Water in a Changing World A6ojooH ul sanssI
Water in a Changing World Issues in Ecology Published by the Ecological Society of America Number 9, Spring 2001
Issues in Ecology Number 9 Spring 2001 Water in a Changing World SUMMARY Life on land and in the lakes,rivers,and other freshwater habitats of the earth is vitally dependenton renewable fresh water,a resource that comprises only a tiny fraction of the global water pool.Humans rely on terorking.indsuses as weasp waterfowl,transportation In many regions of the world,the amount and quality of water available to meet human needs are already limited.The gap between freshwater supply and demand will widen during the coming century as a result of climate change example,accessible runoff of fresh water is unlikely to increase more than 1 percent,yet the earth's population is expected to grow by one third.Unless humans use water more efficiently,the impacts of this imbalance in supply and demand will diminish the services that freshwater ecosystems provide,increase the number of aquatic species facing extinction,and further fragment wetlands,rivers,deltas,and estuaries. Based on the scientific evidence currently available,we conclude that: More than half of the world's accessible freshwater runoff is already appropriated for human use More than a billion people currently lack access to clean drinking water.and almost three billion lack basic sanitation services. Because human population will g ow faster than any increase in accessible supplies of fresh water the amount of fresh water available per person will decrease in the coming century. Climate change will intensify the earth's water cycle in the next century,generally increasing rainfall, urrence of storms.and significantly altering the nutrient cycles in At least 90 percent of river flows in the United States are strongly affected by dams,reservoirs. interbasin diversions,and irrigation withdrawals that fragment natural channels. Globally,20 percent of freshwater fish species are threatened or extinct,and freshwater species make up 47 percent of all federally listed endangered animals in the United States. of water-policy successes. Better monitoring.assessment,and forecasting of water resources would help government agencies allo cate water more efficiently among competing needs.Currently in the United States.at least six federal departments and twenty agencies share responsibilities for various aspects of the water ycle.We believe eithe creation of a single panel with members drawn from each department or else oversight by a central agency is needed in order to develop a well-coordinated national plan that acknowledges the diverse and competing pressures on freshwater systems and assures efficient use and equitable distribution of these resources. Cover(clockwise from top):Homestead,Kalahari Desert of South Africa (R.Jackson):Coastal zone of Serra da Arrabida The W ater (H.Be 1870 everde Cloud ica(R on USA (R. nd C ional Park.Sy and riparian zone.Gardner River of Yellowstone kson：吵and the town of Flores,Guatemala(R. Jackson)
1 Issues in Ecology Number 9 Spring 2001 SUMMARY Life on land and in the lakes, rivers, and other freshwater habitats of the earth is vitally dependent on renewable fresh water, a resource that comprises only a tiny fraction of the global water pool. Humans rely on renewable fresh water for drinking, irrigation of crops, and industrial uses as well as production of fish and waterfowl, transportation, recreation, and waste disposal. In many regions of the world, the amount and quality of water available to meet human needs are already limited. The gap between freshwater supply and demand will widen during the coming century as a result of climate change and increasing consumption of water by a growing human population. In the next 30 years, for example, accessible runoff of fresh water is unlikely to increase more than 10 percent, yet the earths population is expected to grow by one third. Unless humans use water more efficiently, the impacts of this imbalance in supply and demand will diminish the services that freshwater ecosystems provide, increase the number of aquatic species facing extinction, and further fragment wetlands, rivers, deltas, and estuaries. Based on the scientific evidence currently available, we conclude that: • More than half of the worlds accessible freshwater runoff is already appropriated for human use. • More than a billion people currently lack access to clean drinking water, and almost three billion lack basic sanitation services. • Because human population will grow faster than any increase in accessible supplies of fresh water, the amount of fresh water available per person will decrease in the coming century. • Climate change will intensify the earths water cycle in the next century, generally increasing rainfall, evaporation rates, and the occurrence of storms, and