factors refer to predefined storage locations with a known maximum storage volume, i.e. a storage location within a geological trap and an outer storage boundary. However, large aquifers without a geological trap structure are known to exist. If we relax the trap constraint it will be essential to uphold the constraint that the aquifer will need a impermeable top layer to prevent any CO 2 from leaking out through the top of the aquifer. The omission of a trap and the presence of a top seal will require the size of the CO 2 bubble to be controlled during the active injection period as well as in the subsequent period of storage. We performed a limited simulation study to investigate the CO 2 bubble size deve-
1104
and porosities. Next, the percentage of the volume confined by traps was assessed by determining the area occupied by closed structures on available depth maps and extrapolating these data to the entire Dutch onshore. Finally, the storage capacity was calculated from the trapped pore volume, assuming a CO 2 occupation of 2% and a CO 2 reservoir density of 700 kg/m 3. The uncertainty introduced by extrapolation may be considerable. The Triassic structures in the study area, for example, are all related to salt tectonics. Similar structures do not occur elswhere in the Netherlands. In addition, we were not able to define stratigraphic traps (created by facies changes) or very large structures extending beyond the mapped area. This also forms a major uncertainty. The Permian aquifers, for example, are thought to be confined by large fault blocks below a thick package of Zechstein salt. These blocks are expected to form huge traps, but are not included in the storage estimate because they could not be defined. We indentified more than 100 traps in those parts of the Netherlands where suitable depth maps were available. Of these only 50 traps are potentially suited for CO 2 disposal. The remaining structures are either too shallow or do not contain appropriate aquifers. The pore volume in these 50 traps is about 15.7 km 3, of which 2.1 km 3 contains oil or gas. Extrapolation of these results to the entire Dutch onshore leads to a total trapped pore volume of about 35.7 km 3. This corresponds to a CO 2 storage capacity of approximately 0.50 Gt. Previous storage estimates were considerably more optimistic. Van Engelenburg & Blok 3 proposed a capacity of 40 to 82 Gt CO 2. Huurdeman 4 made an estimate of 2.5 to 10 Gt CO2. The discrepancy between these figures and ours can be readily explained by the use of different information and constraints. Van Engelenburg & B lok did not take into account the presence of trapping structures whereas Huurdeman assumed that the entire pore volume in a trap can be saturated with CO 2, an assumption that has to be revised in the light of the results of our simulation experiments.
8. CONCLUSIONS 1) C O 2 storage in aquifers is technically possible. The knowledge about the technology of CO 2 injection in aquifers is adequate, but there is a lack of reliable subsurface data. 2) The C O 2 w a t e r displacement will be dominated by gravity segregation, by channelling, and viscous fingering over the whole subsurface depth range investigated. 3) The C O 2 storage efficiency of a predefined part of an aquifer is small. For practical purposes a CO 2 storage efficiency of 1 to 6 % can be used, depending on the vertical transmissibility of the potential reservoir. 4) The storage capacity of traps on onshore aquifers in The Netherlands is estimated at 0.5 Gt CO 2. 5) The estimated CO 2 storage capacity of quasi-infinite aquifers in general is problematic. It can, however, be stated that they have a large potential. References
1. 2. 3. 4.
van der Meer, L.G.H., Griffioen, J., Geel, IGG-TNO-report OS 92-24-A, February 1992 van der Meer, L.G.H., Paper presented at the ICCDR-2, Kyoto Japan, Oct. 1994. Van Engelenburg, B., & Blok, K. (1991), Report nr. G-91006. Huurdeman, A.J.M. (1992) TNO-report 91-250.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1105
WASTE REDUCTION AND THE STRUCTURE OF THE DUTCH WASTE SECTOR. Paulien de Jong and Maarten Wolsink Department of Environmental Science University of Amsterdam Nieuwe Prinsengracht 130 1018 VZ Amsterdam The Netherlands Abstract Waste is a source of greenhouse gases. In general, waste producers are not encouraged sufficiently to limit waste production. Reduction of waste can not be achieved by just formulating waste policy. The organizational conditions under which removal and processing take place frustrate the achievement of waste reduction. The final objective of this research project is to design a structure for the waste sector which will contribute to the reduction of waste streams for incineration and landfilling. In this paper the results of the first two phases are reported: an analysis of the Dutch waste sector and the identification of key variables in the organizational structure of the waste sector. 1. I N T R O D U C T I O N In many ways waste is a source of greenhouse gases. Waste incineration directly leads to the production of carbon dioxide. Waste on landfill sites also produces CO2, but it is assumed that carbon containing waste decomposes while emitting methane. Since the global warming potential of methane is about 30 times that of carbon dioxide, carbon containing waste disposal in landfills might even be a greater problem than incineration. Furthermore, there is a loss of energy through the energy content of products and materials. Decomposition means the loss of energy previously used for refinement of raw materials and for processing these materials, as well as the energy used in the process of production of goods. In addition, a variety of non-carbonic waste is produced during production processes with sometimes high CO2 emission per kilogram. A reduction of waste which will either be dumped at landfills or incinerated, will be a contribution in controlling the greenhouse problem. 2. WASTE P O L I C Y AND T H E WASTE S E C T O R In Dutch waste policy and in research a great deal of attention has been paid to the formulation of targets and regulations and to the selection of policy-instruments. Nevertheless, the tendency of growing waste streams has not yet been stopped in the Netherlands. The expected waste streams for incineration are growing and a large amount of new incineration capacity is planned. The size and composition of waste streams is determined by several factors like demographic developments (population growth, composition of households), economic developments (prosperity) and technological trends (mixing of materials). The fact that the size and composition is also determined by the conditions under which the removal and processing of waste takes place is too often neglected. In many ways organizations (companies, public services, authorities) have conflicting interests. Their objectives often conflict with the policy goal of waste prevention. Neglecting the importance of organizational conditions is not an uncommon phenomenon. It is not restricted to the waste sector. A conflict of objectives also exists in the energy sector. Saving
1106 energy by means of a reduction of energy consumption is a policy goal, but generally it is not an interest of utilities that gain economic benefits from the supply of more energy. Due to the problem of conflicting objectives is the fact that policy ultimately has to be implemented within society by other actors than the policy formulating bodies. These actors have their own objectives and interests, which may differ from those of the policy makers. These actors try to find ways to seek the fulfilment of their own objectives and they have the capability to frustrate the seeking of fulfilment of others. Most of the actors have no interest in waste reduction or in altering the kind of waste that has to be processed. 3. A C T O R S IN THE W A S T E S E C T O R The best way to get insight in the structure of the waste sector and to understand the dynamics of it, is to adopt the idea that the organizations are linked together in an interorganizational network. Characteristic for such a network is the fact that the individual objectives of the organizations participating in the network are more important than the objective of the network itself (1). It seems as if they cooperate because they share interests and objectives, but more often the real reason for cooperation is mutual dependency (2). Among the public and private organizations which try to influence the circumstances under which the removal and processing of solid waste take place, seven groups can be distinguished. The categorisation is primarily based on the interests and activities that organizations have in common and secondarily on which phase in the material life cycle their activities have impact (3). 1. Waste generators 2. Waste collectors 3. Waste processors 4. Waste disposers 5. Policy makers 6. Research groups and consultancies 7. Interest groups and umbrella-organizations The first four categories of actors are those who are participating on the waste market. The waste sector is much broader: apart from the participants on the waste market other groups of actors are part of the waste sector. These are primarily governmental organizations that have to formulate waste policy. Secondly there are organizations that provide data and ideas to support the policy making process, like research groups and consultancies. Thirdly we distinguish organizations that either defend interests of groups of other actors, or try to influence policy (lobbying). After analysis of the relations between the actors of the seven categories we established several structural impediments for the reduction of waste. 4. I M P E D I M E N T S FOR PREVENTION IN THE WASTE SECTOR The governmental institutions in the Netherlands find themselves in a paradoxical situation. Municipalities invest in disposal plants from the perspective of environmental hygiene. They invest in incinerators which have to be built in accordance to strict environmental standards. The investments are large.and take a long period of time to write off. Therefore there is no interest in source-reduction which can be implemen-ted in the short term. To the contrary, it creates an interest in assurance o f long term waste supply.
1107 The long term of writing off causes short term risks when installed capacity is not fully utilized. Waste processors attempt to guarantee a sufficient flow by tying waste suppliers to long term contracts. As a rule, governmental institutions become tied, while private enterprises remain free to change their supply from one processor to another. Private organizations which are not contractually bound can offer "extra" waste, but then they can negotiate about the rates of the incineration. It is in the interest of the waste generating industrials or smaller enterprises that the above described situation of waste handling and management remains the same. Overcapacity leads to lower incineration rates. In general, it is in the interest of private organizations to keep authorities and policy makers in a situation of uncertainty about the amount, sort and composition of waste that will be released. More information and better planning of facilities will not only limit bargaining opportunities of enterprises and lead to higher processing tariffs. On the other hand exchange of information will give policymakers tools to formulate strict prevention goals in quantitative and qualitative sense. Policy and implementation is connected to a certain level of administration. In the meanwhile other, private organizations operate on a higher level. Those private organizations can not be forced to implement the policy of public authorities. The legal jurisdiction offers authorities a basis for power but it also restricts them. Public bodies are participating in the waste sector with different roles at the same time, which may sometimes cause entanglement: * A role as representative of the law (functions of control and issuing permits) * A role as a participant on waste market. * A role as policy makers, in which general public interests must be served. 5. S T R U C T U R A L E L E M E N T S IN T H E WASTE S E C T O R Based on the analysis of the Dutch waste sector and a rough inventory of twelve different countries, five structural elements determining barriers for waste prevention in the waste sector were indicated. These five variables have been used as criteria for the selection of three cases for a multiple case study (4). This is the next phase of the research, directed at the identification of elements in the organizational structure of waste sectors, which might be implemented in the Dutch waste sector. The five structural elements are: (5) 1. Scale of organizations which are handling waste. 2. Functional separation of tasks and responsibilities between actors 3. Activities directed at input and output attributed to different organizations. 4. The role of the authorities: utility-function vs. market participation. 5. Attribution of responsibilities for waste prevention to existing (or merged, or separated) organizations, or new organizations. 6. FURTHER RESEARCH PLANS The starting point of this research project is that size and composition of waste streams are partially determined by the organizational conditions under which the removal and processing take place. After a first examination of the structure of the waste sector in the Netherlands it became clear that there exist several aspects in the organizational structure of the waste sector that impedes the stimulation of waste minimalization incentives.
1108 To obtain ideas about improvements for (parts of) the structure of the waste sector that may lead to stimulation of waste reduction, a multiple case study has to be carried out. Therefore a broad inventory on significant structural conditions in twelve industrial societies was done. Out of this list of twelve, three cases were selected. Significant characteristics of the waste sectors of New Jersey (US), Nord-Rhein Westfalen (Germany) and Denmark will be investigated. The aim of these studies is to propose improvements for the structure of the Dutch waste sector. Another way to search for ideas for improvement can be found in comparison of the waste sector with other sectors in society. A literature study of this subject will be made. The final objective of this project is to design a structure for the waste sector which will contribute to the reduction of waste streams for incineration and landfilling. The last step in the research project will bring all results about possible improvements together and will have to result in a new concept for the structure of the waste sector. Ex ante evaluation has to be carried out to prospect the possible problems in the implementation of such a new designed structure. Also ex ante evaluation has to be done on the effects of altering (parts) of the existing structure in size and composition of waste streams.
7. REFERENCES: 1. Godfroy, Perspectieven op organisaties: organisatiepsychologie en -sociologie 2, Open Universiteit, Heerlen, 1990. 2. D. Marsh and R.A.W. Rhodes, Policy networks in British Government, Clarendon Press, Oxford, 1992. 3. P. de Jong and M. Wolsink, De structuur van de Nederlandse afvalsector, IVAM/UvA, Amsterdam, 1993. 4. R.K. Yin, Case study research; Design and methods, Sage, Beverly Hills, 1984. 5. P. de Jong, Verkenning afvalsituatie in: Belgie, Denemarken ..... Zwitserland; Internal report, IVAM/UvA, Amsterdam, 1994.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1109
Institutional barriers to waste reduction in Finland J. Hukkinen Maastricht School of Management, P.O. Box 1203, 6201 BE Maastricht, The Netherlands
Abstract Waste reduction, which is the top priority of waste management in industrialized countries, can significantly reduce greenhouse gas emissions. The aim of this study was to improve the capacity of Finnish decision makers to implement long-term waste management policies, such as waste reduction. Focused interviews were conducted in 1992 with 24 researchers, consultants, politicians, government officials, and entrepreneurs. The interviews are the empirical basis of cognitive maps, which represent the causal models underlying expert decisions. The analysis indicates that Finland's environmental institutions integrate conflicting policy interests and systematically prevent decision makers from taking long-term policy action. Administrative and procedural decoupling of conflicting interests characterizes the proposed reforms.
1. INTRODUCTION Waste reduction is the top priority of waste management policy in industrialized countries. Waste reduction can significantly reduce greenhouse gas emissions, because it entails a radical reduction in the material and energy intensity of industrial production, particularly in the burning of fossil fuels. But implementing long-term environmental policies, such as waste reduction, involves significant socio-political and institutional contingencies (1). This study set out to explore the institutional constraints of long-term waste management policy in Finland. The goal of the study was to develop strategies to improve the capacity of decision makers in Finnish waste management to control and adapt to uncertainties of the very longterm future. The main proposition of this study is that there exists an ironclad relationship between Finnish environmental institutions and the expert beliefs that uphold the institutions. Institutions such as laws, regulations, cultural traits, and habits are the rules that organizations follow in the social game. Institutions determine what individuals perceive to be possible to achieve, and these perceptions in turn shape social institutions (2). According to this study, far-sighted waste reduction policy is currently impossible in Finland, because the dominant environmental institutions nurture waste management experts' short-term operating assumptions, which in turn weave together the short-sighted institutional structure.
1110 2. COGNITIVE MAPPING OF EXPERT BELIEFS The perceptions of central decision makers and experts in Finnish waste management were investigated by means of a policy analytical approach known as cognitive mapping, which is based on the notion that it is not the empirically verified reality that determines our decisions, but rather what we perceive to be the reality (3-5). Analysis of the mental models upon which experts and decision makers in waste management base their decisions identifies many institutional constraints of policy making. Focused interviews were conducted in the summer of 1992 with 24 Finnish waste management experts and decision makers on problems that they perceive to become central in the country's waste management in the next 50 years. The interviewees, who were selected through snowball sampling (6), represent various interest groups in waste management, with 4 consultants, 5 researchers, 5 politicians, 6 government officials, and 4 entrepreneurs. The interviewees mentioned 282 different, causally related problem statements in the field of waste management. For analytical purposes the problem descriptions were coded as problem networks (7). Each interviewee's scenario of waste management problems can be represented as a problem network composed of nodes and links, where nodes are problem statements about future waste management and links the causal relationships between them as expressed by the interviewees. Since the interviewees' descriptions of future problems contain some elements in common, individual problem networks can be aggregated into "socially constructed" scenarios. Problem networks reveal mental constructs that an individual expert does not necessarily perceive. The circular network, or loop, is the most interesting one for policy planning, because it obscures the difference between cause and effect. Since the loops are made of statements that the interviewees perceive as problematic, they are unstable, positive feedback loops. Problems included in a loop keep reinforcing themselves (on feedback, see 8-9).
3. ENVIRONMENTAL CORPORATISM IN FINLAND The results of cognitive mapping can be summarized as follows: First, waste management experts typically describe future waste management problems as loops. Fourteen of the 17 loops that emerged in network aggregation by interest group were mentioned by individuals, and half of the 24 individuals mentioned loops. Loops are held together by a cognitive goal conflict between profit maximizing goals, which prioritize shortterm economic profitability, and sustainable goals, which aim at preservation of ecosystems over generations. Second, experts do not let the goal conflict interfere with their day-to-day decision making. All of the loops indicate that expert advise and decisions are guided by profit maximizing, short-term operating assumptions. Third, the loops indicate that the profit maximizing operating assumptions are institutionalized in the administrative, technological, economic, and political structures of the Finnish society. This phenomenon will in the following be referred to as environmental corporatism (on social corporatism, see 10). Its most prominent feature is the systemic integration of conflicting environmental policy interests, to the extent that open conflict resolution is impossible. Finally, the interviewees expect that profit maximizing operating assumptions will lead to troublesome consequences in the long run. The aggregated problem networks have 54 terminal problems, i.e., problems without
1111 outgoing causal links, 40 of which describe threats to the survival of Finnish waste management organizations, society, and ecosystem. Each of the 17 loops identified in the group-level aggregation of problem networks and the terminal problems that result from them support the results. Interviewee no. 1 (a government official) mentioned loop 3, which illustrates the results (Figure 1). The loop describes how Finland's semi-governmental hazardous waste treatment monopoly Ekokem is in a cycle of planning excess treatment capacity only to find the capacity inadequate when environmental regulators order more of the nation's hazardous wastes to be treated at the plant. The dual goals of short-term economic profitability of the plant and long-term ecological safety of waste treatment forces decision makers into a cognitive dilemma, in which they can but alternate their allegiance between the conflicting goals (the first result).
4 Conflict between waste reduction and disposal is polarized.
Environmental officials direct more wastes to Ekokem.
19 Ekokem is designed and redesigned to have excess disposal capacity.
I
~
Ekokemreceives so much waste that it cannot incinerate all of it.
Figure 1. The government officials' loop 3.
Loop 3 also shows interviewee 1 to believe that actual waste management policies will conform with profit maximizing operating assumptions (the second result). Ekokem is described as an automaton, which keeps on expanding as a result of continuous planning for excess capacity. Two features of environmental corporatism secure the operation of the automaton (the third result). First, the corporatist decision making system views waste management purely from a techno-economic point of view, which obfuscates the socioeconomic conflicts of interest between waste reduction and waste treatment. Second, implementors and regulators have intimate linkages in the administration of hazardous waste management - - the Ministry of the Environment is the top regulator of hazardous waste management but also owns a third of Ekokem - - which secures the flow of waste to Ekokem. Finally, the terminal problems emanating from loop 3 support the fourth result. According to interviewee 1, decision making may become systematically irrational, enterprises may lose all interest in sustainable waste management and focus on turning a profit regardless of means, and regulators may end up shifting waste from one environmental sector to another.
4. R E C O M M E N D A T I O N S The central principle of the following recommendations is the administrative and procedural separation of conflicting environmental policy interests. The objective is not to polarize environmental conflicts, but to resolve issues through existing conflict resolution
1112 mechanisms. Where they do not exist, they should be created. First, the close integration of implementation and regulation in Finnish waste management persuades regulators to compromise long-term ecological considerations for the sake of shortterm economics. Regulation should therefore be clearly separated from implementation in waste management. Second, the socio-economic conflict between different technological stages of waste management is a problem particularly in public waste management agencies, which are not just technical implementors but policy makers as well. Administrative separation of waste reduction, recycling, collection, and disposal in the public sector would increase the organizational autonomy of sustainable principles. Third, administrative separation of the technical steps of waste management would not remove the economic friction between them. It would just transform an intra-agency conflict into an inter-agency one. More attention should therefore be paid to political procedures for resolving such conflicts. Environmental impact assessment should be developed into a procedure that would promote scientifically enlightened political discourse on environmental policy (11). Policy choices in waste management would be made after comprehensive public criticism, much like the scientific community selects theories after competition between scientists. Finally, more neutral regulatory mechanisms, such as economic regulation, would dismantle some of the structures of environmental corporatism. The institution of environmental taxes, for example, would require political decisions, which would transfer negotiations from the closed corporatist arena to the open parliamentary one. This would force politicians to make the difficult choices between short-term economics and long-term ecological sustainability. What is more, it would allow the regulators to concentrate on what they do best, namely, monitor and evaluate the effects of regulation on environmental quality.
5. R E F E R E N C E S
1 Bolin, B. (1994). "Science and Policy Making," Ambio, Vol. 23, No. 1, pp. 25-29. 2 North, D.C. (1992). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. 3 J. Management Studies (1992). Special Issue on Cognitive Maps. Vol. 29, No. 3. 4 Berger, P.L. and T. Luckmann (1967). The Social Construction of Reality: A Treatise in the Sociology of Knowledge. New York: Doubleday Anchor Books. 5 Weick, K.E. (1969). The Social Psychology of Organizing. Reading: Addison-Wesley. 6 Goodman, L.A. (1961). "Snowball Sampling," The Annals of Mathematical Statistics, Vol. 32, No. 1, pp. 148-170. 7 Pearl, J. (1988). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. San Mateo: Morgan Kaufmann. 8 von Bertalanffy, L. (1968). General System Theory: Foundations, Development, Applications. New York: George Braziller. 9 Prigogine, I. and I. Stengers (1984). Order Out of Chaos: Man's New Dialogue with Nature. Toronto: Bantam Books. 10 Pekkarinen, J., M. Pohjola, and B. Rowthom (eds.) (1992). Social Corporatism: A Superior Economic System? Oxford: Clarendon Press. 11 Taylor, S. (1984). Making Bureaucracies Think: The Environmental Impact Statement Strategy of Administrative Reform. Stanford: Stanford University Press.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1113
Energy Production on Farms - Sustainability of Energy Crops H. van Zeijts Centre for Agriculture and Environment, P.O. Box 10015, 3505 AA Utrecht, The Netherlands
Abstract
This article reflects the results of a study on sustainability of energy crops. Contribution to the reduction of the greenhouse effect and other environmental effects were investigated for the Netherlands. The study assumed that energy crops are grown on set aside or grain land. Generating electricity and/or heat from hemp, reed, miscanthus, poplar and willow have the best prospects. These crops are sustainable and may in the future be economically feasible. Ethanol from winter wheat has the most favourable environmental effects, but is economically not interesting. Liquid fuels from oil seed rape and sugar beet are not very sustainable.
1. I N T R O D U C T I O N Energy from arable crops may be interesting for both agriculture and the environment. Energy crops can provide new opportunities for agriculture; for the environment it can mean a reduction in the greenhouse effect. The Dutch Centre for Agriculture and Environment (CLM) has investigated the contribution of energy crops on limiting the greenhouse effect. The study was done for Dutch arable farming. CLM has investigated nine energy crops: - winter wheat and sugar beet, for the production of bio-ethanol to replace petrol; - oilseed rape, for the production of methyl-ester to replace diesel; - the annual crops hemp and silage maize and the perennials reed, miscanthus, poplar and willow, for the production of electricity and/or heat. Besides the contribution of these crops on reducing the greenhouse effect attention was also given to other criteria: the emission of nutrients and pesticides, contribution to aridity, erosion, utilisation of by-products and waste, use of space and the contribution to natural and scenic values.
2. E N E R G Y
CROPS AND GREENHOUSE
EFFECT
Presently biomass provides only 2% of the European energy requirement 1. Since the seventies, the role of biomass in the energy supply has regularly been discussed. In the Netherlands this discussion was initially unfavourable for energy crops, principally due to the low profit and the high cost price of Dutch arable products. The attention drawn by the greenhouse effect and the poor economic circumstances of Dutch arable farming has brought the cultivation of energy crops to the fore once again. Recent studies have shown that energy crops can indeed limit the greenhouse effect2.
1114 Energy crops have two effects: - avoiding emissions from fossile energy sources; - fixation of carbon from CO2 in biomass.
Avoiding emissions from fossile energy sources Energy from crops replaces fossile energy. Thus, the emission of CO2 caused by using fossile fuels is lowered. Carbon dioxide from the use of energy crops is part of a cycle: it was taken up during the growth of the energy crops. On the other hand cultivation, transport and processing also causes emission of greenhouse gases. The balance can be calculated in terms of the net avoided emission of CO2. Table 1 indicates the net avoided emission of CO2. Generating electricity by burning reed, hemp, miscanthus and poplar has a great effect on the net avoided emission of CO2 per ha. Producing transport fuels from winter wheat, sugar beet and oilseed rape scores far less favourable on the net avoided emissions.
Fixation of carbon from CO2 in biomass Energy crops temporarily fix CO2 from atmosphere in biomass. This also contributes to a reduction of an increase of the greenhouse effect. The fixed amount of CO2 correlates with the growth stadium of the crop. However, on a longer term the amount of fixed carbon from CO2 remains the same. On a long term scale, the contribution of fixed CO2 on limiting the greenhouse effect is therefore far less important than the contribution of the net avoided emissions from fossiele fuels. Table 1 shows that over a period of a hundred years the contribution of fixed CO2 is only a few percent of the contribution of the avoided emission from fossile fuels.
Table 1 Net avoided CO2 emission and CO2 fixation, for the Dutch central clay area 3 energy crop
reed hemp miscanthus willow maize poplar sugar beet winter wheat oilseed rape
type of energy
electricity, electricity, electricity, electricity, electricity, electricity,
50 MW powerplant 50 MW powerplant 50 MW powerplant 50 MW powerplant 50 MW powerplant 50 MW powerplant ethanol ethanol methyl-esther
net avoided CO2 emission (ton ha -1 yr-1) 20.3 18.0 15.3 14.6 13.2 10.7 6.9 3.1 3.0
CO2 fixation, (% of total net avoided CO2 over 100 yr)
1115 3. O T H E R E C O L O G I C A L E F F E C T S Basic assumption in the study is that energy crops are grown on the area that has been set aside and on part of the area that presently is used for grain production. The environmental consequences of this substitution are divided into direct and indirect effects. Here we just work out a few examples.
Direct effects Direct effects have to do with the following question: is the environmental burden of energy crops higher than those of fallow land or grain cropping? Examples for direct effects are: 9 The use of pesticides in winter wheat is relatively high. Hemp, reed, miscanthus, poplar and willow only need low quantities of pesticides. This means that substituting winter wheat by these crops leads to lower emissions of pesticides. 9 Winter wheat and crops on fallow land require little water. Substitution by energy crops leads to increased water use and therefore contributes to higher aridity of the land. This may have negative effects on nature and on agriculture itself. Indirect effects Indirect effects are effects on other crops at farm level and effects on a regional and national level: 9 At farm level, substitution of fallow and grain land by energy crops has effects on the emission of nutrients and pesticides and on erosion in the rest of the cropping pattern. On Dutch arable farms this concerns sugar beet and potato. In particular, fitting in perennial energy crops leads to intensification of the cropping pattern that may cause problems from an environmental point of view. 9 Fitting in energy crops has consequences for natural and scenic values at farm level as well as at a regional level. For example winter wheat and oil seed rape have great potential natural values and can contribute to natural values at farm and regional level in a positive way. 9 Growing energy crops also influences the Dutch animal breeding sector, because these crops compete with fodder crops for use of land. If arable farmers grow energy crops instead of grain, there will be less native grain on the market. As most of Dutch grain is processed into animal feed, this means that the imports of raw materials for animal feed will increase. The extra transport of raw materials for animal feed increases the CO2-emission and causes an extra disturbance of the Dutch national mineral balance.
4. C O N C L U S I O N S Table 2 summarises overall results on both ecological and economical sustainability of nine energy crops. Each of the used criteria for ecological sustainability (see w1) is given equal weight. Of course this choice is arbitrary: in practice, the ratios in each situation are different. Given unequal weight the ranking may slightly change. The ranking for economical sustainability is derived from results of other studies 2 4.
1116 Table 2 Ecological and economical sutainability of growing energy crops in the Netherlands energy crop
type of energy
ecological sustainability
economical sustainability
reed hemp miscanthus willow maize poplar sugar beet winter wheat oilseed rape
electricity and/or heat electricity and/or heat electricity and/or heat electricity and/or heat electricity and/or heat electricity and/or heat ethanol ethanol methyl-esther
0/+ 0/+ 0 0 -/0 0 + -/0
+ 0 + + 0 + -/0
+ : good long-term perspectives from an ecological and economical point of view - : low perspectives, compared to the other energy crops
From table 2 we can draw the following conclusions: 9 Liquid fuels from oilseed rape and sugar beet and electricity from maize score worst. 9 Production of ethanol from winter wheat scores highest in terms of ecological sustainability. But economic studies reveal that the costs per ton CO2 net avoided are rather high. 9 Generating electricity and/or heat from reed, hemp, miscanthus, willow and poplar can be a sustainable way to reduce the greenhouse effect. Generating electricity from crops will be profitable in the Netherlands in the near future, if the set-aside scheme of the European Union is continued and an environmental tax on energy or a subsidy per ton of avoided CO2 is introduced. But opportunities for energy crops are better in other European countries, due to the intensive Dutch arable farming and high land prices. The long-term perspectives for energy crops are uncertain. Therefore it is advisable to investigate at a European level what other future functions for the land may be supplanted by energy crops.
5. REFERENCES 1. Hall, D.O. (1991). 'Biomass energy'. In: Energy policy, yr. 19, nr. 8. p. 711-737. 2. Lysen, E.H., C. Daey Ouwens, M.J.G. van Onna, K. Blok, P.A. Okken and J. Goudriaan (1992). D e h a a l b a a r h e i d van de p r o d u k t i e van biomassa voor de N e d e r l a n d s e energiehuishouding. Netherlands Agency for Energy and the Environment, Apeldoorn. 3. Zeijts, H. van, E.B. Oosterveld and E.A. Timmerman (1994). Kan de landbouw schone energie leveren? - Onderzoek naar de duurzaamheid van energiegewassen. Centre for Agriculture and Environment, Utrecht. 4. Biomass Technology Group (1994). Conversieroutes voor e n e r g i e g e w a s s e n - Een overzicht van bestaande en mogelijke routes voor de produktie van elektriciteit en transportbrandstoffen. University of Twente, Enschede.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1117
A simple method to estimate regional yields of biomass crops S. Nonhebel, J. Goudriaan, R. Rabbinge, D e p a r t m e n t of T h e o r e t i c a l P r o d u c t i o n Ecology, A g r i c u l t u r a l U n i v e r s i t y Wageningen, PO BOX 430, 6700 AK Wageningen, The Netherlands.
Abstract The use of biomass crops as an energy source is frequently mentioned as an option to reduce CO 2 emissions. To evaluate the possibilities reliable yield e s t i m a t e s of biomass crops are required. In this p a p e r a simple m e t h o d is developed to estimate regional yields of various biomass crops, based on the linear relation between intercepted light and biomass production. The quality of the estimates was studied by using the method to estimate yields of several a g r i c u l t u r a l crops in two regions in The N e t h e r l a n d s . In general a deviation of less t h a n 10 % was found between actual and estimated average yield.
1. I N T R O D U C T I O N One of the options to diminish present CO 2 emissions is replacing a part of the fossil fuels by energy from crops. During their growth crops capture CO2, which is a g a i n released when the crops are used for e n e r g y supply. This m e a n s t h a t CO 2 is recycled. For a s s e s s m e n t of the possibilities of this energy source reliable estimates of the yields are required, since the expected yield of an energy crop determines the outcome of the evaluation. Candidate biomass crops (used for electricity production) like willow and m i s c a n t h u s are not yet grown on a large scale, so t h a t it is difficult to assess their yields. In current energy scenarios and in recent studies of possibilities of using biomass crops as an energy source (1,2,3), the yield estimates are based on a limited n u m b e r of field experiments and data are used for large areas (sometimes even globally). However, the yield of a crop is strongly determined by its growing conditions and large differences in yields between regions and years are observed. This
1118 implies t h a t yields obtained in one region in one year cannot easily be translated into yields in other regions or other years. For energy supply not the yield in one particular field is of interest but the amount of energy t hat can be produced in a region. Therefore yield estimates have to be on a regional rather t h a n on a farm scale. This regional yield cannot be determined in field experiments. Here a simple method is developed to estimate regional yields of biomass crops. The method is based on knowledge obtained in agricultural research.
2. METHOD Research on various agricultural crops has shown t hat a linear relation exists between the amount of solar radiation intercepted by the crop and the above ground biomass produced (4). The slope of the line is the so-called Light Use Efficiency (LUE). Under optimal conditions a LUE value of 1.4 g MJ -1 (global radiation) is found for most agricultural crops, and also for fast growing trees (4,5). This implies that the yield of a crop (Yp) can be calculated by: Yp=Iin t x LUE x HI
(1)
in which Iint is the radiation intercepted during the growing season and HI is the harvest index (fraction of the total biomass that is harvested). Data of both HI and Iint can be derived from literature (6). Yp is the production under optimal circumstances (the crop is growing with ample supply of water and nutrients and free from pests, diseases and weeds), it is a measure of what is potentially possible under given conditions In practice conditions are seldom optimal and and actual yields are generally lower than the calculated potential yield. To obtain actual yields, a correction is needed to account for suboptimal growing conditions. The ratio between actual and potential yield can be interpreted as a characteristic for the type of agriculture in a region. Here this ratio will be called the Yield Correction Factor (YCF). The value of the YCF can be determined by using the above described method for an agricultural crop from which yield data are available and divide actual obtained yield (Ya) by calculated potential yield (Yp): YCF = Ya Yp
(2)
1119
Estimates for regional crop yields under present growing conditions (Yr) can be obtained by using eq 3 Yr=Yp • YCF
(3)
3. RESULTS AND DISCUSSION 3.1. D e t e r m i n a t i o n Y C F f o r t w o d i f f e r e n t regions In The N e t h e r l a n d s potatoes are planted in April and the crop is harvested in September (7). The total amount of global radiation intercepted by the crop is about 1400 MJ m -2. Using eq 1 leads to a potato yield of 15.0 ton ha -1 (HI of a potato crop is 0.75). In 1992 the actual potato yield in Flevopolders (region 1, fig 1) was 10.6 ton ha "1 and in Veenkoloni~n (region 2, fig 1) it was 8.6 ton ha -1 (dry matter) (8). Applying eq. 2 results in a YCF for region 1 (YCF 1) of 0.71 and for region 2 (YCF 2) of 0.56.
S I
O0 km
Figure 1. Location of the regions mentioned in text.
1120
3.2. Regional yields of agricultural crops Since there are no data on average regional yields for biomass crops, the validation possibilities for the method are limited. The only available averages are those from the present agricultural crops. The method described above was used to estimate yields of three agricultural crops and the results were compared with actual average yields obtained in the two regions in 1992 (table 1).
Table 1 Comparison between actual yields (Ya) of agricultural crops and estimated yields (Yr) for two different regions, YCFI=0.71 and YCF2= 0.56. Deviation (dev in % of Ya) is also given. Yields (harvestable biomass) in ton dry matter ha -1. region 1 crop winter wheat sugar beet maize
region 2
Yr
Ya
dev.
Yr
Ya
dev.
7.6 15.1 14.6
7.3 15.4 15.4
4% 2% 5%
6.1 12.3 11.9
5.4 12.0 13.2
13% 3% 10%
The deviation between simulated and actual yields was not large, which shows that the method is a suitable tool for estimating crop yields.
3.3 Regional yields of biomass crops The application of the YCF for biomass crops assumes that knowledge on how to grow such a crop is comparable to that of an agricultural crop. However for the potential biomass crops this is not yet the case. So it is likely that the regional yields of these crops will be lower t han calculated here. F u r t h e r uncertainties exist with respect to values of intercepted radiation and harvest index of the biomass crops. This means that the yield data used for the biomass crops are only indicative. However, it can be concluded t h a t the potential biomass production of 'a' perennial biomass crop lies between 18 and
1121 22 ton ha -1. Present regional averages of these crops would be 13-15 ton ha -1 for the high yielding regions and 10-12 ton ha -1 for the low yielding regions in The Netherlands (table 2). Estimates of biomass crops yields given in literature for present conditions are 10-12 ton ha -1 (1, 9) which agree with values found here. The value of YCF is time and space dependent and must be determined for each region individually. It is likely that YCF values in other regions in Europe will be much lower due to less well developed agriculture which will result in lower yields. To improve yield estimates, detailed field experiments are required to obtain more information on light interception and harvest index of candidate biomass crops.
Table 2 The calculated potential yield (Yp) and the regional yields (harvested biomass, in ton (dry matter) ha -1) of three candidate biomass crops in two regions (Yrl, Yr2)" YCF1 =0.71 and YCF2= 0.56. Yields (harvested material) in ton (dry matter) ha -1.
Crop
harvested parts
Yp
Yr 1
Yr2
Miscanthus Poplar Willow
stems stems/branches stem/branches
21.9 18.0 19.6
15.3 12.6 13.7
12.3 10.0 11.0
4. CONCLUSIONS Based on the result that yields of agricultural crops are estimated with an inaccuracy of 10%, it is concluded that the estimation method described can be a useful tool in research on the possibilities of using biomass crops for energy supply. Estimated biomass crops yields are 10-15 ton ha -1 under present conditions in The Netherlands.
1122 Acknowledgement This work was funded by the Dutch National Research Program on Global Air Pollution and Climate Change Project nr 853117
5. REFERENCES 1 Hall, D.O., et a1.,1993. In: Johansson, T.B., Kelly, H., Reddy, A.K.N., Williams, R.H. (eds) Renewable Energy, Island Press, Washington, pp 593651. 2 0 k k e n , P.A., Ybema, J.R., Kram, T., Lako, P., Gerbers, D., 1994. Energy systems and CO 2 constraints. Netherlands Energy Research Foundation, 3 4 5 6 7 8 9
Petten, The Netherlands. Lysen E.H, et al., 1992. De haalbaarheid van de productie van biomassa voor de Nederlandse energie huishouding. Novem, Apeldoorn. (in Dutch) Monteith, J.L., 1977., Phil. Trans.R. Soc. Lond. 282, 277-294. Cannell, M.G.R. 1989. Scand. J. For. Res. 4, 459-490. Nonhebel, S. 1994.,A simple model to estimate regionally average yields of biomass crops, submitted to Biomass and Bioenergy Jong, J.A.,1985, De teelt van aardappelen,Drachten (in Dutch). PAGV, 1993. Kwantitatieve informatie voor de akkerbouw en de groenteteelt in de vollegrond, 1993-1994, PAGV, Lelystad (in Dutch). Christersson, L.et al, 1993. The Forest Chronicle 69, 687-693.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1123
Energy Accounting on Farms J.A.M. van Bergen Centre for Agriculture and Environment, P.O. Box 10015, 3505 AA Utrecht, the Netherlands
Abstract
This article describes the development of an energy accounting. This is a management tool to give farmers a clear understanding of their energy use and of the emission of greenhouse gases on their farm. Results are given of one year accounting on dairy farms and on pig farms. The results show large differences in energy use and in emissions of greenhouse gases between individual farms. These differences indicate that a substantial reduction of emission of greenhouse gasses is possible.
