* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Climate change and sustainable water resources: placing the threat
Mitigation of global warming in Australia wikipedia , lookup
Fred Singer wikipedia , lookup
Climate resilience wikipedia , lookup
Global warming controversy wikipedia , lookup
Climate change denial wikipedia , lookup
Climate engineering wikipedia , lookup
Citizens' Climate Lobby wikipedia , lookup
Global warming hiatus wikipedia , lookup
Climate sensitivity wikipedia , lookup
Climatic Research Unit documents wikipedia , lookup
Climate governance wikipedia , lookup
Economics of global warming wikipedia , lookup
Instrumental temperature record wikipedia , lookup
Politics of global warming wikipedia , lookup
Climate change adaptation wikipedia , lookup
Global warming wikipedia , lookup
Carbon Pollution Reduction Scheme wikipedia , lookup
Media coverage of global warming wikipedia , lookup
Climate change in Tuvalu wikipedia , lookup
Climate change in Saskatchewan wikipedia , lookup
Attribution of recent climate change wikipedia , lookup
Climate change and agriculture wikipedia , lookup
General circulation model wikipedia , lookup
Solar radiation management wikipedia , lookup
Scientific opinion on climate change wikipedia , lookup
Effects of global warming wikipedia , lookup
Physical impacts of climate change wikipedia , lookup
Climate change feedback wikipedia , lookup
Effects of global warming on human health wikipedia , lookup
Climate change in the United States wikipedia , lookup
Surveys of scientists' views on climate change wikipedia , lookup
Public opinion on global warming wikipedia , lookup
Climate change and poverty wikipedia , lookup
IPCC Fourth Assessment Report wikipedia , lookup
Hydrological Sciences—Journal—des Sciences Hydrologiques, 44(4) August 1999 54] Special issue: Barriers to Sustainable Management of Water Quantity and Quality Climate change and sustainable water resources: placing the threat of global warming in perspective J. A. A. JONES Institute of Geography and Earth Sciences, University of Wales, Aberystwyth, SY23 3DB, UK e-mail: [email protected] Abstract Predicted climate change over the coming decades is likely to add measurable stress to water resources in many regions of the world, including some areas that are currently well endowed. The stresses are likely to involve changes in the frequency of extreme events as well as gradual changes in mean annual net resources. The paper analyses these predictions. It also attempts to place them in context, first, comparing their impact with other major barriers to sustainability, such as increasing demand, wastage, poor water resources assessment and international conflict, and, secondly, considering the limitations of current predictive techniques. Modification climatique et durabilité des ressources en eau: dans la perspective d'une menace de réchauffement planétaire Résumé Les modifications climatiques devant se produire au cours des prochaines décennies devraient aggraver le stress hydrique dans de nombreuses régions du monde, y compris dans certaines d'entre elles actuellement bien pourvues en eau. On doit s'attendre à des modifications concernant aussi bien la fréquence des événements extrêmes que les modules annuels. Cet article s'intéresse à ces prédictions et s'efforce de les remettre en perspective, en comparant leurs impacts à ceux d'autres obstacles à la durabilité comme la croissance de la demande, les rejets, les incertitudes pesant sur l'évaluation des ressources et les conflits internationaux, mais aussi en prenant en considération les limites actuelles capacités de prédiction. INTRODUCTION The International Commission on Water Resources Systems of IAHS has convened a series of symposia and published a Special Issue of this journal on the sustainable management of water resources, which have sought to define aims and measures for future sustainability (Simonovic et ai, 1995; Rosbjerg et ai, 1997; Simonovic, 1997). They have inevitably also identified some of the barriers to sustainability. Falkenmark (1997) noted two such threats: increasing food needs and increasing pollution. This paper is concerned with a third potential threat, climate change, and to adjudge its relative significance. Many recent publications have attempted to predict the impact of an enhanced greenhouse effect upon regional water resources in the twenty-first century (e.g. Arnell & Reynard, 1993; Arnell, 1996; IPCC,1996; Jones et al, 1996). A recent release from the UK Hadley Centre (1997) focuses on the global impact of the climatic predictions from the HadCM2 high resolution coupled atmosphere-ocean General Circulation Model (GCM) transient run on vegetation, agriculture and water resources. In this, Arnell & King (1997) compare the impact on water resources, modelled by inputting Open for discussion until I February 2000 542 J. A. A. Jones the climatic parameters into a daily runoff simulation model, with the calculated increase in demand, based on recent estimates of national population growth rates (WMO, 1997). Three general observations on the results of this recent work seem pertinent. First, GCMs are far less consistent and reliable in their predictions of regional patterns of precipitation and the factors controlling évapotranspiration than they are with temperatures. It is notable that the global patterns of increased or decreased precipitation produced by the recent generation of high resolution transient coupled atmosphere-ocean General Circulation Models developed by the UK Hadley Centre, e.g. the transient run of the original high resolution model (UKTR) and HadCM2, are very different from those of the first generation of low resolution, equilibrium GCMs in which the ocean was represented by a static surface "slab" (Jones, 1996a). The problem for this author is that many of the predictions of the simpler models seem to be generally more climatologically believable than those coming from the newer scientifically superior models. To take two examples, enhanced monsoonal rainfall in Southeast Asia and India might be expected from higher sea surface temperatures (SSTs), higher land surface temperatures in the continental interior and a poleward shift in the Ferrel Jetstream in summer. This often appears in the earlier models, but not in the latest GCMs. Similarly, early American GCMs tended to predict marked reductions in annual rainfall and increased summer drought in the