significantly altering the nutrient cycles in land-based ecosystems that influence water quality. • At least 90 percent of river flows in the United States are strongly affected by dams, reservoirs, interbasin diversions, and irrigation withdrawals that fragment natural channels. • Globally, 20 percent of freshwater fish species are threatened or extinct, and freshwater species make up 47 percent of all federally listed endangered animals in the United States. Growing demands on freshwater resources are creating an urgent need to link research with improved water management, a need that has already resulted in a number of water-policy successes. Better monitoring, assessment, and forecasting of water resources would help government agencies allocate water more efficiently among competing needs. Currently in the United States, at least six federal departments and twenty agencies share responsibilities for various aspects of the water cycle. We believe either creation of a single panel with members drawn from each department or else oversight by a central agency is needed in order to develop a well-coordinated national plan that acknowledges the diverse and competing pressures on freshwater systems and assures efficient use and equitable distribution of these resources. Water in a Changing World Cover (clockwise from top): Homestead, Kalahari Desert of South Africa (R. Jackson); Coastal zone of Serra da Arrábida, Portugal (R. Jackson); The Water Seller (H. Bechard, Egypt ca. 1870); Monteverde Cloud Forest, Costa Rica (R. Jackson); Little Colorado River, Grand Canyon National Park, USA (R. Jackson); Elk and riparian zone, Gardner River of Yellowstone National Park, USA (R. Jackson); and the town of Flores, Guatemala (R. Jackson).
Issues in Ecology Number 9 Spring2001☐ Water in a Changing World Robert B.Jackson,Stephen R.Carpenter,Clifford N.Dahm, Diane M.MeKnight,Robert J.Naiman.Sandra L.Postel,and Steven W.Running INTRODUCTION power,can be achieved only by damming.diverting.or creating other major changes to natural water flows.Such Life on earth depends on the continuous flow of changes often diminish or preclude other instream ben- materials through the air,water.soil,and food webs of efits of fresh water.such as providing habitat for aquatic the biosphere.The movement of water through the hy life or maintaining suitable water quality for human use drological cycle est of these Theecological, cono mic be efits tha delivering an estim neters (km reshwater ystems provide.and the trade-offs between of water to the land each year as snow and rainfall.Solar consumptive and instream values,will change dramati energy drives the hydrological cycle,vaporizing water cally in the coming century.Already,over the past one from the surface of oceans,lakes,and rivers as well as hundred years,both the amount of water humans with- from soils and plants (evapotranspiration).Water vapor draw worldwide and the land area under irrigation have rises into the atmosphere where it cools,condenses.and risen exponentially (Figure 1).Despite this greatly in hwate ad sumption the basic water eds of rld are not b eing met Currently, Renewable fresh water provides many services billion lack basic sanitation services.These deprivations essential to human health and well being.including water cause approximately 250 million cases of water-related for drinking.industrial production,and irrigation,and the diseases and five to ten million deaths each vear.also ent unmet needs limit our ability to adapt to future ch- s in ate supplies and distribu ion.Many nels nextractive or instream benefits).including flood systems des signed topr control,transportation,recreation.waste processing.hy- matic conditions may be ill prepared to adapt to future droelectric power,and habitat for aquatic plants and ani- changes in climate,consumption,and population.While a mals.Some benefits,such as irrigation and hydroelectric global perspective on water withdrawals is important for 00 45 400 30 2 20 1901p01920019001601701602000 Year Figure I-Global data for hum pulation.water withdrawals.and irrigated land area. Redrawn and updated rom Gleick()
2 Issues in Ecology Number 9 Spring 2001 Robert B. Jackson, Stephen R. Carpenter, Clifford N. Dahm, Diane M. McKnight, Robert J. Naiman, Sandra L. Postel, and Steven W. Running Water in a Changing World Figure 1 — Global data for human population, water withdrawals, and irrigated land area from 1900 to 2000. Redrawn and updated from Gleick (1998). INTRODUCTION Life on earth depends on the continuous flow of materials through the air, water, soil, and food webs of the biosphere. The movement of water through the hydrological cycle comprises the largest of these flows, delivering an estimated 110,000 cubic kilometers (km3) of water to the land each year as snow and rainfall. Solar energy drives the hydrological cycle, vaporizing water