1. I N T R O D U C T I O N The Dutch Centre for Agriculture and Environment (CLM) has figured that agriculture in the Netherlands contributes 12 percent to the national emission of greenhouse gases (1). These emissions of carbondioxide, methane and nitrous oxide in Dutch agriculture have also been calculated per sector. Dairy farming has the major contribution with a total emission of 57 percent. Intensive livestock farming contributes 20 percent, greenhouse cultivation contributes 16 percent and field cultivation 7 percent (1). The high score of dairy farming and intensive livestock farming is partly caused by the high use of indirect energy. The importance of the use of indirect energy in animal production has been pointed out in literature earlier (for example 2,3). The CH4-emission by ruminant digestion of feed and the N20-emission by soil processes are the other causes of the high score of dairy farming. So far, no management tool is available to monitor the emissions at farm level. CLM has started a project to develop an accounting system to calculate the use of energy and the emission of greenhouse gases at individual farms. It should give farmers a clear understanding of their energy use and of the emission of greenhouse gases on their farm. The instrument is simply denominated as the Energy accounting. Because of the important contribution of dairy and intensive livestock farming to the greenhouse effect, the energy accounting is first being developed for these sectors.
2. F R A M E W O R K OF AN E N E R G Y ACCOUNTING SYSTEM Setting up an energy accounting system includes developing a registration form, a methodology to calculate emissions and a way to present farmers the results. In addition the project deals with advice on possible strategies and measures to reduce emissions of greenhouse gases. In this article we describe only the framework of the energy accounting.
1124
The energy accounting is based on administrative management data of farms, for example meter readings on energy use or energy bills, data on use of fertilizers and feed concentrates. These farm data are combined with standard factors on energy values and emissions of greenhouse gases. The calculations take place in six modules: - direct energy use and CO2-emission for both dairy and intensive livestock production; - indirect energy use and CO2-emission for both dairy and intensive livestock production; - CO2-emission by mineralization of peat for dairy production; - CH4-emission by feed fermentation for dairy production; CH4-emission by slurry storage for intensive livestock production; under Dutch conditions the CHa-emission by storage of slurry from cows is neglectable; - N20-emission by soil processes for dairy production. -
Table 1 Framework of the energy accounting Module
Basic data per farm
Calculations w i t h standard factors
Results MJ
Results CO2-emission
1. Direct energy use fuel/electricity
meter readings
MJ/I diesel, kWh
MJ
kg CO2
2. Indirect energy use
use of fertilizers, feed concentrates, tools and buildings
MJ/kg N, kg concentrate etc.
MJ
kg CO2
3. CO2-emissionby mineralization of peat
area peat s o i l , drainage
CO2-emission/ha
kg CO2
4. CH4-emissionby feed fermentation
number of c a t t l e , feed ration and level
CH4-emission/cow
kg CO2-equi.
5. CH4-emissionby slurry storage
slurry q u a n t i t y storage days
CH4-emission/ton
kg CO2-equi.
6. N20-emissionby soil processes
area, soil type, fertilization, grazing, drainage
N20-emission/ha
kg CO2-equi.
Totals per farm per product
MJ MJ
kg CO2-equi. kg CO2-equi.
Table 1 gives a schematic view of the framework of the energy accounting. The emissions of CH4 and N20 are converted to an emission of CO2-equivalents. Summation of the emission of the six modules results in a total emission per farm. For comparison of individual farms, this total is expressed in the form of an efficiency-figure, for example kg CO2-equivalent per 100 kg milk. The same applies to the total energy use, for dairy expressed as MJ per 100 kg milk.
1125
3.
F I R S T
R E S U L T S
The system mentioned aboved is now being tested in three study groups of dairy farmers, and four study groups of intensive livestock farmers. Table 2 shows the results of the first testing year of three dairy farms and two pig farms. The farms presented here are selected for the differences in their CO2-equivalent emissions.
Table 2 CO2-equivalent emissions (kg CO2/100 kg milk, kg CO2/100 kg growth) Direct energy Indirect energy
CO>min.
CH4
N20
Total
dairy farm 1 dairy farm 2 dairy farm 3
6 4 6
28 14 27
201 -
27 25 22
100 17 7
362 60 62
pig farm 1 pig farm 2
21 12
204 179
-
81 38
-
306 229
The first results give a good insight in the significance of the emissions and in the differences that were found between individual farms. Concerning the dairy farms, the following results were the most striking: for peat soils, CO2-emission from mineralization as well as N20-emission from soil processes (farm 1) have a large influence on total CO2-emission, compared to total CO2emission from clay (farm 2) and sandy soils (farm 3); the differences in indirect energy related CO2-emission are of much more importance than the differences in direct energy related CO2-emission; on farms with clay and sandy soils, the CO>emissions caused by the use of direct and indirect energy, the CHa-emission and the N20-emission are each of the same importance. The emissions per module vary not only between farms on different soils but also between farms on the same soil. In the study group with sandy soils the extreme values in total CO2emission were 62 and 86 kg CO2 per 100 kg milk. The results indicate that farmers can reduce the emissions of greenhouse gases by improving their efficiency of energy use, of fertilizer use and of concentrate and feed use. The results of the two pig farms show clearly that : the use of indirect energy in pig breeding is very important; the variation in CO2-equivalent emission between the two farms caused by use of direct and indirect energy is of the same magnitude than the variation caused by storage of slurry. The differences indicate that on pig farms there are possibilities to improve use of direct and indirect energy. Farmers can either make energy-saving investments in climate control or improve their general and feeding management. The calculated difference in methane emission is mainly caused by a difference in storage time: technical measures to reduce this emission are developed. -
1126
4. DISCUSSION The results have a provisional character. Changes in the methodology to calculate emissions are possible during the testing years. The purpose of this project is to create a management instrument for farmers. It is in discussion to which extend emissions should be part of this instrument that hardly can be influenced by farmers. Most discussion is about the mineralization of peat (4). Possible changes also depend on the availability of relevant farm management data and on changes in knowledge of emissions. An important source of new knowledge is the NRP research on emission of nitrous oxide from grassland. The variation in the results from individual farms show that farmers do have possibilities to reduce their energy use and emissions of greenhouse gases. These possibilities can lead to a substantial reduction of emission of greenhouse gases. In the second year of the project more research will be done on the contribution of advice on energy-saving investments and on better management practices to reduce the emission of greenhouse gases on farms.
5. R E F E R E N C E S
1 Bergen J.A.M. van and E.E. Biewinga (1992). Agriculture and Greenhouse effect, Survey to reduce the Emissions of Agriculture and Horticulture. (In Dutch with English summary), Centre for Agriculture and Environment, Utrecht. 2 Leach, G.A. (1976), Energy and Food Production. IPC Science and technology Press Limited, Guilford. 3 Pimentel, D. (1980) Handbook of Energy Utilization in Agriculture, CRC Press, Boca Raton. 4 Hanegraaf M.C. and E.E. Biewinga (1994). 'Use of Energy and Emission of Greenhouse Gases at individual dairy Farms'. In : Meststoffen 1994, p. 59-67.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1127
SPACE FOR BIOMASS An exploratory study of the spatial potential for the cultivation of biomass for energy in The Netherlands I. Steetskamp, A. Faaij, A. van Wijk Department of Science Technology and Society Utrecht University Padualaan 14, NL-3584 CH Utrecht, The Netherlands
Abstract The spatial and energy potential in The Netherlands for energy farming is assessed as well as for a number of biomass residues. The future supply of agricultural land is based on closures of farms. Various future claims for infrastructure and nature are taken into account. The net supply of land adds up to 100,000 - 185,000 in 2000 to 245,000 and a theoretical maximum of 465,000 ha in 2015. When this potential is used for energy crops like Miscanthus this land could contribute 20 37 PJ in 2000 and in 2015 62 - 117 PJ. Secondary yields of biomass can contribute a further 32 PJ in 2000, decreasing to approx. 24 PJ in 2015 This implies 2% of the Dutch energy demand in 2000, in 2015 about 3%, provided that energy farming is an economically feasible activity for farmers. 1. I N T R O D U C T I O N The role of biomass as a 'renewable' source of energy is once again the centre of attention for a variety of reasons. Technological developments make it possible to achieve a far higher yield from the conversion of biomass into electricity or fuel than in the past. Developments in the agricultural industry, such as the predicted shedding of agricultural land, also play a role. A study conducted by the Scientific Scientific Council for Government Policy (WRR), entitled 'Ground for choices', outlines a number of scenarios for agricultural use of the land in the European Union depending on the agricultural policy currently in force. Agricultural land is released in each of the described scenarios (1). This land could, however, be used for the cultivation of crops suitable for energy production (energy cultivation). The available surface area is a decisive factor in determining the energy potential of biomass. If agricultural land falls vacant in the Netherlands, there will be several sectors lining up to use it, given the high population density. This has resulted in the formulation of the following question: What is the spatial and energy potential of biomass production in the Netherlands in the long term (2000/2015), with or without other functions, seen together with other claims on the space? This exploratory study focuses primarily on land that is not used for other types of agriculture (any longer). Energy cultivation on this land is possible, provided it is economically viable for agricultural industry (alongside food production). If, however, the yield of energy crops increases in the future, competition with food production may become possible. The spatial potential would then be on a quite different scale.
1128 2. M E T H O D O L O G Y
& RESULTS
2.1 T h e s p a t i a l p o t e n t i a l : s u p p l y a n d d e m a n d for a g r i c u l t u r a l l a n d
The supply of and demand for agricultural land are based on calculations made in the LEI (Agricultural Economic Institute) study 'Regional Land Balances'(2). The base calculation in the study shows a total supply of 280,000 hectares in the period 1990-2000. In a high supply scenario this surface is 410,000 hectares. This LEI study assumes that land that falls vacant comes from closures of farms. An average closure percentage and an average farm hectarage, which incorporate upto-date developments in the sector, were used as a basis for estimating the total surface area of the land which will become vacant. A large part of the available land is grassland with a milk quota. This study assumes that grassland with a milk quota will be used for the same purpose after it goes on offer. We have also assumed that the claims of the intensive livestock farming industry and horticulture (under glass) will be honoured, entailing approximately 8,000 hectares until the year 2000. The demand for non-agricultural land can be divided into 'hard' claims and other claims. Hard claims on land are laid by housing, industry, traffic and military training grounds. Other claims come from forestry, nature and recreation. An analysis of each of these functions has been carried out. Each function was studied to ascertain the expected spatial development and how this translates into a claim on land. A full description of the applied methods is given in (3). Table 1 shows the spatial potential for energy cultivation in the year 2000 for the basic supply of 280,000 hectares and the variant with a higher supply of 410,000 hectares. The basic assumption is that the 'hard' claims will be honoured, and they have been deducted from the supply of agricultural land. This imposes an upper limit on the spatial potential.
Table 1
Spatial potential for energy farming on agricultural land in 2000, calculated as of 1990, in hectares (x 1,000).
Land supply until 2000 l
Hard agricultural claims2
Basic supply 280
105
Higher supply 410
200
Hard nonagricultural claims3
Other nonagricultural claims4
Spatial
potential 100 - 150
24
53
135- 185
The supply of land in the LEI study was based on 1989. This figure is translated to the period as of 1990. Claims from livestock farming (grassland with milk quota) and claims from intensive livestock farming and horticulture (under glass). Claims from housing, industry, traffic and military training grounds. For calculation, see table 9.5. Claims from forestry, nature and recreation.
1129
The spread of the spatial potential depends on whether the other, non-agricultural claims will be honoured wholly, partially or not at all in the future. If all other non-agricultural claims are honoured, the lower limit on spatial potential will be reached. At a supply of 280,000 hectares, at least 100,000 hectares could be available for energy cultivation in 2000, up to a maximum of 150,000 hectares. For the year 2000, this estimate of the spatial potential according to the basic supply seems the most realistic. One must not forget that this potential is based on calculations as of 1990. No part of this potential had been realized by 1994. For 2015, a linear extrapolation was made of the data in the LEI study for 2000. The different kinds of claims were then deducted. In 2015 between 245,000375,000 hectares could be available at a supply of 28,000 hectares per year. According to the author of the LEI study, the linear extrapolation of the data for 2000 produced a conservative estimate of the basic supply in 2015 (4). The LEI study is an approximation at micro level (supply on farm level), and assumes implicitly that the land market wishes of every farmer will be honoured. Developments at macro level (agricultural production ceilings, for example) were not included in the study. This would have made it impossible to meet each individual farmer's wishes. Consequently, the supply after the year 2000 could be considerably higher than in the base calculations, while the claims remain the same. On the other hand, there are other agricultural developments underway (essential reductions in emissions of environmentally harmful substances, biological farming) which could lead to more extensive use of the land, producing in turn a lower supply after 2000 than envisaged in the basic calculations.
x 1.000 ha 500 450
~M.~ 2ooo I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
400
[~ Ml'n. 2015 [ ~'lMax. 2015 1
350 300 X 250 200
X x x
.....................
150 i 00
x N
2~ base supply
higher supply
Figure 1. Spatial potential for energy cultivation on agricultural land in 2000 and 2015 for a basic supply (28,000 ha/yr) and a higher supply variant (41,000 ha/yr), as of 1990.
1130 The crop yields vary according to type of soil, which is why the spatial potential is subdivided according to soil type. Of all agricultural regions in the Netherlands, approximately 700,000 hectares can be counted as high-yield areas and 1.3 million hectares as low-yield areas. The ratio of productive and less productive land released for energy cultivation depends on the extent to which the claims are honoured. Nature claims concern mainly agricultural land of lower quality. At a spatial potential of 150,000 hectares in 2000 for the basic supply, the ratio between high-yield and low-yield land is 1:1.4, resulting in 63,000 hectares with a potential high yield and 87,000 hectares with a potential low yield.
2.2 Energy farming yields on agricultural land Taking the estimated spatial potential and the division according to type of soil as a point of departure, it is possible to calculate the energetic potential. This estimate is based on Miscanthus as the energy crop, because it produces a high net energy yield. The yield figures were derived from data based on the model calculations (5). For the current situation 12.3 tons dm/ha/yr in high-yield regions and 10.6 in low-yield regions is projected. For 2015, 20% higher yields can be expected due to developments in cultivation technology, crop improvement and increasing experience with harvesting methods and maintenance. Using this data as a basis, a calculation was made of the annual yields in tons of dry solids. The net energy yield was calculated by combining the calorific value of the crops (19 GJ/ton din) with the dry solid yield (gross energy potential) and deducting the energy costs of cultivation. For Miscanthus, this produces a net energy yield of 20 to 30 PJ per year for at a spatial potential of 100,000 to 150,000 hectares in the year 2000. At a potential of 245,000 to 375,000 hectares in 2015, a net energy yield of 62 to 95 PJ per year is possible. At a potential of 330,000 to 465,000 hectares, this would rise to between 83 and 117 PJ per year in 2015. 2.3 Energy yields as a secondary function In addition to yields from energy crops on agricultural land, biomass yields can also be generated by non-agricultural activities and by-products of regular agriculture (waste flows such as organic waste and sludge have not been included). Yields from wood produced by thinning activities for forestry and recreation are particularly significant: approximately 15 PJ in 2000. Straw can contribute more than 8 PJ per year on the basis of the current agricultural hectarage. In the future, this contribution will fall as the hectarage for food production decreases. A yield of approximately 32 PJ per year is possible from secondary activities up to the year 2000. After 2000, this figure will fall to approximately 24 PJ per year by the year 2015 due to smaller straw yields on the one hand, and a lower proportion of thinning wood in the total volume of wood cut on the other. These figures assume that all claims for forestry have been honoured. If this is not the case, forestry hectarage will be smaller and the energy yield lower as a result. Table 2 shows a brief summary of the secondary yields. There is no data available for a number of biomass flows, and these have not been included in the table. In the case of these secondary yields, it should be pointed out that these flows already have an alternative application. For example, straw is sold to livestock farmers and the bulb cultivation industry, turf is composted and reeds are used for roofing.
1131 Table 2
Overview of secondary yields of biomass
Function
Type of material
Yield (ton ds/ haJyr 1)
Hectarage 1993/2000 (x 1,000)
Gross energy yield 1993/2000 (PJ/yr)
Hectarage 2015 (x l,O00)
Gross energy yield 2015 (PJ/yr)
i,~i ~ i i
1 Forestry I
Thinning wood
2.0
447/460
15.8
4472-497
11.2-13.1
2 Nature
Cut sods
1.4
35
0.8
35
0.8
!
Reed
4.0
0.1
!: 3 Traffic
Verge grass
5.1
!i il il
!i Parks and Gardens
Residual wood
i, Agriculture
Straw
'i
i Total
0.1 37
2.6
37
2.6
+16
4.4
+16
44
"
!i Jl !i
3.7
149
8.3
753
4.2
695/708
32.0
621-671
23.3-25.2
~t
Including forestry designated for recreation, nature and military training grounds. Lower limit if none of the claims for forestry are met (and a consequent maximum spatial potential is achieved). Assuming that the hectarage of grain falls by 50% due to the increase of spatial potential of energy farming.
Table 3
An overview of spatial potential and total energetic potential in 2000 and 2015. Basic supply
:' Spatial potential 2000 (x 1,000 ha) Energy yield 2000 (PJ/yr)
Energy farming
100-150
Secondary yields Energy farming
Energy farming
Energy yield 2015 (PJ yr)
Energy farming I
135-185
i',
i
20-30
27-37 32
52-62
Spatial potential 2015 (x 1,000 ha)
Ii,
708
Secondary yields Total energy yield 2000 (PJ/yr)
i! Higher supply
!
245-375
59-69 330-465
670-620
Secondary yields 62-95
Secondary yields
i
83-117
25 -23
!I .....
:i Total energy yield 2015 (PJ/yr)
87-118
108-140
This figure assumes higher yields of dry solids (20% increase) than in 2000. Taking the same yields of dry solids as in 2000, the figures would read 50-84 PJ/yr for the basic supply and 67-94 PJ/yr for the higher supply.
1132 3. DISCUSSION AND CONCLUSION The spatial potential of biomass is made up of two components: energy cultivation on agricultural land and potential biomass yields on land with another function. The net supply of land adds up to 100,000 - 185,000 in 2000 to 245,000 and a theoretical maximum of 465,000 ha in 2015. When this potential is used for energy crops like Miscanthus this land could contribute 20 - 37 PJ in 2000 and in 2015 62 - 117 PJ. Secondary yields of biomass can contribute a further 32 PJ in 2000, decreasing to approx. 24 PJ in 2015. In the year 2000, the total potential contribution of these two flows of biomass can contribute approximately 2% to the primary energy demand (as estimated in the Follow-up Paper on Energy Conservation (6)). In 2015, this total can be about 3%. The expected growth in energy consumption has already been calculated into these percentages. A linear extrapolation is made up to 2015 for the potentially available land. It should be pointed out, however, that the WRR study entitled 'Ground for Choices' does support the theory that more land than estimated in the base calculations may become vacant. It may then become possible to achieve the results of the greater supply variant (410,000 hectares per year), which was conducted in the LEI study as a sensitivity analysis: a spatial potential in 2015 of between 330,000 and 465,000 hectares. Compared to the study 'The feasibility of biomass production for the energy system in the Netherlands'(7), the estimate of energy potential on agricultural land is clearly lower. The study calculated a yield of 140 PJ. The fact that the estimates in this study lag behind has a variety of causes. Firstly, the calculation of the spatial potential in this study according to the basic supply is significantly lower (245,000-375,000 hectares in 2015) than the maximum estimated long-term spatial potential of 500,000 hectares in the mentioned study. Secondly, this study has assumed lower yield figures and differentiated according to land quality. Thirdly, this study has deducted the energy costs of the cultivation from the potential. It does, however, employ a higher calorific value based on data from recent research material.
ACKNOWLEDGEMENTS This study was sponsored by the Netherlands Agency for Energy and the Environment (NOVEM) REFERENCES 1. Scientific Council for Government Policy (WRR), Ground for choices, four perspectives for rural areas in the European Union; Vol. 42; Sdu uitgeverij, Den Haag, 1992 2. F.H. Bethe, Regional ground balances. Exploration of demand and supply of land until the year 2000, report 83, Landbouw-Economisch Instituut, Den Haag, 1991. 3. I. Steetskamp, A. Faaij, A. Van Wijk, Space for biomass, An exploratory study of the spatial potential for the cultivation of biomass for energy in The Netherlands, Department of Science Technology and Society, Utrecht University (in Dutch), December 1994. 4. F.H. Bethe Personal communication, October 1994. 5. S. Nonhebel, Potential yields of woody crops, in: Fuelwood, perspectives for forestry and energy production, Lelystad, January 1994. 6. Ministry of Economic Affairs, Follow up paper energy conservation, Sdu, Den Haag, 1993. 7. E.H. Lyssen et. al., The feasibility of production of biomass for the Dutch energy system, Netherlands organization tbr energy and environment, 1992.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1133
CONVERSION ROUTES FOR ENERGY CROPS; INTEGRATING AGRICULTURAL AND ENVIRONMENTAL OPPORTUNITIES IN EUROPE Eric J.M.T. van den HEUVEL, BTG Biomass Technology Group B.V.
Introduction Energy crops are interesting for both the agricultural and energy sector. Several conversion routes exist for the transformation of these crops into electricity, heat and/or transport fuels.The major environmental advantage lies in avoided CO 2 emissions to the atmosphere in comparison to the use of fossil fuels. Those conversion routes are analyzed with respect to technical/financial and environmental characteristics.
Conversion routes Energy crops are considered as an alternative for food crops in the agricultural sector. After harvest, they can be converted into electricity and/or heat and into transport fuels. For energy crops with high cellulose contents, like poplar, willow and miscanthus, thermo-chemicai conversion routes which convert the crop into electricity are most viable. Important thermochemical conversion technologies are combustion, gasification and pyrolysis. For high oil-content or sugar-content energy crops physical-chemical (extraction) and biochemical (anaerobic digestion or fermentation) routes are available. Those routes produce respectively gaseous and liquid fuels that can be used in transport applications. Upgrading of the secondary products of the gasification and pyrolysis technology processes also can lead to the production of methanol or bio-diesel. In Figure 1 these routes are presented.
I
hot w i r e r
oils
up=
,din
Figure 1: Available conversion routes for energy crops
1134
The technical, financial and environmental aspects of energy crops based electricity generation and production of transport fuels has been investigated. Large scale power generation based on combustion technology and application of steam cycles is technically mature. The same holds for the use of biogas in gas engines. Also the production of ethanol from sugar and grain crops as well as the production of rapeseedmethylester (RME) is technically viable. Biomass gasification integrated with a combined cycle (gas turbine and steam turbine utilization), cocombustion of pulverized or gasified biomass in conventional large-scale coal or gas fired electricity plants and production of methanol through gasification are currently demonstrated and expected to become technically mature around the year 2000. Newer technologies like the use of pyrolytic oil in a combined cycle applications or the use of synthesis gas from biomass gasification as a fuel source for fuel cells still need further research and will certainly not be commercial before 2010.
Some options for conversion routes
Several conversion options have been analyzed through spreadsheet models in order to determine: o the specific production costs (FI/kWh or FI/I fuel); o the specific amount of avoided CO 2 (ton CO2/ha) generated; and o the specific costs of avoided CO 2 (FI/ton CO2), calculated as the difference between the annual financial returns and the conversion option costs divided by the amount of avoided CO 2. The options (numbers correspond to numbers in Figures 2 and 3) considered are: Conversion routes for electricity .qeneration: 1 Small-scale co-generation of heat and power with combustion technology (5 MWel/ 10 MWth); 2 Medium-scale electricity generation with combustion technology (50 MWel); 3 Co-combustion of pulverized miscanthus (max 10% on energy basis) in conventional powder coal electricity plant of 500 MWel; 4 Small-scale co-generation of heat and power with circulating fluidized bed gasification and gas turbine technology (5 MWel/lO MWth); 5 Medium-scale electricity generation with integrated circulating fluidized bed gasification and combined cycle utilization (50 MWel); 6 Gasification of energy crops and combustion of the resulting producer gas in coal fired conventional electricity plant; 7 Gasification of energy crops and combustion of the resulting producer gas in natural gas fired conventional electricity plant; Conversion routes for production of transport fuels: 8 Fermentation of wheat for ethanol production (100 million I/year), with combined heat and power generation through combustion based on wheat straw; 9 Fermentation of sugar beet for ethanol production (100 million I/year), with purchase of required heat and power; and 10 RME production based on oil-extraction and esterification of rapeseed (1 million I/year).
Results
All technologies mentioned above are technically mature, except for the integrated gasification combined cycle (gas and steam turbine) technology, which is still in a demonstration phase. The specific production costs figures for electricity and transport fuels and the specific amount of CO 2 avoided are presented in Figure 2. The specificproduction costs for electricity are lowest for the large scale gasification option and the co-combustion options in large scale coal or gas fired electricity plants.
1135
50-
--I
combustion
t ~ ]
gasification
t
~
t transport, fuels I
45-
180
40-
160
r : x~ vo
35-
,a
3oI
4~
25-
.O 4-1
20-
o
200
o
-140
r
-120
G) ::~
-IOO
tO
-8o
. . . .
15-
(D
I
.........
i
60 -40
10-
0
c~ r,/) r (~i
-20
1
2
3
4
5
6
7
8
I
i 9
(0 10
options (see text for expl.) I L~
electricity
~
transport, fuels !
Figure 2: Specific production costs of several conversion routes
At the moment, with current fossil fuel prices, none of these technologies are economically viable. The annual returns are lower than the annual costs. Utilization of sustainably grow energy crops for electricity generation results in avoided CO 2 emissions, because fossil fuel based electricity is replaced. For each option the amount of annually avoided CO 2 emission is calculated and from this the specific costs of avoided CO 2 emission are determined. The results are summarized in Figure 3. It can be concluded that thermochemical conversion routes, like combustion and gasfication processes, for electricity production have the highest potential for reducing CO2-emission (given as tonnes CO2/ha). This indicates that per ha of land used for energy crops through thermochemical conversion routes most CO 2 emission will be avoided. At the same time the specific costs for CO 2 emission are also much lower for electricity production as compared to the production of transportation fuels. It is therefore concluded that future research activities must concentrate on high-efficiency electricity production, from energy crops. The most promising routes seem to be large scale gasification and the co-combustion in existing power plants. This justifies research to further develop these technologies. The major problems with gasification lies in the required producer gas cleaning for gas turbine utilization and in the adaptation of gas turbines to low heating value gasses. For co-combustion it must be researched whether to co-combust producer gas or pulverized biomass. A study on the environmental impacts of the Netherlands Center for Agriculture and Environment (CLM) has shown that utilization of the energy crops miscanthus, poplar, willow, hemp and reed, under the Dutch agricultural system, will probably have the lowest negative impacts. Therefore, these crops will be considered in future research.
1136
! 0as'''~176 / i /trans ~
,07] [com us,,on I
2OO -180 -160
~" 35r
-140
04 30O O 25-
-120 O O -100
04
..........
~= 20cr
-80
:4i
1510-
-60
,-, o o
-40 20
0 1
2
0
-~ ..... 3
4
5
6
7
8
9
10
options (see text for expl.) 1~
quaniity ~
costs
!
Figure 3: Avoided CO 2 emission
References Johansson et al. (eds.), 1993, Renewable energy, sources for fuels and electricity, Island Press, Washington DC. USA. Lysen E.H. et al., 1992, The feasibility of the production of biomass for the Dutch energy sector, Novem, Utrecht, the Netherlands. Heuvel, E.J.M.T. van den, Stassen, H.E.M., 1994, Electricity from biomass, a comparison between combustion and gasification, Novem, Utrecht, the Netherlands. Siemons, R.V. 1993, Electricity generation through co-combustion of straw and grass residues in conventional power plants, BTG, Enschede, the Netherlands.
Acknowledgement This research has been carried out within the Framework of the National Research Programme on Global Air Pollution and Climate Change (NOP/MLK) of the Ministry of Housing, Urban Planning and Environment. The project has been co-financed by the Netherlands Company for energy and the Environment (Novem). The project has been executed in co-operation with the Centrum voor Landbouw en Milieu - which used information on technical and environmental aspects of the conversion routes for their study on the environmental impacts of energy crops cultivation - and the Department of Theoretical Production Ecology, Wageningen Agricultural University.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
Forests and wood consumption
1137
on the carbon balance.
Carbon emission reduction by use of wood products R. Sikkema (SBH a) and G.J. Nabuurs (IBN-DLO b) Institute for Forest and Forest Products (SBH), PO Box 253, NL-6700 AG Wageningen, The Netherlands.
Abstract Until now studies on the greenhouse effect paid much attention to carbon fixation by forests, while the entire CO2 cycle of forests and forest products remained underexposed. Utilization of wood products instead of energy-intensive materials (plastics/steel) and fossil fuels (coal) proves to play an important role as well. The effect of utilization is even greater t h a n that of fixation. In all, additional forests together with the multiple use of trees can contribute substantially to the reduction of CO2 emissions. The contribution can run from 5.3 ton CO2/ha/yr for a mixed forest of oak~eech to 18.9 ton CO2/ha/yr for energy plantations (poplar).
INTRODUCTION
The greenhouse effect is a problem acknowledged worldwide. The increasing concentration of greenhouse gasses in the atmosphere (carbon dioxide, methane, nitrogen oxide and others) may in time lead to an unwanted t e m p e r a t u r e rise. Forests and the use of wood contribute to the fight against the greenhouse effect in three ways: 1. Carbon fixation (during their growth trees convert CO2 into timber) 2. CO2 avoidance through substitution by wood of energy-intensive materials such as plastics, aluminium and steel. Processing timber uses relatively little energy (fossil fuels). After use it can easily be reused, e.g. as particle board. 3. CO2 avoidance by using timber instead of fossil fuels for generating energy. When recycling has become technically or economically impractical, wood may be used for energy purposes. The same is valid for timber from special energy plantations. Only the previously sequestered CO2 will be released when burning the woody material. This makes the use of timber 'CO2 neutral'. In 1989 the CO2-emissions in the Netherlands amounted to 183 million tons. These increase by 3.5 million tons annually. The Dutch government is striving for stabilization with respect to the 1989 emissions level. This corresponds to a reduction of 21 million tons CO2 by 1995.
1138 FIXATION OR AVOIDING? Recently a number of reports are published concerning the carbon sequestering potential of various forest types. These studies primarily examine carbon sequestration in biomass (the tree), soil and timber products. Findings suggest that long rotation forests provide a greater contribution than short rotation forests (Table A, No.l). Besides the average carbon sequestration by trees and in timber products, an important CO2 reduction effect is created through substitution for nonwood products (product substitution) and fossil fuels (fuel substitution). This CO2 avoidance was included in the NOVEMC-report 'Bossen en hout op de koolstofbalans'. Table A Potential contribution to CO2 reduction (in tons CO2/ha) of several forest types Oak/Beech
Spruce
Poplar 15 year
Poplar 5 year
1. CO2 fixation in biomass, humus and products (average) 2. CO2 avoidance through replacement of non-timber materials 3. CO2 avoidance through replacement of fossil fuels
432
394
104
-106
182
784
653
0
966
1289
1560
5788
Total CO2 reduction in 300 years
1580
2467
2317
5682
*)
*)Including processing energy of fertilization (0.44 ton CO2/ha/yr) Substituting wood for plastic, aluminium and steel leads to an important reduction in emissions. This is because the production of wood materials uses far less fossil energy than the mentioned alternatives. The effect is most evident in saw and packing timber products, such as frames, construction timber and pallets. Norway spruce is a forest type that produces relatively many saw logs and packaging timber products (Table A, No.2). The greatest contribution to CO2 reduction, however, results from substituting wood for coal in energy production. This is especially true for energy wood plantations (short rotation poplar), from which all wood produced, such as increment, is used as fuel (Table A, No.3). The effects ofboth product and fuel substitution are repeatable. During cultivation, harvest, use and renewed cultivation, no additional CO2 is released into the
1139 a t m o s p h e r e . T h e a v e r a g e s e q u e s t r a t i o n clearly h a s a once-time effect, b e c a u s e in t i m e t h e CO2 is r e l e a s e d again, e i t h e r t h r o u g h decay or t h r o u g h c o m b u s t i o n ( F i g u r e
A) F i g u r e A Total CO2 r e d u c t i o n effect* of N o r w a y spruce s t a n d h a v i n g a r o t a t i o n of 75 y e a r s (tons CO2/ha) C02-reduction ( tons C02/ha ) 2.500 [-
C02-avoidance through fuel substitution
~
C02-avoidance through material substitution
2.000 -
C02-sequestration
~
inproducts
-~
1.500
C02-sequestration in biomass and litter
-
1.000 -
500
0
0
75
150 Year
225
300
Source: Stichting Bos en Hout, Wageningen, The Netherlands
*) Carbon sequestration in biomass, litter and products is subject to fluctuations during every rotation. An average sequestration, which reaches a constant value of 394 tons CO2/ha after several rotations, was used for calculations in the text. Table B C o n t r i b u t i o n by forest type to 1995" goal for n a t i o n a l CO2-emission reduction Oak/Beech A n n u a l r e d u c t i o n (in tons CO2/ha) R e d u c t i o n p e r 100,000 h a (in 1000 tons CO2) C o n t r i b u t i o n to policy '95
Spruce
P o p l a r 15 year
Poplar 5 year
5
8
8
19
530
820
770
1890
2.5
3.9
3.7
9.0
*)National CO2-emissions were 183 million tons in 1989. Emissions increase 3.5 million tons annually. 1995 goal: Stabilization with respect to 1989 correspondends to a reduction of 21 million tons CO2.
1140
In the study CO2 balances of long rotation forests (oak/beech, 150 years) were compared with those of short rotation forests (poplar, 5 and 15 years), and supplemented with those of Norway Spruce, a wood species ideal for recycling. The contribution to CO2 emissions reduction (see Table B) proves to be substantial: 5 to 19 tonnes COJha/yr. The contribution to the Dutch reduction goal can run from 2.5% for a mixed forest of oak~eech to 9% for energy plantations of poplar.
FOREST TYPES
Calculations were made for four forest types (see Table C). An average tree-species specific increment was assumed. The average CO2 sequestration in the tree itself and in the upper soil layer (litter) of each forest type over a 300 year period was modelled. Sequestration in the stable humus was not considered, because this factor is greatly dependent on soil types and previous use of soil (e.g. agriculture). In the Netherlands the extra fixation by afforestation amounts to a small quantity. Table C Important characteristic figures of considered forest types Oak/Beech Rotation time (yr) N u m b e r of rotations in 300 yr Mean increment (m3/ha/yr) Density air dry (kg/m 3) Amount of carbon in dry m a t t e r weight (%)
150 2 5.4 700 50
Spruce
Poplar 15 year
75 4 11.5 460 50
15 20 15.9 450 50
Poplar 5 year 5 60 29.5 450 50
F I X A T I O N O F CO2 IN T R E E AND S O I L With each rotation, forests 'produce' wood that becomes available during thinning and during the final felling at the end of the cycle. Most of the wood ends up being used outside the forest as industrial wood. Another part of the felled trees remains in the forest as dead wood. This wood ends up in the litter layer and breaks down during decomposition to CO2 and water. Thus the fixed CO2 in wood is gradually released into the atmosphere. No wood is left behind in short rotation poplar forests. The entire above-ground biomass (the t r u n k including the branches, but excluding the leaves) of this forest is destined to serve as fuel for energy production. Thus, hardly any sequestration occurs in the litter layer. Removal of the t r u n k and branches also means the
1141 disappearance of nutrients. Fertilizing compensates for this effect. The processing energy of the fertilizer (0.44 ton CO2/ha/yr) such as lime and K20 is subtracted from the net sequestration. Total sequestration as a result has a negative value.
FIXATION OF CO2 IN P R O D U C T S Depending on the diameter and tree species, harvested forest products are assigned to be used as fuel wood, pulp wood, wood based panels, packing or sawn timber. The research model assumes optimal utilization of the available amount of raw material. This means sawing residues (bark, sawdust, chips, etc.) are assigned to the most durable uses, like chipboard and paper, or when not possible, to fuel. Fuel wood is delivered to the power station in chipped and dried form. Assumed is the possibility to allow the wind to dry the wood in the forest naturally (to a maximum moisture content of 15%). All timber products run through a so-called 'cascade model' (figure B). Where possible this calls for wood to be reused at the end of its technical life span. In this model packaging material and other timber products will be reused for chipboard or paper. Ultimately, when written off or replaced, particle board and waste paper are used for energy generation. Thus all sequestered carbon will again be released to the atmosphere as CO2. Figure B All wood and timberproducts end at the stage of energy-generation ~
/
~
~ .
f~ r~
C02
Ill ( ) llIV wood
\\X.."-"JJf
i
wasted
,---
i-z-
ood
L..2
wasted p a p e r / ~ (board) ~ L ~=~>"~
~
timber
{ ,=_~-, L._
!~
~j.v
JJ II II f J IIII I / /
recycling particle board paper(board)
1142
CO2 A V O I D A N C E T H R O U G H MATERIAL S U B S T I T U T I O N The first reduction of CQ-emission occurs when wood replaces non timber products. Consideration is made of the use of fossile fuels. An important factor is the energy applied during production and transportation of raw materials, semifinished products and finished products. Net energy applied is calculated in the model. This means t h a t r e m n a n t wood is utilized for drying other wood. Some production processes are therefore COJenergy neutral. It is assumed that mineral oil is the only fossil fuel used in this application.