Canadian prairies and Great Plains. This caused major political concern that a reduction in food aid from North America could destabilize many African countries (e.g. Tickell, 1991; Jones, 1997). Yet the problem seems to have been substantially reduced or has even disappeared in the latest GCM output (cf. Hadley Centre, 1995; Jones, 1996a; Arnell & King, 1997). Secondly, whilst analyses of recent instrumental records suggest trends that are consistent with some predicted by the GCMs, there are also many cases of opposite trends. Jones (1996b) collated evidence from recent instrumental analyses in Europe that was broadly in line with the tendencies indicated by most GCMs, in particular, a drying trend in southern Europe, increased water resources in northern Europe and long-term reduction in snowfall, snow and ice cover, and snowmelt floods. However, most of the trends observed in the recent worldwide collection of historical analyses in Jones et al. (1996) are at variance with the HadCM2 climate change scenario for 2080 (Table 1). Three conclusions are possible: (a) the recent hydrometric data do not yet Table 1 Comparison of HadCM2 derived runoff with recent observed trends. Recent observations drying in north and west, lower discharges, declining glaciers and lakes USA increase in southeast, decrease in northeast Japan decrease in discharges Australia increase discharge in east, if at all Europe decreased discharge, if at all, including Scandinavia Argentina increased precipitation in north Source of observed trends: Jones et al (1996). China HadCM2 2080 +10-50% runoff decrease in south, increase in north +10% discharge decreased discharge everywhere -10-25% discharge in south, increases in Scandinavia and UK decreased precipitation in north Climate change and sustainable water resources 543 indicate a significant global warming signal; (b) the model scenario is misleading; and (c) the progress of greenhouse gas driven climate change may not be unidirectional in specific regions. The latter possibility is climatologically feasible in many regions, for example, if higher SSTs result in greater oceanic evaporation and land downwind receives more precipitation before a general shift in storm trajectories, e.g. mid-latitude depressions, carries that precipitation to a different latitude. DEMAND AND WASTAGE AS BARRIERS TO SUSTAINABILITY The threat of climate change in the twenty-first century aggravates, and is also aggravated by, projected trends in population growth and demand. The single most fundamental barrier to sustainable use of water resources is population growth. UN (1991) estimates suggest a "most likely" rise of about 50% in world population by 2050. Unfortunately, most of this expansion will be in the lesser developed countries, which as a group are also more prone to water shortage. The WMO's (1997) assessment of global freshwater resources calculated that, whereas a third of the world's population currently lives in countries suffering moderate to high water stress, by 2025 this will have risen to two thirds of a much larger population. Key sources of demand The trend to urbanization further aggravates the problem by increasing per capita demand. Between 1990 and 2025, the percentage of the world population living in urban areas could rise by 50%, accounting for nearly two thirds of the total world population in 2025. Whereas most rural populations consume no more than 100200 1 per capita per day, and within the last decade nearly two thirds of the world population survived on less than 50 1, cities typically consume 300-600 1. Centralized sewage systems, industrial processing and the accoutrements of modern urban living are largely to blame. Perhaps fortunately for water demand, but sadly not for human health, much of the new urbanization in the developing world is unplanned and often unsupported by central supply systems and sewerage. Nevertheless, it is estimated that by 2025, 79% of the world urban community could be in the least developed countries (LDCs), compounding the stress on the least resilient economies. A very crude but conservative estimate of the "multiplier effect" of urbanization suggests that by 2025 the world will have over two billion more city dwellers than in the early 1990s, almost double the 1990 urban community, each consuming over twice the current rural average. To this must be added the demand for irrigation, which is rising in direct response to the need to feed the growing population. The peak wave of expansion may have passed, but irrigation remains the largest consumer of water worldwide, consumption has increased by nearly 20% in the last two decades, and a further 20% expansion in the area of irrigated agriculture is still regarded as environmentally feasible (Jones, 1997, p. 14). 544 J. A. A. Jones Inefficiency and wastage To the effects of expanding urbanization and irrigation must be added the problems of inefficiency and wastage in all spheres of consumption. Here the problems are very much a question of water quality as well as quantity. Lvovich's (1979) doomsday scenario of pollution essentially destroying global water resources by 2000 may have been averted, but the legacy of the decades of neglect that influenced Lvovich's viewpoint was clearly apparent in much of former communist eastern Europe in the early 1990s. Current fears focus on urban wastewater pollution in Southeast Asia, where rapid urbanization is likely to triple the amount of urban wastewater during the 1990s, and where increasing amounts receive little or no treatment. The current financial crises in the "tiger economies" will only exacerbate urban squalor and delay the provision of treatment plants. As tragically illustrated during 1998 in Bangladesh, China and Central America, floods combined with poor urban sanitation can be the source of major threats to human health from water-borne diseases. Similar health problems followed the River Oder floods in Poland in 1997, although petroleum products and industrial chemicals were possibly more important ingredients than human sewage in those floodwaters. There is a potentially important interaction here with climate change, if flood frequencies increase. Pollution will remain an important barrier to global sustainability for the foreseeable future. Even in the most advanced countries, where point sources of pollution, like sewage and industrial outfalls, are now largely under control, nonpoint