from the surface of oceans, lakes, and rivers as well as from soils and plants (evapotranspiration). Water vapor rises into the atmosphere where it cools, condenses, and eventually rains down anew. This renewable freshwater supply sustains life on the land, in estuaries, and in the freshwater ecosystems of the earth. Renewable fresh water provides many services essential to human health and well being, including water for drinking, industrial production, and irrigation, and the production of fish, waterfowl, and shellfish. Fresh water also provides many benefits while it remains in its channels (nonextractive or instream benefits), including flood control, transportation, recreation, waste processing, hydroelectric power, and habitat for aquatic plants and animals. Some benefits, such as irrigation and hydroelectric power, can be achieved only by damming, diverting, or creating other major changes to natural water flows. Such changes often diminish or preclude other instream benefits of fresh water, such as providing habitat for aquatic life or maintaining suitable water quality for human use. The ecological, social, and economic benefits that freshwater systems provide, and the trade-offs between consumptive and instream values, will change dramatically in the coming century. Already, over the past one hundred years, both the amount of water humans withdraw worldwide and the land area under irrigation have risen exponentially (Figure 1). Despite this greatly increased consumption, the basic water needs of many people in the world are not being met. Currently, 1.1 billion people lack access to safe drinking water, and 2.8 billion lack basic sanitation services. These deprivations cause approximately 250 million cases of water-related diseases and five to ten million deaths each year. Also, current unmet needs limit our ability to adapt to future changes in water supplies and distribution. Many current systems designed to provide water in relatively stable climatic conditions may be ill prepared to adapt to future changes in climate, consumption, and population. While a global perspective on water withdrawals is important for
Issues in Ecology Number 9 Spring 2001 ensuring sustainable water use.it is in sufficient for regional and local stabil- ity.How fresh ter is m ed in pa h sheds is the key to sustainable water management. The goal of this report is to describe key features of human-induced changes to the global water cycle.The on on wate availabil ity and on purifi on costs have been addressed previously in /ssues in Ecol ogy.We focus instead on current and potential changes in the cycling of wa- ter that are especially relevant for eco ng the global wate igure The renewable freshwater cycle inunits kmy cluding it current state and historic io000rnd xes (olac hPreciptationoS context. We next examine the extent er recy to which human activities currently al 40.000 evap ter the water cycle and may affect it in t15,000.0 the future.These changes include di- tive exchange wit rect actions.such as dam construction andindire ct in cts such as hos result from human-driven climate change.We examine human appropriation of fresh water of the so-called greenhouse gases (others include carbon globally.from both renewable and non-renewable sources. dioxide.nitrous oxide.and methane)that warm the earth The report ends by discussing changes in water use that by trapping heat in the atmosphere.Water vapor contrib- may be especially important in the future.We highlight utes approximately two-thirds of the total warming that greenhouse gases supply.Without these gases.ther rface te ture of the ould be well belov freezing,an THE GLOBAL WATER CYCLE the planet.Equally important for life,atmospheric water turns over every ten days or so as water vapor condenses Surface Water and rains to earth and the heat of the sun evaporates new supplies of vapor from the liquid reservoirs on earth. Most of the earth is covered by water,more than one billion km of it.The Solar energy typically vaporates about 425.000 ma water km of ocean wate ach year Most of this rain however,is in forms una ed or fres back directly to th e oceans,but approximately 10 pe water ecosystems.Less than 3 percent is fresh enough cent falls on land.If this were the only source of rainfall to drink or to irrigate crops,and of that total,more than average precipitation across the earth's land surfaces two-thirds is locked in glaciers and ice caps.Freshwater would be only 25 centimeters (cm)a vear.a value typical lakes and rivers hold 100,000 km3 globally,less than for deserts or semi-arid regions.Instead.a second.large one ten-thousandth of all water on earth (Figure 2). urce of water is rec cled fron plants and the tant ine vapor in impor through evapo thi ate and on the cyc eve source creates a direct feedback between the land face and regional climate.The cycling of other materials atmosphere at any time.This tiny fraction,however,is such as carbon and nitrogen(biogeochemical cycling)is vital for the biosphere.Water vapor is the most important strongly coupled to this water flux through the patterns