CO2 A V O I D A N C E T H R O U G H F U E L S U B S T I T U T I O N Secondly, the application of energy wood, r e m n a n t wood, waste wood and waste paper for energy purposes is considered. Timber products (in contrast to fossil fuels) are CO2 neutral. Fuel wood does produce carbon dioxide emissions, but the emissions occur within a closed cycle. After all, wood originates and grows by extracting an equal amount of CO2 from the atmosphere. In this study, a comparison was made with coal, one of the most used fossil fuels for generating electricity in the Netherlands. This comparison is the most realistic, because electricity producers have serious plans to use wood in coal-fired power stations in the short term.
CONCLUSIONS Studies on CO2 reduction pay too much attention to sequestering of CO2 in biomass, soil and products. Thus, the total CO2 cycle of forests and forest products remains underexposed. Utilization of wood (multiple use of products and energy-generation) proves to play an important role. The influence on the CO2 balance (avoidance through product substitution and fuel substitution) is even greater than carbon sequestration by trees. Forests and the multiple use of wood (including energy-generation) can contribute substantially to the reduction of C02 emissions. The contribution to the Dutch government's policy (reduction of the annual emission by 21 million tons) can run from 2.5% for a mixed forest of oak and beech, to 9% for short cycle (5 years) poplar. These percentages are based upon an additional 100,000 ha of forest.
Notes aSBH:
bIBN/DLO: cNOVEM:
Stichting Bos en Hout, Wageningen, the Netherlands Instituut voor Bos- en Natuuronderzoek, Wageningen, the Netherlands Nederlandse Onderneming voor Energie en Milieu, Utrecht, the Netherlands
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1143
Potential of controlled anaerobic wastewater treatment in order to reduce the global emissions of methane and carbon dioxide Marjo J. Lexmond and Grietje Zeeman Department of Environmental Technology, Agricultural University of Wageningen P.O. Box 8129, 6700 EV Wageningen, the Netherlands
Abstract An estimation is made of the current global methane and carbon dioxide emissions from wastewater treatment and disposal. Furthermore, the potential of controlled anaerobic treatment to reduce these emissions is investigated. Considered wastewaters are: domestic wastewater and wastewater from the Food & Beverage and Pulp & Paper industry. The current methane emission is estimated to be about 5 Tg/y, and is mainly the result of uncontrolled degradation of untreated wastewater in developing countries. Carbon dioxide emission is estimated to be 1 5 Tg/y, which is mainly due to aerobic wastewater treatment in the developed countries. Anaerobic wastewater treatment, provided a minimization of the percentage methane loss and an optimal reuse of biogas, can significantly reduce the current emissions.
1. INTRODUCTION
Little is known about the quantities of the atmospheric CH4 emission from the treatment, storage, and discharge of domestic and industrial wastewaters. Thorneloe (1993) roughly estimated this to be about 26-40 Tg/y, thereby representing about 8-11% of the total anthropogenic CH4 emissions. However, large uncertainties concerning this estimation are recognized. In most CH4 emission reduction technologies the aim is to avoid uncontrolled anaerobic degradation and to promote aerobic degradation. However, by promoting aerobic treatment the CO2 emission due to fossil fuel consumption will strongly increase. Aerobic treatment, as conventionally applied in most wastewater treatment systems, is rather energy consuming since this process depends on a more or less intensive aeration. A very interesting alternative is formed by anaerobic treatment. Under anaerobic conditions organic material can be completely degraded into CO2, CH4, water, and a small amount of biomass. The produced biogas can be used as a fuel. By doing so, anaerobic degradation has the following advantages over aerobic degradation: 1) Production of a valuable fuel, the use of which can lead to a reduction of the amount of fossil fuel consumed, 2) no energy requirement for aeration, and 3) significantly less sludge production. On the other hand, if anaerobic degradation occurs in an uncontrolled way, CH4 can be emitted to the atmosphere where it can enhance the greenhouse effect.
1144
In our study we made an estimation of the present emissions of CH4 and CO2 from wastewater treatment. Furthermore, we estimated the potential production of CH 4 from anaerobic wastewater treatment and the possible reduction of the CO2 emission due to the use of this CH4. The treatment of the produced sludge is not yet taken into account.
2. ESTIMATION METHODS
In our estimations the following cases are considered: 1) Complete anaerobic treatment. Within this case, three options are regarded: * all CH4 is flared (FLARING) * CH4 is partly used, only for the maintenance of the wastewater treatment plants. The excess is flared (PARTIAL) * CH4 is completely used for energy production (COMPLETE) 2) Complete aerobic treatment (AEROBIC) 3) Current situation (CURRENT) In each case the calculated emissions are the energy related CO2 emission and the CH4 emissions from controlled treatment systems and from uncontrolled degradation in the environment. The world was divided into underdeveloped (UND) and developed (DEV) countries. Relevant data concerning the amount, composition, and degradability of the different types of wastewater were collected from literature and queries. The same method was applied for the information on the treatment systems (efficiency, sludge growth, energy demand, methane emission factors and frequency of operation). All data were assembled in QUATTRO-spreadsheets. Models were developed for the estimations of CH4 and CO2 emissions at different assumptions (Lexmond & Zeeman, 1994). Different percentages of CH4 loss were used: For aerobic systems all CH4 formed (Czepiel et aL, 1993) is emitted to the atmosphere, and for anaerobic systems a fixed percentage of loss (due to leaking, low concentrations, etc.) is assumed. Finally, the influences on the total global warming potential (GWP) due to wastewater treatment were calculated. The most important assumptions made were: 1) The presence of oxidizing compounds, such as nitrate, oxygen, and sulphate, in the wastewater (which can result in lower CH4 production), as well as possibly toxic compounds is ignored. 2) The influence of the temperature is ignored. 3) A certain percentage of CH4 produced within anaerobic treatment systems is lost (020%). 4) The energy requirement (in kWh per unit of organic material removed) of anaerobic treatment systems is about 25-30% of that of aerobic systems. We calculated our CO2 emissions based on an energy requirement of one third of that of the Dutch aerobic systems (CBS, 1992). 5) We used an average carbon dioxide emission factor (in m3/kWh) for the conversion of energy (viz. electricity) into CO2 emissions (Blok, 1994). 6) For the current situation of emissions we had to estimate the extent of the use of different treatment systems. Due to the scarcity of information about this, we had to assume these values. Because the large influence of these values on the results, we give
1145
the assumed values in table 1. Especially the percentage of uncontrolled aerobic and anaerobic degradation of discharged wastewater in underdeveloped countries, is a very important parameter in the model. We assume that a large part of the wastewater is discharged on surface waters. Disposal at sea is assumed not to result in significant CH4 emissions, disposal on large waters and on land is assumed to result in some CH4 emissions, and the disposal on small waters and lagoons will result in significant CH4 emissions. Furthermore, we assume that in underdeveloped countries average temperatures are higher and consequently the percentage of anaerobic degradation for untreated wastewater will be somewhat higher. Table 1. Assumptions concerning the treatment of wastewater and the division in aerobic and anaerobic degradation treated (%)
untreated (%)
total
aerobic
anaerobic
total
aerobic
anaerobic
underdeveloped countries * domestic * industrial
10 50
70 85
30 15
90 50
75 75
25 25
developed countries * domestic * industrial
90 95
90 85
10 15
10 5
80 80
20 20
3. RESULTS
AND
DISCUSSION
In figure 1 the CH4 and CO2 emissions and the resulting GWP for the treatment of the total amount of wastewater at the 5 different conditions are summarized, assuming a total CH4 loss percentage of 10% and a time horizon of 100 years. For the current situation a CH4 emission of 5 Tg/y is estimated from the investigated wastewater streams, which is considerably lower than the estimated 14-20 Tg/y by Thorneloe (1993).
.~
120
120
100"
L~
,_., 14' >,
Z
_o 03 W
"
T 0 N 0 ~
-40
z
__ 1o. 03
N
LU 0 0 0
-eO -8O
9100 ~"
12.
_H_o
60'
FLARING' PARTIAL C;OMPLETEAEROBIC 'CURRENT CH4
~
CO2
~
GWP total
Figure 1. CH4 and CO2 emissions and the resulting GWP (time horizon 100 years and 10% CH4 loss)in different cases
2-
98 0
"~ ._
"60
0
-40 Q.
?: -20
~
TOTAL 'DEV.IND, bEV.DOM~UND.IND.bND.DOM.
m cH.
~
co2
~
GwPtot,s
Figure 2. Current CH 4 and CO 2 emissions and the resulting GWP (time horizon 100 years) from the different sources
1146
Figure 2 presents the CH 4 and CO2 emissions and the resulting GWP from the 100" different sources in the current situation. 50 The amount of treated and untreated oJ wastewater, the applied treatment systO O 0 em, and the conditions at which the untreated wastewater is disposed of ~ -50. determine the current emissions. From figure 2 it can also be seen that the FLARING PARTIALCOMPLETEAEROBIC CURRENT total GWP of wastewater treatment and 0% LOSS ~ 7 ~ 5% LOSS r ~ 10% LOSS disposal in the current situation is mainly ~ - ~ 15% LOSS ~ 20% LOSS determined by CH4 emissions from domestic wastewater in underdeveloped counFigure 3. GWP due to wastewater treattries. The extent of anaerobic digestion of ment at different percentages of CH4 loss untreated wastewater in underdeveloped in the different cases countries is highly affecting the estimated GWP of the current state. If we, for instance, assume that the amount of uncontrolled anaerobic degradation is not 25 % but 50%, the total CH4 emission will increase from 5 to about 11 Tg/y. Unfortunately very little data are available on this subject for most countries. To be able to estimate the present emissions more accurately, more information is required.
Ji n
In figure 3 the effect of the percentage CH4 loss in anaerobic treatment systems is shown. It is clear that the percentage CH4 loss should be minimized. In the current situation this is not so important because only a small percentage of the wastewaters is actually treated anaerobically (see table 1). The current CO2 emission is about 15 Tg/y and is mainly the result of aerobic degradation of wastewater in the developed countries. So, not only from the human health point of view treatment of wastewater should be encouraged. Our results show that, providing a minimization of the CH4 loss and an optimal reuse of the produced CH4, anaerobic treatment should be stimulated in order to reduce the emissions of greenhouse gases.
5. REFERENCES
Blok, K.: 1994, Utrecht University, written communication, January 7. CBS: 1992, 'Water Quality Control, Part b: Purification of Wastewater 1990', Environmental Statistics, Voorburg/Heerlen, the Netherlands (in Dutch). Czepiel, P.M., Crill, P.M., and Harriss, R.C.: 1993, 'Methane Emissions from Municipal Wastewater Treatment Processes', Environmental Science and Technology 27 (12), pp.2472-2477. Lexmond, M.J. and Zeeman, G.: 1994, 'Potential of controlled anaerobic wastewater treatment in order to reduce the global emissions of methane and carbon dioxide', from Ham, J. van et aL (eds): 1994, 'Non-CO 2Greenhouse Gases, pp.411-419, Kluwer Academic Publishers, the Netherlands. Thorneloe, S.A.: 1993, 'Wastewater treatment', from Amstel, A.R. van (Ed.): 1993, 'Methane and Nitrous Oxide', pp. 115-130, National Institute of Public Health and Environmental Protection (RIVM), Bilthoven, the Netherlands.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1147
A S S E S S M E N T R E P O R T ON S U B T H E M E
"MOBILITY A N D M O T O R I S E D T R A N S P O R T I N R E L A T I O N TO S U S T A I N A B L E D E V E L O P M E N T "
C.A.J. Vlek Department of Psychology, University of Groningen Grote Kruisstraat 2/1 9712 TS Groningen The Netherlands
With contributions by: P. Kooreman, J. Rouwendal, L. van Staalduinen
LUW, Agricultural University of Wageningen
A.J. Rooijers, E.M. Steg
RUG, University of Groningen
I.M. de Boer, D. van Kreveld, P.G. Swanborn, G. Tertoolen, E.C.H. Verstraten P. Nijkamp, S.A. Rienstra, J.M. Vleugel
RUU, University of Utrecht VUA, Free University of Amsterdam
1148 Contents Abstract 1.
Introduction
2.
B e h a v i o u r a l d e t e r m i n a n t s of t r a v e l m o d e c h o i c e
3.
E f f e c t s of cost a n d / o r e n v i r o n m e n t a l f e e d b a c k on car u s e
4.
Problem awareness, measures
0
Car user differences susceptibility
willingness in
to c h a n g e , e v a l u a t i o n
environmental
pollution
of policy
and
6.
E c o n o m e t r i c a n a l y s i s of p r i v a t e car u s e by h o u s e h o l d s
7.
S u s t a i n a b l e t r a n s p o r t a n d traffic s y s t e m s for t h e 21st c e n t u r y
8.
General observations, conclusions and suggestions
9.
References
policy
ABSTRACT The seemingly irresistible growth of motorised transport and its environmental effects have led the NRP to also put mobility and transport (M&T) on its agenda. NRP questions are focused on psychosocial factors and mechanisms underlying the popularity of motorised transport, and on technical as well as behavioural measures and strategies to reduce global air pollution stemming from mobility and the use of motor vehicles. In Phase 1 of the programme, five NRP-funded M&T projects have been conducted. Together with one or two related projects, these will be briefly summarised and commented upon. General observations, conclusions and some suggestions will be provided at the end of this paper. 1.
INTRODUCTION
In western societies, the second half of the twentieth century is characterised by rapid expansion of h u m a n mobility and motorised transport. Motor-cars, vans and trucks have become available in large numbers. Many airlines are being operated through a multitude of airplanes. Although transport markets in the wealthiest countries are not even satisfied yet, the opening-up of Central and Eastern Europe and further "economisation" elsewhere in the world imply that motorised transport of persons and goods will further intensify, particularly through the air. "The private car," say Rouwendal, Van Staalduinen and Kooreman in their project
1149 s u m m a r y (this volume), "is an ambiguous symbol of western society in the late twentieth century. On the one hand it reflects its success in providing a high level of material wellbeing to a large majority of the population. On the other hand, it brings out the failure of the same society to solve the environmental problems evoked by its success." Some p e r t i n e n t data - quoted from (Vlek et al., 1992) are as follows. Over six h u n d r e d million cars, vans and trucks populate the world's roads of today: about 70% in the Western industrialized countries, and roughly 10% in Japan, 12% in the less industrialized countries and 8% in Eastern Europe and the Soviet Union (Low, 1990 and Bleviss et al., 1990). Worldwide some 35 million cars are being produced each year; a net total of about 20 million cars is added to the world's fleet of automobiles (Low, 1990, MacKenzie, 1990 and Walsh, 1990). Currently there are more than five million motor vehicles in the Netherlands and the projected figures for the year 2010 lie around 8 million (EZ & VWS, 1987 and VWS, 1989). A third of the Dutch population of 15 million now owns a car; this might well be 50% in 2010; there currently are about 130 motor vehicles per s q u a r e km. For the w e a l t h y w e s t e r n p a r t of G e r m a n y a figure of 70% car ownership (700 cars for 1000 inhabitants) is foreseen for the year 2010. Not only does the n u m b e r of available cars go up steadily, also the size and the engine capacity of the average car are increasing (EZ & VWS, 1987 and Lenz, 1990). If a third of all the world's population in the year 2010 would drive cars, there would be two billion motor vehicles altogether. At 50% and 70% worldwide car ownership in 2010 these total numbers of cars would be three and four billion, respectively, for the projected world population of six billion people. This seems unbelievable. But it does not seem unrealistic to expect a doubling of the n u m b e r of motor vehicles worldwide within the next decade; around 2030 there might well be two billion of such vehicles altogether (Bleviss et al., 1990, MacKenzie et al., 1990 and MacKay, 1990). By now, the negative external effects of all this are well known. Motor vehicles involve large amounts of direct and indirect (embodied) energy consumption and greenhouse gas emissions. Their use periodically contributes to urban smog, and it is dominantly responsible for urban noise and traffic accidents. The enormous volume of motor vehicles in m a n y countries has necessitated the construction of extensive systems of roads, motorways, parking places and traffic regulation. The growth in air t r a n s p o r t is involving more and bigger airports, more and more intensely used air routes and an increasing i n f r a s t r u c t u r e of various service industries, which have their own patterns of energy consumption and greenhouse gas emissions. All this has already significantly reduced the peace and quiet of natural and urban areas, and it is threatening basic environmental qualities. These developments are worrying policy makers and citizens alike. The quality of life in and around cities, the protection of our n a t u r a l environment, and the accessibility of i m p o r t a n t destinations together may well require t h a t the use of motor vehicles be significantly reduced in the years to come, so t h a t , e.g., motorized t r a n s p o r t is limited to serving only the essential needs of society. This would fit into the concept of sustainable development as applied to t~e area of h u m a n mobility and transportation: "(This) involves more than growth. It requires
1150
a change in the content of growth, to make it less material- and energy-intensive and more equitable in its impact" (WCED, 1987). NRP-questions concerning mobility and transport are aimed at obtaining a better view of the social and psychological determinants of travel mode choice, of the physical and technical options for environmentally less harmful mobility and transport, and of the relative effectiveness of various policy strategies for encouraging people to move around and transport their goods in socially and environmentally sustainable manners. In the following, five NRP-funded projects and related research are briefly presented and discussed. In commenting upon the various projects, the author has attempted to draw general conclusions about sustainable-transport policies and to pinpoint further research and policy questions. Table 1 provides an overview of the various research projects to be discussed. Table 1.1 List of projects in NRP subtheme "Mobility and motorised transport in relation to sustainable development" Title
Project leader
Number
Attitudes and behaviours toward the environment
D. van Kreveld
850013
A behaviour analysis of private car use by households
J. Rouwendal
852081
Environmentally relevant differences among L. Hendrickx car user groups and the effectiveness of policy measures
852092
P. Nijkamp Comparative analysis of options for sustainable transport and traffic systems in the 21st century
853102
Non-NRP project Problem awareness & behaviour change
2.
Steg et al.
BEHAVIOURAL DETERMINANTS OF TRAVEL MODE CHOICE
De Boer, Van Kreveld and Swanborn, (NRP project no. 850013) collected questionnaire responses from about 500 regular car drivers in the city of
1151 Hilversum, who first recorded their own transportation behaviour for a period of four days. Respondents filled in a comprehensive questionnaire containing items m e a s u r i n g their personal attitude, perceived social norms and the "opportunity structure" - t h e i r needs and m e a n s - for using a motor-car, p a r t i c u l a r l y for commuting. Questions about attitude salience, knowledge about pros and cons of car use and their personal image of using a car were also asked. It appeared t h a t travelling time is crucial for explaining car use preferences, while commuting distance and income explain a great deal of habitual t r a n s p o r t a t i o n behaviour. Personal attitudes add significantly to this, in contrast to perceived social norms which do not seem to be very influential. In a second survey, the authors studied the relationship between transportation behaviour and several personal variables such as feelings of alienation and powerlessness, self-enhancement with respect to environmental behaviour, and willingness to care for the environment under certain "pulling" or "pushing" policy measures. Completed questionnaires were r e t u r n e d by some 330 i n h a b i t a n t s of the city of Utrecht, half of them "always", the other half "sometimes" going to their work by car; within each group half of the respondents lived further away, the other half closer by t h a n 15 kms from their job location, a distance which (in The N e t h e r l a n d s ) conditions the possibility of considering to use a bicycle for commuting. It was revealed t h a t the habit of commuting by car goes along with a self-enhancing view of one's behaviour toward the environment CI am not polluting very much myself'). Respondents seemed willing to reduce their car use if they would be encouraged to do so on a voluntary basis. No correlations could be observed between personal feelings of alienation and powerlessness on the one hand, and habitual car use and willingness to reduce this, on the other. In view of the modest study results, the authors conclude that ingrained personal habits play a significant role in t r a n s p o r t a t i o n behaviour, which they consider to originate primarily in various external, i.e., sociostructural and organisational factors. C o m m e n t s o n D e B o e r et airs p r o j e c t The design of De Boer et al's two survey studies typically reflects the cognitivistic (as opposed to behaviouristic) psychological view t h a t h u m a n b e h a v i o u r is "reasoned" and emanates from beliefs and evaluations, social norms and perceived i n s t r u m e n t a l i t i e s for a c h i e v i n g p e r s o n a l goals. F o r a c u l t u r a l l y a n d socio-economically strongly embedded behaviour domain such as mobility and transport, this view may well be overpersonalised. That is, given the enormous a v a i l a b i l i t y of m e a n s and facilities, e v e r y d a y needs and desires and the socioeconomic system pressures related to the use of motor-cars, little additional variance in people's habits, preferences and choices with respect to using the car can be explained by variations in personal attitudes, perceived social norms and feelings of alienation and powerlessness. On the contrary, as the authors conclude, the massive use of cars has developed into a social and economic and (therefore) cultural habit for most people. And behaviouristic - not cognitivistic - psychologists well know t h a t deeply ingrained habits are a u t o m a t e d behaviour mechanisms which can only be modified via significant changes in the incentive structure of the physical and social environment which provokes such habits and p e r p e t u a t e s them.
1152 3.
EFFECTS CAR U S E
OF C O S T A N D / O R E N V I R O N M E N T A L
FEEDBACK
ON
In a carefully designed field experiment supervised by Van Kreveld, Tertoolen and V e r s t r a t e n (NRP project no. 850013-2), also in Utrecht, have e v a l u a t e d the effects on travel mode choice and frequency of car use, of providing feedback information on the financial costs of one's own car driving, of providing information about one's "own" environmental effects of car travel, and of providing both kinds of feedback simultaneously. Control groups of respondents received no information whatsoever; one group did (like the three experimental groups) and one did not s y s t e m a t i c a l l y record their own travel behaviour for a certain period of time. Altogether 350 people participated in the experiment; all of them were asked to use t h e i r car as little as possible during the experimental period. This approach, tailored to the individual, contrasts with current government approaches designed to address larger segments of the general public via mass media campaigns and general pricing measures. Results of both a pilot study and the main experiment revealed t h a t subjects did express changes in attitude as a result of feedback information. However, no significant behavioural effects (as recorded in transport diaries) could be observed in r e l a t i o n to feedback about e i t h e r cost and/or e n v i r o n m e n t a l effects. Respondents rated speed, comfort and independence to be the most i m p o r t a n t advantages of using a motor-car and they stated that neither the financial nor the environmental costs of car driving weighed heavily when they were travelling. The a u t h o r s conclude t h a t a t t e m p t s to influence car use b e h a v i o u r a r o u s e psychological resistance, often expressed t h r o u g h dissonance reduction and reactance (counter-behaviour). As a result of dissonance-enhancing information about e n v i r o n m e n t a l effects, intensive car users originally having a positive environmental attitude, may s t a r t thinking t h a t environmental pollution is not t h a t bad after all, and t h a t others bear a greater responsibility for environmental problems t h a n they themselves do. Information about the financial costs of car use especially leads to reactance. Car users seem to experience financial policy measures as a restriction of their individual freedom. They therefore tend to have a dim view of both such measures and the authorities contemplating to implement them. Comments on Tertoolen and Verstraten's project The field experiment by Tertoolen and colleagues is unusual both in its purpose and scope, and in the care with which it has been prepared and conducted. As the authors expected themselves, the intensive and personalised procedure followed should have had greater effects than any generalised mass media campaign urging car users to moderate their behaviour. The fact that, nevertheless, respondents in all t h r e e feedback conditions did not reduce the use of their car, gives little e n c o u r a g e m e n t for public authorities contemplating mass media a t t e m p t s at motorists' behaviour change. This conclusion is in line with the earlier one about the difficulty of changing frequently reinforced, habitual car use without modifying socio-economic system characteristics. Another conclusion worth noting is t h a t little progress on the way toward less rnotorised mobility can be expected from policy m a k e r s having too much respect for individual car users' freedom of travel mode choice. Individual behaviour change of everyone's own free will seems only plausible to the extent t h a t a general positive a t t i t u d e toward p r e s e r v i n g
1153 environmental qualities becomes central in most people's view of the world and their own lives.
4.
PROBLEM AWARENESS, WILLINGNESS E V A L U A T I O N OF POLICY M E A S U R E S
TO
CHANGE,
F u n d e d by the University of Groningen and the Ministry of Housing, Physical Planning and Environmental Affairs, Steg, Vlek and Rooijers (see project s u m m a r y and Steg et al., 1995) have carried out a related project on personal mobility and possible behaviour change. Steg et al. collected home interview responses from 539 r e g u l a r car drivers in and around the cities of A m s t e r d a m , Eindhoven and Groningen. A field-experimental design was followed in which respondents were categorised a priori by region (as indicated above), distance to their city centre (0-7, 7-15 and over 15 kms) and whether or not they had been presented with a brochure providing balanced information on the pros and cons of the massive use of motor-cars in The Netherlands. Respondents kept personal transport diaries for four days prior to the interview. The latter was conducted by a trained interviewer and contained item sets designed to assess respondents' problem awareness, their willingness to reduce car use and their evaluation of 17 different policy measures all regarding the use of private motor-cars. It appeared that, on average, massive car use was perceived to be "a problem" (not a small nor a particularly big one), but significantly less so in and around G r o n i n g e n t h a n in the more densely motorised areas of A m s t e r d a m and Eindhoven. Also, city dwellers are more problem-aware t h a n people living at 7 or more kms away from the city centre. The information brochure did not affect respondents' problem awareness very much, probably because the pros and cons of massive car use were quite well known already through regular media coverage over the years. With regard to behaviour change, less t h a n one third of all r e s p o n d e n t s declared to be willing to reduce their car use. In this respect Eindhoven stood out more positively than either Amsterdam or Groningen, as did city dwellers compared to people living beyond 7 kms from the city centre. Of seventeen actual or contemplated policy measures to reduce the use of private motor-cars, none was rated as significantly effective, Groningers proving to be even more sceptical t h a n inhabitants of the Amsterdam and Eindhoven regions. So-called push measures such as increasing fuel prices or parking rates, were evaluated as h a r d l y acceptable, while pull measures such as improving public t r a n s p o r t or bicycling facilities, were judged to be acceptable. A post hoc categorisation of the 539 respondents into a "low", "middle" and "high" problem awareness group, respectively, yielded the conclusion that the extent of problem awareness correlates significantly with people's willingness to reduce their own car use a n d t h e i r e v a l u a t i o n of the r e l a t i v e e f f e c t i v e n e s s (or r a t h e r : non-ineffectiveness) and acceptability of policy measures aimed at a reduction of p r i v a t e car use. The a u t h o r s conclude t h a t increasing collective problem awareness is a pre-requisite for getting motorists to change their behaviour for the common interest. Apart from this, clear policy goals and consistent government strategies are essential.
1154 C o m m e n t s o n S t e g et al.'s project This r e s e a r c h d e m o n s t r a t e s again t h a t using a private motor-car is highly attractive and i m p o r t a n t to most people. Nevertheless people t h r o u g h o u t The Netherlands perceive the collective disadvantages of car use to be a problem about which the government should do something. This problem awareness, which varies both between and within diverse regions of the c o u n t r y - as well as between sex and age groups of respondents - significantly covers with people's (limited) willingness to change t h e i r t r a n s p o r t behaviour and with t h e i r (sceptical) evaluation of various policy measures. According to the authors, increasing public problem a w a r e n e s s and providing feasible t r a n s p o r t a l t e r n a t i v e s would be essential ingredients of any serious government policy designed to reduce the intensity of car traffic. Official Dutch policy goals are: to save energy and diminish environmental pollution, to enhance the accessibility of important destinations and to keep cities worth while to live in or visit. Given t h a t policy m a k e r s currently know fairly well why and how to effectively influence individual people's t r a n s p o r t a t i o n b e h a v i o u r , the e s s e n t i a l q u e s t i o n now is: u n d e r w h a t environmental, social and/or economic conditions could the car driving population (top politicians included) be "moved" to vote for and accept restrictive policy m e a s u r e s aimed at reducing the use of motor-cars in order to preserve and promote vital collective goods and qualities?
5.
C A R U S E R D I F F E R E N C E S IN E N V I R O N M E N T A L P O L L U T I O N A N D POLICY SUSCEPTIBILITY
Building upon previous research by Rooijers (1990) on speed differences among different types of car drivers, Cavalini, Hendrickx and Rooijers (NRP project no. 852092) at the University of Groningen have analyzed and described systematic differences among distinct car user groups, with respect to their usual speed, fuel consumption and environmental effects. The latter were taken as depending upon various types of decisions or behaviour. Five categories were distinguished: car purchase, choice of car type, car use, timing and routing of trips, and driving behaviour. The idea here is that net environmental effects may be directly as well as indirectly related to current behaviour, so t h a t emission reductions could be a t t e m p t e d at various points in the above sequence. Through questionnaire, interview and field observation research, the authors were able to classify car drivers in terms of car possession, car characteristics, n u m b e r of kilometres driven annually, average occupation rate, type of (either or not congested) roads used, and driving speed and style. Three main car user groups could be identified, viz. private drivers, commuters and business drivers. The latter were subdivided into business drivers using their own car and those having a company car at their disposal. The first field study addressed four issues: the usefulness of the distinction among the four car user groups, the size of these groups in the total Dutch population, environmental parameter differences among the four groups, and the combination of user group and environmental p a r a m e t e r type so as to identify significant possibilities for emission reductions. Raw data were collected via mailed questionnaires returned by an apparently representative sample of 1150 respondents.
1155 It a p p e a r e d possible and useful to s e g m e n t the driver population in t e r m s of car u s e r group, as defined above. Forty-two percent of all car users drives for private r e a s o n s a n d r e p r e s e n t s 23% of all car kilometres driven. C o m m u t e r s r e p r e s e n t 37% of all drivers a n d cover 40% of all car kilometres. B u s i n e s s drivers w i t h private car and business drivers using a company car m a k e up 13% and 8% of the driver population, respectively, and t h e y each r e p r e s e n t 18% of all k i l o m e t r e s driven by car. W i t h respect to all p a r a m e t e r s studied t h e r e a p p e a r to be large differences among car user groups. Business drivers using a company car t u r n out to be the most energy-consuming and environment-polluting group, while private drivers are doing the (relatively) least h a r m to the environment. By focusing on p a r t i c u l a r c o m b i n a t i o n s of u s e r group and type of b e h a v i o u r or decision (see above), it s e e m s possible to achieve CO2 emission reductions of about 5%, per c o m b i n a t i o n , so t h a t the overall CO2 emission reduction p o t e n t i a l should be substantial. The second field study by Cavalini et al. was aimed at clarifying the potential of various policy strategies w h e n applied to different car user groups in relation to different t y p e s of b e h a v i o u r or decision (see above). Six categories of policy i n s t r u m e n t s were distinguished for inducing changes in environmental p a r a m e t e r s of car use. These are, respectively, physical a l t e r n a t i v e s a n d r e a r r a n g e m e n t s , regulation and enforcement, financial-economic stimulation, providing information a n d c o m m u n i c a t i o n , social m o d e l l i n g a n d s u p p o r t , a n d i n s t i t u t i o n a l a n d organisational change. The m a i n research questions were: 1. To w h a t extent are the different car user groups able to change various types of behaviour or decision regarding car use? 2. How sensitive are the various behaviours and decisions of the four user groups to the application of different policy instruments? 3. To significantly reduce CO2 emissions, which type of policy i n s t r u m e n t m a y best be applied to which group and in connection with which type of behaviour or decision concerning car use? To collect r a w data, first some 4000 roadside observations were m a d e of passing cars, whilst car type, license n u m b e r and speed were recorded. Then, car owners were identified via the national car registration s y s t e m and t h e y were invited by telephone to participate in the study. W h e n 50 confirmations for each user group (see above) h a d been obtained, personal interviews were conducted. The l a t t e r were focused on possible changes in car use and on the respondent's evaluation of various policy measures. Some m a i n results are the following. All car drivers indicated t h a t they would have less personal control over possible changes - like giving up their car, changing type of car or reducing car kilometrage to the extent t h a t these would have far-reaching consequences for t h e i r mobility a n d daily life. The m a j o r i t y of drivers could t a k e a s m a l l e r car, decrease t h e i r n u m b e r of "private" kilometres or drive more slowly and quietly. P r i v a t e drivers and c o m m u t e r s are generally more inclined to change their behaviour t h a n either group of b u s i n e s s drivers. For achieving reductions in h a r m f u l car emissions, communicative and educational m e a s u r e s seem to be less effective t h a n legal and financial m e a s u r e s as well as i n f r a s t r u c t u r a l a n d o r g a n i s a t i o n a l m e a s u r e s . Financial m e a s u r e s would be more effective in changing the behaviour of private drivers a n d c o m m u t e r s t h a n t h e y would be affecting the b e h a v i o u r of business
1156
drivers. Also, financial measures are most effective in making people to give up their car and to make them drive fewer "private" or "commuting" kilometres. Finally, the reasons given for not changing behaviour or decisions about car use, under any of the presented policy measures, reveal similarities with the reasons provided in relation to personal control, as mentioned above. C o m m e n t s o n C a v a l i n i et al.'s p r o j e c t The methodological approach ventured in this project proved to be successful in identifying distinctly different car user groups and demonstrating their differential energy consumption and environmental pollution, as well as their differing sensitivities to various policy measures. Moreover, a comparison of the two studies yields the conclusion that those who consume the most energy and produce the most harmful emissions, also are the least sensitive to current policy measures. Such information provides a useful basis for designing and targeting specific measures and strategies for reducing the harmful effects of mobility and motorised transport. The major policy conclusion from this research is t h a t m a n a g e m e n t policies for motorised mobility should be designed to fit the personal motives, habits and mobility needs of possible target groups of car users. Cavalini et al.'s respondents clearly signalled t h a t reducing the environmental effects of their mobility would also reduce the (perceived) "control" over their daily lives, if it would involve giving up their car or driving significantly less t h a n usual. Thus, f u n d a m e n t a l thought must be given to policies designed to compensate for this feared "loss of control" when the use of private motor vehicles is to be diminished for reasons of collective importance. This means that target groups should be investigated a priori, to determine the potential impact of contemplated policy measures on their daily lives and the extent of "cooperative power" that they could, or would, have. "Cooperative power" (i.e. sufficient- r e m a i n i n g - personal control) u n d e r changing conditions for mobility and t r a n s p o r t could be enhanced by decreasing people's structural needs and desires for motorised mobility, by helping people to better organize their daily or weekly travel patterns, and/or by providing a l t e r n a t i v e t r a n s p o r t modes known to be socially and e n v i r o n m e n t a l l y less harmful. 6.
ECONOMETRIC HOUSEHOLDS
ANALYSIS
OF
PRIVATE
CAR
USE
BY
At the Agricultural University of Wageningen, Rouwendal, Van Staalduinen and K o o r e m a n (NRP project no. 852081) have systematically looked into the dependence of car ownership and car use upon variations over time in the price of cars and of car fuel. Their aim was to find out the extent to which car users are actually sensitive to the "price mechanism", and in what respect (e.g., type of car, kind of trip or driving style) such sensitivity would be manifested in their behaviour. Knowing this is important for applying financial policy measures to reduce the volume of car traffic and/or the purchase and efficient use of smaller cars. On the basis of econometric analyses of some 3759 observations about 1379 motorists from the "private car panel" of the Central Bureau of Statistics (a continuous, time-variable sample of respondents), the authors have estimated various model parameters. This research is still in progress. Some preliminary results and conclusions are as follows.
1157 Automobile drivers do change their short-term "demand" for car kilometres in response to changes in fuel prices, especially when they do not receive an employer c o m p e n s a t i o n for automobile costs. Demand-price elasticities (behavioural sensitivities) appeared to be different for different age groups, male versus female car users, and for summer versus winter periods. Older, male and "winter" drivers appear to be less price-responsive t h a n younger, female or "summer" drivers. It also appeared t h a t higher income, greater commuting distance, being a company director, holiday driving and getting a car-use allowance, constitute circumstances under which car driving is intensified and less price-sensitive. The authors state that, although short-term demand-price elasticities are significant, the demand for cars and car kilometres has steadily grown over the past 15 years. This seems due to powerful other factors t h a n the variable costs of car driving. For instance, an increased general income level has made car ownership and car use relatively cheaper, women's increased participation in the labour market has enhanced their share of the car driving population, and backward developments in public transport have "forced" many people to equip themselves with private motor-cars. Comments
on Rouwendal
et al.'s project
In the past, demand-price elasticities for car ownership and car use have hardly been studied systematically in sufficient detail to u n d e r s t a n d which type of behaviour change occurs in response to certain price changes. Rouwendal and colleagues have demonstrated that fuel price changes affect particular kinds of car use (e.g., social-recreational trips) more than others, and that certain categories of people (e.g., middle-aged men) are less price-sensitive than others. In this respect Rouwendal et al.'s project goes nicely along with the work on distinguishing car user groups, conducted by Cavalini et al. (see above). Naturally, a fuel price change means different things to different people, to the extent that their "substitution behaviour" - what they can and will do instead of their higher-priced current car use - turns out to be different. Looking more closely (and perhaps prospectively i n s t e a d of retrospectively as m a n y econometrists do) into subjects' likely substitution behaviour may reveal their reasons for manifesting different patterns of reactions. Another problem in demand-price elasticity research lies in the distinction between short-term and long-term behavioural adaptations to price changes. A sudden change of price may yield a short-term behaviour change all right, but what happens on the longer term is often revealing of more fundamental driving forces u n d e r l y i n g a given category of behaviours, as the a u t h o r s themselves acknowledge. Factors discussed by the Dutch Physical P l a n n i n g Service (Allsop, 1993), for example, are the increased physical separation of living and working locations in the 1970s and 1980s, the growth in the labour market for women, the larger n u m b e r of one- and two-persons households emerging from "individualisation", and growing i m m i g r a t i o n from abroad. Such long-term developments and trends raise the question of the significance for car ownership and car travel of the variable costs of car driving as a factor by itself. With reference to Tertoolen and Verstraten's project (see above) we might say t h a t altering car ownership and car use via the "price mechanism" would require fairly drastic financial policy measures. The project by Steg et al. (see above) has revealed t h a t this would be unacceptable for most people, unless certain key conditions for feasible behaviour change would be fulfilled.