sources such as agriculture, urban drainage and landfills are not. Nearly two thirds of all pollutants entering US rivers come from nonpoint sources, of which agriculture accounts for 60%. Again, water efficiency is improving rapidly in the most advanced economies, but there is little evidence of this in the developing world, where per capita wastage remains high and rising in most of the main areas of increasing population (Jones, 1997, p. 15ff). In part, this is due to the nature of their industries and the affordable level of technology, including irrigated agriculture. In part, it is also due to the level of public education and the dissemination of information which agencies like WMO and UNESCO are still fighting to improve. To take just one example from the developing and two from the developed world: modern methods of drip or sprinkler irrigation can reduce evaporative losses by up to 75%. In the UK, leakage rates of 30-40% from public water supply systems are rapidly being reduced following the National Leakage Control Initiative of 1994, but the best values are still 10-20% and therefore comparable to the probable average effect of climate change next century. Experiments in England have also indicated that domestic metering may save up to 20% of public water supply (Jones, 1997, p. 322). Finally, an important point must be made on the efficiency of management systems. Wastage, inefficiency and even damage to the fundamental resource have occurred all too frequently, especially in developing countries, as a result of piecemeal developments and fragmented institutional responsibilities (Hufschmidt & Tejwani, 1993). At the grandest scale, the era of rapid dam-building, especially in the 1960s and 1970s, resulted in numerous cases where lack of understanding of environmental Climate change and sustainable water resources 545 impacts or of the socio-political infrastructure has either impaired the resources or reduced the estimated benefits (Takeuchi, 1997; Jones, 1997). In many instances, the repayments on loans taken out to fund such schemes have added unnecessary financial stress to developing countries. In the worst instances, they have led to probably irrecoverable damage to the environment and human health, as witnessed in the irrigation schemes of the Aral Sea basin (Glazovsky, 1990). The UN is actively promoting "integrated water resource management" (IWRM) to link all resources within a drainage basin under one overarching management structure. Bonell et al. (1993) note the importance of this approach for management in the humid tropics. The vertical reorganization of water industries to combine all stages of water use and reclamation began in the West in the 1970s as recommended by the UN (1970), but the concepts are now being extended to cover ever broader aspects of environmental management on the land as well as within the water cycle. This trend recognizes the intimate linkage between land management and runoff (Kundzewicz, 1997). Among the biggest obstacles to adoption of basinwide IWRM are international boundaries. Whilst international cooperation is reasonably well established in North America and western Europe, the greatest number of international basins occur in the developing world where collaboration has been sadly lacking (McCaffrey, 1992). Over 60% of Africa and Asia is covered by shared basins. The internationalization of the Aral Sea basin since the demise of the USSR is already an obstacle to environmental rehabilitation. Recent but qualified successes may be found in the international accords on the Mekong (1994), in southern Africa (1995) and on the Ganges (1996), although the Mekong agreement excludes the important headwater nation, China. In contrast, the Middle East and North Africa is essentially an agreement-free zone. The Middle East is also a region under considerable stress, where the "demographic explosion" is likely to add 34 million to the population by the 2020s and push demand well above the conventionally available resources (Anderson, 1991). It is also a region where shared aquifers and "groundwater mining" can be as important as shared rivers, where there is severe inter-community enmity, and where the prospect of "water wars" is still real (Jones, 1997, p. 330ff). CLIMATE CHANGE AS A BARRIER TO SUSTAINABILITY Effective water management requires a knowledge of the likelihood of extreme events as well as averages. Current GCMs are beginning to give usable future scenarios for averages, but the estimation of what might happen to patterns of extreme events over the coming decades is still very much a research frontier. Results from the HadCM2 GCM The second high resolution coupled atmosphere-ocean GCM developed by the UK Hadley Centre showed a significant improvement in the hindcasting experiment over its predecessor (Johns et al, 1997) and has formed the basis for a wide-ranging study 546 J. A. A. Jones of global impacts (Meteorological Office, 1997). Amell & King's (1997) estimation of the impact of greenhouse gas warming (the HadCM2GHG scenario) on global runoff by 2080 (Fig. 1 ) suggests increases in equatorial regions, peaking at over 50% in parts of East Africa and eastern Brazil, much of Southeast Asia, the USA (except the Great Plains) and the south-central Asian republics of the CIS (Commonwealth of Independent States), as well as generally north of 60°N. The main areas with reduced surface water resources occur in the sub-tropical and Mediterranean climates, especially between the North Atlantic and India and from Argentina, across southern Africa to Australia. In general, the reductions are under 25%, but in India and the Southern Hemisphere there are large areas with deductions over 25% and even over 50%. The critical feature of these calculations is that most of the increase will occur in regions that are not currently stressed and most of the decreases will occur in those that are already stressed under the present climate. The main beneficiaries, countries currently using more than 20% of their potential resources and therefore considered "under stress", that will experience fewer problems by the 2020s, include the USA, the broad belt from the Caspian to China and Mongolia, the Sudan, Somalia and Peru. The main losers include Mexico, the belt from Mauritania through Egypt, Turkey and the Middle East to India and Bangladesh, South Africa and western continental Europe from Spain to Poland and