1158 7.
SUSTAINABLE TRANSPORT AND TRAFFIC SYSTEMS FOR THE 21ST C E N T U R Y
The mobility and transport research results so far may leave the reader with reserved feelings about the possibilities to control the growth of motorised traffic and reduce harmful emissions. One long-term policy strategy, therefore, could be to go more deeply into the structural determinants of mobility and attempt to adapt or to change social systems so as to reduce the inherent demand for mobility. An other policy strategy could be to acknowledge the need and the desire for greater mobility of persons and goods, and to design sophisticated transportation modes and systems whose environmental effects stay within ecological limits. Working towards the latter policy strategy, economists Nijkamp, Rienstra and Vleugel (NRP project no. 853102) at the Free University of Amsterdam have conducted a "comparative analysis of options for sustainable transport and traffic systems in the 21st century". Other research associates are at the Technical University of Delft, the University of Groningen, the Energy Research Centre in Petten, and University College London. The project has been conducted in two parts, one focused on exploring separate transport modes, the second directed at the construction and evaluation of diverse national transport scenarios. In their report of Part 1, after an analysis of various problems of transport in relation to environmental quality, the authors systematically describe current trends in transport demand and supply. They point at the quest for higher transport quality and discuss various factors underlying the growth in mobility, such as rising incomes, spatial spreading of homes and work places, population growth and individualisation (leading to more and smaller households). They also discuss several types of market failure yielding undesirable "externalities", and they indicate failures in government policy to manage societal demand for mobility. On the basis of interviews and workshops with international experts, the authors t h e n p r e s e n t a list of technical, economic, spatial, i n s t i t u t i o n a l and socio-psychological factors that would be important for future developments in t r a n s p o r t a t i o n . Subsequently, selected (new) t r a n s p o r t a t i o n modes are systematically evaluated against these various factors. The authors' analysis goes along the advanced automobile, the high-speed train, low-speed Maglev (magnetic levitation) systems for urban transport, the electronically guided vehicle, subterranean transport and liquid hydrogen aircraft. Conclusions from Part 1 are that there are many possibilities for future reductions of greenhouse gas emissions from transportation. Three general strategies for emission-reduction are: cleaner transport technology, changing the modal split between polluting and (relatively) clean transport modes, and reducing the demand for mobility. Focusing on the first and second strategies mentioned, the authors conclude that the most likely technologies seem to be improvements of the private car, the high-speed train and the use of telematics for increasing transport efficiency. It seems unlikely that Maglev high- and low-speed transport will be introduced at any large scale. The third general strategy, reducing mobility demand, has received less attention in this project.
1159 In the second stage of the project, two reference scenarios, a "regulatory" and a "market" scenario, have been constructed which reflect extreme profiles in a "spider model" comprising eight major dimensions. The latter are categorised in pairs as spatial, institutional, economic and social/psychological, respectively. Against this background an "expected" and a "desired" scenario were constructed on the basis of questionnaire responses from various Dutch t r a n s p o r t experts. Subsequently, these two scenarios were discussed by an international group of experts. In the "expected" scenario which comes close to the "market" scenario just mentioned, it is assumed that current trends will continue, and that therefore the private motor-car will remain the dominant transport mode. Its freedom of use would, however, be restricted by regulatory measures such as fuel price increases and higher parking rates. In the "desired" scenario more stringent policy measures to discourage the use of private cars would be introduced, together with policies aimed at developing higher-quality means of collective transport. The authors conclude t h a t the expected scenario is, of course, more plausible, but that it could not be called "sustainable" unless an environmentally much less harmful mode of private t r a n s p o r t would be developed t h a n seems technically feasible for quite some time. The "desired" scenario, on the other hand, would involve considerable social behaviour change, together with stricter government policies and fairly big investments in rather different infrastructure than the sort which is underlying the private car system. C o m m e n t s on N i j k a m p et al.'s project Although the scenario study has not yet been definitely reported, it m a y be concluded t h a t this project has been a useful exercise on the possibilities of sustainable mobility and transport. With a focus on The Netherlands and with i n p u t s from experts from E u r o p e a n countries facing s i m i l a r t r a n s p o r t developments, a multidisciplinary picture has been sketched of the main technical options and several distinct societal scenarios for mobility and transport in the early 21st century. It has become clear - once again, we might say - t h a t western industrial society is strongly tuned toward the free-market system and toward meeting the demand for individual transport by motor-car for any person at any time and in almost any place. Reducing the social, economic and environmental costs of this t r a n s p o r t system, which m a n y people find no longer sustainable, would seem to require principal decisions by government policy makers. It would also require far-reaching social a t t i t u d e and behaviour changes, which are conditioned by problem perception and the availability of behaviour alternatives in relation to mobility and t r a n s p o r t (see e.g., the projects by Tertoolen and Verstraten and by Steg et al., above). Methodologically, Nijkamp et al.'s project relies heavily on expert assessments and opinions (e.g., about the "desired" scenario). It was not designed to analyze the fundamental societal factors and individual motives underlying the increased demand for motorised transport. Nor was the project aimed at collecting attitude and behavioural data from diverse sectors and groups in society, so that an assessment could have been made of the social and economic viability of particular transport options in relation to specific behaviour changes. Finally, it appeared that the experts themselves, too, may be of different opinion when it comes to designing and recommending "sustainable" transport scenarios for the near future. One point of discussion, for example, was to what extent government could at all come to grips with the collective problems
1160
of private car use, given the degree of organization and the economic capabilities of the car industry. 8.
G E N E R A L OBSERVATIONS, CONCLUSIONS AND S U G G E S T I O N S
NRP-funded research on mobility and transport so far has primarily addressed the massive use of private cars. Research on freight transport on the roads has not been undertaken in phase 1 of the NRP, nor have any studies been started on air transportation. The private-car mobility research carried out has, on the one hand, been focused on the individual car user, to assess his or her motives, attitudes, behaviour and sensitivity to policy measures and/or feedback information. On the other hand, long-term physical and technical alternatives to the private motor-car have been explored and evaluated, under the premise that the demand for mobility is there and could not (or should not) be influenced. Both lines of investigation seem to underrate the importance of social and economic system factors underlying the demand for mobility and the need for motor vehicles. System factors strongly influence and shape individual motives and preferences to the extent t h a t individuals may be brought in a forced position to acquire and use a motor-car. Also, the fate of physical and technical alternatives for the private motor-car seems strongly dependent upon the nature of economic and social activities and upon the way in which social and economic interaction is organised. In this respect, there seems to be room for more fundamental studies into non-transport factors residing in various social and economic domains, whereby mobility and the need for motorised transport may be generated, or may be reduced. At the Dutch national level, the NRP research on mobility and t r a n s p o r t complements the research initiated and funded by the Advisory Service for Traffic and Transportation of the Ministry of Traffic and Waterways. This Service's programme of "anticipating research" for 1995 (VWS, 1995), for example, lists such topics as "green" transport scenarios, effects of changes in government administration, possibilities and consequences of improved transport informatics, electronic vehicle guidance, fast waterborne transport and improvement of government communication strategies. This government-directed research still largely evolves from the Second Structural Scheme on Traffic and Transportation (1988-1990) and it is designed to yield results potentially supporting current government policy. It is therefore still very much in line with Nijkamp et al.'s "market scenario" (see above) which reflects a societal as well as large-scale individual preference for privately organised mobility and transportation. Internationally, NRP research in phase 1 links up with the concerns and intentions expressed in various programmes, conferences and workshops. For example, in 1992 the European Commission published a "Green paper on the Impact of Transport on the Environment" (CEC, 1992); also, its Directorate-General XII funds several projects on "the integration of environmental concerns into transport policy" EC, 1994), perhaps a modest beginning but something that could fly-wheel itself up. The Human Dimensions of Global Environmental Change Programme (HDP, 1994) of the International Social Science Council has not yet identified a separate line of research on mobility and transport, although it has indicated "industrial transformation and energy use" to be a key area for research. More importantly, ISIRT, the International Scientific Initiatives on Road Traffic group
1161 since 1988, has conducted three international "round tables" on mobility and t ra ns po r t in relation to environmental problems. A s u m m a r y report entitled "Agenda for safe access to a stable environment" was prepared by Allsop (1993). His final conclusion - on behalf on the ISIRT steering committee - reads: "Radical changes in road traffic and its uses, w h e t h e r the changes be technical, institutional, behavioural, regulatory, financial or fiscal, are likely to be uncomfortable at least for some people in the short term. But they should be brought about because the alternative is to continue to put up with the many and severe adverse effects of road traffic in its present form, and thus fail to use it to the best advantage" (Allsop, 1993, p. 6). Another pertinent meeting, organised in September 1992 by the European Ministers of Transport (ECMT) Conference, was held at the OECD headquarters in Paris. On page 237 of the conference proceedings, titled "Transport policy and global warming" (ECMT, 1993), a summary and conclusions section is phrased in ten points. The first four of these "messages for ministers" read as follows. "(1) C u r r e n t trends are clearly inconsistent with the Rio (UNCED 2; Ch.V.) aspirations. (2) New technology can improve matters, but there is no complete technological fix immediately available. (3) Nevertheless it is possible to reduce transport's contribution to global warming; what is necessary is the political will to introduce the necessary measures. (4) Existing technology is not being put to best advantage because of the freedom of transport users to adapt their behaviour to convert potential environmental amelioration into more transport service. (5) Controlling this adaptive behaviour should begin immediately by seriously addressing the issues of reducing the specific power, performance and speed of vehicles" (ECMT, 1993, p. 237). In view of this assessment of NRPI-research and the wider conclusions mentioned above, mobility and transport remain on the NRP agenda. For phase 2 of the NRP, covering the period of 1995 through 2001, societal causes and solutions of potential climate problems will again also be sought in cleaning up and/or reducing mobility and t r a n s p o r t by motor vehicles. Relevant themes are: economic and social-cultural determinants of mobility and transport, options for limiting the need for mobility and transport, and sustainable mobility and transport policies and strategies for society. Thus more attention is being asked for mobility-generating developments and trends in society, such as, e.g., i n t e r n a t i o n a l tourism, development of the labour market and upscaling of the educational system. Also, a focus is being laid on social implementation and acceptance problems in relation to sustainable-transport options and strategies. Finally, mobility and transport are being linked to consumption patterns and lifestyles, in an attempt to clarify the potential effects on the quality of producers' and consumers' life of low-mobility activity patterns and collective- transport scenarios for society as a whole. 9.
REFERENCES
Allsop, R.E., 1993. "Agenda for safe access to a stable environment; issues for decision m a k e r s as identified at ISIRT Round Tables 1989-1991". International Association of Traffic and Safety Sciences (IATSS), Tokyo, 29 pp.
1162 Bleviss, D.L. and Walzer, P., 1990. Energy for motor vehicles. Scientific American, 263 (September): 55-61. CEC, 1992. Green paper on the Impact of Transport on the Environment. COM (92)46. Commission for the European Community, Brussels. EC, 1994. Project summaries; research on economic and societal aspects of environmental issues. European Commission, Directorate-General XII, Brussels. ECMT, 1993. Transport policy and global warming. Proceedings of a European Ministers of Transport seminar in Paris. ECMT Series 75 93 10 1, OECD Publications. EZ&VWS: Dutch Ministries of Environmental Affairs and Traffic and Waterways, 1987. Verkeer en Milieu (Traffic and Environment) Policy Document for Second Chamber of Parliament. Ministry of VROM, Division of Public Information, The Hague, 66 pp. EZ, Ministry of Environmental Affairs, 1991. Ruimtelijke verkenningen. (Spatial explorations). Yearbook of the National Physical Planning Service of The Netherlands (Rijksplanologische Dienst). Ministerie van VROM, The Hague. HDP, Human Dimensions of Global Environmental Change Programme, 1994. Work Plan 1994-1995. International Social Science Council, Paris. Lenz, K.-H., 1990. Motorization and trends in road traffic. In V~ig- och Trafikinstitutet LinkSping: Proceedings of "Road safety and traffic environment in Europe". September 1990, VTI-Report no. 362a. Lowe, M.D., 1990. Alternatives to the automobile: transport for livable cities. Paper 98, Worldwatch Institute, Washington D.C.,49 pp. MacKay, M., 1990. Towards a unified traffic science. IATSS Research: Journal of the International Association of Traffic and Safety Sciences 14: 19-26. MacKenzie,J.J. and Walsh, M.P., 1990. Driving forces; motor vehicle trends and their implications for global warming, energy strategies, and transportation planning. World Resources Institute, Washington D.C. Rooijers, A.J., 1990. Drivers' attitudes and beliefs towards speed limits and speeding on Dutch motorways. In V~ig- och Trafikinstitutet LinkSping: Proceedings of "Road safety and traffic environment in Europe", September 1990. VTI-Report no. 363a. Steg, L., Vlek, C.A.J. and Rooijers, T., 1995, in press. Gedragsverandering ter vermindering van het autogebruik: probleembesef, verminderingsbereidheid en beoordeling van beleidsmaatregelen. (Behaviour change for diminishing car use: problem awareness, willingness to change and evaluation of policy measures). In: F. Siero, E., van Schie, D. Daamen and A. Pruyn (Red.). Sociale psychologie en haar toepassingen. Deel IX, Eburon, Delft. Vlek, C.A.J. and Michon, J.A., 1992. Why we should and how we could reduce the use of motor vehicles in the near future. IATSS Research: Journal of the International Association of Traffic and Safety Sciences, 15: 82-93. VWS: Ministry of Traffic and Waterways: Advisory Service for Traffic and Transportation, 1995. Anticiperend onderzoek; projecten 1995. (Anticipating research; projects-1995). Directoraat-Generaal Rijkswaterstaat, AVV, Rotterdam. VWS: Dutch Ministry of Traffic and Waterways, 1989. Tweede Structuurschema Verkeer en Vervoer. Deel A: Beleidsvoornemen. (Second Structural Scheme on Traffic and Transportation: Part A: Policy Intentions). SDU, The Hague.
1163 Walsh, M.P., 1990. Global trends in motor vehicle use and emissions. Annual Review of Energy 15: 217-243. WCED: World C o m m i s s i o n on E n v i r o n m e n t and D e v e l o p m e n t , 1987. Brundtland-Report: Our common future. Oxford University Press, Oxford/New York.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1165
Discussion on the NRP assessment reports "Mobility and motorised transport which fit in sustaineble development" and "Culture, consumption and lifestyles" M.M. Berk Introduction In this session prof. Vlek gave an overview of the main results of the 11 projects within the subthemes "Consumption and lifestyles" and "Mobility and Transport" of NRP-phase I. Due to time restrictions their was limited time for discussion. During Vleks' presentation some correcting / additional remarks were made. These have been used for finalizing the subtheme reports. Here, only the main discussion items are reported as the main results of the projects are covered fully in the subtheme reports. General remarks Vlek started with presenting a few general notions of relevance for both studying Lifestyles and Consumption pattern and Mobility and Transport.
One general notion is the formula commonly used in environmental studies, according to which environmental impacts can be described as the product of Population(P), Economy (E) and Technology(T). Traditionally, policy efforts to restrict environmental impacts have focused on changing technologies used. Given the magnitude of environmental problems the question has been raised whether not also the level of economic activity- both production and consumption - needs to be reduced. Addressing the Population factor is being viewed as difficult for either social or political reasons. From the perspective of studies on consumption and lifestyles, in addition to the above mentioned factors, it is important to look at how culture (C) and institutions(I) interact with population, economy and technology. They influence population development, consumption needs and technological development. In studying consumption it is useful to take the so-called production-consumption cycle into account, in which, on the one hand, consumers in turn for wages deliver labour to the production process and at the same time influence production by buying products and services. Both at the production and consumption side there are environmental impacts due to use of land, materials and energy and the production of waste and pollution. Production and consumption are interrelated. Therefore, one should study the environmental effects of consumption and production in an integrated wayand not just focus on the consumption side only.
1166
In t h e context of the issue of Global E n v i r o n m e n t a l C h a n g e the s t u d y of t h e h u m a n dimensions is relevant for studying: the socio-economic impacts and risks of environmental change; t h e causal m e c h a n i s m s or causes b e h i n d h u m a n induced e n v i r o n m e n t a l changes an the h u m a n responses to these changes. Global E n v i r o n m e n t a l Changes generate large collective risks in situations which can be defined as social d i l e m m a s . In t h e s e s i t u a t i o n s it is not r a t i o n a l for individual actors to do w h a t m a y be rational to serve the interest of all. Needed for handling collective risks are: a clear description of the risks and its sources; a w a r e n e s s of the risks with major actors; an weighting of the risks against the cost and benefits of action a p e r c e i v e d n e e d for c h a n g e l e a d i n g to an a s s e s s m e n t of b e h a v i o r a l alternatives setting of risk limits and translated into behavioral objectives m e a s u r e s and i n s t r u m e n t s to change behaviour policy implementation and evaluation feedback on collective risk reduction and total benefits Lifestyles and Consumption patterns On t h e basis of the r e s e a r c h on Lifestyles and C o n s u m p t i o n p a t t e r n s w i t h i n NRP-I Vlek presented the following conclusions:
* * * * *
*
*
there is still a need for an operational definition of lifestyle, not j u s t in t e r m s of behaviour but also of attributes and values; r e s e a r c h into the possibilities for s u s t a i n a b l e b e h a v i o u r is m u c h m o r e p r o m i s i n g if n a t u r a l science, technological and socio-economic expertise is linked in joint research efforts and research planning; it m a y be wise for policy m a k e r s to a s s o c i a t e s t a t u s a n d p r e s t i g e to s u s t a i n a b l e c o n s u m p t i o n , b u t it s h o u l d be c o m b i n e d by o t h e r policy instruments; s u s t a i n a b l e lifestyles should be p r o m o t e d by positive, r e w a r d i n g a n d attractive policy strategies, emphasizing the "desirable" not the "undesirable"; in r e s e a r c h on s u s t a i n a b l e lifestyles the meso and micro-level should be a d d r e s s e d condordantely, because of the links b e t w e e n c o n s u m p t i o n a n d production; it seems t h a t m u c h is possible technically if there would be enough problem a w a r e n e s s . W i t h o u t sufficient problem a w a r e n e s s people will not accept the policy m e a s u r e s a n d b e h a v i o r a l c h a n g e s n e e d e d to m a k e use of t h e s e technological opportunities; problem a w a r e n e s s depends on "visible" e n v i r o n m e n t a l effects of household m e t a b o l i s m (giving feedback on e n v i r o n m e n t a l behaviour). Because of the absence of directly visible effects of the impacts of and behavioral response to m a n y global environmental changes this is a very i m p o r t a n t issue.
Discussion * The question was raised if it was important to distinguish different lifestyles as
1167 these were viewed to be only minor deviations of the overall abundant "western lifestyle'. It was replied by Aarts that social research gives insight into the social processes that constitute different lifestyles and how these may change or be changed. Moll of the University of Groningen responded that in practise lifestyles can be rather easily defined on the basis of two main dimension:, socio-economic status and level of education. Ester emphasized that from the (theoretical) sociological perspective lifestyles are much more complex than as defined in empirical research. They embrace also different values. The importance of distinguishing different lifestyles is t h a t they are an integrating explanatory variable for different sets of environmentally relevant behaviour. Vlek remarked that from a policy perspective it is only relevant to make distinctions between different lifestyles as far as these have implications for the use of policy strategies: when people with different lifestyles have to be approached in different ways. From a policy perspective the concept of lifestyles should not only be used in a descriptive way but also in a prescriptive way- exploring desirable sustainable lifestyles. It was noted that the distinction of different lifestyles can only be of any practical relevance if it is actually possible to clearly define and distinguish these. For that reason it was suggested to make only broad policy relevant distinctions. In response to the conclusions presented by Vlek van Kreveld of Utrecht University remarked that, as confirmed by the outcome of the NRP-research on private car use, environmental awareness itself is not sufficient to make people change their behaviour. To make people change their behaviour it is e.g. also necessary that people are offered real behavioral alternatives. Vlek acknowledged that with respect to global environmental change problem awareness is, indeed, only but the first prerequisite for changing h u m a n behaviour. However, he liked to stress the importance of problem awareness in response to presentations on technological options often ignoring the question how to bring about the development, implementation and social acceptance of these technologies.
Mobility and Transport By introduction Vlek stated that, presently, transport is responsible for about 20% of the global emissions of greenhouse gases and constitutes a fast growing source. Not only transport markets in the developed countries are not yet satisfied, also much growth is to be expected in eastern Europe and in industrializing developing countries, like in Asia. Besides its contribution to the emissions of greenhouse gases, transport is causing many other problems like air pollution, noise and space demands. An important question is whether all these problems can be solved by better technologies or that the demand for mobility and transport itself should be addressed. Both directions were researched within NRP-I.
1168 After giving an overview of the main result of the research projects Vlek presented the following policy oriented conclusion: * *
* *
* * *
motor car use is a very suitable and attractive mode of transport for citizens, companies and governments alike; car use is very individualistic, but it is socially and culturally regarded as an obvious thing to do. Restricting car use would provoke strong resistance. One of the important reasons for that resistance is that there is a lack of (socially) acceptable alternatives; neither personal nor environmental costs are dominant factors in determining car ownership and car use; private car drivers, commuters and business drivers differ systematically in their environmental impacts and their sensitivity to various policy measures. So for policy makers it is very useful to distinguish between these different target groups; F u t u r e options for more sustainable transport systems seem r a t h e r modest: improved car technology, high speed t r a i n s and more intensive use of telematics (e.g. regulating traffic). Changing the demand for mobility should get more research attention. There is little research into the underlying socio-economic system characteristics t h a t provoke the (growing) demand for mobility; The environmental impacts of, demand for and governmental policies related to air traffic have been neglected in the NRP sofar and need more attention.
Discussion * It was noted t h a t the results of the research seemed r a t h e r obvious and not very surprising. Were these outcomes not known already from previous research? According to Vlek this is not the case. The research did however confirm the hypothesis that it is very difficult to get people out of their car. Van Kreveld added t h a t often results of social sciences seem obvious in retrospect, but were not known or commonly accepted when the research started. By illustration Vlek r e m a r k e d t h a t m a n y measures t a k e n by the Dutch M i n i s t r y of T r a n s p o r t are based on a s s u m p t i o n s which differ from the conclusion presented here, like approaching car drivers indifferently, focusing policy on the effects of the price mechanism and persuading the public to drive less by public information campaigns without paying due a t t e n t i o n to structural causes of the growing demand for mobility and transport.
With respect to the reported lack of change in car driving behaviour it was r e m a r k e d that in historical perspective a profound decrease in the growth of car use can be detected. The general idea that things remain the same is not well founded. Vlek noted that, although such changes can be noticed, these are not enough in the light of the change needed to arrive at a sustainable transport system. While car use may not growth that fast any more, the same does not apply for e.g. air travel which is ever growing faster.
1169 Also with respect to car drivers the question was raised how r e l e v a n t knowledge of the differences in sensitivity of different car driver groups for policy measures is given the potential contribution of these kind of measures to the reduction of greenhouse gases. Cavalini from the University of Groningen r e m a r k e d t h a t their research clearly stated the relevance of differentiating between different driver groups. As an example he mentioned that while, relatively, business drivers pollute most, measures influencing private car use make a bigger contribution to limiting emissions as private car drivers are responsible for half the total milage driven, including many relatively polluting short trips for which car use is less necessary. It was asked w h e t h e r the research also took foreign efforts to cope with growing problems of t r a n s p o r t a t i o n and mobility into account, like the experiences with restricting car use in Singapore. Their experiences m a y indicate other ways to approach car use. Furthermore, it was noted t h a t incremental solutions to mobility and transportation problems may postpone real solutions, thereby making problems only worse in the end. Making the present system crash might open the road for more fundamental solutions. With respect to the research by Midden et. al. at the University of Eindhoven on the effectiveness of emotion-oriented c o m m u n i c a t i o n compared to cognitive-oriented communication strategies it was r e m a r k e d t h a t in the United States the use of fear arousal in the communication on global environmental issues eventually backlashes as the stated effects did not occur a n d were contradicted. So, fear which does not hold w o r k s contra-productive.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1173
Private car mobility. Problem awareness, willingness to change, and policy evaluation: a national interview study among Dutch car users Linda
Steg a, Charles Vlek a and Ton Rooijers b
Department of Psychology, University of Groningen, Grote Kruisstraat 2/1, 9712 TS Groningen, The Netherlands. a
b Traffic Research Centre, University of Groningen, P.O. Box 69, 9750 AB Groningen, The Netherlands.
Abstract This paper reports on a field study, based on personal interviews with 539 car users. Problem awareness appears to be an important condition for any attempts to make people voluntarily reduce car use. Problem awareness also is an prerequisite for the acceptance of policy measures aimed at reducing car use. Problem awareness is higher the more people are confronted with the problems of car use. The provision of information in a brochure did not influence respondents' problem awareness.
1. THEORETICAL BACKGROUND The social dilemma paradigm is a useful model to understand and to manage problems in which numerous individual benefits are running up against cumulative collective costs and risks, such as from car use [1]. In large scale social dilemmas it is attractive to continue to act in one's own interest. Individual contributions to collective costs and risks, as well as to their reduction, seem negligible. Moreover, most people are pessimistic about the cooperation of others. So, individuals tend not to feel responsible for collective problems. This makes individual contributions to collective solutions unlikely. Members of the public as well as policy makers will only contribute to resolving largescale social dilemmas if two conditions are fulfilled. First, people must perceive motorised traffic as a source of serious societal problems. This requires a clear and unambiguous description of the various negative consequences. Second, people have to balance the collective disadvantages against the personal advantages of car use, and they must be convinced that the problems need to be solved. Thus, p r o b l e m a w a r e n e s s is an important condition for any attempts to make people voluntarily reduce car use [2-3]. For this study, we hypothesised that the higher people's problem awareness, the more they are willing to reduce car use, and the more favourably they evaluate relevant policy measures. Furthermore, we expected that the more people are confronted with problems of car use (in densely populated areas, in city centres, or by reading information about these problems), the higher their problem awareness would be, the more they would be willing to reduce their car use, and the more favourably they would evaluate poficy measures.
1174
2. METHOD We studied problem awareness, possible behaviour change, and the evaluation of policy measures for reducing car use through in-depth interviews with 539 car users selected as living within 7 kilometres, between 7 and 15, and further than 15 kilometres away from the centre of Amsterdam, Eindhoven, and Groningen, three cities having rather different mobility profiles. The collective problems of car use are most visible in the Amsterdam region, because of the high traffic volume, while in the Groningen region traffic volume is low and a lot of problems are not visible yet. The Eindhoven region takes a middle position. Within each geographic condition, a few days before the interview two thirds of the respondents received systematically different amounts of prior information in a brochure about the most important societal problems of the massive use of cars and possible solutions for them. One third of the respondents received information about the present problem situation. Another third received information about the present and future problem situation. The remaining respondents received no information. Twenty people were interviewed in each (19 in one) research condition. Structured interviews were conducted at respondents' homes by trained interviewers. The questionnaire contained, amongst other things, several items measuring the key concepts of 'problem awareness', 'willingness to reduce car use', and 'evaluation of policy measures'. Prior to the interview, respondents were given a travel diary in which they recorded all movements on the Friday, Saturday, Sunday and Monday prior to the interview. Interviewers checked to what extent the respondents had actually studied the brochure.
3. RESULTS We will concentrate on subjects' problem awareness, their willingness to reduce car use, and their evaluation of policy measures. Only differences which are statistically significant at p < .05 will be reported. On average, the respondents perceive various collective consequences of car use as 'a problem'. The scores on 'problem awareness' could vary from -10 ('not a problem at all') to +10 ('a very big problem'). The mean score (M) was 3.1. As hypothesised, on average people living in Groningen (M = 2.5) do have a lower score on 'problem awareness' than people living in the Eindhoven (M = 3.5) and Amsterdam (M = 3.2) region. People living in or near the city centre (M = 3.6) do have a higher score on 'problem awareness' in comparison to people living outside the city centre (M = 2.8). No significant differences were found between the information conditions. Only 30% of the respondents appear to be actually willing to reduce their car use. People living in the Eindhoven region (38%) have a greater willingness to reduce their car use in comparison to respondents living in the regions of Amsterdam (25%) and Groningen (24%). Among the 'distance' groups, also, there is a significant difference in 'willingness to reduce car use'. Respondents living within 7 kilometres of the city centre (34%) are more willing to reduce car use than people living between 7 and 15 kilometres of the city centre (24%). No significant differences were found between the information conditions. Respondents were asked to evaluate the effectiveness and acceptability of 'push' and 'pull' measures. Push measures are directed at making car use less attractive, such as
1175 through higher fuel prices. Pull measures are aimed at improving the alternatives for car use, such as improving the quality of public transport. Scores could range from -10 ('not at all effective' or 'not at all acceptable') to +10 ('very effective' or 'very acceptable'). On average, people evaluate neither push measures (M = -4.3) nor pull measures (M = -3.7) as effective. Respondents evaluate pull measures as 'acceptable' (M = 4.4). Push measures were evaluated as 'not acceptable, nor unacceptable' (M = -0.1). Again, people living in the (quieter) Groningen region evaluate push measures as well as pull measures as less effective and less acceptable in comparison to the respondents living in the more populated regions of Eindhoven and Amsterdam (see table 1). There are also significant differences in the evaluation of policy measures among the distance groups. This only pertains to the evaluation of the acceptability of pull measures: respondents living within 7 kilometres of a city centre evaluate pull measures more favourably (M = 5.1), especially in comparison to respondents living between 7 and 15 kilometres of the city centre (M = 3.9).
Table 1 Evaluation of push measures and pull measures per region I
effectivity 'push' effectivity 'pull' acceptability 'push' acceptability 'pull'
Amsterdam -4.3 a -3.6 a -0.1 4.8 a
Eindhoven -3.9 a -3.0 b 0.4 a 4.5 a
Groningen -4.9 b -4.3 c -0.5 b
3.7 b
1 Means with unequal superscripts differ at p < 0.05.
The 539 respondents were divided into three equal groups, on the basis of their scores on the concept of 'problem awareness'. Table 2 shows that respondents with a higher 'problem awareness' are more willing to reduce their car use in comparison to people with a lower problem awareness. Moreover, respondents with a higher score on 'problem awareness' evaluate policy measures more favourably.
Table 2 Willingness to change and evaluation of push measures and pull measures for groups differing in problem awareness (all percentages and means differ at p < 0.05) problem awareness
low
middle
high
willing to reduce
18%
29%
39%
effectivity 'push' effectivity 'pull' acceptability 'push' acceptability 'pull'
-5.7 -4.7 - 1.7 3.5
-4.0 -3.6 -0.2 4.2
-3.2 -2.7 1.7 5.4
1176 4. D I S C U S S I O N On average people perceive car use as 'a problem'. However, most people are not willing to reduce car use. Respondents evaluate current Dutch push measures as well as pull measures as rather ineffective. They judge pull measures to be acceptable, while push measures are evaluated as 'acceptable nor unacceptable'. So, on average people believe that policy measures aimed at reducing car use are acceptable, but not very effective (or they think the measures are acceptable because they are not very effective). There are several explanations for the perceived ineffectiveness of policy measures. First, problem awareness may not be as high as to make people actually do something about it. Second, problem awareness by itself is not a sufficient condition for reducing car use. People also must have the impression that the collective problems c a n be solved, that their own contribution is useful, and that others will also contribute to the solution of the problems [3]. As hypothesised, there is a positive relationship between problem awareness, willingness to reduce car use, and the evaluation of policy measures. Heightening problem awareness, therefore, seems a useful strategy, provided there are sufficient feasible alternatives available to reduce car use. Our expectation that the more people are confronted with the problems of car use, the higher would be their problem awareness, is only party confirmed. Respondents living within 7 kilometres of a city centre do indeed have a higher problem awareness, are more willing to reduce car use, and evaluate policy measures more favourably. Moreover, respondents living in the quieter Groningen region have a lower score on problem awareness, are less willing to reduce car use, and evaluate policy measures less favourably. However, there were no differences between respondents who did or did not receive prior information. Maybe the information, which can regularly be read in the newspaper, was not new to the respondents. It is also possible that people perceive the information as unreliable, and deny or downplay the information as valid. Collective costs and risks of car use are difficult to control. Effective solution strategies require, besides problem awareness, clear policy objectives, and a forceful and consistent government policy, based on several different policy instruments.
5. REFERENCES
1 Ch. Vlek, L. Hendrickx and L. Steg, A social dilemmas analysis of motorised-transport problems and six general strategies for social behaviour change, in: ECMT: Transport policy and global warming, Paris: European Conference of Ministers of Transport (ECMT), OECD Publication Service, 1993, pp. 209-225. 2 D.M. Messick and M.B. Brewer, Solving social dilemma's: a review, in: L.Wheeler and O. Shever (eds.), Review of Personality and Social Psychology, 4 (1983), Beverly Hills, Calif.: Sage. 3 B. Klandermans, Persuasive communication: measures to overcome real-life social dilemmas, in: W.B.G. Liebrand, D.M. Messick and H.A.M. Wilke (eds.), Social dilemmas: theoretical issues and research findings, Oxford: Pergamon, 1992.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1177
A Behavioral Analysis of Private Car Use by Households Jan Rouwendal, Lanie van Staalduinen en Peter Kooreman Department of Household and Consumer Studies, Wageningen Agricultural University, P.O. Box 8060, 6700 DA Wageningen, The Netherlands. Abstract The relevance of econometric studies of ownership and use of private cars for environmental issues is sketched and some recent results are reviewed.
1 INTRODUCTION The private car has been in the centre of concern about environmental issues ever since it became a mass consumption good. Exhaustion of oil resources, air pollution, acid rain and the greenhouse effect are all related to ownership and use of private cars by a large number of households. The private car is an ambiguous symbol of western society in the late twentieth century. On the one hand it reflects its success in providing a high level of material well-being to the large majority of the population. On the other hand it brings out the failure of the same society to solve the environmental problems evoked by its successes. The widespread anxiousness about the future sustainability of western societies has given rise to much concern about the continued growth of automobile ownership and use. However, the popularity of the private car has increased over the years and should be expected to do so in the future. This apparent paradox has been interpreted by social scientists as a social dilemma: everybody knows how to improve society, but apparently nobody is willing to take the necessary actions himself. 2 AN E C O N O M I C A P P R O A C H Social scientists seem to agree on the proposition that human behavior with respect to automobile ownership and use can be fruitfully considered as being driven by the desire to reach some purposes, for instance being able to reach the work location fast and comfortably. The driver is then assumed to act deliberately on the basis of a consideration of the benefits and costs associated with the various alternatives available to him. This plausible vision may be termed 'rational,' although that term should be interpreted in a limited sense. It does not suppose, for instance, that de actor takes into account the effects of his behavior on other people in the same way as the effects that concern himself. Nor is it required that future consequences should be given the same weight as immediate consequences. It is precisely because of these characteristics of human decision making that widespread concern about environmental problems coexists with increasing popularity of the private car. Welfare economics has developed a recipe for this problem. The basic trick is to introduce a tax to be paid by the actor that has the same effect on his decision making
1178 as a proper calculation of all present and future effects of his behavior on society as a whole would have. The environmental costs that are neglected by the decision maker because they do not concern him, or do not concern him immediately, are in this way brought to his attention and the balance between individual and societal rationality is restored. With respect to the contribution of the private car to the greenhouse effect, this prescription would require an estimate of the costs of adding one additional unit of carbondioxide to the environment and charging them to each driver. Since the emission of greenhouse gases is closely related to the amount of fuel used, a fuel tax would be the appropriate policy instrument. Since such a tax has been introduced in all western countries, the required increase of this tax should not be expected to give rise to implementation problems. Moreover, the widespread concern for environmental problems may be expected to offer the necessary political support for such a measure. The real problem for following this strategy is the determination of the marginal environmental costs of the emission of greenhouse gases. The present state of knowledge only allows one to think about the effects of such gasses in general and imprecise terms. Nevertheless, it may be said that enough is known to justify a policy directed at a reduction of the further emission of such gasses. In order to see what policy efforts are required in order to reach the desired effect, it is necessary, among other things, to investigate the determinants of automobile ownership and use by households. 3 A DECISION CHAIN It is useful to distinguish a number of steps in decision making with respect to the private car: - The most elementary decision concerns ownership.. Should one buy one (or more) private car, or make use of public transport? If a car is purchased, what ~ should it be? A large number of brands and makes are available. Fuel type, cylinder volume and weight are important characteristics for the environmental aspects of automobile use. - Which use will be made of the car? How many kilometers should be driven for homework interactions, for business purposes, for social purposes and on holidays? - The driving ~ influences fuel use significantly. Car users who like to accelerate fast and drive at high speeds cause more environmental damage than others. - The decisions taken at all steps determine the emission of greenhouse, gases by automobiles. The various steps in this decision chain show a certain hierarchy in the sense that the ones made earlier are more basic. For instance, the type of car is only relevant if a car will be bought. It should be kept in mind, however, that the various steps should be considered as interrelated. For instance, decisions with respect to car ownership are made on the basis of, among other things, preferences with respect to car use for different purposes. One important consequence of the many facets of decision making with respect to automobile ownership and use is that the introduction of, for instance, a higher fuel price may be expected to have a number of different effects that may operate at different time scales and possibly in different directions. For instance, an increase in the fuel tax may have the immediate effect of a decrease in the number of kilometers driven for social purposes. When a new car is bought, fuel efficient makes will be bought more often. This results in lower costs per kilometer, which mitigates the immediate effect of -
1179 the higher. Moreover, the higher costs of mobility may induce people to consider the possibility of living in the neighbourhood of his work relation more intensely than he would have done otherwise. This may result in a shorter commute, which strengthens the original effect of the tax measure. 4 A REVIEW OF RESEARCH RESULTS Research on the various aspects of automobile ownership and use dates back to the early history of econometrics. The early studies concentrated on time series of numbers of automobiles owned or produced. Gradually research shifted towards the micro economic aspects of car ownership and use. In the eighties econometric models that enabled a researcher to study the ownership and use of one or more cars by individual households became available (see Mannering and Winston [1985]. For the Netherlands this type of model was introduced by De Jong [1990] who found substantially larger effects of changes in fuel prices on both the number of kilometers driven and the decision to own a car than were suggested by earlier studies: a change in variable costs of 1% would in the short run give rise to .65 % less kilometers driven, while in the long run the effect would increase to 1.11%. Since fuel costs are the major component of variable costs, this suggests that drivers are sensitive to changes in these prices. De Jong used cross section data and did not take into account differences between car types. His results were therefore not based on observed reactions to changes or differences in fuel costs per kilometer driven. In the period from 1980 to 1993 fuel prices changed significantly only in 1986 and 1991. Use of time series would therefore offer only limited opportunities for measuring the effects of fuel prices in a more direct way. De Jong's work provided a good starting point for the work that is currently being done at the Department of Household and Consumer Studies of Wageningen University. The aim of this research is to provide a more detailed picture of household behavior in the various parts of the decision tree. Although this work is still in progress, we can mention some preliminary results here. One part of the project is a more careful investigation of drivers reaction to the decrease in fuel prices occurring at the beginning of 1986. Monthly data concerning the year 1986 were analyzed by Van Staalduinen and Rouwendal [1994] who found gasoline price elasticities for the number of kilometers driven that are of the same order of magnitude as those found by De Jong. The short run sensitivity of the demand for automobile kilometers for changes in fuel prices is mainly due to the social motive, as commuting and business travel are usually harder to change. The demand for commuting kilometers is determined to a considerable extent by the choice of the residential and work locations. In Rouwendal and Rietveld [1994] a search model that explains these choices is developed and estimated. Empirical application a this model enables one to estimate the required compensation for an additional kilometer of commuting for various types of workers. Part of the required compensation consist of fuel prices. Preliminary estimates of the model confirm the existence of such a trade off, suggesting that the long run effects of higher fuel prices on commuting distance should not be ignored. Another study concerned the determinant of the driving style. Rouwendal [1994] regressed the fuel use per kilometer, as indicated by the main drivers, on characteristics of the car, characteristics of the driver and on monthly data about fuel prices and average temperature. Driver characteristics were included because of their presumed effect on driving style. A significant coefficient for the gasoline price indicates that
1180 drivers respond to changes in fuel prices by driving in a more or less fuel efficient way. Continuation of this line of research may be expected to contribute to a more detailed and coherent picture of the determinants of car ownership and the behavioral reactions to changes in fuel prices. 5 OUTLOOK In the recent past increases in the fuel tax have not been able to slow down the increasing popularity of the private car substantially. There are several reasons that explain this state of affairs. It may be reasonably expected that variable cost per kilometer is the crucial variable influencing driver's behavior. These costs are influenced by the price of crude oil and by the fuel efficiency of the motor, as well as by the fuel tax. Over the past 15 years there has been no significant overall increase in the variable costs, despite several increases in the fuel tax. Moreover, income growth and increased participation of women in the labor force have contributed significantly to the rising number of cars. It must be expected that these forces will still be effective in the near future. It is therefore important to consider the effect of fuel taxes within a broad framework that incorporates the major economic and social trends. In this way the study of the determinants of car ownership and use may be expected to contribute to a better insight into the possibilities and limitations of reducing the emission of greenhouse gases. 6 REFERENCES
de Jong, G.C. [1990] An Indirect Utility Model of Car Ownership and Private Car Use, European Economic Review, 34, 971-985. Goodwin, P.B. [1992] A Review of New Demand Elasticities with Special Reference to Short and Long Run Effects of Price Changes, Journal of Transport Economics and Policy, 26, 155-169. Mannering, F. and C. Winston [1995] A Dynamic Empirical Analysis of Household Vehicle Ownership and Utilization, Rand Journal of Economics, 16, 215-236. Oum, T.H., W.G. Waters and J.-S. Yong [1992] Concepts of Price Elasticities of Transport Demand and recent Empirical Estimates, Journal of Transport Economics and Policy, 26, 139-154. Rouwendal, J. [1994] An economic Analysis of Fuel Use per Kilometer by Private Cars, research paper, Wageningen Agricultural University. Rouwendal, J. and P. Rietveld [1994] A Structural Model of Commuting Distances and Spatial Job Search, research paper, Wageningen Agricultural University. Rouwendal, J. and L. van Staalduinen [1994] A Panel-Data Analysis of Short-Term Changes in Travel Demand, research paper, Wageningen Agricultural University.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1181
Differences among car user groups regarding CO2 emissions and sensitivity to policy measures P.M. CavalinP, L. Hendrickx ~ and A.J. Rooijers b aCenter for Energy and Environmental Studies, University of Groningen, P.O. Box 72, 9700 AB Groningen, The Netherlands bTraffic Research Centre, University of Groningen, P.O. Box 69, 9750 AB Haren, The Netherlands
Abstract Two field studies revealed large differences among various subgroups in the population of car drivers. Private drivers, commuters, and business drivers differed strongly with respect to current decisions and behaviour which affect CO2 emissions, and with respect to their sensitivity to various policy instruments. Several promising policy targets were identified" combinations of user groups and behaviours where substantial CO2 reduction may be achieved. The sensitivity of different car user groups to various policy measures showed whether and how desired behavioral changes may be realised.
1. I N T R O D U C T I O N Motorised traffic contributes to an important extent (app. 15%) to CO2 emissions. For the major part (app. 10%) passenger cars are responsible for these emissions. The number of cars has almost doubled between 1970 and 1992 from 2.8 to 5.3 million. The annual total number of kilometres driven by these cars has increased from 36 to 86 billion in this period. Decreasing (the negative effects of) car mobility has become a societal priority. Policy measures with regard to passenger car mobility, may be categorized according to: (1) the behaviour or decision they aim to alter, (2) the (sub)group of drivers they are directed at, and (3) the policy instrument used to achieve the desired change. 1
Type
of decisions
or behaviour
Decisions - car purchase - type of car - car use - timing and routing - driving behaviour
--- > ---> ---> --- > --->
System parameters affected car possession car park characteristics number of kilometres, occupation rate congestion, road use driving speed, driving style
1182
2 Type of car users - private drivers - commuters - business drivers with private car - business drivers with non-private car (company or leased cars) Type of policy instruments (as distinguished by Vlek and Michon, 1992) - physical alternatives and (re)arrangements - regulations and enforcement strategies - financial and economic strategies - information and communication strategies - social support strategies - institutional and organisational strategies Differences in current behaviours of the various user groups implicate that measures aimed at altering behaviour may have different potential effects upon various user groups. Moreover, users groups may differ in their sensitivity to various policy measures. For designing policies which will effectively reduce energy use and adverse emissions by passenger cars two steps, taken in two studies, are necessary.
Study 1" who and what? This study focuses on the various types of car users and types of decision or behaviour, distinguished above. The aim of this study is to identify combinations of behaviour type and car user type, where at least in principle substantial CO2 emission reductions are possible. Study 2: whether and how? Whether these CO2 reductions may actually be achieved depends on two factors: personal control and sensitivity to policy instruments. Personal control refers to the extent to which drivers are able to change their decisions and behaviour. Due to eg. infrastructural or organisational factors, the degree of personal control may differ among user groups. The extent to which drivers are sensitive to various types of policy instruments may also differ among the user groups. The aim of study 2 is to determine differences among user groups with regard to personal control over their decisions and behaviour, and their sensitivity to various policy instruments, in order to determine - for each of the 'behaviour and user group' combinations identified in study 1 whether behavioral changes are possible and how these changes may best be achieved.
2. METHOD In study 1 a large and representative sample of Dutch car drivers ( n = 1150) filled in a postal questionnaire in which information was collected about various types of decisions and behaviour, as indicated above. The sample was drawn from car registration files. In study 2 interviews were held with app. 50 representatives of each user group, in which the degree of personal control over decisions and behaviours and the respondent's sensitivity to different types of policy instruments were assessed. The sample was drawn by observing cars on motorways.
1183 3. R E S U L T S The results of study 1 indicate that user groups differ strongly with respect to almost every CO2 relevant decision or behaviour. For most behavioral parameters, the betweengroup differences have a similar pattern. Private drivers (42% of the total population of car drivers) score less negatively on all CO2 relevant parameters (except car age). They drive relatively light, old, and fuel-efficient cars. Private drivers have the lowest kilometrage. They report to drive more slowly and in a more energy-efficient way than the other groups. Hence, their driving style results in fewer CO2 emissions per kilometre driven~ Business drivers with non-private car (8% of the population) score most negatively on all parameters (except car age) and contribute disproportionably to CO2 emissions: on average, they have the highest kilometrage, they drive heavy cars with a low fuel efficiency, and their speed choice and driving style result in relatively high CO2 emissions per kilometre driven. The commuters (37%) and business drivers with private car (13%) fall in between with regard to all CO2 relevant parameters. On the basis of these results several promising combinations of user group and type of decision or behaviour with regard to the reduction of CO2 emissions, were identified (for details see Cavalini, Hendrickx, and Rooijers; 1993). The results of study 2 reveal that the amount of personal control drivers (perceive to) have varies for the different decisions and behaviours studied. Many respondents report that, even if they would be willing to do so, they would not be able to give up their car, to drive fewer commuting kilometres, and/or to drive fewer business kilometres. With regard to other behaviours (take a smaller car with the next purchase, decrease the number of private kilometres, drive on other times, drive more slowly, and drive more responsibly) the respondents report to have a considerable amount of control. In general, the drivers view that they have less personal control over decisions which, if altered, would have more far reaching consequences, and vice versa. Moreover, it was found that user groups differ in the amount of control they have over specific decisions or behaviours. For instance, business drivers think they have less freedom to give up their car than private drivers and commuters. Compared to the other groups, fewer private drivers could decrease the number of private kilometres. Only a minority of commuters could drive on other times, whereas both groups of business drivers may have more freedom to do so. Subjects were asked why it was not possible to change a certain type of decision or behaviour. Three sorts of reasons were given: 1) reasons referring to organisational circumstances or conditions, 2) from the subjects' point of view their behaviour is already 'optimal', and 3) anti-public transport reasons. With regard to the drivers' sensitivity to the various policy instruments, the results demonstrate the following. In general, private drivers and commuters are more inclined to change their behaviour than both groups of business drivers. All user groups are more willing to change behaviour which does not alter their mobility life style (eg. smaller car, drive slower). Drastic behavioral changes are less likely to occur (eg. give up car, drive less). On average, the drivers are less sensitive to communicative measures (education and information) than to legal, financial, infrastructural, and organisational measures.
1184 As expected, user groups differed with regard to their sensitivity to different policy instruments. For instance, private drivers and commuters are to a larger extent than the business drivers, willing to give up their car, to buy a smaller car, to drive fewer commuting kilometres and to drive on other times~ The former groups are more sensitive to infrastructural, organisational, legal, and financial measures than business drivers. The difference is largest for financial measures. Especially the business drivers with nonprivate car are not sensitive to this kind of measures. They are more sensitive to hffrastructural~ organisational, and legal measures~ Of all groups, private drivers are most sensitive to information measures~ Financial measures appear to be the best type of policy measures to induce people to give up their car and to decrease the number of private or commuting kilometreso Infrastructural and organisational measures may best be used to affect decisions about type of car~ drive on other times, and drive more slowly. Legal measures could best be utilised to diminish business kilometres, to induce drivers to drive on other times and to drive more slowly~ Information measures may best be used to change routing behaviour of drivers.
4. C O N C L U S I O N This research project has indicated that segmenting the total car users population into several user groups, and distinguishing various types of decisions or behaviour regarding car mobility, may enable policy makers to design more effective policy programs which intend to decrease (the negative effects of) car mobility. The first study revealed large differences among user groups with regard to current COz relevant behaviours. Combinations of behaviours and user groups, where in principle substantial COz reductiolm are possible, were identified. The second study of the project demonstrated that the user groups distinguished differ in the extent to which they are able to change various types of decision or behaviour. This study also showed which policy measures would be effective to induce the desired changes in different user groups.
5. REFERENCES Cavalini, P~ Hendrickx, Lo and Rooijers, A.J. (1993). Differences among car user groups regarding CO 2 emissions. IVEM-OR Noo 65, Rijksuniversiteit Groningen. Cavalini, P.M., Hendrickx, L. and Rooijers, A.J. (1995). Differences among car user groups regarding their sensitivity to policy measures. Rijksuniversiteit Groningen. In press. Vlek, C.A.J. and Michon, J.A. (1992). Why we should and how we could reduce the use of motorvehicles in tile near future. Journal of the hzternational Association of Traffic and Safety Sciences, 15, 2, 82-93.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1185
Choosing a means of transportation: Two inquiries into situational and personal determinants of moving behaviour
I.M. de Boer, D. van Kreveld, P.G. Swanborn Vakgroep Sociale en Organisatiepsychologie, University of Utrecht, Heidelberglaan 1, 3584 CS Utrecht, The Netherlands
Two survey research projects on situational and personal determinants of moving behaviour have been carried out. The Hilversum survey defines the problem of choosing a means of transport in relation to private mobility behaviour and the models and situational variables used for explaining this choice. We discuss the models which are normally applied to behaviourial science investigations in order to explain behaviour. Using various traditions we discuss attitude, social environment and the opportunity structure as separate complexes. These three are the determining factors in the basis model in which behaviour is the factor to be defined. Attitude is determined, at least according to the much used Ajzen and Fishbein model, by a person' s expectations for the consequences of his/her behaviour and the value he/she places on those consequences. The (pressure from the) social environment can, in turn, also be explained. We also discuss the discrepancy between reasoned action - on which most explanatory models are based - and habitual behaviour. Finally, we describe a few other psychological factors which can influence human behaviour, such as the salience and the strength of the attitude. More than 600 car drivers over a 4-day period monitored their movements in a journal and filled in a comprehensive questionnaire. The response was 78 resp. 79%. By means of this journal the most significant dependent variable was determined: relative car use (this is the number of movements made by car divided by the total number of movements). This same variable is also measured specifically for commuter travel and shopping trips. We determined, by means of the questionnaire, attitude(s), pressure applied by the social environment and aspects of the opportunity structure. Habitual behaviour was measured by means of a question about the means of transport used most within a household in general and for seven specific categories of mobility behaviour. The investigators tested different elements other than those specified in the Ajzen-Fishbein model, while measures of salience, knowledge questions and questions relating to the image of car use were included in the questionnaire as well. Firstly we examine the aspects of the opportunity structure, such as distance, time difference, income, stage of family development, because these are situated the most to the left in this model of causal variables. Furthermore, they are the most concrete and recognizable. The variance in car use explained by these variables was generally narrow, less than 10%. The variable playing the most important role here is always the time taken to travel by
1186 car versus the time taken to travel by an alternative means of transportation. Distance does not matter (provided that the time difference is taken into account) nor does income level. An exception to this is individual habitual-behaviour; here the structural variables accounted for 21% of the variance. For all dependent variables, the explained variance increases considerably when the attitude is added to the model (thus habitual behaviour reached 33% of the explained variance). Subsequently adding perceived pressure from the social environment hardly leads to a rise in the explained variance. We mention in the context of the Ajzen-Fishbein model, which specifies that the product sum of other people's opinions and the tendency to be influenced by them (a common procedure but one statistically not accountable), can as well be replaced by the generalized tendency to be influenced by what others think and say. The results proved disappointing with respect to the environmental salience: when attitude is already included in the model, no extra explained variance is recorded. The measure of environmental knowledge did not prove to be a useful instrument. The measures of misperception yielded clearly no data in the predicted direction. The measures of image in relation to cars, bicycles and public transport revealed interesting results. We also tried to assess the strength of attitudes, in the way described by Fazio et al., in order to increase the prediction of mobility behaviour. The strength of attitudes was individually determined under laboratory conditions. This was performed on only 55 participants in the Hilversum survey because of its time-consuming nature. It turned out to be impossible to determine their strength due mainly to too many errors made by the respondents, far more in fact than reported by Fazio et al. As we found this unsatisfactory we repeated the experiment on a sample of 34 respondents with an academic background. Here too, so many errors were made that it proved impossible to determine the strength of attitudes. The Utrecht survey was focused on the relationship between personality variables and mobility behaviour, especially commuting behaviour. The personality variables were: (1) if the person feels alienated or comfortable in society, (2)if the person feels powerless, especially concerning environmental problems, (3) if the person distorts reality, especially the self-enhancing illusion of one's own behaviour towards the environment (this means he/she perceives his/her own behaviour as being more environment-friendly than that of others), and (4) if the person expects he/she will behave in a more environmentally acceptable manner as the result of either voluntary or enforced changes. For the elements of mobility behaviour to be predicted we used the following: (1) individual habits when choosing a means of transport, (2) the intention to reduce car use, and (3) the tendency towards cooperative behaviour. The variables used were measured by means of a written questionnaire. The respondents were given questions and statements, partly taken from other research and partly developed for this particular study. All respondents were inhabitants of the city of Utrecht and were chosen at random from the telephone directory. Only people were chosen who had access to a private car and worked outside the home. The sample was compiled so that half of the respondents always went to work by car and the other half went to work, sometimes by car, and sometimes by an alternative means of transport. In both categories half of the respondents lived 15 kilometres at the most from work, which meant that it was possible for
1187 them to use a bike as an alternative means of transport. The other half lived more than 15 kilometres from work which meant that the most important alternative means of travel for them was public transport. The questionnaire was completed by 329 respondents (the response was 82%). The scale for measuring alienation proved to be only fairly reliable. The scale for measuring powerlessness was sufficiently reliable. Self-enhancing illusions were found to be present and reasonably reliable to measure. Voluntary or coerced behaviour change did not prove to be a reliable variable to measure. The analysis was, therefore, carried out with a number of separate items of the scale. The habit of commuting by car and the intention to reduce this practice was measured by posing direct questions. The scale for measuring the tendency towards cooperative behaviour proved insufficiently reliable. We therefore decided to use the two most important questions of it relating to driving speed. The habit of going to work by car proved to be predictable from a high self-enhancing illusion. The intention to reduce commuting by car can be predicted, though less reliably, if a person expects this to be achieved on a voluntary basis rather than as a forced measure. When the respondents thought of restraint, they thought primarily in terms of the probability of receiving a fine for exceeding the speed limit. The tendency towards cooperative behaviour can be predicted if one expects this to be achieved on a voluntary basis rather than by force; this also applies, but to a lesser extent, to a high self-enhancing illusion. The question of voluntariness or restraint applies here in particular to the intention to reduce car use for the benefit of the environment and in order to limit the number of fines. This latter variable, however, is conceptually related to cooperation. No correlation was established between the variables of alienation and powerlessness (that were at least reasonably reliable to measure) and the three aspects of mobility behaviour. Relationships were also found between the dependent variables and other data obtained from the respondents, but these connections were weak. All in all, the results from the Utrecht survey were negative if one believes that by using these measuring instruments, a clear increase in the predictability in the choice of travel mode for commuting would appear. In as far as the personality characteristics were reasonably and reliably measured, no relationship is shown to exist between them and mobility behaviour. This, however, can be seen as supporting the already established fact determined in the Hilversum survey, that deeply ingrained individual habits play a significant role in mobility behaviour. Although this particular behaviour proved to be correlated to a general environment attitude, still it is not considered to be the result of a configuration of personality characteristics. This behaviour originates primarily from a number of external factors. Finally, several policy recommendations are described, partly related to the results from both inquiries.
1188 Reference
I.M. de Boer, D. van Kreveld, P.G. Swanborn (1994) De keuze van een vervoermiddel. Twee onderzoeken naar situationele en persoonlijke determinanten van verplaatsingsgedrag. (Choosing a means of transportation: Two inquiries into situational and personal determinants of moving behaviour. With a summary and policy recommendations in English.) Utrecht: Vakgroep Sociale en Organisatiepsychologie, University of Utrecht.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1189
Changing Attitudes and Behaviour by Means of Providing Information. A Study on Private Car Use G. Tertoolen and E.C.H. Verstraten Section of Social and Organizational Psychology, University of Utrecht, Heidelberglaan 1, 3584 CS Utrecht, The Netherlands
Abstract In a field experiment we attempted to stimulate car users to come to a more selective use of their vehicle by means of providing information and feedback about different negative consequences of their car use. Attitude change was observed but the experimental treatments did not lead to behavioural changes. Attempts to influence car use arouse psychological resistance. Therefore, effects opposite to those intended occurred. We discuss the possible implications of the results for policymaking.
Introduction One of the emerging objectives of the Dutch environmental policy is to modify behaviour on a voluntary basis. As the car is a means of transportation that is rather damaging for the natural environment, one of the objectives of the present environmental policy in The Netherlands is to restrict private car use. 'Using a pricing policy' and 'influencing behaviour via communication and education' play an important part in the policy strategy. The goal is to achieve a social situation within which there is room for considerable structural changes and whereby traffic participants make a conscious choice between the different means of transportation. Implicitly it is assumed that the effects of the various measures will reinforce each other. In this paper an investigation is summarized on how private car use can be reduced by applying influence techniques based on behavioural science. We also attempt to gain an insight into the psychological resistance that is aroused when these influence techniques are applied to private car use (see for an extensive report: (1) and (2)).
Research Design Our study focuses on two research topics: how car use can be restricted by (1) emphasizing the negative collective environmental consequences or by (2) emphasizing the individual financial consequences. In a field experiment (N= 350) we attempted to stimulate car users to come to a more selective use of their vehicle by means of the following manipulations: providing information about the negative consequences of car use, self-monitoring of own transport behaviour and giving feedback on the negative consequences of personal car use. By means of a random procedure the respondents were assigned to five different conditions: three experimental and two control conditions. In the experimental conditions the respondents received information: in condition (1) about the environmental effects of car use, in condition (2) about the individual financial consequences of car use, and in condition (3) about both types of consequences of car use. Subsequently the subjects registered their own transport behaviour for eight weeks. Every two
1190
weeks they received feedback from the researcher's assistant about the consequences of their car use in a person-to-person talk. The content of the feedback referred to the particular kind of information received in the respective experimental conditions. The other respondents participated in the experiment without receiving any information or feedback about driving behaviour from a researcher's assistant (the control conditions). We asked all respondents if they were prepared to use the car as little as possible during the study period. In the experimental conditions, the respondents who gave a positive reply were requested to restrict their car use and thereby making a commitment to a research assistant. In all conditions, at the beginning and at the end of the experimental period questionnaires were filled in to measure the various attitudes with regard to car use and the environment. Results
The target group was chosen in such a way that it consisted of regular car users. They turned out to be more or less "attached" to using their vehicle. Speed, comfort and independence are mentioned as the most important advantages of the car (see figure 1). The respondents state that when they travel, neither the environment nor the costs are of much interest to them. Apart from the car, drivers make frequent use only of the bicycle; public transport is used sporadically by them. Attitudes play an important role in the perceived possibilities of the reduction of car use. However, these attitudes (including those related to the environment) appear to play hardly any role in the actual (reported) car use. In our study attitude change was observed but the experimental treatments did not lead to behavioural changes; i.e. no decline of car use was observed.
RAPIDITY INDEPENDENCE COMFORT HEALTH
COSTS ENVIRONMENT SAFETY
/
OTHERS
mm 0
5
lO
15
20
25
30
:35
%
F i g u r e I. M o s t transportation
important according
a s p e c t s in r e l a t i o n to to the p a r t i c i p a n t s .
1191 Information about the environment leads to a greater general concern about the environment but does not convince people to alter the way they use their car in order to create a cleaner environment. Unexpectedly, information about costs leads to less worry, not only about the environmental effects, but also about the financial consequences of car use. A combined environmental and costs information programme leads in many cases to results similar to those obtained with the control conditions; as in the pilot study, the effects often neutralize each other. Attempts to influence car use arouse psychological resistance. Information about the environment leads to dissonance reduction by means of attitude change. As a result of dissonance-enhancing information about the environment, car users who drive a lot, yet have a positive attitude towards the environment, start thinking that the environment is less important and point out that others are more responsible for the problems than themselves. They also become irritated with the behaviour of fellow road-users. Information about financial costs of the respondents car use leads to resistance as well. Car users experience financial measures as a restriction of their individual freedom and as a result they have a dim view of both the measures and the authorities responsible for implementing them. In addition, when car users react to these measures in a contrary manner, effects opposite to those intended often occur. The respondents who committed themselves to drive less, did in fact not keep their agreement. Instead they tended afterwards to displace the responsibility for environmental problems on to others. Those respondents who received information about the environment had a greater appreciation of environmental policy after the research. They probably have a greater understanding of the necessity for environmental protection and of the problems that can arise from a good environmental policy. Those respondents who received financial information had (slightly) less appreciation of environmental policy. By emphasizing how costly a car actually is, a reduction in appreciation of the policy was achieved. After all, it is the authorities who are responsible for the high costs of running a car. Our study received the lowest rating from the respondents who received financial information only. Discussion In our research some of the respondents were approached personally during a fairly long period and confronted with information specific to the individual about the effects on the environment of their car use. Such an intensive and personalized procedure should have more effect than a superficial, generalised attempt to influence via mass media, which the authorities often make use of. The environmental information we directed at individuals led to an increased general environmental awareness, but respondents did not become more aware of their own part in pollution. This result gives little encouragement for the authorities' publicity campaigns about the environment. With those respondents who were relatively well environmentally aware before the research and who used the car more than they judged, the information they received actually caused a reduction in environmental awareness. When the discrepancy between attitude (environmental awareness) and behaviour (car use) is pointed out then apparently people are more likely to alter their attitude than their behaviour. Even if the message is formulated so that the receivers cannot avoid the fact that it relates to their own individual behaviour, it still would not automatically lead to a change in behaviour. Just as disonnance theory forecasts, if attitude
1192 and behaviour are not in line, it is more likely that attitudes will change (which can mean that the environment becomes less important, as is shown in our research). Individually relevant information about the costs of running a car lead in our research to a higher estimate of the individual's car costs. However, the awareness that one's own car use has negative consequences (both individual financial and collective environmental consequences) was reduced as a result of the cost information. This is interpreted as a form of psychological reactance. Car drivers turn against the measures and those who implement the measures more when an attempt is made to reduce car use with financial methods than when environmental information is provided. We assumed that this reactance arose from a motivational state directed towards the re-establishment of free behaviour. If the car user holds the uncompromising view that he or she has a right to pollute for the very reason that he or she pays excise duty, reactance could be conceived as a form of protest as well. If taken from such an 'exchange' point of view other policy measures to restrict pollution (such as an appeal to change behaviour) could provoke irritation, since the person has already paid a compensation for the damaging behaviour. This would seriously harm the policy strategy "the polluter pays", which has the intended purpose to stimulate people to pollute less. Our research shows that in some cases the results of combined cost and environmental information are comparable with the results of the control group, i.e. respondents who did not receive any information at all. The environmental policy, as mentioned above, assumes that the effects of various measures will reinforce each other. Although it cannot be concluded on the basis of our research results that this assumption is essentially wrong, care is probably justified. Summing up it can be stated as a result of our reserch that little progress can be expected by requesting individual drivers to voluntarily reduce car use. The method of influence used in our research was based on some of the most powerful instruments available in psychology (giving individually directed feedback, self-registration and commitment). Nevertheless, there was no change in behaviour. Drivers will not leave their cars of their own free will, the car is too strongly linked to feelings of independence and convenience for that to happen. There are several positive attitudes, which are linked to various individual advantages where car use is concerned, whereas there are only limited negative attitudes, which are linked to the disadvantages of car use. With such a balance the dissonance theory forecasts that the negative attitudes will change in the direction of the most prominent attitudes. The environment is seen to be important, but we can not talk of a central attitude that is so important that car use is equivalent to it. Mass media publicity campaigns do not seem able to develop such a central environmental attitude on a large scale. References 1 Tertoolen, G. (1994). Uit eigen beweging...?.t Een veldexperiment over be~nvloedingspogingen van het autogebruik en de daardoor opgeroepen psychologische weerstanden (With a summary in English). Dissertation: Universiteit Utrecht. 2 Tertoolen, G. (1995). Free to Move... ?.t A field experiment on attempts to influence private car use and the psychological resistance it evokes. A policy oriented report. Utrecht: University of Utrecht, VSOP.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1193
Energy and environmental issues as choosing elements for selecting options in the transportation sector aimed at reducing CO 2 emissions" an application to the italian case D. Barbieri, A. Nucara, M. Pietrafesa and G. Rizzo Istituto di Ingegneria Civile ed Energetica, Facolt/l di Ingegneria, Universit/t di Reggio Calabria Via E. Cuzzocrea, 48, 89128 Reggio Calabria, Italy Abstract
A new transportation demand model is described showing a simple data-base structure. It only requires input data referring to the fleets and the engine characteristics of the transportation park. The main characteristic of the model is its expertise in analysing the effects of different policies oriented to the reduction of the pollution levels and to the energy savings in the transportation sector. The results are provided both in terms of energy consumption and quantities of pollutant released to the environment. The effects of different transportation scenarios can be easily analysed using a simple "electronic sheet" way of representation.
1. INTRODUCTION In this paper we will present a new transport demand model, showing a simple data-base structure. It is founded on generally available information about the structure of the transportation park and on the size and type of the used engines and allows the obtaining of a desegregated view of the system and the evaluation of the pollutant emissions. The model, built-up for the whole Italian sector, is easily applicable, with minor modifications, to any country for which the required input data is available. It is also suitable for analysis regarding smaller areas, even to a regional scale. The main feature of the model is to provide results both in terms of energy consumption and quantities of pollutants released in the environment, as effects of the assumed scenarios. Starting from the "zero" scenario, referring to the system when all the requirements of the Italian government' s rules are accomplished, some alternative options are analysed.
2. DESCRIPTION OF THE MODEL The structure of the model is essentially founded on the following four points: 1. the transportation demand of the analysed region is organised with respect to three components: the object of the transportation (people or goods), its spatial domain (urban or
1194 non-urban) and the way with which the transportation is accomplished. The units employed are the passengers per kilometre and per year (pkm/y) or the tons of goods per kilometre moved per year (tpk/y); 2. a distributive model assigns the total energy consumption to each fuel source, by means of the specific values of the consumptions, available in the literature. The units here utilised are the kilotons equivalent of oil (ktoe); 3. for each pollutant component released by the transportation means, the emission factor, intended as the quantity of pollutant released for unitary energy (t/ktoe) is calculated. These parameters link the energy consumption with the CO2 emissions [ 1-2]; 4. the quantities of the CO2 emissions are then computed for each component of the transportation demand and for each energy source. A description of the main features and potentialities of the model can be offered by analysing one of the tables that constitute the way of representation of the results obtained. Fig. 1 contains an example of these print outs: along with the modal distribution of the transportation demand, subdivided into the urban and extra-urban components. The figure reports the specific and total energy consumption by each component. The share of the energy demand covered by the fuel sources is also shown. The yearly increasing rate, for each modal component of the whole transportation sector, represents the most important parameter in order to characterize the scenarios. In any case, the numerical quantities reported in Fig. 1 are to be considered only as an example, since the main purpose of this paper is the presentation of the structure and the potentialities of the model. The model also allows evaluations of the emissions of the main environmental pollutants linked to the transportation sector: carbon monoxide (CO), nitrogen oxides (NOx), volatile organic compounds (VOC), including hydrocarbons and suspended particles (SP), in terms of yearly tons of quantities given off. The types of pollutants chosen depend on the kind of 1992
(ESTIMATED YALUES) SnerW rote by sources
Forms of transportetlon
Yearly~ncreaslng Accomplished M o d a l Specific Energy rate (%) demand split consumption consumptson 1989-1992 Mrd pKm % gep/pKm Kte~
r
emi.ians
Energy sources Gas~lne D,asel
Jet fuel
(ktr Energy sources
LPG
El energy CNG
0.073
0.011
Gasoline
Diesel
5269
1892
Jet fuel
LPG
El energy CNG
Total
Passengers urban traffic Cars
.3.14
166.85
75.12
52
8676.17
0.674
Motorcycles
3.89
35.51
15.99
Buses
0.56
15.13
6.81
21
745.74
I
18
272.36
Underground
0.56
2.60
1.17
10
26.03
Other collective means
056
2.00
0.90
15
30.05
Urban passengers total
3.02
222.10
100.00
3.14
860,48
0.242
571
62
245
245 48
1 0.024
7793
6T2
672 1 0.976
1
i
9750.86
55
5941
2188
571
62
8814
6150
4229
864
90
11332
Extraurban pass traffic Cars
70 9'3
35
Motorcycles
3,89
17,73
3,49
22
390.01
Buses. trams
4.05
76.35
15.02
15
1145.30
krcralts
8.51
7.32
1.44
313
2291.44
8
Railways
1.41
46,35
9,12
Extraurban pass total
321
508,23
100,00
12616,85
0,541
0,872
0,076
0.011
1
351 0.947
0.053
351 977
9"E
2065
370,77
0,25
16814,37
0,75
84 6501
5290
517 2065
864
517
600 90
15326
Fre4ght traff~c T rue,ks 451on
3,03
9,88
g0
Trucks > 5 ton
3,0,3
20.57
8,57
82
1686,92
I
1520
1520
Trucks. long vehicles
3.03
138.61
23,71
57 74
35
4851,46
1
4371
4"~71
Ships
0,95
34,98
14.57
6
209,87
I
189
9
199,81
0.25
Railway
1,39
22,20
9,25
Total freight traffic
2,56
240,08
100,00
Total
.~,00
970,40
2133,99
0,088
0,912
9082,05
Figure 1. Example of the output structure of the model.
169
1754
1923
189
45
279
324
169
7879
279
8326
1L~11
15306
2065
1435
898
151
32466
1195 analysis required [4]. Energy consumptions and pollutant emissions represent the selecting criteria in order to judge the effects produced by an assigned modal and structural distribution [5-6].
3. APPLICATION TO SOME TRANSPORTATION SCENARIOS Three scenarios have been assumed here in order to show the features of the method. 9 The "zero" scenario. This scenario has been identified as "zero" because it is considered the
reference point for the whole analysis [7]. It is characterised by the absence of specific interventions and therefore it appears as simply driven by the demand of mobility, for which an incrasing tendency is supposed. 9 The "modal split" scenario. This option takes into account the effects of some interventions that modify the modal split of transportation. From 1995 until 2020 the transportation demand is supposed to shift toward the public means, with rates of 20% for the urban passenger, 25% for the extra-urban passenger and 20% for the freight movement. Moreover, an increase of 15% in the use of bicycles in the urban context is also supposed. 9 The "car p o o l i n g " scenario. Within this alternative we suppose that the italian occupancy coefficient rises from 1.3 to 2.0 persons per car in the urban context and from 1.7 to 2.5 persons per car in the extra-urban context by the year 2020. The model provides a grafic representation of the compared effects of the assumed scenarios, but also details numeric results, within each scenario under analysis, referring to the other environmental pollutants. An analysis of the results provided by each of the previous mentioned scenarios is beyond the purposes of the present paper. Fig. 2 depicts the estimated trends of CO2 emissions of the whole Italian transportation sector from 1995 until 2020. Curves refer to the different scenarios. Table 1 contains the percentage of reduction for the considered emissions and for the energy consumption recovered by means of the "car pooling" scenario referred to the base case, that is the "zero" scenario.
~-, 55 50-II Zero Scenario o 45-r,r r~ .,..~
[] Modal Split Scenario 40--
D Car Pooling Scenario
r
9
~
35)f
.."
30 1992
,
~
I
l
I
I
I
1996
2000
2004
2008
2012
2016
2020 years
Figure 2. Estimated trends of CO 2 emissions for the whole italian transportation sector.
1196
Table 1 Reductions percentages obtained with the car pooling scenario with respect to the zero one. CO2 CO NOx VOC PS Final energy consumptions Urban
30.9
26.2
29.8
Extra-urban Total
19.3
29.8
31.2
23.1
21.3
19.4
23.3
19.6
18.9
26.9
23.5
8.7
17.1
17.5
19.7
4. CONCLUSIONS As it is possible to note, even within the summary here presented, the reliability of the approach strongly relies on the accuracy of the available data. Data on the car and truck fleet and on the freight movements are, as matter of fact, capable of affecting in a remarkable way the results obtained. This data, in fact, is employed as multiplier parameters by the algorithm of the model. On the other hand, the main assumptions of the methodology, especially concerning the evaluation of the emission factors, could introduce some simplifying features within the frame of approach. These considerations suggest a need for further attention when analysing the transportation sector and the complex links between energy consumption and environmental emissions. But the "electronic sheet" structure of the model and its capability of investigating different phenomena, such as pollutant releases and fuel use, make it a suitable tool in order to explore the effects of different scenarios referring to the policies to be selected in the transportation sector.