Italy. Table 2 shows the combined effects of population increase and climate change over the coming century, assuming that per capita demand remains static. Although Table 2 Predicted trends in global precipitation, runoff and per capita water resources based on the output from HadCM2GHG, with some additions from HadCM2SUL. 2020s Present 2050s +3.0% +4.5% Precipitation*(HadCM2GHG) Precipitation* (HadCM2SUL) +2.0% +2.5% Runoff (HadCM2)+ +2.9% +4.0% Additional runoff (km3) (HadCM2GHG) 1126 1553 Population increase 9759 Population in millions* 5266 8121 +54% +85% % increase Per capita resources -35% -46% Due to population increase only -44% Due to population increase + climate -34% change Population in stress Billions using >20% of available water 1.9 5.1 5.9 resources (AWR) + 168% +211% % increase 66 extra millions due to climate change* 3.1 Billions using >40% of available water 0.454 2.4 resources (AWR) % increase +429% +583% 1]7!t)170«) extra millions due to climate change 26 * Data from Hadley Centre and Climatic Research Unit, University of East Anglia, LINK ' Data from graphs and text in Arnell & King (1997). * UN ( 1991 ) "most likely" estimates. 5 Arnell (1998). 2080s +6.25% +4.2% +6.5% 2524 10 672 + 103% -51% ^8% 6.5 +242% 3.6 +693% 98 Programme. Climate change and sustainable water resources 547 548 J. A. A. Jones population growth is seen to be the overwhelming source of stress, climate change alone could bring at least 100 million extra people within the realm of "extreme water stress", without allowing for any increase in demand due to climate change. Estimates for the 2050s vary from 117 million (graphs in Arnell & King, 1997) to 170 million (Arnell, 1998), but perhaps with a slight reduction by 2080. Results from HadCM2SUL The rerun of HadCM2 with sulphate aerosols produced significantly cooler temperatures than the "greenhouse gas only" run over large areas downwind of the main industrial sources of sulphates in North America, Europe, Russia and Japan (Mitchell et al., 1995; Hadley Centre, 1995). It also reduced the global increase in precipitation by 2080 by one third (Table 2). There is still considerable debate over whether sulphates will be such an important factor by 2080. The issue centres on the probable reduction in sulphate emissions from fossil fuel burning and on the relatively short residence time of the aerosols in the atmosphere. Increased rainout and washout in a higher rainfall environment in the future could complicate the already difficult task of transforming emissions into actual atmospheric concentrations. It remains, nevertheless, a scenario worth examining, especially given the predicted high level of increases in sulphate emissions in Southeast Asia over the next half century. The broad pattern of change in HadCM2SUL is similar to HadCM2GHG, but there is a markedly lower increase in global precipitation and runoff by 2080, and a commensurate reduction in the degree of change in regional water resources (Fig. 2; Table 2). The area of negative trend in surface water resources is reduced, but many of the same areas of reduction can be identified, e.g. the Middle East to Bangladesh, the Ukraine, Argentina, south-central Africa, east-central Australia and parts of the Sahel. The only area of significant increase outside the Greenland icecap lies in east-central China. There is also some evidence for "seesaw" trends in precipitation patterns, for example, a drier Britain and wetter Spain in 2020, reversing by 2080 (Fig. 3). A few notes of caution are needed here. First, the Hadley Centre emphasize that these estimates are based on a single experiment and cannot therefore incorporate the range of uncertainties. Hence, they are not "predictions" so much as possible realizations of how the climate system may respond to a given forcing. They do, nevertheless, represent the results from one of the few transient experiments with coupled atmosphere-ocean models, with 19 atmospheric and 20 oceanic levels. Second, the runoff in Fig. 2 is derived via the "on-line" runoff calculation within HadCM2, whereas that used in Fig. 1 was derived "off-line" by inputting HadCM2 hydrometeorology into a separate water balance model. Thirdly, there is some concern about the model itself and the emissions scenario used. Although the sulphate model produced a hindcast over the last 100 years that fitted the observed temperature trend better than with greenhouse gases only (Hadley Centre, 1995), the scenario used overestimates present-day sulphate emissions and may well overestimate future emissions (Conway, 1998). The Hadley Centre has recently been running HadCM3 Climate change and sustainable water resources 549 550 J. A. A. Jones Annual precipitation change (mm/day) % t// ',/,% 4*~ .,, ) M/// •f, Vf //,. 20*0 2069 , . *7-sr. / x y „, % ^ <,A< ^W^yWé$^%^^ 2070 2099 Fig. 3 C hanges in mean annual daily precipitation for the years around 2020, 2030 and 2080 (mmday~') according to the HadCM2SUL transient simulation. Redrawn with permission from a map provided by the LINK programme. Climate change and sustainable water resources 551 which assumes only half the sulphate emissions of the previous model and these results are eagerly awaited. Conversely, HadCM2SUL did not consider the indirect effects of sulphate aerosols on clouds and rainfall, only the radiative effects, and, in this respect, the net effects are likely to be underestimated from the hydrological viewpoint. Effects on snowmelt and extreme events Of all the hydrological extreme events, the trend in snowmelt floods is likely to be most readily and accurately predicted. This will centre on the seasonal altitudinal and latitudinal shift in the "freezing line". The result is likely to be generally decreased snowfall over most regions, combined with more rapid melting at lower altitudes and latitudes, reducing snow cover durations. Flood risk is likely to increase on rivers like the Rhine issuing from the Alps due to more rapid melting, despite some reduction in snowfall and snow cover in France and Switzerland (Kwadijk, 1993). Higher temperatures could also maintain spring flood levels on the Danube despite less snow cover, but the floods will occur slightly earlier (Gauzer, 1993). Hence, earlier spring melt events, more rain-on-snow events and more mid-winter melt events are to be expected. Thinner ice covers on rivers and lakes could also reduce the frequency of ice jam floods. Despite a general rise