5. REFERENCES 1
2
3 4 5
6 7
S. Unnasch, C.B. Moyer, D.D. Lowell and M.D. Jackson, , Comparing the Impact of Different Transportation Fuels on the Greenhouse Effect, Acurex Corporation Environmental Systems Division, California, Mountain View, (1989). M.A. De Luchi, Emissions of Greenhouse Gases from the Use of Transportation Fuels and Electricity, Center for Transportation Research, Argonne National Laboratory, United States Department of Energy, Illinois, Argonne, (1991). Ministero dei Trasporti, Conto Nazionale dei Trasporti, Italy, Roma, (1992). G. Rizzo, Trasporto su Strada di Merci e Persone: Aspetti Tecnologici ed Ambientali, La Nuova Ecologia, Vol. 80, Italy, Milano, (1990). M. Fergusson and H. Claire, Atmosferic Emissions from the Use of Transport in the United Kingdom, Volume two: The Effect of Alternative Transport Policies, Earth Resources Research, United Kingdom, London, (1990). Peeters, The Netherlands Travelling Clean, Netherlands Agency for Energy and Environment, Netherlands, Utrecht, (1989). W. Leontieff and P. Costa, I1 Trasporto Merci e l'Economia Italiana: Scenari di Interazione al 2000 e al 2015, Piano Generale dei Trasporti, Italy, Roma, (1988).
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1197
C O M P A R A T I V E ANALYSIS OF OPTIONS FOR SUSTAINABLE T R A N S P O R T AND T R A F F I C SYSTEMS IN THE 21st CENTURY
Peter Nijkamp
Sytze Rienstra
Jaap Vleugel
Economic and Social Institute, Free University, De Boelelaan 1105, 1081 HV Amsterdam, The Netherlands. In association with: Marina van Geenhuizen Edith van der Heijden/ Ton Rooijers Richard Smokers Johan Visser
UCL, London VSC, Haren ECN, Petten OTB, Delft.
Abstract In this project on future sustainable transport altematives a two-step search process has been followed. First an analysis of critical success and failure factors of new technological options in passenger transport is made. These factors are found in the spatial, institutional, economic and social/psychological environment of the transport system. Next, systematically structured and expert based scenarios are constructed in order to achieve a sustainable transport system in the year 2030 in which possible, expected and desired developments in the distinct fields are analyzed. Finally some policy conclusions are drawn.
1.
INTRODUCTION
The current trends in almost all fields show a continuing shift in the modal split towards individual modes and a rising mobility rate. Therefore, the externalities (social costs) caused by transport are still rising, which makes it necessary to bend these trends in order to achieve a more sustainable transport system. In this project we have investigated which technical options are developed (or are being developed nowadays), which may reduce these externalities. The emphasis is here put on the resulting reduction of CO2-emissions and other greenhouse gases, since this provides a direct link to sustainable development. In the first phase (state o f the art) of the project, success and failure factors for the introduction of new technological options have been identified. In the second phase (scenarios for a sustainable transport system), several scenarios have been constructed in which these options have been filled in for the transport system. Finally, an assessment of policy choices has been made which may influence the future of transport.
1198 2.
RESEARCH STRATEGY
In the first phase, an extensive literature search has been carried out supplemented with an international workshop in order to identify the success and failure factors of several new options (new fuels, electric car, people mover, subterranean transport, telematics, HST, maglev and shuttles through vacuum tunnels), which may contribute to the reduction of externalities when they would replace current modes. In the second phase of the project reference scenarios, which describe contrasting future developments in the field of transport have been constructed by using the recently developed spider model. In this stage also a questionnaire has been sent to Dutch transport experts. The results of this survey have been presented to a second international workshop organized towards the end of the project. With this information 'expected' and 'desired' scenarios have been constructed, based on these expert opinions. This has also allowed us to asses the resulting environmental implications.
3.
PHASE 1" STATE OF THE ART
There are many critical success and failure factors which influence new technical options. The most important of them have been summarized in Figure 1. SUCCESS
ECONOMIC
SPATIAL
"
LEVEL OF MOBIUTY MODAL SPLIT EXTERNAL EFFECTS PER OPTION
Spatial organisation of living and working areas
"
1
Government support P r e s s u r e of n a t i o n a l I n d u s t r y
IN~IlITUTIONAL
SOCIAl./ PSYCHOLOGICAL
TECHNICAL
AND FAILURE FACTORS
Competitiveness Profitability Financing
TOTAL EXTERNAL EFFECTS OF TRANSPORT
Acceptance (society) Adoption (individual)
-
Direction of R & D Environmental criteria Technical Inertia
I
SUSTAINABLE MOBILITY
I
Figure 1. Success and failure factors of new technological options
It appears that especially modes and options which are compatible with existing modes have the best chance to succeed since they may use (temporarily) existing infrastructure and may easily be incorporated in prevailing institutions and existing transport systems. Therefore, the High Speed Train and improvements of the current car - and to a lesser extent also new fuels (and electricity) in cars - have the best chance of being introduced and accepted. In general however, a principal choice has to be made between an emphasis on individual or collective modes, since both modes require an entirely different spatial organisation, technical development and institutional environment. These also have a significant influence on important economic and social/psychological factors. For a more detailed analysis of these phenomenon and their solution strategies we refer to Nijkamp, Rienstra and Vleugel (1994).
1199
4.
PHASE 2: S C E N A R I O S F O R 2030
4.1
The methodology of the spider model Based on the phase 1 study and various scenario experiments developed by others, we have identified eight main factors, which influence the future transport system; these are to be found in four distinct scientific fields. These factors are presented in the so called spider model (see Figure 2). ~ c o m p a c t l ......
speelalisation _ "~." ~ ~ and concentratigjrl~ ; ~
" "
chalns~and'zones~\ in d i v i d ~ . i a ~ ~ soci,I
. - ~
I r:7
|
/
/
I
/
".. -
publlc
management
9
I
rolm~gement
profltTIble I~ansport 9 I
in~p~ subsidlsed transport
coor centralisation
Figure 3. The expert based scenarios Legend ....... expected developments ........ desired developments Scenarios can be constructed by connecting characteristic points on the distinct axes. In this way in principle thousands of scenarios may be constructed. On the basis of assumptions on the developments in the several axes the resulting transport system can be identified and described. We constructed two scenarios which form the inner and outer circle of the spider and are used to analyze the scenarios designed by means of expert opinions; therefore these are called reference scenarios. The regulatory scenario forms the outer circle of the spider. A compact and concentrated spatial development is combined with an emphasis on equity and regulatory measures. In this scenario a transport system occurs which is largely based on collective modes, while individual modes largely disappear. The market scenario on the other hand is formed by the inner circle. In this scenario diffuse spatial developments have been combined with market-oriented measures. In this way a transport system occurs in which individual modes are dominant.
4.2
The expected and desired scenario Both an expected and desired scenario have been constructed next by means of the questionnaire and the information gathered from the second international workshop. In the expected scenario (see Figure 2) it is expected that current trends will largely continue. Therefore, mainly improved versions of the current car will be dominant, while also measures which are common nowadays (reducing parking places, raising parking tariffs and fuel prices) will be introduced to a larger extent. Also road pricing may be introduced to some extent. It is expected that also other main trends in
1200 society and economy will largely continue. In the desired scenario an entirely other transport system is found. Measures which will be introduced to make the car more unattractive are introduced at a much larger scale, while collective modes are supported much more than expected. Also many trends in society are reversed in order to favour such a transport system. It is clear that the expected scenario will only lead to more sustainability, if a much larger improvement of the current car will occur than is now expected by technical experts. In the desired scenario much more sustainability may be achieved. In this scenario however, many more changes and measures are necessary, which will have a much larger impact on the life of individuals and the society at large. It is noteworthy that the latter phase generated many new insights into the feasibility and desirability of transport systems alternatives, in particular from the viewpoint of global environmental changes. More details can be found in a forthcoming publication (see Nijkamp, Rienstra and Vleugel, 1995).
5.
CONCLUSIONS
The current trends in transport are not expected to lead to a more sustainable transport system. Therefore, a change in the behaviour of individuals and a stricter government policy seem to be necessary. Several consistent and effective policy choices have to be made. The most important concern is the one between an emphasis on an individual or a collective transport system, because both systems require an entirely different policy in many fields, which have a profound impact on many other aspects in society, like individual lifestyles, the level of equity and individualisation, the choice of housing locations etc. Other issues related to this choice are the necessary reduction of the mobility growth, the investments in transport infrastructure, the direction of technical development, the way in which transport may be regulated etc. From the expert opinions it may be concluded that government policy alone may not be sufficient for achieving a more sustainable transport system, for example in the expected scenario the CO2-emissions will probably not be reduced in a sufficient way. The policy solution chosen by experts appears to favour a more collective transport system (as is shown in the desired scenario). It may be concluded that the road towards a (more) sustainable transport system will be very hard, but that with sufficient behaviourial changes and other changes in society such a (more) sustainable transport system seems to be socially and technically feasible.
REFERENCES -Nijkamp, P., S.A. Rienstra and J.M. Vleugel, 1994, Comparative analysis of options for sustainable transport and traffic systems in the 21st century, phase 1: state of the art, report of the Dutch National Research Programme on Global Air pollution and Climate Change, theme E: Sustainable solutions (policy research), ESI, Free University, Amsterdam. -Nijkamp, P., S.A. Rienstra and J.M. Vleugel, 1995, Comparative analysis of options
for sustainable transport and traffic systems in the 21st century, phase 2: Scenarios for a sustainable transport system in 2030, idem.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1201
ASSESSMENT REPORT ON SUBTHEME "CULTURE, CONSUMPTION AND LIFESTYLES IN RELATION TO SUSTAINABLE DEVELOPMENT" C.A.J. Vlek Department of Psychology, University of Groningen Grote Kruisstraat 2/1 9712 TS Groningen The Netherlands
With contributions by: B. Breemhaar, P. Ester
KUB, Catholic University Brabant, Tilburg
C. Wilke, M.A. Mentzel
RUL, University of Leiden
W. Biesiot, E.M. Kamminga, H.C. Moll, G. Slotegraaf, A.J.M. Schoot Uiterkamp, A.W.L. van der Veen, H. Wilting K. Blok, C. Vringer R.M. van Aarts, J. Goudsblom, F. Spier, C. Schmidt C.J.H. Midden, W.A. van Gool, A.L. Meijnders
RUG, University of Groningen
RUU, University of Utrecht UvA, University of Amsterdam
TUE, Eindhoven University of Technology
1202
Contents Abstract 1.
Introduction
2.
The e n v i r o n m e n t a l b u r d e n of h o u s e h o l d c o n s u m p t i o n
3.
The p r o d u c t i o n - c o n s u m p t i o n cycle
4.
Population, affluence and technology
5.
E c o n o m i s a t i o n a n d ecologisation
6.
Rise of t h e e n v i r o n m e n t a l p r o t e c t i o n m o v e m e n t
7.
S t a t u s seeking t h r o u g h m o d e r a t i o n
8.
CO2 emissions r e d u c t i o n via lifestyle c h a n g e s
9.
Low-energy, low-CO2 emissions scenarios
10. Lifestyles a n d domestic e n e r g y c o n s u m p t i o n 11. F e a r - a r o u s a l a n d a r g u m e n t a t i o n in risk c o m m u n i c a t i o n 12. Social welfare an e n v i r o n m e n t a l quality 13. T o w a r d s s u s t a i n a b l e h o u s e h o l d m e t a b o l i s m 14. G e n e r a l conclusions, suggestions a n d r e c o m m e n d a t i o n s 14.1 'Economisation' and environmental exploitation 14.2 Some policy recommendations 14.3 Lifestyles and behaviour change strategies 14.4 International research efforts 14.5 Further NRP intentions 14.6 Religions on consumption 15. R e f e r e n c e s
ABSTRACT Many people believe that 'sustainable solutions' to global air pollution and climate change should include significant changes in human consumption and lifestyles. Under this heading six different NRP projects have been conducted. This chapter gives a review and assessment of these projects, supplemented with discussions of related research. The paper starts with a general statement of the environmental
1203 problem of household metabolism as a key component of the socio-economic production- consumption cycle. It summarises and comments upon nine different (sub)projects. And it ends with general observations, conclusions and suggestions for r e s e a r c h and policy in relation to s u s t a i n a b l e c o n s u m p t i o n p a t t e r n s . C o n c e p t u a l problems, m u l t i d i s c i p l i n a r y perspectives and i n t e r n a t i o n a l implications are given special consideration. 1.
INTRODUCTION
Household consumption is at the beginning and at the end of industrial production. H u m a n needs and desires, habits and decisions, norms and rights in modern society materialise in an enormous 'household metabolism'. This involves the transformation, sooner or later, of many different kinds of energy, materials and products into various kinds of positive fulfilment, of course, b u t also in environment-polluting kinds of gaseous, liquid or solid waste. Through the direct and indirect use of fossil-fuel energy for household activities, including transport, and through the exhaust gases from landfills and waste incinerators, households contribute significantly to global air pollution and the risks of climate change. In this paper, brief reviews and commentaries are given of six different projects on 'culture, consumption and lifestyles', as conducted during Phase 1 (1990-1994) of the Dutch National Research Programme on Global Air Pollution and Climate Change (NRP). These will be supplemented with a few related but otherwise funded studies on household consumption vis-a-vis environmental resource use. Table 1.1 offers an overview of project codes, titles and principal investigators. For more extensive project descriptions and for full references to complete project reports, the reader is referred to the relevant project summaries elsewhere in this volume. Table 1.1 List of projects in the NRP subtheme "Culture, Consumption and Lifestyles" Title
Project leader
Number
Conditions for a moral code of moderation
J. Goudsblom
851038
Reduction of CO2 emissions by lifestyle changes
W. Biesiot/ H.C. Moll
852086
Analysis of the social significance of long-term lowenergy/low CO2 scenarios for The Netherlands
W. Biesiot/ H.C. Moll
852085
Toward a sustainable lifestyles
P. Ester/ C.J.M. Midden
853119
Cognitive vs emotion oriented information on sound C.J.M. Midden Environmental behaviour
852093
1204
Non-NRP-projects
Social welfare and environmental quality
M.A. Mentzel
Sustainable household metabolism
A.J.M. Schoot Uiterkamp
This section will be concluded by a general discussion of household consumption in view of sustainable development, followed by conclusions and suggestions for research and policy regarding household consumption and consumer lifestyles. 2.
T H E E N V I R O N M E N T A L B U R D E N OF H O U S E H O L D C O N S U M P T I O N
The throughput of energy, materials and products in households of varying size and style has grown impressively during the last fifty years. Some pertinent figures for The N e t h e r l a n d s are as follows. During the period of 1950-1990 the Dutch population has increased from 10 to 15 million inhabitants. In the same period, the percentage of Dutch land area (a total of about 34.000 square kilometres) used for buildings, roads and recreational facilities, increased from 8.4 to 16.1. Around 1950 there existed about 2 million household dwellings; this number had risen to 6 million in 1992. The average annual income, corrected for inflation, of heads of households in 1990 was twice as high as in 1950. Between 1965 and 1992 water consumption in Dutch households has increased from 100 to 135 liters per person per day; today about twice as much water is being used for bathing and showering and for textile washing t h a n 30 years ago. The ownership and use of motor vehicles - especially p a s s e n g e r cars, but also vans and lorries - has grown very strongly since the 1950s. In 1960 some 670,000 four-wheeled motor vehicles populated the Dutch roads and streets. In 1980 there were about 4 million and in 1990 about 6 million motor vehicles (Vlek et al., 1993). The Dutch fleet of motor vehicles is expected to approximate the figure of 10 million in 2010, an average of about 300 motor vehicles per square kilometre of land area. The number of airplane starts and landings at Schiphol Amsterdam Airport rose from about 90,000 in 1960 to some 235,000 in 1990. The Schiphol authorities expect (and stimulate) t h a t between 1990 and 2010 the n u m b e r of 'passenger movements' will triple from 16 to 50 million annually. From an international perspective it may suffice to quote Corson (1994) who - in a recent special issue of F u t u r e s - outlined various strategies for a sustainable future. The author starts his paper by describing current 'unsustainable trends': "Between 1950 and 1990, the world's h u m a n population more than doubled (from 2.6 billion to 5.3 billion), domestic livestock population grew 1.8-fold (from 2.3 billion to 4.1 billion), grain consumption rose 2.6-fold, water use nearly tripled, fish consumption grew 4.4-fold, and energy use quintupled. Over the same period, global consumption of wood and copper roughly doubled; steel production quadrupled;
1205 economic output nearly quintupled; industrial production grew sevenfold; aluminium output and the use of chemical fertilizers increased roughly 10-fold; world production of organic chemicals, major sources of air and water pollution, rose 20-fold; and global air travel, which causes significant atmospheric pollution, soared nearly 70-fold. On average, resource use per person nearly tripled between 1950 and 1990. This growth, coupled with a doubling of human population, resulted in roughly a sixfold increase in human impact on the global environment during the four decades. H u m a n activity is now altering the Earth's basic life-support systems and cycles, including the atmospheric system and the carbon, nitrogen, sulphur, biologic and hydrologic cycles" (Corson 1994, p. 206-207). Taking this together, we may conclude that the households sector since 1950 has developed as an environment-burdening consumer of energy, water and materials, of meat, fish and agricultural produce, and of motorised transport and land area, and it has become a major producer of diverse kinds of waste. Household metabolism, therefore, is an important focus for scientific research and for government policies aimed at reducing global air pollution and the risks of climate change. 3.
T H E P R O D U C T I O N - C O N S U M P T I O N CYCLE
To understand household metabolism, its driving forces and its potential for change toward sustainable conl~umption patterns, it is necessary to appreciate the interwovenness of house!lold consumption and industrial production. Figure 3.1 r e p r e s e n t s w h a t m a y be called the p r o d u c t i o n - c o n s u m p t i o n cycle, as institutionalised in a social, i.e., government-regulated market economy. The figure reflects the siJnple truth that consumers and producers need each other for different reasons, and that both parties need some government regulation for which the government :n turn needs them, again for different reasons. The relationships among consumers, producers and government are expressed in flows of money, products, labour, taxes and subsidies. Main system functions for consumers are feeding, clothing, housing, education and recreation. Major functions for producers are energy provision, industrial production, agriculture and stock-breeding, product distribution and commerce. Inputs from outside the socio-economic system are formed by various environmental resources such as energy, raw materials and land area. External outputs or derivatives occur in the form of various kinds of waste materials as well as transport.
1206
-" " "
/
land - ~ ~ ~ _ . ~ .
mobility~l
~
waste
~.
~
___
" " ,, ,,,
~
",%.
,
"~
,, "
energy ~
income
.~/
,,~,
~
7 // / / " / .
.
.
. '
......... " > \\\
i
9
I CONSUMER I . . . . . . . . . -, I
[ I
-'.,""'"'" ~~ housing feeding clothing education
\
N ~
\
i
.// products ~'- / ~ ~ ~ turnover----
~
~
I ~ /
/
land
raw materials
PRODUCER I : ~ " " ..... ]~--~-"---"~* waste
// "~ ~
energy
,,~l
/
/
recreation
"'''''.~ utilities industry agriculture transport
transport
commerce/ services
Figure 3.1 Social m a r k e t economy 4.
POPULATION, AFFLUENCE AND TECHNOLOGY
An other w a y of positioning household consumption is to follow E h r l i c h and Holdren's (Ehrlich et al., 1971) formula for e s t i m a t i n g the total e n v i r o n m e n t a l impact from h u m a n activities: I = P x A x T, or: environmental impact (I) equals the product of population size (P), the degree of affluence (A) per person, and the e n v i r o n m e n t a l d a m a g e from the technology used (T) to produce one u n i t of affluence. The formula implies t h a t there are three different fronts on which the battle for sustainable development is to be fought. The formula also reveals the s u b s t i t u t a b i l i t y of one component by the other. To illustrate, total e n v i r o n m e n t a l i m p a c t m i g h t r e m a i n constant under considerable population growth, as long as average personal affluence and/or the technological impact per unit of affluence is/are p r o p o r t i o n a l l y reduced. Also, while total e n v i r o n m e n t a l i m p a c t s t a y s c o n s t a n t , the degree of affluence per person m a y well increase significantly, provided t h a t the n u m b e r of people and/or technological impact per unit affluence is/are proportionally reduced. The 'IPAT-formula' enables one to explain and predict and to eventually m a n a g e the size and seriousness of environmental impact, for different geographic regions or countries of the world, to d e t e r m i n e the most i m p o r t a n t i m p a c t g r o w t h factor(s) a n d to d r a w conclusions on o p t i m a l environmental m a n a g e m e n t policies.
1207 Recently, Goodland, Daly and Kellenberg (1994) have systematically examined the potential for change in the three areas covered by the IPAT-formula: (1) limiting population growth, (2) limiting affluence and consumption growth, and (3) reducing the environmental impact of production and consumption technology. Like Corson (1994), these authors generally conclude on a number of policy priorities, which are different in character for high-income and low-income nations of the world. For example, high-income nations are advised to work on "transforming the culture of consumerism (..) into an ethics of sufficiency and environmental sustainability", and on "internalizing environmental costs in energy prices and accelerating the transition to renewable energy sources" (Goodman et al., 1994, p. 153). In contrast, the authors advise low-income nations to give priority to: "accelerating the transition towards population stability (..), supporting technologies which provide increased employment opportunities for unemployed and underemployed individuals (..), and improving efforts towards poverty alleviation .." (Goodman et al., 1994, p. 154). One general conclusion by Goodland et al. is: "Technological change and population stabilization cannot suffice to move the world towards an environmentally sustainable future. Instead, a reduction in per capita consumption in high-income nations and a decrease in environmental throughput are required" (Goodman et al., 1994, p. 154). Whether and how this could be accomplished is a major question underlying this review of recent Dutch research on household metabolism. Let us now look at the separate projects. 5.
ECONOMISATION AND ECOLOGISATION
Three interrelated NRP-funded projects on consumption and lifestyles (NRP project no. 851038) have been carried out at the University of Amsterdam. Together with sociologist-psychologist Goudsblom, researchers Schmidt, Spier and Aarts have looked into the possibilities for developing a 'morality of increasing moderation' among household consumers. All three projects started from the premise that environmental degradation, including global air pollution and climate change, is a problem of h u m a n civilisation (see also Thoenes, 1990 and Vermeersch, 1990), which strongly resides in ecologically unbridled economic growth. Reviewing available literature, Schmidt (NRP project no. 851038-1) - interacting with Goudsblom - critically analyses and describes the historical process of 'economisation' in western society. In his view, this has come to replace more traditional, often military ways to acquire, maintain and expand the wealth of nations. Through strong inherent growth tendencies economisation has also led to a decline of 'ecological regimes' for production and consumption. With this less violent, thus 'more civilised' economisation also came the translation of 'everything worthwhile' into equivalent monetary values. By implication, non-valuable goods and services tended to be seen as 'worthless'. Effective economisation required greater military self-control from powerful individuals, and it gave greater technical control over nature. Economisation also led to a significant increase in the division of labour, and to the development of industrialised cities. Thus, many people came to live at a distance from the natural environment and they lost sight of their natural living conditions. And gradually, with increasing affluence came a relaxation of traditional norms of frugality and thrift.
1208 The solution to 'economisation', according to Schmidt, lies in a g r a d u a l 'ecologisation' of society. This would amount to developing a strong environmental a w a r e n e s s and a t t e m p t i n g to keep h u m a n activities and h u m a n populations within the limits of the earth's carrying capacity. Such a 'next step in h u m a n civilisation' would require new forms of self-control and it would demand new kinds of intelligent control over the environmental effects of human action. Comments on Schmidt's project
Agriculturation, militarisation, economisation and ecologisation might well be distinguishable periods in h u m a n civilisation. It seems important to note t h a t these developmental stages occur(red) in correlation with a growing and spreading h u m a n population and with an ever more intensive exploitation of the earth's surface. Was economisation actually driven by war-weariness, or were military efforts only too supportive of economic expansion? Or was it population growth and increasing agricultural uncertainties that have led to economic industrialisation? Other socio-cultural analysts of environmental degradation point at the role of C h r i s t i a n i t y (White, 1967) in which m a n is positioned above, not among other living beings. Or they indicate the fundamental roles of science, technology and capital (Vermeersch, 1990) in creating a technologically violent 'here and now' consumer culture. In the latter, privacy, power and freedom are key values for the individual (Thoenes, 1990). Environmental economists (Opschoor, 1989), point at the conflict between short-term individual and long-term collective interests, which is inherent in a free-market economy; through the accumulation of the external costs of n u m e r o u s individual activities, society as a whole is burdened with collective costs and risks for which individual actors tend to feel little responsibility. Schmidt's analysis is i m p o r t a n t in so far as it draws our a t t e n t i o n to the drawbacks and the possible excesses of economisation as a civilisation stage. As a socio-cultural analysis of 'the environmental question', however, it needs to be supplemented by views from other social-science disciplines. 6.
RISE OF THE ENVIRONMENTAL PROTECTION MOVEMENT
What societal powers exist that are or could be pushing the needed 'ecologisation' deemed necessary by Schmidt? Spier (NRP project no. 851038-2), supported by Goudsblom, has explored the rise and the effectiveness of the environmental protection movement (EPM) in The Netherlands. Started in the early twentieth century as an elite form of appreciation and care for nature, the Dutch EPM now consists of various professional organisations of monitors, publicists, advisors and activists, working towards the protection of natural areas and warning against careless industrial pollution, the unlimited growth of motorised traffic and the inconsiderate use of open space for siting industries, house-building and/or road construction. EPM organisations, however, differ in effectiveness. Spier makes a distinction between the more traditional societies for the protection of birds and for the maintenance of natural monuments on the one hand, and the newer, more critically operating organisations for general environmental protection on the other. He notes that the, more successful, traditional organisations have always a t t e m p t e d to realise their relatively modest ambitions in a positive and prudent way, taking care to appeal to the general public and to be acceptable for policy makers. The more general EPM organisations, however, are pursuing their more
1209 ambitious programmes for social behaviour change and societal restructuring in a less discreet and more activistic m a n n e r , while e m p h a s i s i n g negative developments in environmental conditions. This may explain why many people don't like them and refuse to heed their advice. Spier concludes that "sounding the alarm is a necessary component of efforts to stimulate ecological awareness, (but) positively phrased campaigns to stimulate specific forms of moderation will in my view prove to be more successful than alarmist approaches, and should clearly be kept separated. In addition, the ability to exercise influence at the highest level of decision making (..) may be very helpful to spread forms of ecological moderation". C o m m e n t s o n Spier's p r o j e c t For environmental policy to be effective, there should be sufficient social understanding of the problem and acceptance of policy measures. This is often dependent upon the activities of pioneering organisations. Spier's project demonstrates that their public following and their political influence significantly depend - of course - on the relative attractiveness or painfulness of their message, but also upon the style and the way in which this message is being distributed. For environmental protection organisations, being stereotyped as a noisy minority preaching unachievable ideals, would be lethal. Alternatively, the stigma of the well-to-do land owner who wants to preserve natural peace and quiet for himself and his friends, would be similarly killing. The EPM would be well advised to think hard about the fundamental conditions that should be fulfilled for their messages to get t hr o u g h and to be transformed into actual policy measures and social behaviour changes. Using principles of commercial product marketing, taking account of basic social-behavioural mechanisms, and continually working on image-building might prove to be effective. The government, as the guardian of public environmental qualities, could assist here: by supporting diagnostic environmental research, by sharing the use of its distribution channels, and by being clear and consistent in its environmental policy goals. Currently, both the traditional and the 'modern' EPM organisations are dissatisfied with Dutch policy making - and more so with policy implementation - concerning environmental qualities and conditions. They agree with one a n o t h e r t h a t long-term e n v i r o n m e n t a l protection also is an economic necessity and t h a t various short-term economic priorities reflect the short-sightedness of government departments and industrial organisations. 7.
STATUS SEEKING THROUGH MODERATION
Obviously, the basic attitude and 'lifestyle' of the environmental protection movement is considerate for the environment, moderate in consumption, reserved towards industrial expansion and restrained in the use of motor vehicles. If such moderation in household consumption would appear to be necessary on any large scale, in order to limit greenhouse gas emissions and diminish the risks of climate change, how could it be accomplished? In her study on consumption and social stratification, Aarts in collaboration with Goudsblom (NRP project no. 851038-3) lists four general approaches. Economic policy, legislation and enforcement, and environmental information feedback combined with moral appeals are three of
1210 them. The fourth one involves encouraging spontaneous shifts in social behaviour through increasing the status and prestige derived from voluntary self-restraint in consumption. Aarts's project is focused on the latter strategy. Via in-depth interviews and questionnaire surveys she has investigated if and how moderate and self-controlled lifestyles among higher-income, higher-status groups (who have a choice) may 'trickle down' into environmentally less aware and more consumptive segments of society. "Affluence is an i m p o r t a n t condition and point of d e p a r t u r e for moderation," says Aarts, "as it liberates people from the struggle for subsistence and increases their opportunities to plan ahead." Moderation may enhance social prestige, because it reveals the 'distinctive' ability to restrain oneself, and it therefore may also be sought after by other population groups. Thus moderation and the 'new' social status it provides may trickle down into society as a whole and ultimately lead to the strong collective awareness of environmental processes and effects, which is needed to achieve an 'ecologisation' of society. The s t u d y by Aarts reveals t h a t among the higher-income groups, the better-educated are more sensitive to the environmental effects of consumption. They eat healthier food and less meat, buy more bottled instead of tetra-packed milk, save more fossil-fuel energy, and are more aversive to producing waste, than the average lesser-educated and/or lower-income group member. However, better-educated higher-income group members show little restraint in cultural and holiday travelling (often by car or airplane), which they see as being socially prestigious par excellence and as therefore putting the prestige from self-restraint into its shadow. C o m m e n t s o n Aarts's p r o j e c t This research capitalises on the importance of social comparison processes, in which individuals continually try to identify themselves among others, in terms of income, talent, education, power, and status. A basic principle of social comparison theory (Suls et al., 1997 and Masters et al., 1987) is t h a t people feel most comfortable when they are just a little better (off) - in various respects - than other people in their social environment. Thus, striving to reach a position of 'being slightly better (off)' is an almost daily activity for most everyone. And if 'being better (off)' is largely defined in terms of material consumption and possessions (Dittmar, 1992), then we have a powerful explanation for permanent consumption growth. Aart's thesis is that a re-definition of 'being better (off)' is necessary and seems feasible, w h e r e b y m a t e r i a l c o n s u m p t i o n m a y be reduced for the improvement of environmental quality.
One r a t h e r tricky and perhaps depressing property of the social processes described by Aarts is that new 'distinctive' and 'prestigious' forms of (moderate) consumption among high-status groups arise only when the old, to-be-rejected consumption behaviours are experienced as 'too common' and therefore no longer status-giving. This and other considerations lead Aarts to conclude t h a t other strategies such as economic measures, legislation and normative appeals should also be relied on, in order to bring about environmentally sustainable patterns of h u m a n consumption.
1211 8.
CO2 E M I S S I O N S R E D U C T I O N VIA L I F E S T Y L E C H A N G E S
Nicely complementary to the previously discussed projects by Schmidt, Aarts and Spier is the energy-technological study by Vringer, Wilting, Biesiot, Blok and Moll (NRP project no. 852086). This project team from the universities of Utrecht and Groningen has first developed and refined an input-output energy analysis (IOEA) methodology for determining the direct and indirect energy requirements of various household consumption patterns. In their first subproject, indirect energy use for consumption was determined by assessing the cumulative energy intensity, i.e., the total amount of energy needed for one financial unit of production, of 56 Dutch production sectors. This measure, of course, also includes the energy intensity of exported goods, but with the help of available export statistics it can be corrected for this. Also needed, then, are estimates of the cumulative energy intensity of imported goods, which constitute part of total domestic consumption. The primary aim of Vringer et al.'s project is to analyse if and how CO2 emissions can be reduced by changing lifestyles. Thus, in their second subproject, the authors attempted to identify and describe distinct 'lifestyles' in terms of systematically different patterns of household consumption in regard to their energy intensity and 'CO2 content'. Lifestyles are identified by correlating income, time budget and consumption variables with total household energy requirement. The latter was assessed by d e t e r m i n i n g the energy intensities of about 350 consumption categories and combining these using data from a recent national household expenditure survey. In a third subproject, the consequences of possible future technological developments on the energy intensities and the CO2 content of lifestyles are being modelled via a scenario approach. Some substantive results from the first two subprojects are as follows. Although the energy intensity per financial unit of production has significantly declined between 1969 and 1988 in most sectors of the Dutch economy, the embodied energy of total production has increased. This is because the growth of production has dominated the decline in energy intensity. Between 1969 and 1988 the embodied energy of Dutch exports has exceeded t h a t of imports; The Netherlands now is a net exporter of embodied energy of materials and products. This partly explains the rise in total Dutch energy consumption in this period. While m a n y energy conservation programmes are focused on reductions in direct energy consumption, several production sectors and the households have a higher indirect (embodied) than direct energy consumption. Therefore, reducing indirect energy consumption should get more attention in government energy policy. As far as households are concerned, their total average energy demand in 1990 was 240 gigajoules, of which 46% consisted of direct energy consumption (for heating, lighting and car fueling) and 54% reflects indirect energy use (as embodied in m a t e r i a l s , goods and services purchased). There appears to be a strong relationship between household expenditure and total energy demand, expenditure level being strongly correlated with net income. One additional factor is household size; one-person households use significantly less energy t h a n households consisting of two or more persons. Large differences in energy intensity were observed among different consumption categories, as well as among households
1212 having the same expenditure level. This is indicative of the fact t h a t lifestyle changes may result in significant reductions in energy use and CO2 emissions.
C o m m e n t s on Vringer et al.'s project The IOEA-methodology is an important development in energy analysis research. A tool like IOEA is indispensable if one wishes to assess the cumulative energy intensity of household consumption and to identify energy-relevant differences in lifestyle. For identifying and distinguishing lifestyles themselves, however, household-economic and demographic data, such as income and household size, m a y be greatly insufficient. Despite the fairly strong correlation between household income and total energy demand, there appear to be relatively energy-intensive low-income households as well as energy-thrifty high-income households. This is obviously due to different patterns of expenditure. Would these be explained by differences in 'lifestyle'? How could lifestyle be independently defined and assessed? And to what extent would the lifestyle of an household be d e t e r m i n e d by personal or family-cultural factors on the one hand, and by situational factors inherent in that household's physical and social environment on the other? At this point one would like to see research inputs from sociology, social psychology and cultural anthropology, where differences in lifestyle have been the subject of study for quite some time already. The lifestyle subproject on energy and CO2 emissions reduction is still under way (until mid-1995). By means of scenario construction possible changes in industrial practices and consumer behaviour will be modelled and evaluated with regard to their consequences and implications for fossil-fuel energy consumption.
9.
LOW-ENERGY, LOW-CO2 EMISSIONS SCENARIOS
Related r e s e a r c h is being conducted at the University of Groningen in a multidisciplinary project by Kamminga, Slotegraaf, Van der Veen and others (NRP project no. 852085), on the social significance, feasibility and acceptability of low-energy, low-CO2 emissions scenarios for The Netherlands. Here, recent macro-economic scenarios for the development of the Dutch economy in an international context formed the starting point (CPB, 1992). The investigators argued that prospective modelling by macro-economists insufficiently indicates the meaning and implications of the relevant scenarios for various social and economic groups in society, and that their acceptability as possible futures is empirically unclear. Also, the r e s e a r c h e r s wished to explicate the a s s u m p t i o n s and presuppositions underlying the scenarios and to inspect the way in which predictive elements of the scenarios - such as, e.g., employment rates or energy price levels - had been handled. As a beginning, the project team of sociologists, economists, psychologists and environmental scientists has critically evaluated the CPB scenarios. These had been published by their composers as an explorative means to shake up mental maps of policy makers and to provoke and guide public debates about the future of socio-economic life in The Netherlands. The CPB scenarios were designed on the basis of three different views on economic development, viz. the equilibrium perspective, the co-ordination perspective and the free market perspective. A next
1213 step in their elaboration was a comparative-strength analysis of seven different economic regions of the world. Finally, analyses were made of seven long-term trends, such as population growth, environmental qualities, world food supply and international co-operation. Eventually three different scenarios for the Dutch economy in international context emerged: (1) 'Balanced Growth', an optimistic scenario, (2) 'Global Shift', a pessimistic scenario, and (3) 'European Renaissance', a crisis-overcoming scenario. All three scenarios involved policy measures and expected effects with regard to energy, housing, agriculture, industry, transport and health care. None of the scenarios, however, clearly stood out as a 'low-energy, low-CO2 emissions scenario'. Because of this, because no formal scenario construction methodology had been followed, and because various assumptions and predictions that had been made, could not easily be validated, the University of Groningen team decided to develop its own policy scenario for a low-energy, low-CO2 emissions future for the Dutch social market-economy. After carefully studying available documents and interviewing relevant experts, the team has constructed an overall package of general energy-savings and emissions-reduction measures for The Netherlands as a whole, plus four subsets of sector-specific packages directed at industry, households, greenhouse horticulture and freight transport. The scenarios and subscenarios involve policy measures such as a general energy tax, energy-savings information campaigns, subsidies for energy-efficient technology, application of energy consumption standards and quota, subsidies for low-energy lighting equipment, promoting efficient transport management, discouraging air transport, and speed limitations for road traffic. In three subsequent empirical studies, the macro- and meso-economic significance and effects, the evaluation and assessment by meso-level social and economic actors and decision makers, and the evaluation and acceptance by micro-level (i.e., household) representatives were investigated by an economist, a sociologist and a social psychologist, respectively. Some substantive results are the following. Economically, the significance of a 'low-energy, low-CO2 emissions scenario' hinges upon: (1) its distributional effects in terms of income, employment and economic growth, (2) its structural effects in terms of new opportunities at the supply side of the Dutch economy, and (3) institutional changes necessary to support the restructuring and redistribution involved in a sustainable economic development. Sociologically, it appeared possible to specify the socio-political plausibility of major policy measures reasonably well, via a modelling of key meso-level actors' preferences and power positions in the overall political decision-making process. For example, it turned out to be 'quite probable' that a gradually increasing energy tax for 'small' consumers will actually be introduced, while the probability of significant car-use reduction measures was assessed to be a moderate 50% on the short term. Social-psychologically, it became clear that some 1200 Dutch household representatives are fairly well informed about the global greenhouse effect and judge it desirable that something be done about it. Also, on the average they evaluated a number of household energy-savings measures as reasonably effective and sufficiently acceptable in view of expected changes in quality-of-life. Women, higher-educated persons and non-motorists appeared to find
1214 mobility-directed energy-savings measures to be more acceptable t h a n men, lower-educated persons and regular car-drivers.