in snowlines, however, not all basins will respond in the same way. Collins (1989) found that basins with high levels of glacierization are more sensitive to temperature changes, whereas those with less glacier ice respond more to changes in the total snowfall receipts. Some increase in snowfall at higher altitudes is possible given increased oceanic evaporation, increased air temperatures and vapour pressures and increased instability, and some larger ice masses could grow (e.g. Woo, 1996). In other cases, meltwater floods could reduce after an initial increase, as ice masses contract (e.g. Quintela et al, 1996). Although actual snowfall amounts are likely to be lower, the dominance of the Bergeron precipitation-forming process is likely to be maintained in mid to high latitudes as convective instability and higher vapour pressures will probably outweigh the rise in the altitude of the zero isotherm (Jones, 1996b). This could raise rainfall intensities and increase local summer floods. This leads to a consideration of storminess and rain-generated floods. There is a general perception that storm frequency and intensity will increase globally as a result of increases in air temperatures, SSTs, vapour pressure and relative humidity, convection and land-sea contrasts. In northern Europe this could be intensified during the earlier phases of global warming by a temporary increase in the poleward thermal gradient in high mid-latitudes, whilst higher SSTs coexist with Arctic sea ice in the North Atlantic. But this phase should pass as the ice retreats and the equilibrium scenario should see a net reduction in the poleward thermal gradient worldwide. The limited number of studies that exist in this field, tend to support these predictions. They remain largely statistical estimates rather than physically-based predictions (cf. Wigley, 1989; Gregory et al, 1990; Beven, 1993; Holt & Jones, 1996a), although Gellens (1991) found an increase in winter floods and summer droughts in Belgium using a conceptual hydrological model. 552 J. A. A. Jones Drought and low flow frequencies are of major concern to water managers and there are widespread fears that these may increase. Many regions are predicted to experience a reduction in summer precipitation combined with increased évapotranspiration losses. Holt & Jones (1996b) found that the scenarios from the UKTR transient run year 75 would cause regular autumn droughts in Wales, with flows perhaps reduced by 25%. Pilling et al. (1998) calculated that the HadCM2GHG scenario for 2080 is likely to reduce summer rainfall by 14% in central Wales, and Pilling & Jones (1999) predict that large areas of southeast England could see late summer flows reduced by 50% or more. In Greece, Mimikou & Baltas (1997) concluded that lower reservoir inflows under the UKTR scenario would require increases of 25-50% in reservoir storage volumes to maintain the current level of risk, and 12-38% under the Hadley Centre high resolution equilibrium scenario, UKHI. Lower and earlier meltwater discharges are likely to increase summer drought problems in regions like the Canadian prairies and central Europe, although the exact response may change over time and will be different for nival runoff regimes than for pluvial and mixed regimes (cf. Gauzer, 1993; Kwadijk, 1993; Woo, 1996). UNCERTAINTIES AND RESEARCH PRIORITIES FOR IMPROVING PREDICTIONS Major advances have been made in the last few years in the physical basis and spatial resolution of GCMs and we are beginning to emerge from the era of "sensitivity studies" and approach the stage of true prediction. Crucial improvements have been made in introducing and modelling oceanic circulation and atmosphere-ocean interactions, in cloud physics, and in vertical and horizontal resolution. The resolution of the Alpine Low in the high resolution European window in UKTR was an important advance (Hadley Centre, 1995). Valuable progress is also being made outside GCMs in upscaling and downscaling procedures to link GCMs with hydrological and biological models, especially through the IGBP's BAHC project (Biospheric Aspects of the Hydrological Cycle) (Bass et al, 1996). Future experiments introducing variations in solar activity and volcanic eruptions planned for the new HadCM3 model will further improve simulations of the atmospheric system. Nevertheless, there are still several difficult and critical scientific problems currently occupying the modellers that the "user" hydrologist needs to appreciate. First, both temperature and precipitation are very sensitive to cloud processes, especially upon the altitude, thickness, updraft intensity, temperature and density of ice crystals. Experiments at the Hadley Centre have shown reductions of up to 2°C in global warming if ice crystals are allowed to fall out of clouds preferentially (Rowntree, 1990; Mitchell, 1991). However, increased convective activity and vapour pressure in a warmer climate should increase the vertical development of clouds and probably boost the Bergeron process and increase rainfall intensities. According to the Clausius-Clapeyron theory, vapour pressure should increase by 6% per °C of warming. It is thought that just under half of this is likely to be converted into precipitation. This percentage seems to be confirmed by the average results from HadCM2. However, Probable Maximum Precipitation formulae suggest that extreme rainfall events use Climate change and sustainable water resources 553 vapour content more efficiently, i.e. a higher percentage of vapour is transformed into rainfall, leading to higher rainfall intensities. Secondly, both precipitation and temperatures on land, especially in maritime regions, are sensitive to SSTs and ocean currents. Doubts have been expressed about the future stability of the ocean thermohaline circulation or "conveyor belt", especially in the North Atlantic, given an expected increase in the discharge of fresh meltwater into the Arctic Ocean. This could stifle the production of dense "deep water" that drives the circulation and so cause cooling in western Europe. The latest AOGCMs are now capable of testing this hypothesis. Interestingly, experiments with the new Hadley Centre model suggest that any associated southward shift in the Gulf Stream and slowdown in the rate of circulation will still cool western Europe less than the amount of direct warming from greenhouse gases (Meteorological Office, 1998). Further experiments of this nature have become realizable with the increased oceanic resolution (1.25° x 1.25°) in the new HadCM3 model. Thirdly, there is still a tremendous amount to be learnt about interactions with the biosphere. This includes the oceans. It has been known for some years that the role of the oceans in the carbon balance is not simply one of chemical solution and release. Phytoplankton are an important carbon sink. More recently, phytoplankton have emerged as an important source of dimethylsulphide (DMS) which can form into atmospheric aerosols that act as condensation nuclei and encourage cloud formation, just like the sulphate aerosols derived from SO2 emissions. More DMS is likely to be produced in a warmer world, which would mean more condensation nuclei competing for moisture and smaller droplets. Since the reflectivity of clouds is inversely proportional to the mean droplet radius, these changes could have a marked effect on the heat balance (Jones, 1993). Unfortunately, there is still a lot to be learnt about how phytoplankton will react to global warming and how much their genetic material might be damaged as a result of the ozone hole. On land, estimating evapotranspirational losses is still difficult. Far more variables are involved than with precipitation: indeed, the complexity increases from temperature through precipitation to évapotranspiration. Radiation, temperature, soil moisture content, wind, relative humidity and atmospheric CO2 all affect the physiological responses of plants. Carbon dioxide-fertilization is likely to increase biomass and absorb C0 2 , but also increase interception and evapotranspirational losses. The type of plant carbon fixation biochemistry also affects response: C3 plants like wheat will tend to grow more than C4 plants like maize and rice, but their response is particularly sensitive to soil moisture content. Indeed, soil moisture itself is still a difficult parameter to estimate, yet it could have important climatic feedback especially on rainfall. Since all trees have C3 metabolisms, their accelerated growth is likely to increase interception and canopy evaporation losses. This alone could substantially reduce runoff, as indicated by the results of the UK Institute of Hydrology paired catchment experiment in mid-Wales, where yields from the forested basin are 15-20% lower mainly because of interception losses (e.g. Hudson & Oilman, 1993). There is also evidence that temperature and CO2 concentrations may act in opposite directions upon stomatal resistance. Increased temperature should reduce stomatal resistance and increase transpiration, but CO2 may stimulate closure or even 554 J. A. A. Jones cause a long-term reduction in the number of stomata (Woodward, 1987; Jones, 1996b). The balance could be difficult to resolve. Finally, two observations on the actual output from recent GCMs: actual evapotranspirational losses and soil moisture are sensitive to the frequency and intensity of rainfall, but current GCM output is not giving this information. Also, current GCMs are giving important information on the seasonal pattern of warming. They are indicating that most warming will occur at night and in the winter in Europe, in which case the effects upon actual evapotranspirational losses could be less than expected from the average increase in temperature. These complexities suggest some obvious research priorities, not least for hydrologists (Jones, 1996c). More information is needed on the magnitude/frequency distributions of events. Stochastic Weather Generators seem to offer a promising way forward here (cf. BAHC, 1993; Bârdossy et al, 1993; Wilby et al, 1998; Pilling et al, 1998). More needs to be done to link GCM output with hydrological simulation models at all scales, and this will involve more work on physically-based models for large basins, e.g. the Rhine and Mississippi (cf. Parmet & Mann, 1993). The interactive responses of vegetation and soils to climate change need to be built into hydrological models, and this must include the important question of including the effects of changes in land use and agriculture, which may be partly in response to climate change. CONCLUSIONS There are likely to be reduced resources in the subtropics and mid-latitudes equatorwards of 50°N or 40° S, and an increase in resources polewards of 55°N or 40°S. Seasonal shortages are likely even where there is a net annual increase, e.g. in the UK. Most of the loss in resources will be due to reduced precipitation. Increases in precipitation are likely to be mainly in rainfall and the percentage in snowfall will be reduced. This means reduced volumes and durations for glaciers, snow cover and freshwater ice. Extreme events are likely to occur more frequently in most climatic regions, though meltwater and ice jam floods could eventually decrease. Some areas show current trends that fit well with the predicted climate change trends, but many do not. Few regions show widespread and consistent trends in discharge (cf. Chiew & McMahon, 1996). However, some confusion may be introduced by different local sensitivities to climatic variability and cycles, e.g. El Nino, and to human modification of catchments. Even so, there is some GCM evidence that precipitation trends may not be as unidirectional as temperature trends. Placed in perspective, climate change is only one of a wide range of barriers to sustainability, many of which are likely to be problematical long before there is real pressure from global warming. Population growth, urbanization and profligate use of resources are the main hindrances. Nevertheless, the effects of climate change are likely to become first apparent in extreme events, which could hit "net winners" as badly as, if not worse than, regions where average water stress will increase. In western Europe, increased seasonality in precipitation, with wetter winters and drier summers, Climate change and sustainable water resources 555 will provide a background for increased flooding and reservoir spillage in winter and drought and low flow events in summer (Holt & Jones, 1996a). At the very least, this will require substantial increases in the sizes of reservoirs to maintain public water supply (Cole et al, 1991) and it may even be a critical factor in decisions to undertake long-distance inter-basin transfer (NRA, 1994). There are still substantial uncertainties in climate change prediction and the best