C o m m e n t s on K a m m i n g a et al.'s project "Macro-economists tend to see and contemplate things at a high level of aggregation. What certain future events and policy measures actually mean for the people and the organisations concerned, does insufficiently enter their functional range of vision, and so does the potential degree of social acceptability." This critical viewpoint has fruitfully stimulated the investigators to explore the essence, the meaning, the feasibility and the (differentiated) acceptability of socio-economic and energy-savings scenarios for The Netherlands. It is important to learn t h a t this macro-economic scenario construction by the Central Planning Bureau and associated institutes (CPB, 1992) was not based on formal concepts and an explicit methodology. It was a drawback for the project team to note that a significant 'low-energy, low-CO2 emissions scenario' was not available at the outset. But then it t u r n e d out to be highly instructive to go around energy documents and experts in an attempt to compose one's own package of feasible energy-savings measures. And it is worth-while to learn that 'social acceptability' has a different meaning for an economist (thinking about distributional effects), a sociologist (thinking about the preferences and political influence of meso-level actors) and a social psychologist (thinking about changes in quality-of-life and people's potential for adaptation). The multidisciplinary co-operation which has been established, needed its time to develop. A period of two years may be too short for operationalising the original research plan, getting the team to function effectively and to conduct the field research neccessary to test your hypotheses and underpin your conclusions. With a little more manoeuvring space this multidisciplinary project could have yielded even more useful and interesting results. For example, the separate evaluation of sector-specific packages of policy measures by meso- and by micro-level actors could have been extended from the households to all four sectors covered in the scenario design phase. Also, a further differentiation of socio-economic sectors could have been made, in order to obtain a more comprehensive picture of policy m e a s u r e s and their acceptability. The innovative thing about this project is its basic approach of exploring and describing lower-level social effects and responses related to energy-relevant conditions and policy measures, and of subsequently a s s e s s i n g t h e i r social acceptability in t e r m s of economic, sociological and social-psychological considerations. Such an approach m a y c o n s t i t u t e an important counterpart to macro-economic efforts at 'scanning the future'. 10. L I F E S T Y L E S AND D O M E S T I C E N E R G Y C O N S U M P T I O N A final consumption and lifestyles project is carried out at the universities of Tilburg and Eindhoven, by Breemhaar, Van Gool, Ester and Midden (NRP project no. 853119). Here, an exploration is made of the measurability of the concept of lifestyle which appears to be somewhat difficult to define. Also, an attempt is made to specify 'sustainable consumption patterns' with regard to household energy use. B r e e m h a a r et al. seek to define 'lifestyle' in terms of means-end chains, i.e., hypothetical strings of a particular consumer product, its perceived attributes, the
1215 consequences associated to the attributes and the basic (implicit)values t h a t are ultimately served when a consumer experiences those consequences. For example, a sports bike (a means) m a y be perceived as light, sturdy and dependable (its attributes), so t h a t one m a y reach a not-too-distant destination fast, without hassles and along a quiet route, while having some exercise (the consequences), all of which is valued for its goal-effectiveness, 'naturalness' and healthiness (the ends). Such cognitive means-end chains are assessed via in-depth interviews with consumers. A 'laddering technique' is used to lead r e s p o n d e n t s along the hypothetical links in a means-end chain. Such chains are likely to be different for different products. They may also be different for different domains of consumer behaviour, such as feeding, clothing and transportation. And means-end chains may be categorised into distinct groups which are characteristic of different groups of consumers. The authors' p r i m a r y research question reads: "Is it possible to group means-end chains concerning a p a r t i c u l a r behavioural domain with r e g a r d to energy consumption, and are the groups interpretable as lifestyles concerning energy consumption?" An answer to this question is being sought via consumer interviews on means-end chains regarding home-work commuting, home heating, living-room lighting and using a freezer, a washing machine and a washing-dryer. Through content analysis and cluster analysis, the investigators arrive at graphical r e p r e s e n t a t i o n s of adjectives describing attributes, consequences and values associated to particular consumer behaviours. Their project s u m m a r y elsewhere in this volume contains the example of home-work commuting, based on interviews with a small n u m b e r of respondents. Here, it appears t h a t 'motorists' could be clearly d i s t i n g u i s h e d from 'cyclists', and t h a t the general as well as the commuting-specific context variables were differentially clustered for these groups. Results for the other types of consumer behaviour are still being analysed. As 'lifestyles' may be strongly context-dependent, the researchers are also probing into the relationship between observed clusters of (personal) means-end chains a n d (more collective) clusters of context v a r i a b l e s for c e r t a i n groups of respondents and for given domains of consumer behaviour. They state t h a t "it is difficult to conclude whether or not the similarities in classification of respondents on the basis of their means-end chains and on the basis of context variables constitute a causal relationship." Another unresolved issue is the generality of clusters of consumption consequences and consumer values across different types of household consumption. For example, in what way and to what extent would the goal-effectiveness, 'naturalness' and healthiness of the bicycling commuter also show up in his or her means-end chains for living-room lighting, home heating and using electric household machinery? C o m m e n t s o n B r e e m h a a r et al.'s p r o j e c t
This research is methodologically explorative and it proves to be labour-intensive. D e t e r m i n i n g means-end chains for specific consumption behaviours requires a s e r i o u s a n d r e l a t i v e l y long i n t e r v i e w w i t h a t t e n t i v e r e s p o n d e n t s . Content-analysing recorded responses and cluster-analysing coded elements of means-end-chains demands sophisticated data analysis and careful interpretation of results. Someone's 'lifestyle' may emerge as a certain clustering of means-end chains across different types of a consumer's behaviour. A group of consumers
1216 sharing a particular lifestyle may show up, when it appears that their generalised means-end chains are sufficiently similar, in contradistinction from other groups of consumers who cherish other 'lifestyles'. This seems much to expect, and researchers m u s t have some luck to obtain the commonalities underlying the lifestyle concept. Too much differentiation of lifestyles with regard to types of consumer behaviour and/or with regard to subgroups of consumers, would weaken the use of any concept of lifestyle. Also, too much emphasis on cognitive elements such as perceived attributes, consequences and values, may detract from the policy relevance of the lifestyle concept ("de gustibus non est disputandum"). Finally, it m u s t eventually become clear to what extent lifestyles are person- or household-specific , and to what degree they depend upon characteristics of a consumer's physical and social context. The present project is still under way, until mid-1995. Since not all data have yet been analysed and a full report is not yet available, it is still unclear to what extent lifestyles provide an a p p r o p r i a t e c o n c e p t u a l i s a t i o n of domestic e n e r g y consumption. However, if they do, opportunities exist to alter energy-intensive lifestyles into more sustainable ones. The investigators are continuing their search for 'sustainable consumption' and are attempting to define this concept in terms of patterns of energy-extensive household behaviours. Eventually, such patterns will be presented to small consumer panels for their evaluation. 11. F E A R - A R O U S A L COMMUNICATION
AND
ARGUMENTATION
IN
RISK
For environmental policy in general and for climate policy in particular it is crucial t h a t the risks of global warming be presented such that respondents accept the need for remedial actions. Adequate diagnostic research on environmental change and climate processes is one condition for this. An other condition is effective communication of diagnostic results. To study the effects of problem information on energy-savings attitudes, Meijnders, Midden and Wilke at the universities of Eindhoven and Leiden (NRP project no. 852093) have performed a series of experimental studies. In the first experiment their goal was to observe the effects of fear-arousal and argument quality on people's attitudes toward purchasing 'a new type of energy-saving light bulbs'. Four experimental conditions were created by crossing a low- versus high-fear arousing problem-information variable with a weak versus strong purchase-argument variable. Low-fear information was given in a concise description of global warming; in the high-fear condition this information was supplemented with photographs illustrating potential w a r m i n g effects. W e a k versus strong a r g u m e n t quality was varied via selection of arguments previously rated for their 'convincingness'. On the average, the four groups of 19 subjects each (inhabitants of Eindhoven) reflected no overall (main) effects of fear level and of argument strength on their a t t i t u d e s towards purchasing the new light bulb. In the low-fear information condition, however, the average subject's attitude proved to be more favourable after strong arguments' presentation than after weak arguments. At the same time subjects, in a 'thought-listing' task, responded by giving more issue-relevant responses to the high-fear message t h a n to the low-fear message. These partly
1217 unexpected results are provisionally interpreted as a possible suppression, in the high-fear condition, of systematic information processing. The authors generally conclude t h a t "fear m a y have a positive effect on (people's) motivation to elaborate relevant information, but at high levels of fear, this positive effect may be overruled by a negative effect on information processing capacity". This project is being continued and therefore results of further experimentation are still to become available. C o m m e n t s on M e i j n d e r s et al.'s p r o j e c t In view of apparently serious problems of climate risk communication, one may wonder what conclusions and recommendations would emerge from the voluminous literature on fear arousal and information processing in the face of risk. So far in this project the impression is given that almost exclusive reference is made to the social-psychological literature, and not to the many chapters and articles on 'risk communication' t h a t have appeared since the mid-1980s, in several books on technological risk management, and in journals like Risk Analysis, J o u r n a l of Communication and Journal of Social Issues. Against the background of much of t h a t literature the question arises whether the sort of fear-arousal, the kind of argumentation and the type of 'action' used in this project's first experiment may hit the point hard enough. Methodologically, this study has been designed and conducted in a convincing manner, about which it is enjoyable to read. The theoretical basis of the project is interesting and important, but it could be broadened so as to incorporate sensitive elements from the multidisciplinary debate on technological-risk communication. For a policy-supporting research p r o g r a m m e like NRP the question is w h e t h e r theory-directed, high-quality e x p e r i m e n t a t i o n will indeed yield the sort of useful results the p r o g r a m m e committee is hoping for. Perhaps a more daring kind of field experimentation, based on a multidisciplinary effort to formulate problem information, select type of communication and design environment-protective actions, could provide the kind of conclusions and recommendations that would be both theoretically justified and practically useful.
12. SOCIAL W E L F A R E AN E N V I R O N M E N T A L QUALITY "The currently dominant idea of material welfare is at odds with a lifestyle t h a t does justice to basic h u m a n values. M e a s u r e m e n t of welfare needs to attach importance to a good environment." This dual thesis forms the starting point of a non-NRP project which fits into the debate on sustainable consumption and lifestyles, conducted by Mentzel at the University of Leiden (see Table 1.1.). The author critically reviews the concept of social welfare as used by economic policy makers and he puts this in contrast with perceived well-being and quality-of-life as experienced by individuals. 'Economic' welfare is expressed in terms of ownership and consumption of material goods and of access to high-energy activities such as in transport. Aggregate economic welfare is quantified into a country's gross national income (GNI), and economic growth in terms of GNI is believed to be crucial. It is becoming clear that material economic growth is damaging basic environmental qualities and ultimately threatens the earth's life support systems. Therefore, particularly in the industrialised consumer societies of the northern hemisphere, a search for a 'sustainable lifestyle' is necessary. This, says Mentzel,
1218 should cover the main spheres of life: at home, in the shopping-mall, in the workplace, in transport and traffic, and in recreational activities. To delineate what is needed, a reconceptualisation of what we mean by 'the good life' is required, as well as empirical research yielding people's own conceptions and dimensions of welfare and quality-of-life. Various empirical studies on perceived well-being and quality-of-life have been performed which reveal basic dimensions of perceived welfare. For example, an important empirical dimension appears to be the capacity to control and consciously direct one's own living conditions. 'Having', 'loving' and 'being' are useful central labels for characterising essential conditions for human development and existence. Good health is highly valued in present-day society, while societal improvements are being sought in better interpersonal relationships and a higher quality of the natural environment. The author keenly puts his finger on the role of national and international institutions by which socio-economic developments towards sustainable consumption and production patterns are to be promoted and co-ordinated. Two questions are sensitive here: how to arrive at a just (re-)distribution of welfare, and how to increase the socially perceived, and thus (also) the political value of nature and its resources. C o m m e n t s on Mentzel's project
The problem of unsustainable economic growth and ecologically unbridled consumption (see also Schmidt's, Spier's en Aarts's projects above) necessitates a fundamental reconsideration of classical notions of welfare and quality-of-life. To conduct this debate in a fair manner, it seems useful to keep in mind that the currently dominant concept of economic (i.e., material) welfare is rooted in people's natural motivation to be safeguarded against poverty, discomforts and diseases. It is the 'overshoot' of an economic system originally designed to meet such essential h u m a n desires, which has put many (though by far not all) of us up with a consumer society where personal satisfaction, power and prestige have become strongly associated to material possessions and consumption. The critical goal of sustainable development, therefore, should be a set of economic (i.e. material) conditions which could be considered 'sufficient' and 'fair'. In a shortlist of recommendations for a sustainable lifestyle, Thoenes (1990) indicates the necessity of, among other: a guaranteed satisfaction of basic needs, the creation of a basic social equality for everyone, and expansion of possibilities for energy- and material-extensive behaviours. Official present-day economic views of welfare are not as materialistic as Mentzel seems to suggest. The Organisation of Economic Co-operation and Development (OECD, 1982), for example, considers productivity, employment rate, purchasing power, balance of payments, government deficit and rate of inflation as basic economic indicators. The OECD recommends, however, that governments also pay attention to such dimensions as health, education, work conditions, social life, and the quality of public environmental goods such as air, water and natural areas. In a systematic review of social indicators research, Henderson (1994) searches for new indicators of wealth and progress and for changes in the meaning of 'development'. For example, for some time already the city of Jacksonville in Florida, U.S.A. evaluates its 'progress' in terms of nine categories of indicators, viz. 'the economy', public safety, health, education, natural environment, mobility (transport), government/politics, social environment and culture/recreation.
1219 According to Henderson, a clarification of the confusion of m e a n s (e.g. material consumption, economic growth) with essential e n d s of human development is badly needed. It would seem that this could best take place in a public debate among policy makers, supported by relevant specialists from social philosophy, economics, sociology, psychology and cultural antropology. 13. T O W A R D S S U S T A I N A B L E H O U S E H O L D M E T A B O L I S M
A final project deserving attention is funded by N.W.O., the Netherlands Organisation for Scientific Research. In 'HOMES: HOusehold Metabolism Effectively Sustainable', Schoot Uiterkamp at the University of Groningen co-ordinates a multidisciplinary group of environmental scientists, economists, spatial scientists, social psychologists and administrative scientists. Since early 1994 these investigators first of all attempt to diagnose and explain developments and trends in household consumption between 1950 en 1990 and as far into the future as 2030. Because household consumption encompasses a multitude of goods, services and activities and therefore must be delineated, the project's focus is on housing and transportation, home heating and lighting, and durable household equipment, whereby a distinction is made between strategic (mostly: purchasing) decisons and the operational use of electricity, water and different fossil fuels for daily activities. Secondly, the project group is determining the environmental impacts of household consumption and assessing its (un)sustainability, both in terms of descriptive variables such as various kinds of resource use and waste materials and in terms of subjective judgements collected from household representatives. Thirdly, the HOMES team will systematically analyse and describe possible technical as well as behavioural options and strategies for changing household consumption such that it may be considered 'sustainable' in the long run. To this end, specific technical options and behaviours will be considered, and consumption patterns and lifestyles will be designed and evaluated in collaboration with consumer groups and individuals. Also, various different policy strategies for encouraging households to adopt sustainable consumption patterns will be described and evaluated for their potential effectiveness. By doing all this in a multidisciplinary fashion, the HOMES team aspires to cover and integrate the physico-chemical and the technical aspects and possibilities of household consumption as well as the social and behavioural components and opportunities for sustainability. The project as a whole is to be concluded in 1998. Comments on Schoot Uiterkamp's project 'HOMES' is a problem-oriented, multidisciplinary endeavour to assess and understand household metabolism and to indicate ways and means for modifying this into a sustainable direction. Such an approach is explicitly stimulated by N.W.O. (see above) which - in its priority research programme on 'Sustainability and environmental quality' (1993-1998) funds altogether three such pluralistic projects (the other two deal with five major metal flows through the economy and with international river basin management, respectively). Considering the social and economic opportunities and motives for household consumption, looking into its relation to demographic developments and to physical infrastructure and government policies, and charting its various environmental effects, requires wide-ranging exploration and assessment as well strong co-ordination and overall
1220 modelling. Furthermore, household metabolism and industrial metabolism are strongly interwoven (see Section 3 above). Hence both a diagnosis of current consumption and the design of sustainable metabolism would sooner or later have to cover both the households and various relevant production sectors of the economy. The la tt er perspective is already taken in K a m m i n g a et al.'s NRP-project on 'low-energy, low-CO2 emissions scenarios' (see Section 8 above), and it is also adopted in a newly started NRP phase II project on emissions reductions via lifestyle changes by Biesiot (Groningen), Blok (Utrecht) and others, which links up directly with the environmental-science subproject of HOMES. Whether the broad-ranging and ambitious HOMES-project will succeed is a matter of prudent delineation, effective co-operation, personal enthusiasm and some luck in designing and conducting data collection and overall modelling of results. A scientifically hazardous approach like this, however, seems badly required for u n d e r s t a n d i n g costly and harmful developments in society and for designing sustainable patterns of social and economic behaviour. 14. GE N E R A L CO N C L US I O N S, RECOMMENDATIONS
SUGGESTIONS
AND
14.1 'Economisation' and e n v i r o n m e n t a l e x p l o i t a t i o n The NRP-research on consumption and lifestyles conducted so far has been a mixture of social-science, technological and environmental-science studies. These investigations have yielded important data and conclusions about societal motivations, developments and trends about consumption and lifestyles. By virtue of this, an interesting and useful picture of 'household metabolism' is emerging. During several decades now, strong increases in consumer purchasing power, in technological potentialities and in the market supply of a great variety of products, services and facilities, have met with social-cultural changes in individual and social motives of consumers. 'Economisation' has gradually led consumers, who are always partly driven by producers and advertisers, to adopt or aspire a prevailing lifestyle of high-quality material possessions and facilities, and of fast, short-term consumptive behaviours, whereby some basic goal or sense of life is easily obscured. For a long time the economic system of western industrialised countries has been truly successful in combatting human poverty, ignorance, discomfort and diseases. In recent times, however, it seems to be overshooting its original goals and to be developing into an energy- and material-intensive monster which gradually eats up its own existential conditions and seriously diminishes various qualities-of-life. The 'modern consumer lifestyle' inherent to this system is increasingly expansive, mobile and environmentally harmful. Many technical options seem to be available for increasing energy- and materials-efficiency and for reducing the amount and variety of household waste. But the social implementation of these, as well as the possible occurrence of 'unsaving' substitution behaviours and further consumption growth, deduct from the environmental effectiveness of technology. Therefore, behaviour change and particularly 'moderation' are becoming the key words for policy makers who - on behalf of society as a whole - are trying to steer away from unsustainable household consumption.
1221
14.2 Some policy recommendations In their co-ordinated project summary Goudsblom, Aarts, Schmidt and Spier (NRP project no. 851038, see Sections 4-6 above) present a number of useful conclusions a n d policy r e c o m m e n d a t i o n s . For example, as ' i m p o r t a n t obstacles to ecologisation' Goudsblom et al. mention: the strong social pressures to produce, inherent to industrial market regimes; the constantly rising productivity of labour as a result of competition; the fact that economic growth also is to create, or at least maintain, sufficient job opportunities; and the boosting effect on consumption of the s t a t u s h i e r a r c h y in industrial m a r k e t regimes. After listing various 'facilitating conditions for ecologisation', these authors also provide a n u m b e r of policy recommendations, for example: to use and exploit the s t a t u s motive in environmental policies, for instance, by associating social prestige to energy- and material-extensive behaviours; to stimulate further research into fossil-fuel energy savings techniques and their social implementation; to utilise the m a r k e t mechanism through levies, taxes and subsidies for an 'ecologisation' of production, commerce and consumption; and to develop specific campaigns with regard to car-driving, holiday air travel, meat consumption and other energy-intensive social behaviours. Goudsblom et al.'s conclusions and recommendations fit in with the fifth policy direction: 'institutional and cultural change', that emerged from a multidisciplinary and multi-party series of specialists' workshops conducted by Klabbers (Nijmegen) and Vellinga (Amsterdam); see (Klabbers et al., 1994). This strategic policy option came out of intense deliberations as one that might be inevitable to select if it would appear t h a t 'no regrets', 'least regrets', 'acceleration' and 'technological innovation', the other four policy directions, are not effective enough to secure a sustainable development of society. Some focal policy actions under 'institutional and cultural change' would be: to initiate a debate on improving the quality of society; to intensify the care for residential environments; to encourage consumers to select higher-quality food products, to buy local products and to follow local cuisine; to promote active participation in cultural activities; and to use trendsetters as examples of behavioural change. Such a view also links up with recent ideas about a 'greening of the economy', about which a key author remarks t h a t "we should be aiming to maximise the welfare obtained from economic activity while minimising the volume of matter and energy which flows through the economy" (Jacobs, 1991, p. 114).
14.3 Lifestyles and behaviour change strategies In none of the NRP-studies conducted so far has the concept of lifestyle been given a theoretically convincing and methodologically sound definition. Therefore the notion of lifestyle is up for further improvement and operationalisation to support continuing research aimed at delineating sustainable lifestyles. It would seem that such research should be interdisciplinary in nature; consumption patterns might be defined as 'lifestyles' to the extent t h a t they meet certain sociological, economic-psychological and ecological criteria. Candidate variables for these are family background, education, income, energy consumption, amounts of waste, degree of mobility, major life goals, habits and attitudes toward energy-savings, and appreciation of nature and natural living conditions. In a t h r i v i n g consumer society, changes in lifestyle or in prevailing social behaviours, in order to achieve energy savings and emission reductions, are hard to
1222 explain and to bring about. It is an underestimated problem that such changes need to rest upon a sufficient awareness of environmental problems, that they cannot occur without the availability of feasible behaviour alternatives, that policy instruments for inducing behaviour changes may, if wrongly selected and tuned-in, be ineffective or even counter-productive, and that the subject of the desired behaviour changes to whom the policy instruments are applied, wants to have an idea of 'what the future will bring'. The social and behavioural sciences have much in store to clarify this problem and to support the design of effective policy instruments. A major kind of conceptual tools are models for analysing and explaining consumer behaviour. Attitude- intention-behaviour models (Ajzen, 1991), o r i e n t a t i o n - p u r c h a s e - u s e - d i s c a r d models (Van Raaij, 1994), and motivation-opportunity-ability models (Oelander et al., 1994) have proven to be suitable means for coming to grips with consumer behaviour and its potential for modification. 14.4 I n t e r n a t i o n a l r e s e a r c h efforts Household consumption, its stimulation by and its implications for various production sectors, and its gross environmental impact in terms of resource use, land exploitation and waste, is a topic of increasing international interest. For example, Oelander and colleagues in Denmark, with the support of the Danish government, are conducting a multidisciplinary project on ' u n d e r s t a n d i n g consumer behaviour as a prerequisite for environmental protection' (Oelander, 1994). At the International Institute for Applied Systems Analysis in Laxenburg, Austria, Nakicenovic and colleagues are conducting and forming a network around their 'Environmentally Compatible Energy Strategies Project' (Nakicenovic et al., 1994 and Grfibler, 1991).So do Schipper and colleagues at Lawrence Livermore Laboratories in Berkeley, California (Schipper et al., 1989 and 1992). Also in the U.S.A., Stern [see Stern, 1994 for a review] has long investigated the psychological determinants of energy- and material-intensive behaviours and possible strategies for achieving environment-saving behaviour.
In The Netherlands, the Netherlands Energy Research Foundation in Petten has organised and published the results of several national workshops on 'lifestyle and energy consumption' (Perrels 1993 and 1994), where various motives, types of behaviour and strategies for behaviour change have been critically discussed. Again, it appeared that technical options are to be supplemented with behavioural options, and that the acceptability of any behaviour changes significantly depends upon their feasibility and their (perceived) environmental effectiveness. ECN-editor Perrels also recommends a multidisciplinary a t t e m p t at better defining the concept of lifestyle, and to carefully consider what different types of actors in society (e.g., consumers, producers, retailers, utility companies and government policy makers) actually do and could do to stimulate energy- and material-extensive behaviour patterns. Like the Dutch National Institute for Public Health and Environmental Protection RIVM (1991) in its National Environmental Survey 1990-2010, Perrels (1994, p. 73) also concludes that, in order to arrive at sustainable household metabolism, our cherished concept of economic growth may have to be differently filled-in (i.e., rather more qualitatively than quantitatively) and that international re-distribution of economic potential and wealth would be important for realising world-wide sustainability. Vivid stimulation of international comparative studies on consumption and lifestyles is
1223 to be expected from the H u m a n Dimensions of Global E n v i r o n m e n t a l Change programme (HDP), initiated by the International Social Science Council in Paris. In HDP's Work Plan for 1994-1995 (HDP, 1994) household metabolism is not m e n t i o n e d as such, but energy consumption, household r e s o u r c e use and individuals' attitudes and behaviours towards the environment are somehow incorporated in several 'major research areas', such as 'industrial transformation and energy use', 'demographic and social dimensions of resource use' and 'public attitudes, perceptions, behaviour and knowledge'. One problem with the HDP research programming so far, however, seems to be the predominance of general explorative questions as contrasted with specific research hypotheses about reasonably delineated (potential) policy issues. Another problem, it would seem, is the r u d i m e n t a r y development of an interdisciplinary perspective on global e n v i r o n m e n t a l change, whereby component research tasks might be usefully allocated in a multidisciplinary fashion. 14.5 F u r t h e r N R P i n t e n t i o n s In the second phase on the NRP (1995-2001), the problem of energy- and m a t e r i a l - i n t e n s i v e household consumption and the search for s u s t a i n a b l e consumer lifestyles remain high on the programme's agenda. Study topics are, for instance: the relationship between (total) household metabolism, population development and trends in household formation and household activity patterns; the identification and explanation of 'unsustainable' lifestyles; and methods and i n s t r u m e n t s for designing and implementing 'sustainable' consumption patterns. Investigations concerning household consumption and lifestyles will be deliberately linked with studies on climate-problem awareness and with research on personal mobility and the diverse use of motor vehicles (see the review chapter on mobility and transport, elsewhere in this volume). Also, the NRP committee will promote international co-operation and exchange of ideas and research findings, as a way to improve i n t e r n a t i o n a l u n d e r s t a n d i n g and policy making regarding household metabolism. 14.6 R e l i g i o n s on c o n s u m p t i o n To conclude this review chapter on culture, consumption and lifestyles in view of global environmental change, it may be appropriate to cite Durning (1992) who after documenting, characterising and criticising western-industrial consumption styles much like Vermeersch (1990) does - provides a t a b u l a r overview of ideological statements on h u m a n consumption and wealth, as derived from nine major world religions. The Buddhists, for example, profess that "who in this world transcends his selfish desires, his worries drop from his shoulders as dew-drops from a lotus flower". The Hindi like to say: "He who is fully free of desires and without craving .. reaches peace". The Christians cherish their biblical quote: "It is easier for a camel to go through the eye of a needle than it is for a rich man to enter the kingdom of God", while the Muslims repeat after their prophet Mohammed: "Poverty is my pride". But perhaps the most applicable statement in view of the present review comes from the Confucianists: "Excess and want are equally bad". Would 'the middle way' be truly sustainable?
1224 15. R E F E R E N C E S
Ajzen, I., 1991. The theory of planned behavior. Organizational Behavior and Human Decision Processes 50: 179-211. Corson, W.H., 1994. Changing course: an outline of strategies for a sustainable future. Futures 26: 206-223. CPB: Centraal PlanBureau, 1992. Nederland in Drievoud. Een scenariostudie van de Nederlandse economie 1990-2015. SDU Uitgeverij Den Haag. Dittmar, H., 1992. The social psychology of material possessions. Harvester Wheatsheaf, Hemel Hempstead, U.K.; St. Martin's Press, New York. Durning, A.T., 1992. How much is enough? W.W. Norton Company, New York~ondon. (In Dutch: Hoeveel is genoeg? De konsumptiemaatschappij en de toekomst van de aarde. Pauli Publishing, Worldwatch Institute Europe, Berlaar Belgium). Ehrlich, P.R. and Holdren, J.P., 1971. Impact of population growth. Science 171: 1212-1217. Goodland, R., Daly, H. and Kellenberg, J., 1994. Burden sharing in the transition to environmental sustainability. Futures 26: 146-155. Grfibler, A., 1991. Energy in the 21st century: from resource to environmental and lifestyle constraints. Entropie 164/165: 29-33. HDP-committee, 1994. H u m a n dimensions of global environmental change programme. HDP Work Plan 1994-1995. Occasional Paper no. 6. Paris: International Social Science Council at UNESCO. Henderson, H., 1994. Paths to sustainable development; the role of social indicators. Futures 26(2): 125-137. Jacobs, M., 1991. The green economy: sustainable development and the politics of the future. Pluto Press, London. Klabbers, J., Vellinga, P., Swart, R., Van Ulden, A. and Janssen, R., 1994. Policy options addressing the greenhouse effect. NRP, Bilthoven, The Netherlands. Masters, J.C. and Smith, W.P., (eds), 1987. Social comparison, social justice and relative deprivation: theoretical and policy perspectives. Volume 4. Hillsdale (N.J.), Erlbaum. Nakicenovic, N., Nordhaus, W.D., Richels, R. and Toth, F.L., Eds, 1994. Integrative assessment of mitigation, impacts and adaptation to climate change. International Institute for Applied Systems Analysis Laxenburg (Austria). OECD, 1982. The OECD list of social indicators. Organisation for Economic Cooperation and Development Parijs. Oelander. F. and Thogerson, J., 1994. Understanding of consumer behaviour as a prerequisite for environmental protection. Keynote address presented at 23rd International Congress of Applied Psychology.Aarhus School of Business, Aarhus, Denmark. Opschoor, H., 1989. Na ons geen zondvloed. Voorwaarden voor d u u r z a a m milieugebruik. Kok/Agora, Kampen. Perrels, A.H., 1994. Slotbeschouwing. In: De Paauw, K.F.B., A.H. Perrels and A.F.M. van Veenendaal, (Red.): Leefstijl en energie; van intentie naar actie? Petten: Energie Centrum Nederland, Rapport ECN-C-94-068. Perrels, A.H. (Red.), 1993. Leefstijl en energie" waar moet dat heen, hoe zal dat gaan.. Een interdisciplinaire kruisbestuiving. Rapport ECN-C-93-049, Energie Centrum Nederland, Petten.
1225 RIVM, 1991. Nationale Milieuverkenningen 2: 1990-2010.RijksInstituut voor Volksgezondheid en Milieuhygiene, Bilthoven en Samson/Tjeenk Willink, Alphen a/d Rijn. (in Dutch). Schipper, L., Bartlett, S., Hawk, D. and Vine, E., 1989. Linking life-styles and energy use: a matter of time? Annual Review of Energy 14: 273-320. Schipper, L. and Meyers, S., 1992. Energy efficiency and human activity; past trends, future prospects. Cambridge University Press, USA. Stern, P.C., 1992. Psychological dimensions of global environmental change. Annual Review of Psychology 43: 269-302. Suls, J.M. and Miller, R.L., Eds, 1977. Social comparison processes: theoretical and empirical perspectives. Hemisphere Publishers, Washington D.C.. Thoenes, P., 1990. Milieu en consumptie: blijft meer steeds beter? In Commissie Lange Termijn Milieubeleid: Het milieu: denkbeelden voor de 21ste eeuw. Kerkebosch, Zeist. Van Raaij, W.F., 1994. Consumentengedrag en milieu. In: Midden, C.J.H. and G.C. Bartels (Red.). Maatschappelijke aspecten van het milieuvraagstuk Bohn Stafleu Van Loghum, Houten. Vermeersch, E., 1990. Weg van het WTK-complex: onze toekomstige samenleving. In Commissie Lange Termijn Milieubeleid: Het milieu: denkbeelden voor de 21ste eeuw. Kerkebosch, Zeist. Vlek, Ch., Hendrickx, L and Steg, L., 1993. A social dilemmas analysis of motorised-transport problems and six general strategies for social behaviour change. In ECMT: Transport policy and global warming. European Conference of Ministers of Transport, OECD Publication Service Paris, p. 209-225. White, L., 1967. The historical roots of our ecologic crisis. Science 155: 1203-1207.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1229
NRP-project LIFESTYLE Reduction of CO2 emissions by lifestyle changes Vringer, K., Wilting, H.C., Biesiot, W., Blok, K. & Moll, H.C. ~* poster presentation International Conference on Climate Change Research, Evaluation and Policy Implications 8 December 1994, Maastricht, The Netherlands
Abstract The aim of the Lifestyle project is to analyse the C O 2 emission reduction potential of lifestyle change. The analysis is carried out by examining the direct and the indirect energy contents of the average Dutch household consumption. An overview of the past developments of Dutch sector energy intensities is produced and its consequences for the average household energy requirement are studied. Also differences in energy requirement related to differences in lifestyle are assessed. Calculations of the Dutch household expenditure survey has resulted in an overview of the energy requirement per income and spending subcategory. The correlations between some relevant household factors are determined and discussed.
Introduction The Lifestyle project succeeds preliminary studies about the direct and indirect energy contents of an average household consumption pattern. The aim of the project is to analyse if and how COz emissions can be reduced by changing lifestyles or by changes within lifestyles. Six research stages are discerned. First, - t o serve this goal- it is necessary to enlarge the scope of the methodology to calculate the energy content of consumption patterns and to improve the quality of data (research stages A1 - A3). Next differences in CO2 emission related to differences in lifestyle and possibilities of lifestyle changes are to be assessed and evaluated on their potential to reduce the CO2 emission (research stages B1 - B3). The six research stages are: A1. An improvement of the input output energy analysis methodology (including CO2 emissions) by correction of possible biases and by an assessment of its scope by application on several generic lifestyles.
# Harry Wilting, Wouter Biesiot and Henk Moll, Centre for Energy and Environmental Studies (IVEM) University of Groningen, Nijenborgh 4, 9747 AG Groningen, The Netherlands * Kees Vringer, Kornelis Blok, Department of Science, Technology and Society Utrecht University, Padualaan 14, NL-3584 CH Utrecht, The Netherlands
1230 A2. A3. B1. B2. B3.
An assessment of structural trends with regard to the energy intensities of economic sectors from a time series of Dutch energy consumption data. Further and deeper analysis of consumption and construction of a database concerning the energy and CO2 content of consumption activities. Identification of different lifestyles by correlating income, time budget and consumption with energy requirement and trend analysis on these lifestyles. Description of the lifestyles in terms of financial and energy/CO2 costs and trend analysis on these costs. Assessment of the effects of possible technological developments on energy intensities and on energy and CO2 content of lifestyles by a scenario approach.
This paper discusses the results of two subprojects: Energy consumption in relation to economic activities, 1969- 1988, addressing the research stages A1 and A2, and The direct and indirect energy requirements of households in the Netherlands, addressing the research stages B1 and B2.
Energy consumption in relation to economic activities, 1969 - 1988
1
Economic activities, production and consumption, are closely related. Production, in fact, occurs on behalf of consumption (exports included). Therefore, the total energy use of an economy can be attributed to the consumption sectors. So, the indirect energy requirements of households, as a consequence of the purchase of goods and services, are not only determined by the consumption patterns of the households, but also by the cumulative energy intensities of the production sectors. The cumulative energy intensity gives for each sector the total amount of energy, direct and indirect, that is needed for one financial unit of production of that sector. We aim to obtain an overview of the developments of the cumulative energy intensities for the Dutch production sectors over a period of twenty years (1969-1988). Besides, we attempt to determine the historic trends of the embodied energy of imports and exports and the indirect energy requirements of households. The cumulative energy intensities of the production sectors are calculated by using inputoutput energy analysis, which makes use of economic input-output tables. These tables, published by the Netherlands Central Bureau of Statistics annually, describe the transactions in an economy in financial terms. To gain insight in the development in the energy consumption of the whole production system, we calculated the cumulative energy intensity of the total production of the economy. Using the energy intensities, we calculated the energy flows in the economy, especially the embodied energy of the imports and the exports, and the indirect energy requirements of the households. The energy data required are taken from the Dutch Energy Statistics.
Results In the period 1969-1988, the cumulative energy intensity decreased for 40 of the 56 Dutch production sectors. 31 sectors showed a decrease by more than 10%. This points to an energy efficiency increase for these sectors. The direct and the cumulative energy intensity of total production decreased both by about 20% in the period 1973-1988.