informed water management authorities are opting at most to keep a "watching brief on the predictions, perhaps considering the possible future need to raise the height of dams and dykes and designing new works accordingly, rather than taking immediate constructive action. This is reasonable: forewarned is forearmed. Meanwhile, there is considerable work to be done in curbing demand, increasing public awareness, controlling leakage and protecting surface water and groundwater from pollution. There is also an urgent need to improve national planning and water resources assessment over most of the globe (WMO/UNESCO, 1991) and to maintain the effectiveness of its foundation, hydrological observations, within a general context of declining networks (Burn, 1997; Kundzewicz, 1997). Beyond this, there is a widespread requirement for water agencies, and the governments that give them their powers and their briefs, to appreciate more the intimate relationships between water resources and the wider environment (cf. Falkenmark, 1997). This may be summarized in two dicta: (a) Water should be managed in such a way as to minimize the interference with nature and to maximize the benefits for nature. (b) The environment should be managed in such a way as to minimize adverse impacts on and maximize benefits for water resources or flood hazard. Without following these dicta, there is no hope of achieving true sustainability. Acknowledgements The author wishes to thank Dr David Viner and Ruth Dogherty of the University of East Anglia and The Hadley Centre for kindly providing the rainfall and runoff maps from HadCM2SUL, under the Climate Link Programme, and Dr Nigel Arnell of the University of Southampton for permission to reproduce the runoff map from HadCM2. REFERENCES Anderson, E. W. (1991) The violence of thirst. Geogr. Magazine 63(5), 31-34. Arnell, N. W. (1996) Global Warming, River Flows and Water Resources. John Wiley, Chichester, UK. Arnell, N. W. (1998) Impacts of climate change on water resources. In: Climate Change and its Impacts: some highlights from the ongoing UK research programme—a first look at results from the Hadley Centre's new climate model, 7. Dept of the Environment, Transport and the Regions/The Met. Office, Met. Office Communications, UK. Arnell, N. W. & King, R. (1997) The impact of climate change on water resources. In: Climate Change and its Impacts: a Global Perspective, 10-11. Dept of the Environment, Transport and the Regions/The Met. Office, Met. Office Graphics Studio, UK. Arnell, N. W. & Reynard, N. S. (1993) Impact of climate change on river flow regimes in the United Kingdom. Report to the Dept of the Environment, Institute of Hydrology, Wallingford, UK. BAHC (Biospheric Aspects of the Hydrological Cycle) (1993) Development of the Weather Generator (Chair: B. Bass). BAHC Focus 4: The Weather Generator Project, Activity 4.2. BAHC Core Project Office, Freie Universitat Berlin, Germany. Bârdossy, A., Duckstein, L., Bogardi, I. & Muster, H. (1993) Stochastic precipitation modeling using fuzzy classification of atmospheric circulation patterns. In: Proc. Conf. on Stochastic and Statistical Methods in Hydrology and Environmental Engineering (ed. by K. Hipel), vol. 3. Kluwer, Dordrecht, The Netherlands. 556 J. A. A, Jones Bass, B., Akkur, N., Russo, J. & Zack, J. (1996) Modelling the biospheric aspects of the hydrological cycle: upscaling processes and downscaling weather data. In: Regional Hydrological Response to Climate Change (ed. by J. A. A. Jones, C. M. Liu, M.-K. Woo & H.-T. Kung), 39-62. Kluwer, Dordrecht, The Netherlands. Beven, K. J. (1993) Riverine flooding in a warmer Britain. Geogr. J. 159(2), 157-161. Bonell, M., Hufschmidt, M. M. & Gladwell, J. S. (eds) (1993) Hydrology and Water Management in the Humid Tropics. Cambridge University Press, Cambridge, UK. Burn, D. H. (1997) Hydrological information for sustainable development. Hydro!. Sci. J. 42(4), 481^492. Chiew, F. H. S. & McMahon, T. A. (1996) Trends in historical streamflow records. In: Regional Hydrological Response to Climate Change (ed. by J. A. A. Jones, C. M. Liu, M.-K. Woo & H.-T. Kung), 63-68. Kluwer, Dordrecht, The Netherlands. Cole, J. A., Slade, S., Jones, P. D. & Gregory, J. M. (1991) Reliable yield of reservoirs and possible effects of climate change. Hydrol. Sci. J. 36(6), 579-598. Collins, D. N. (1989) Influence of glacierisation on the response of runoff from Alpine basins to climate variability. In: Conf. on Climate and Water, vol. 1, 319-328. Academy of Finland, Helsinki. Conway, D. (1998) Recent climatic variability and future climate change scenarios for Great Britain. Progr. Phvs. Geogr. 22(3), 350-374. Falkenmark, M. (1997) Society's interaction with the water cycle: a conceptual framework for a more holistic approach. Hydrol. Sci. J. 42(4), 451-466. Gauzer, B. (1993) Effect of air temperature change on the Danube flow regime (in Hungarian). VITUKI, Budapest, Hungary. Gellens, D. (1991) Impacts of a C02-induced climatic change on river flow variability in three rivers in Belgium. Earth Surf. Processes Land/. 16, 619-625. Glazovsky, N. F. ( 1990) The Aral Crisis: the Source, the Current Situation and the Ways to Solving it. Nauka, Moscow. Gregory, J. M., Jones, P. D. & Wigley, T. M. L. (1990) Climatic change and its potential effect on UK water resources. Report to the Water Research Centre, Medmenham, UK, Parts 1 and 2. Hadley Centre (1995) Modelling climate change 1860-2050. Meteorological Office, Bracknell, UK. Holt, C. P. & Jones, J. A. A. (1996a) Sensitivity analyses on the impact of global warming on water resources in Wales. In: Regional Hydrological Response to Climate Change (ed. by J. A. A. Jones, C. M. Liu, M.-K. Woo & H.-T. Kung), 317-335. Kluwer, Dordrecht, The Netherlands. Holt, C. P. & Jones, J. A. A. (1996b) Equilibrium and transient global wanning scenario implications for water resources in Wales. Wat. Resour. Bull. 32(4), 711-721. Hudson, J. A. & Oilman, K. (1993) Long-term variability in the water balances of the Plynlimon catchments. J. Hydrol. 143,355-380. Hufschmidt, M. M. & Tejwani, K. G. (1993J Integrated water resource management—meeting the sustainability challenge. UNESCO International Hydrological Programme Humid Tropics Programme Series no. 5, UNESCO, Paris. 