1231 The energy flows are determined by the energy intensities. Figure 1 shows the embodied energy of the 2500 imports versus the embodied = _ energy of the exports during the = = = 2000 = = period 1969-1988. Since 1971 the embodied energy of the exports has increased more than the embodied energy of the imports. In 1988 the embodied energy of _ ~_ the exports was 28% higher than 500 I the embodied energy of the !/-/ / I / imports. 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 Figure 2 shows the direct and year indirect energy consumption of the Figure 1 Embodiedenergy of imports, for production and households during the period consumption, and exports (PJ). 1969-1988. In this period, in energy c o n s u m p t i o n (P J) which the number of households 2500 [] indirect [] direct increased with about 50% the total energy consumption of the 2000 households grew with about 30%. In 1988, the total energy --- Z -1500 consumption of the households was about the same as the total z z 1000 energy consumption in 1973. The xx • energy consumption per household decreased by about 10%, partly 500 caused by a decline of the number of persons per household. Changes in the indirect energy 69 70 71 72 73 74 75 76 77 7B 79 80 81 82 83 65 86 B7 88 year consumption of the households are Figure 2 Directand indirect energy consumption households caused by changes in the energy in 1969-1988. intensities of the production sectors and volume and structure changes in the consumption patterns of the households. Figure 3 shows the progression of the indirect energy requirements of the households due to the changes mentioned above with regard to the base year 1969. The progress in the indirect energy requirements mainly resulted from volume changes and the improvement of the energy efficiency of the economic sectors. Changes in the structure of the consumption pattern, i.e. shifts in the purchases from production sectors to other production sectors, hardly affect the indirect energy consumption of the households. 3000
embodied energy (P J)
imports (production) [~ exports
'0001 00 !IIi"
[~ imports (consumption)
F
II-!
fli:
t lll
Conclusions Many energy conservation programs consider only direct energy consumption. Several production sectors and the households have a higher indirect energy consumption than
1232 direct energy consumption. Therefore, the indirect energy consumption should get more 160 attention in energy policy. We have found a significant negative correlation between 140 energy prices and energy intensities. A negative correlation 120 means that an increase in energy >~ '--x-- x ""k'J x "~O prices coincides with a decline in 100 -o- - o - _ D . o _ o _ e _ a _ , 3 _ a"'~-"~.--o--o--o--o-'~176 energy intensities. The strongest "x ~ -x ~ -x correlation is between the energy intensities and the energy prices 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 three years before. This is year consistent with the fact that the Figure 3 Indices indirect energy consumption households due to changes in volume and structure o f the consumption, due energy intensities of some energy sectors started to to changes in energy intensities of the production sectors, and in intensive decrease some years after the first total. oil crisis. Sectors need some time to react to energy price changes.According to goods and services, the Netherlands has become a net exporter of energy. This means that the CO2-emission caused by the use of fossil energy in the Netherlands on behalf of other countries is higher than the CO2emission in other countries on behalf of the Netherlands. The Dutch production system has concentrated more on exports. 180
index indirect energy consumption 0969 = 100) - - total
-~ volume
-o- structure --~ intensity
+
The direct and indirect energy requirements of households in the Netherlands 2
One way of reducing C O 2 emissions is to reduce direct and indirect household energy requirements by influencing the household consumption pattern. A household not only uses direct energy in the form of gas, electricity and petrol, but it also uses indirect energy embodied in consumer goods such as food, furniture and services. Before discussing the ways in which the household consumption pattern should be influenced, one needs to have quantitative information about these energy requirements. We aim to obtain an overview of the total energy requirement of households and the energy requirement per consumption category. Also we attempt to quantify the relation of household expenditure, net household income and number of household members to the total energy requirement of households. To obtain an overview of the cumulative energy requirement of Dutch households, we analysed the total consumption package for its cumulative energy requirement. The energy intensities (in MJ per Dutch guilder (Dfl)) of about 350 basic consumption categories are determined, using a hybrid energy analysis method. The energy requirement of Dutch households is calculated by combining the 350 energy intensities with data from the Netherlands Household Expenditure Survey of 1990. This
1233
survey gives the expenditure of 2767 representative households in the Netherlands in 1990. The result is an overview of the total energy requirement of Dutch households. Results The total average energy demand per household in the Netherlands in 1990 was 240 GJ, of which 54% was indirect. Table 1 gives the average energy requirement and energy intensity
of
the
Dutch
households,
Energy requirement (GJ) (% of total) Total
240
100
Indirectenergy requirement
Energy
intensity (MJ/Dfl) 6.3
130
54
3.5
Food
41
17
5.6
categories.
Household effects
19
8
5.5
Figure 4 shows the relation between the total energy requirement and the household expenditure. We give the 10, 25, 50 (median), 75 and 90
Clothing&footwear
3 5 2
2.7
Hygiene
8 12 5
4.1
Education&recreation
24
10
3.0
percentile
Transport& communication
11
5
2.8
110
46
45.0
28 60
12
25
46.5
22
9
22.4
aggregated into 11 main consumption
lines
in
this
figure
to
demonstrate the variance of the energy requirement for the spending subcategories. The 10 percentile line represents the levels for which 10% of the households of the corresponding spending subcategory requires less e n e r g y given by this line.
than
the
level
House
9
Medicalcare
Directenergy requirement Electricity
Heating
Petrol
4
1.4
3.4
57.8
Table 1 Total energy requirement and energy intensity of an average Dutch household in 1990 per main category.
Figure 4 shows - as expected - that the energy requirement increases with household expenditure. But also substantial variance within the spending subcategories is observed: e.g. 10% of the households use 22% less energy than the energy requirement of an average household with the same expenditure. 700 go pete, The relation between energy 75 pete. requirement and net household ..(125%) 50 I)erc. income shows also an ~50025 pete. increasing relationship of the E ..... (100"/,,) net household income and the 10 I~rc, ~400(BIPI.) energy requirement. But, the (TB*/.) variance is larger than the >'300variance shown in figure 5 (9 2oobecause of differences between o income and expenditure. 100In Figure 5 we plot the total energy requirement versus the ~ i'o ~o ~o do ~o ~o #o ~o ~;o 1oo net income for various household expenditure (Dfl x 1000) household sizes to investigate a ..o"
9"
Figure 4 expenditure,
(112"/o)
Total household energy requirement versus household
possible dependence of these factors
apart
from
the
1234 500
dependence related to differences in net income. Figure 5 demonstrates that only a significant difference in energy requirement, independent of the net household income, is observed between one-person households and several-person households (approx. 45 GJ).
450-
.-.400~350-
."5 .... .'.-'-..............2
e-
~30o-
"_~
250-
~200r 150-
o
o
"" 100-
500
0
io
2o
3o
4o
so
6o
~o
80
~
1oo
total net income (Dfl. x I000)
Figure 5 T o t a l h o u s e h o l d e n e r g y r e q u i r e m e n t versus net household income for 1 to 4 household members.
Conclusions Because the indirect energy requirement amounts at least 54% of the total requirement of households, further research is needed into the indirect household energy requirement. Future energy policy must pay attention to the indirect energy requirement of households. The strong relation between income and total energy requirement suggests that, with further increases in income levels, the average household energy requirement will probably rise as well. However, the large differences between the energy intensities of the various consumption categories indicate that the total household energy requirement can be reduced by a change of our consumption patterns.
References Wilting H.C., Biesiot , W., Moll, H.C., Economische activiteiten vanuit energetisch perspectief" Veranderingen in Nederland in de periode 1969-1988, IVEM onderzoeksrapport no. 72, juli 1994, Groningen. Vringer K., Blok, K., The direct and indirect energy requirement of households in The Netherlands, NW&S, 1993, Utrecht
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1235
Life styles and domestic energy consumption: a pilot study B. Breemhaar 1, W. van Gool 2, P. Ester ~, C. Midden2
1Institute for Social Policy Research and Consultancy, Tilburg, The Netherlands
2Eindhoven University of Technology, Dept. of Philosophy and Social Sciences, The Netherlands
Abstract The contribution of households to C O 2 production is still increasing. To alter patterns of energy consumption for example with respect to commuter traffic, using the freezer, and warming the house, changing life styles related to domestic energy consumption is considered. In our study, we have operationalized life style as m e a n s - e n d chains, that link perceived benefits of a particular behavior to basic values that people pursue. In this paper, preliminary results are presented of the study that is aimed at empirically establishing the feasibility of the concept of life style in relation to domestic energy consumption.
1. INTRODUCTION Domestic consumption of energy contributes considerably to CO2 production in the Netherlands. The proportion of CO 2 produced by households has increased strongly over the last 30 years, a tendency which is unlikely to change in the future without policy modifications. To promote sustainable development, a substantial reduction is required in domestic energy consumption. A promising approach to alter patterns of energy consumption may be to change life styles related to domestic energy consumption. Various authors [1, 2, 3] suggested that individuals' separate consumption behaviors constitute a more or less coherent pattern. It has been supposed that a person's particular acts of consumption are guided by basic values [4]. Values are consumers' mental representations of important end states they are trying to achieve in their lives [5]. These values guide individual choices with respect to consumption, causing characteristic differences or similarities between individuals. If this conception is correct, its implication would be that the focus of change must be patterns of energy consumption, rather than separate consumption behaviors. Attempts to change these energy consumpti-
1236 on patterns may eventuelly lead to change in particular energy consumption behaviors that make up the complete pattern [6]. The concept of coherent patterns of consumption within individuals is consistent with the concept of life styles, put forward by Weber [7]. According to Weber, individual tastes and preferences in behavior conform to socially determined structm'es. He maintained that groups of citizens can be distinguished by their socio-economic and status position. That is, persons sharing a similar status position, enjoy equal prestige in society, and are characterized by a common life style. Weber defines life style as a collection of explicit and expressive modes of behavior or behavioral preferences, in which consumption of material goods plays a dominant role. 1.1 An alternative conceptualization of life style In this study, the concept of life style is operationalized as so called means-end chains. This operationalization stems from economic psychological theories applied to marketing [8]. It is used to link products to individuals, according to the product's attributes that the individual considers valuable. According to the conception of meansend chains, people consider a particular product attractive, because it has attributes that implicate particular desirable consequences. In turn, these consequences are desired, because they serve to realize basic values which an individual considers of vital importance to pursue. Individuals can be classified according to differences and similarities in their means-end chains. This innovative operationalization of life style as means-end chains meets several of the criticisms with respect to the traditional concepts of life style. First, it is a domain specific approach: it recognizes that a person's life style may differ between different behavioral domains. Second, it does not limit itself to observable behaviors, but also includes a person's attitudes, socially determined normative beliefs, and basic values. To end this theoretical part, we formulate the research question of this study: is it possible to describe domestic energy consumption adequately in terms of life styles?
2. METHOD In order to trace means-end chains, we interviewed thirty-four consumers about their household energy behavior regarding six behavioral domains, namely commuter traffic, heating the house, lighting the living room, using the freezer, using the washingmachine, and using the washing-dryer. Every respondent has been interviewed on three of the six behavioral domains, according to a predetermined scheme. In the first part of each interview, we determined the common household context of the respondent, namely household situation, residential situation, and employment situation. Subsequently, we determined the context variables that were specific to a particular domain of energy consumption. For example, in the case of commuter traffic, the distance to work, the means of transport, and receipt of allowance were recorded. This part of the interview was ended by asking for the perceived benefits of the behavioral domain. In the second part of each interview, we used the laddering depth-interview method, that is used in consumer behavior research to trace means-end chains of products [8]. In our study, we have replaced products by perceived benefits of the behavioral domain. This implies that the attribute level in the means-end chain will be skipped. The laddering interview consists of determining why the most important perceived benefits are so important to the respondent. At best, each means-end chains ends at value level. In Figure 1, an example of one of our interviews on using the freezer
1237 is depicted. The percived benefit of using the freezer is the functional consequence
always food in stock. Through a number of steps, this means-end chain ended on the value level gives me a feeling of hospitality. GIVES A FRRTJNG OF HOSPITALITY
PEOPLE GET THE IDEA THEY ARE REAILY WELCOME
PRESENT S O n G
TO UNEXPECTED G ~
ANTICIPATE IN UNEXPECTED SITUATIONS
ALWAYS FOOD IN STOCK
Figure 1. Example of a means-end chain on using the freezer
2.1 Analysis The initial task of the analysis is to make a content-analysis of all elements from the laddering interviews [8]. All responses were divided as functional consequences, psychosocial consequences, instrumental values, and terminal values. This process resulted in a codebook of thirty-eight codes. Next, all individual ladders were rewritten in these number codes. In this case, we have made our data suited for the data-analysis. In order to detect groups of individuals with common characteristics, we have made use of a tandem use of correspondence analysis and cluster analysis [9]. This results in a graphical representation, which contains both the consequences and values, and the clusters of respondents.
2.2 Research questions To answer our research question whether domestic energy consumption can be described adequately in terms of life styles, we have made the following operationalization: 1. Is it possible to group means-end chains concerning a particular behavioral domain with regard to energy consumption, and are the groups interpretable as life styles concerning energy consumption? 2. Is it possible to group common context variables and context variables that are specific to a particular domain of energy consumption behavior? 3. Do groups of means-end chains regarding a particular behavioral domain of energy consumption overlap with groups of common context variables and with groups of context variables that are specific to that behavioral domain?
1238 3. PRELIMINARY RESULTS Currently, we merely have preliminary results at our disposal. We have enlarged our sample, but we have not assimilated it in the present results. Due to lack of space, we will only discuss the results of one of the six behavioral domains, namely commuter traffic. In Figure 2, the results of the data-analysis on the means-end chains on commuter traffic is depicted. It shows both the consequenes and values (the name codes), and the clusters of respondents (the numbers). The closer a respondent is situated near a name code, the more that name code applies to that respondent. In the case of commuter traffic, the cluster analysis resulted in three clusters. We can see that two large clusters (n=7 and n=6) and one very small cluster (n=l) were formed. Respondents in the first cluster, which were mainly cyclists, emphasized in the laddering interviews the healthy fresh air, physical movement, saving money, feeling pleasantly and at ease, and a better environment. Respondents in the second cluster, which were mainly motorists, emphasized saving time, ambition, being independent of external facors, freedom, functionality, and safety. The respondent in the third cluster, who travels by train, emphasized the atmosphere and the possibility to relax and dream.
~/ III
/
33
heze
better e n ~ t healthier 21 24 sa.ve money \ clean fresh ai~ 34 20 move~en t pleasuzable health health/ 7 save e ~ g y ol
2 ~, zng
Figure 2. Clusters of the means-end chains on commuter traffic
When considering the common context, three clusters were formed, mainly on the basis of the respondents' education, working situation, income, age, size of the house, and the presence or absence of school attending children. When considering the specific context, one large cluster and three small clusters were formed, mainly on the basis of distance to work, means of transport, the amount of travel-expenses, and receiving of a pay for expenses.
1239 When looking at the overlap between the clusters of the means-end chains, the common context, and the specific context, we notice that two groups of respondents are clustered together every time. This means that they show a strong correspondence in their means-end chains, and that they have a common and specific context that is very identical. Due to lack of space, we will not go into the content of these groups with an overlap in the means-end chains and context.
4. DISCUSSION In this article, we only discussed the results of the analysis on commuter traffic. The other behavioral domains - warming the house, lighting the living room, using the freezer, the washing-machine, and the washing-dryer - were left out of consideration in this contribution. In each domain, we found different clusters of similar means-end chains, and different clusters of similar demographic and relevant contextual variables, which partly overlap with clusters based on means-end chains. So, the preliminary results of this study offer indications that different groups of individuals can be distinguished with respect to a single domain of domestic energy consumption, based on the consequences attached by each individual to that particular behavior and the basic values he or she attains by that behavior. However, as yet no definite conclusions can be drawn about the relationship between on the one hand the consequences and values of a particular energy behavior, and on the other hand the relevant context of their living circumstances. It is difficult to conclude whether or not the similarities in classification of respondents on the basis of their means-end chains and on the basis of context variables constitute a causal relationship. Another unresolved issue concerns the relationship between structures of consequences and values with respect to one domain of domestic energy consumption and another. For example, are a person's positively valued consequences and related values with respect to his or her mode of commuting related to positively valued consequences and values with respect to heating the house, or are they unrelated? Further analyses of the data will have to provide answers to these questions. First, additional data will be collected in order to obtain information about means-end structures with respect to each domain of domestic energy consumption from approximately 35 respondents. This will provide a sound basis for further establishment of the reliability and validity of clusters of consequences and associated values regarding various domains of domestic energy consumption. Further, by means of discriminant analysis, we will explore the relationship between clusters of valued consequences and associated values, and clusters of general (demographic) and domain specific context variables. That is, we will examine how respondents clustered in a particular group on the basis of appreciated consequences and values differ from respondents clustered in a second group, with respect to general and domain-specific context variables.
1240 5. REFERENCES A. Mitchell, The Nine American Lifestyles, New York, Warner Books, 1984. H. Ganzeboom, Leefstijlen, in Jaarboek '90-'91 Nederlandse Vereniging van Marktonderzoekers, Haarlem, De Vrieseborch, 1989. P.A.F. de Bruijn and R.J.T. Custers, Voorwaarden voor Consumptieverandering, 's Gravenhage, SWOKA, 1993. M.J. Rokeach, The Nature of Human Values, New York, The Free Press, 1973. J.P. Peter and J.C. Olson, Consumer Behavior and Marketing Strategy, 3rd edition, Hollywood, Irwin, 1993. C.A.J. Vlek, Leefstijlen, gedragsverandering en energiebesparing: een conceptuele en methodologische beschouwing, in K. de Paauw, A. Perrels, en A. van Veenendaal (eds.), Leefstijl en energie: van intentie naar actie?, Petten, ECN2, 1994. M. Weber, Wirtschaft und Gesellschaft, Tfibingen, 1972. T.J. Reynolds and J. Gutman, Laddering Theory, Method, Analysis, and Interpretation, J. of Advertising Research, 28 (feb/mar) (1988), 11-31. P.E. Green, C.M. Schaffer and K.M. Patterson, A reduced space approach to the clustering of categorical data in market segmentation, J. of the Market Research Society, 30 (3) (1988), 267-288.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1241
NRP-project SCAN (SCenario ANalysis) Analysis of the social significance, acceptability and feasibility of longterm low energy/low CO2 scenarios for The Netherlands Kamminga, K.J., Slotegraaf, G., van der Veen, H.C.J. & Moll, H.C. poster presentation International Conference on Climate Change Research, Evaluation and Policy Implications 8 December 1994, Maastricht, The Netherlands
Abstract Long term low energy/low CO2 scenarios are developed for an interdisciplinary research about the significance, acceptability and feasibility of such scenarios. The social psychological research - to measure the acceptability - is directed at three determinants of cooperative behaviour in a social dilemma situation: knowledge, trust and morality. On the basis of this triad, research variables have been formulated and a research model has been developed. The sociological r e s e a r c h - to measure the feasibility - concerns an assessment of resistances, blockades, interests, and conditions for cooperation of the involved organisations with regard to the (package of) measures and the expected effects of the measures. The economical research- to measure the significance - will be mainly directed to acquire qualitative insight into effects of sectoral measures and packages of measures at a macroeconomic level.
Introduction
1
The aim of the research in the SCAN-project is to supplement general long-term lowenergy/low-CO2 scenarios and to clarify these in terms of their social significance, acceptability and feasibility. This research is conducted by an interdisciplinary team (psychology, sociology, economic and environmental sciences). The following research questions in the SCAN-project are discerned: A. 1. What is the expected quality of life within the variants of these scenarios? A.2. What is the acceptability of the. proposed scenario variants? A.3. What is the feasibility of the prospective elements of the proposed scenario variants?
Cor Kamminga and Henk Moll, Centre for Energy and Environmental Studies (IVEM), Nijenborgh 4, 9747 AG Groningen; Goos Slotegraaf, Institute for Social and Organisational Psychology (S&O), Grote Kruisstraat 2/1, 9712 TS Groningen; Henk van der Veen, Department of Sociology, Grote Rozenstraat 31, 9712 TG Groningen. All authors are working at the Groningen University of The Netherlands.
1242 Firstly, a new scenario is devised, i.e. packages of measures aiming at a (substantial) reduction of the energy consumption and the CO2-emission by The Netherlands. This scenario has been revised and evaluated by experts with regard to energy conservation and CO2-emission reduction, and to the behaviourial and economic effects of the considered measures. The scenario consists of measures derived from different relevant categories, i.e. technical measures, regulatory measures, measures providing financial-economic incentives, educative and communicative measures and measures aiming at organisational and institutional change. It is supposed, that an effective scenario should cover all these categories. The scenario measures are directed at four sectors: industry, greenhouse-horticulture, freight transport and (household) consumption. About 80% of the total energy use of The Netherlands is consumed by these sectors. This scenario is the starting point for the research, concerning the social significance, acceptability and feasibility of low energy/low CO2 scenarios. The relationships with these issues and the different disciplines are presented in figure 1. The results of these scenarios are predicted with help of the
Package of Measures SIGNIFICANCE economy environmental studies
economic and environmental
analysis. The results are described in terms of reduction Soc~a~ Psychology of the total Dutch energy consumption, of reduction of Results the Dutch CO2 emission, and of the changes of relevant Energy economic parameters e.g. C02 employment for industry, greenhouse horticulture and freight transportation. These ACCEPTABILITY results will demonstrate the Social Psychology environmental and economic Sociology significance of these scenFigure 1 Integrationof the 3 SCAN lines of research. arios. The issue of acceptability of the scenarios is primarily elaborated by social-psychological research by a postal survey among a large sample (several thousands) of households. Also the significance of individual and economic effects is measured for the households by this research. The issue of the feasibility of these scenarios is examined by a sociological field research. Political and institutional actors are interviewed about their interests, opinion, position, resources and influence. In this way major resistances and barriers concerning the acceptance of CO2 emission reducing measures will be determined and the feasibility of these measures will be estimated. FEASIBILITY
Soc,ology
1243 The survey results about the acceptability of CO2 emission reduction measures by households may influence also the feasibility of these measures. Political and institutional actors will be confronted with these acceptability judgments of households.
The social psychological line of research The social psychological line of research, aims at understanding and predicting the acceptability of policy measures at the individual level. For this purpose, specific policy measures aiming at the energy saving behaviour of individuals and households have been selected. This kind of behaviour can be characterised as a social dilemma; long term benefits can only be achieved by the cooperative behaviour of others.
Theoretical background Dawes (1980) 2, who elaborated on the social dilemma paradigm, earlier introduced by Hardin (1968) 3, argues that the three most important determinants of cooperative behaviour in a social dilemma situation, are best described by the psychological constructs I knowledge, trust and morality. ! On the basis of this triad, research variables have been : ._ , lull formulated and a preliminary i ' model has been developed. The model can be characterised, as demonstrated in figure 2, by the two 'routes', by which the acceptability is li ! UNCERTAINTY influenced. The first route ~dlllHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIimllllWIHIIIIIIHIIIIIIIIIIIIIIl~lll concerns the causation of the problem and the uncertainty about environmental processes. Knowledge, problem awarei , ` " ~ ~ ~ ~ ] ness, responsibility for causing the problem and the perceived solvability of the problem are 1 1 the key variables. The other route affects the solutions proposed to tackle the problem and the uncertainty about the (cooperative) behaviour of Social psychological determinants of the acceptability of policy measures others. The impact of the pro@ Stimulus , , Predictor 9 Dep. var. posed solutions (or policy Figure 2 Model and variables for the social-psychological measures) on the individual research approach to assess the acceptability of energy reduction 'quality of life', the amount of measures and scenarios.
IIIllllill lllllllllllllJ tJ
_ll l U UNCERTAINTY nl lP-
11111
L
llff
r
1244 trust in others, the responsibility for solving the problem and perceived effectiveness of the policy measures are the key variables here.
Methodology In a large scale postal survey research amongst 3000 households, a package of future policy measures presented is presented as a short scenario. The model parameters are supposed to have a predicting value in relation to the acceptability of the policy measures in the scenarios. This will be tested by means of a multiple regression model. Furthermore, differences between certain groups will be inspected. Next to the variables mentioned above, attention will be paid to individual and group differences, regarding income level, education, age, gender, type of household, car use and ownership, etc. Final results can be expected in the spring of 1995.
The sociological line of research The research objective of the sociological research line is to determine the short term and long term social political feasibility of extensive and controInvolved Organizations versial energy saving measures on four social sectors each having large I I energy saving potentials. Quantitative Data Qualitative Data The defined package of measures for the sector Households will be completely covered and for Descriptive Statistics Content Analysis Acceptability Statistic the other sectors some Policy Decision Making Models measures are selected according to controversiality, effect size, concreteness and the Organizational Acceptability Resistance and Blockades number of involved orgCollective Acceptability interests predicted Outcome of Policy Process Conditions for Cooperation anisations. For each Power Distribution of Organizations Effects and Side Effects Impact of Policy Position Charge sector the acceptability and feasibility of the introduction of an energy I tax will be examined. Acceptability of Energy Saving Measure For the sector GlassFeasibility of Energy Saving Measure house Industry the estabFigure 3 Sociologicalapproach to study the feasibility and accepta- lishment of usage quota bility of energy use reducing measures and scenarios. Measure
1245 of natural gas and for the sector Freight Transport of the realisation of the Betuwespoorlijn are analysed specifically.
Methodology The methodological framework is demonstrated in figure 3. The first stage of the research consists of the specification of the measures and the determination of the organisations who participate in the political decision making process concerning that measure. In the second stage qualitative data as well as quantitative data will be gathered in expert interviews for each involved organisation. In the third stage the data will be analysed. Content analysis of the qualitative data will result in resistances, blockades, interests, conditions for cooperation of the involved organisation with regard to the (package of) measures and the expected effects and side effects of the measures. The quantitative data will be analysed using several statistics and political decision making models. This will result in estimates about the acceptability, predictions about the outcome of the decision making process, in an overview of the power distribution of the involved organisations and an assessment of the impact of policy position change of key organisations. The final results are judgments of acceptability and feasibility of the (package of) energy saving measures. The other (not selected) measures will be dealt with in a merely qualitative way. Though the survey is not yet completed, it seems that most energy saving measures will be supported at the institutional level. The notion appears to exist that energy saving measures are necessary and inevitable. These initial results make the question about the political feasibility of the measures very salient.
The economic line of research
The main research question of the economical research line is about the economic significance of the considered low energy/low CO2-scenarios, i.e. packages of measures concerning the sectors Glasshouse Industry, Industry and Freight Transport in the Netherlands. Before this main question can be answered, another question arises: How, from an economic point of view, has the present situation arisen concerning energy use and CO2emission in the Netherlands? In answering these research questions two phases can be distinguished. The first phase has already been finished. These results will be applied in phase 2. In the first phase attention was paid to the description of the relations between the two basic economic elements 'production' and 'consumption'. In doing so emphasis was laid upon a qualitative approach. Within this framework the following question was answered: How, from an economic point of view, has the present situation arisen concerning energy use and CO2-emissions in the Netherlands? The accent was on establishing the connection between the economic development during the 2 0 th century and the economic sector structure. The importance of this phase lies in the fact that it generates a diagnosis by
1246 which possible economic bottle-necks can be demonstrated to get to a low-energy/low CO: future. Supplemented with elementary quantitative data on energy-use, CO2emissions and economic performance, it also offers a first general insight into possible ways a measure or a package of measures might work out in social-economic terms during the time needed to get from the present situation to a low-energy/low CO2 society (the transition period). This in turn offers the opportunity to recognise possible socialeconomic problems in time and to think of strategies to avoid them.
Conceptual framework In phase 2 the main research question will be answered based on the general concept shown in figure 4. In answering this question interviews will be held with both general economic experts and experts from the relevant sectors. Concerning the economic significance two angles can be distinguished. Firstly, attention will be directed towards the meaning of a sectoral package of measures for the economic sector structure. In this way it is tried to generECONOMICAL
ENVIRONMENTAL
I
REQUIREMENTS - innovation - strenghtening sectorstructure
REQUIREMENTS - energy saving - reduction CO2 emission
I ECONOMICAL
I I .
I
INSTITUTIONAL CHANGES
I I
IOot::UoOellem~
CHANCES
enwronmental I production sector
I MEASURES RES iI
v ECONOMIC TRANSITION
" ALTERING SECTORSTRUCTURE
iore eveI I ':n ntoflI
II
ate some qualitative insight into effects of sectoral measures and packages of measures at a macro-economic level. Next to that, the significance of the packages at a sectoral level will be examined. In that connection more quantitative research is carried out.
Figure 4 Conceptual framework regarding environmental and economic requirements.
References This paper is partly based on the interim report of the SCAN-project of July 1994. A final report will be produced in the spring 1995. 2.
Dawes, R.M. (1980). Social Dilemmas. Annual Review of Psychology, 31, 169-193
3.
Hardin,G. (1968). The tragedy of the commons. Science, 162, 1243-1248
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1247
Toward a morality of increasing moderation W. Aarts, C. Schmidt and F. Spier Amsterdam School for Social science Research, Oude Hoogstraat 24, 1012 CE Amsterdam, The Netherlands
Abstract This paper sumlnarizes some provisional conclusions of three interrelated historicalsociological studies on 'economization' and 'ecologization'. Special attention is paid to the role status aspirations play in these processes. The focus is on ecologization, its conditions and obstacles.
1. E C O N O M I Z A T I O N AND E C O L O G I Z A T I O N AS CIVILIZING PROCESSES Current environmental problems are to a considerable extent caused by ecologically unbridled economic growth. Central problem of the first study is what social driving forces are behind this type of growth and to what extent they obstruct the control of environmental degradation. The answer to this question requires the elaboration of two theoretical concepts: 'economization' and 'ecologization'. Economization refers to a long-term social process in the course of which a growing number of societies turned into relatively peacefully competing regimes for generating wealth. For the societies involved this meant that 'economic' ways of doing and thinking gradually penetrated more and more spheres of life including that of the state. Economization might be considered a civilizing process for two related reasons. First, 'economic' activities were increasingly looked upon as more 'civilized' and prestigious than the extraction of surplus under the threat of violence, predation or war. Second, the process of economization brought about a growing social pressure towards self-control as well as an increasing control over nature. In the course of time, however, the resulting increase in affluence led to a relaxation of standards of frugality in the sphere of consumption. Economization implied an enormous increase in the division of labour. This meant in practice that a growing number of people were living and working in cities where they were not immediately confronted with the ecological effects of their activities. They could even cherish the illusion not to be dependent on nature any more. In reality, however, the increasing control over nature that made the urban-industrial way of life possible implied a growing, though less directly felt, dependence on the environment. The illusion of being released from ecological constraints explains the long-term short-sightedness of 'economized' societies with respect to the ecological effects of unbridled intensive growth. The term 'ecologization' refers to a re-awakening to these effects - the development of
1248 what came to be called 'environmental awareness' - as well as to attempts to keep the nature of human activity and the numbers of the human species within constraints considered 'ecologically acceptable'. In a way, the process can be looked upon as a continuation of economization because ecologization implies striving for optimum welfare within ecological constraints. In the twentieth century, social pressures toward more 'ecological self-control' have increased considerably. The second study deals with an important aspect of this long-term development.
2.THE RISE AND EFFECTIVENESS OF NON-GOVERNMENTAL ENVIRONMENTAL ORGANIZATIONS IN THE NETHERLANDS
From the beginning of this century, private organizations made efforts to protect specific parts of the Dutch landscape, flora and fauna, such as the Organization for the Protection of Birds and, most notably, the Organization for the Conservation of Nature Monuments. They focused on limited goals, the conservation of specific sites and/or of certain biological species. They were largely made up by members of the higher classes, whose rather effective political lobbying was mostly done in a discreet way. At the same time, they promoted their goals publicly by trying to get attached to it high status and prestige, as the name 'nature monuments' already suggests. This image-tbrming strategy can be summarized by saying that they sought to project a positive, 'high culture' image. Nature was beautiful, and should consequently be preserved. This motivated many people to associate with their cause. Although today the leadership of Nature Monuments expresses discontent with the current situation, the organization has been highly successful in terms of its original goals. The idea of protected areas has ahnost completely been accepted by the Dutch public (which explains why they are so easily overlooked). Such sites have steadily grown in size and numbers. In the 1980s, membership sharply increased and by 1994, its paying following is the largest of all ecological organizations in the Netherlands. By contrast, many sections of the ecological movement that came up in the 1960s had very wide-ranging aims, which included major changes in consumption as well as incisive societal change. Their campaigns were characterized by a rather informal code of conduct. Such activists tended to sound the alarln and projected an image of their goal which up to today is seen by many as an abhorrent example (the 'goat's woollen socks' image). For instance, the organization Environmental Defense (Milieudefensie) continually prophesied doom and gloom if its advice would not be heeded. Yet, their positively phrased 'Action Plan Sustainable Netherlands ' attracted a great deal of attention at home as well as abroad. This leads to the conclusion that those organizations which addressed tar-reaching issues like personal general ecological awareness and moderation chose a rather ineffective strategy to attract followers to their cause. By contrast, their not so spectacular predecessors reached their less ambitious goals by a rather effective strategy. Although sounding the alarm is a necessary component of efforts to stimulate ecological awareness, positively phrased campaigns to stimulate specific forms of moderation are likely to be more successful than alarmist approaches, and should clearly be kept separated. In addition, the ability to exercise influence at the highest level of decision making, including
1249 getting public support of highly-placed citizens, not only verbal but also in practice, may be helpful to spread forms of ecological moderation. The third study deals with clues for these and other forms of moderation in consumption, especially in the Netherlands.
3. CONSUMPTION AND STRATIFICATION The striving for the maintenance and improvement of social status is one of the primary driving forces underlying the continuing increase in consumption. The same drive, however, may also lead to an increasing moderation of consumption. In search of feasible solutions for environmental degradation, the third study focuses on the counterforces to the growth of consumption. Broadly speaking, sociological research reveals a positive relationship between power, wealth and prestige on the one hand, and the quantity of consumption on the other. In addition, a 'trickle down-effect' has been frequently observed. Patterns of consumption and consumer goods that were initially reserved for the members of privileged groups spread out to society at large. Ii1 this way holidays by air, cars and eating meat every day trickled down as did less tangible elements such as sensitivity to art and nature. High status, however, does not always coincide with conspicuous consumption. Historical-sociological research indicates that the members of privileged groups have time and again imposed restrictions on each other and on themselves. For example, in situations of rivalry between groups with economic power on the one hand and groups that possess principally cultural power on the other, the latter frequently tend to distinguish themselves by consumption that bears witness to self-control, tact and good taste. Moreover, whenever consumer goods become more widespread, their status-conferring character diminishes and from that moment on moderation might become prestigious. Closer analysis of research into the development of smoking and eating habits since the Second World War demonstrates that status has played an important part in pushing back smoking and eating fat food in industrial societies. The spread of nonsmoking and health food are typical examples of the effectiveness of the trickle down-effect. Interviews with members of high-status groups who practise forms of restraint which are not (yet) common indicate that they meet with growing social esteem. However, various sorts of environmentally friendly restraint do not seem to add much to social prestige. This may change, though, as a result of extensive attention in the media and the efforts of government and industry. People's preferences for moderation in different areas are interrelated. They are part of a more general status-related morality in which striving for self-control, responsibility and quality are at the centre. In most cases environmental concerns appear to be not the main motivation. Analyses of secondary resources demonstrate, for instance, that the educated hardly refrain from consumption that causes excessive emissions of carbon dioxide. Their environmental concern in this respect is still mainly symbolic though communicatively significant. But then, the anxiety about the greenhouse effect is relatively recent and still controversial. The fact that educated people are over-represented among the members of environmental organizations and buyers of environmentally more friendly products shows at least their willingness to do something for the environment. So there seems to be a potential for change here. However, as the report shows, the problem remains that the political-economic regime,
1250 which created the conditions for the beginning ecologization of society, continues to be permeated with strong social pressures obstructing that very same process. Nevertheless, some support has been found for the hypothesis that under specific conditions an appeal to status may be effective in strengthening ecological regimes.
Climate Change Research: Evaluation and Policy Implications S. Zwerver, R.S.A.R. van Rompaey, M.T.J. Kok and M.M. Berk, (Eds.) 9 Elsevier Science B.V. All rights reserved.
1251
C l i m a t e C h a n g e , L i v i n g E n v i r o n m e n t and W a y s o f Life M. J~irvel~i and M. Wilenius Research Institute for Social Sciences, University of Tampere, P.O. Box 607, SF-33101 Tampere, Finland
Abstract
Our empirical material based on interviews with influental actors in environmental policy in Finland suggested that possible future climatic changes illustrates the greatest single environmental threat on a global scale. The influential actors did not hesitate to consider as an issue of high certainty a kind of man-induced climate change. In mapping out social resources among actors to tackle climatic risks we have utilizised a teleological reasoning of rational action as an ideal model.
1. Introduction
From the point of view of global social and political regulation, the most complex and challenging issue in present-day ecological policies can be seen the question of climate change. Furthermore, within scientific world, there is a widening consensus about necessity to carry out social science research to contribute, on the side of natural and technical sciences, our knowledge about global environmental issues like climatic changes (see Morrisette & Plantinga 1991, Buttel & Taylor 1992,). In this presentation we will draw attention to the issue of climate change as a special case of social and environmental conflict in late modernity. In our complex societies, experts seem to gain ever more influence over issues like social regulation as the problems themselves grow more compicated (Fores et al. 1991, pp. 83-84, Parsons 1958, 34). In the field of environmental protection, the task to create rational modes of thinking and political action strategies is easily left to few highranking experts (Sundqvist 1992). The present part of our research focuses on socially influential groups that have a significantly important status in determining the interests, knowledge and morality in the definition of problems in environmental policies. Our empirical research sample includes various environmentally influential experts found in major industrial companies in Finland, politicians active in environment policy, experts in public administration and in the field of science, and journalists interested in environmental issues. We have also interviewed some eminent civil activists. The empirical material consists of sociological theme interviews. Rather than outlining different viewpoints of interest, our research focuses on the idea and knowledge resources
1252 that project different ways of thinking. The following figure illustrates the various social dimensions and structures embedded in the handling of the issue: The social resources of climate change politics
Theoretical Teleologies
f life
economicalworldview/value ; ~ ; e n d e s X ecological worldview / ,,l., \ sdentiflc/ i I ~ ~ tTaditJonal/mythical knowledaeT" "/
interpretaUgnof reality ~.1/ -
~
~
and sden~lflc
knowledge
fields of "conflict
I elites
>