1PCC (Intergovernmental Panel on Climate Change) (1996) Climate Change 1995 (ed. by J. T. Houghton, B. A. Callander & S. K. Varney). Cambridge University Press, Cambridge, UK. Johns, T. C , Carnell, R. E., Crossley, J. F., Gregory, J. M., Mitchell, J. F. B., Senior, C. A., Tett, S. F. B. & Wood, R. A. (1997) The second Hadley Centre coupled atmosphere-ocean GCM: model description, spinup and validation. Climate Dynamics 13, 103-134. Jones, A. (1993) Cloud microphysics. In: Progress Report 1990-1992 and Future Programme of Research, 49. Hadley Centre, Bracknell, UK. Jones. J. A. A. (1996a) Predicting the hydrological effects of climate change. In: Regional Hydrological Response to Climate Change (ed. by J. A. A. Jones, C. M. Liu, M.-K. Woo & H.-T. Kung), 1-20. Kluwer, Dordrecht, The Netherlands. Jones. J. A. A. (1996b) Current evidence on the likely impact of global warming on hydrological regimes in Europe. In: Regional Hydrological Response to Climate Change (ed. by J. A. A. Jones, C. M. Liu, M.-K. Woo & H.-T. Kung), 87-131. Kluwer, Dordrecht, The Netherlands. Jones, J. A. A. (1996c) The impact of global warming on hydrology and future research priorities. In: Regional Hydrological Response to Climate Change (ed. by J. A. A. Jones, C. M. Liu, M.-K. Woo & H.-T. Kung), 419^422. Kluwer, Dordrecht, The Netherlands. Jones, J. A. A. (1997) Global Hydrology: Processes, Resources and Environmental Management. Addison Wesley Longman, Harlow, UK. Jones, J. A. A., Liu, C. M., Woo, M.-K. & Kung, H.-T. (eds) ( 1996) Regional Hydrological Response to Climate Change. Kluwer, Dordrecht, The Netherlands. Kwadijk, J. C. J. (1993) The impact of climate change on the discharge of the River Rhine, Netherlands Geographical Studies no. 171. Kundzewicz, Z. W. ( 1997) Water resources for sustainable development. Hydrol. Sci. J. 42(4), 467-480. Lvovich, M. 1. ( 1979) World water resources and their future. American Geophysical Union, Washington, DC. McCaffrey, S. C. (1992) Water, politics, and international law. In: Water in Crisis: a Guide to the World's Fresh Water Resources (ed. by P. H. Gleick), Chapter 8, 92-104. Oxford University Press, Oxford, UK. Meteorological Office (1997) Climate change and its impacts: a global perspective. Met. Office & Dept of the Environment, Transport and the Regions, Bracknell, UK. Meteorological Office (1998) Climate change and its impacts: some highlights from the ongoing UK research programme: a first look at the results from the Hadley Centre's new climate model. Met. Office & Dept of the Environment, Transport and the Regions, Bracknell, UK. Mimikou, M. A. & Baltas, E. A. (1997) Climate change impacts on the reliability of hydroelectric energy production. Hydrol. Sci. J. 42(5), 661-678. Climate change and sustainable water resources 557 Mitchell, J. F. B. (1991) The equilibrium response to doubling atmospheric CO2. In: Greenhouse-gas-induced Climatic Change: a Critical Appraisal of Simulations and Observations (ed. by M. E. Schlesinger), 49-61. Elsevier, Amsterdam, The Netherlands. Mitchell, .1. F. B., Johns, T. C , Gregory, J. M. & Tett, S. F. B. (1995) Climate response to increasing levels of greenhouse gases and sulphate aerosols. Nature 376, 501-504. National Rivers Authority (NRA) ( 1994) Water—Nature's Precious Resource. An Environmentally Sustainable Water Resources Development Strategy for England and Wales. HMSO, London. Parmet, B. W. A. H. & Mann, M. A. M. (1993) Influence of climate change on the discharge of the River Rhine—a model for the lowland area. En: Exchange Processes at the Land Surface for a Range of Space and Time Scales (ed. by H.-J. Bolle, R. A. Feddes & J. D. Kalma) (Proc. Yokohama Symp., July 1993), 469^177. IAHS Publ. no. 212. Pilling, C. & Jones, J. A. A. (1999) High resolution climate change scenario implications for British runoff. Hydro!. Processes 13. Pilling, C , Wilby, R. & Jones, J. A. A. (1998) Downscaling of catchment hydrometeorology from GCM scenarios using airflow indices. In: Hydrology in a Changing Environment (ed. by H. S. Wheater & C. Kirby), vol. 1, 191-208. John Wiley/British Hydrological Society, Chichester, UK. Quintela, R. M., Scarpati, O. E., Speschia, L. B. & Capriolo, A. D. (1996) The probable impact of global change on the water resources of Patagonia, Argentina. In: Regional Hydrological Response to Climate Change (ed. by J. A. A. Jones, C. M. Liu, M.-K. Woo & H.-T. Kung), 389^107. Kluwer, Dordrecht, The Netherlands. Rosbjerg, D., Boutayeb, N.-E., Gustard, A., Kundzewicz, Z. W. & Rasmussen, P. F. (eds) (1997) Sustainability of Water Resources under Increasing Uncertainty (Proc. Rabat Symp., April-May 1997). IAHS Publ. no. 240. Rowntree, P. R. (1990) Estimates of future climatic changes over Britain. Part 2: Results. Weather 45(3), 79-89. Simonovic, S. P. (ed.) (1997) Sustainable Development of Water Resources. Special Issue Hydrol. Sci. J. 42(4). Simonovic, S. P., Kundzewicz, Z. W., Rosbjerg, D. & Takeuchi, K. (eds) (1995) Modelling and Management of Sustainable Basin-Scale Water Resource Systems (Proc. Boulder Symp., July 1995). IAHS Publ. no. 231. Takeuchi, K. ( 1997) Least marginal environmental impact rule for reservoir development. Hydrol. Sci. J. 42(4), 583-597. Tickell, SirC. (1991) The human species: a suicidal success? Geogr. J. 159(2), 219-226. UN (1970) Integrated river basin development: report of a panel of experts. UN Dept of Economic and Social Affairs, New York. UN ( 1991 ) Long-Range World Population Projections: Two Centuries of Population Growth 1950-2150. United Nations Population Division, New York. Wigley, T. M. L. (1989) The effect of changing climate on the frequency of absolute extreme events. Climate Monitor 17, 44-55. Wilby, R. L., Hassan, H. & Hanaki, K. (1998) Statistical downscaling of hydrometeorological variables using general circulation model output. J. Hydrol. 205(1998), 1-19. WMO (1997) Comprehensive assessment of the freshwater resources of the world Report to the UN Commission on Sustainable Development, World Meteorological Organization, Geneva, Switzerland. WMO/UNESCO ( 1991 ) Report on water resources assessment. WMO, Geneva, Switzerland. WMO/UNESCO (1997) Assessment of Water Resources and Water Availability in the World. World Meteorological Organization, Geneva, Switzerland. Woo, M.-K. (1996) Hydrology of northern North America under global warming. In: Regional Hydrological Response to Climate Change (ed. by J. A. A. Jones, C. M. Liu, M.-K. Woo & H.-T. Kung), 73-86. Kluwer, Dordrecht, The Netherlands. Woodward, F. 1. (1987) Stomatal numbers are sensitive to increases in CO2 from pre-industrial levels. Nature 327, 617618.