3 Issues in Ecology Number 9 Spring 2001 ensuring sustainable water use, it is insufficient for regional and local stability. How fresh water is managed in particular basins and in individual watersheds is the key to sustainable water management. The goal of this report is to describe key features of human-induced changes to the global water cycle. The effects of pollution on water availability and on purification costs have been addressed previously in Issues in Ecology. We focus instead on current and potential changes in the cycling of water that are especially relevant for ecological processes. We begin by briefly describing the global water cycle, including its current state and historical context. We next examine the extent to which human activities currently alter the water cycle and may affect it in the future. These changes include direct actions, such as dam construction, and indirect impacts, such as those that result from human-driven climate change. We examine human appropriation of fresh water globally, from both renewable and non-renewable sources. The report ends by discussing changes in water use that may be especially important in the future. We highlight some current progress and suggest priorities for research, emphasizing examples from the United States. THE GLOBAL WATER CYCLE Surface Water Most of the earth is covered by water, more than one billion km3 of it. The vast majority of that water, however, is in forms unavailable to land-based or freshwater ecosystems. Less than 3 percent is fresh enough to drink or to irrigate crops, and of that total, more than two-thirds is locked in glaciers and ice caps. Freshwater lakes and rivers hold 100,000 km3 globally, less than one ten-thousandth of all water on earth (Figure 2). Water vapor in the atmosphere exerts an important influence on climate and on the water cycle, even though only 15,000 km3 of water is typically held in the atmosphere at any time. This tiny fraction, however, is vital for the biosphere. Water vapor is the most important of the so-called greenhouse gases (others include carbon dioxide, nitrous oxide, and methane) that warm the earth by trapping heat in the atmosphere. Water vapor contributes approximately two-thirds of the total warming that greenhouse gases supply. Without these gases, the mean surface temperature of the earth would be well below freezing, and liquid water would be absent over much of the planet. Equally important for life, atmospheric water turns over every ten days or so as water vapor condenses and rains to earth and the heat of the sun evaporates new supplies of vapor from the liquid reservoirs on earth. Solar energy typically evaporates about 425,000 km3 of ocean water each year. Most of this water rains back directly to the oceans, but approximately 10 percent falls on land. If this were the only source of rainfall, average precipitation across the earths land surfaces would be only 25 centimeters (cm) a year, a value typical for deserts or semi-arid regions. Instead, a second, larger source of water is recycled from plants and the soil through evapotranspiration. The water vapor from this source creates a direct feedback between the land surface and regional climate. The cycling of other materials such as carbon and nitrogen (biogeochemical cycling) is strongly coupled to this water flux through the patterns Figure 2 — The renewable freshwater cycle in units of 103 km3 and 103 km3 /yr for pools (white numbers) and fluxes (black numbers). Total precipitation over land is about 110,000 km3 /yr. Approximately two-thirds of this precipitation is water recycled from plants and the soil (evapotranspiration = 70,000 km3 /yr) while one-third is water evaporated from the oceans that is then transported over land (40,000 km3 /yr). Ground water holds about 15,000,000 km3 of fresh water, much of it fossil water that is not in active exchange with the earths surface.
Issues in Ecology Spring 2001 of plant growth and microbial decc adtiomalecback osition,and this betwee anssthe Colorado iverr nth the size he Ar times smaller. Simila variati occurs contributes two-thirds of the 70 cm of precipitation that continental scales.Average runoff in Australia is only 4 falls over land each year.Taken together,these two cm per year,eight times less than in North America and sources account for the I 10.000 km3 of renewable fresh- orders of magnitude less than in tropical South America water available each year for terrestrial,freshwater,and As a result of these and many other disparities.freshwater estua availability varies dramatically worldwide. greater than the amount of water that evaporates from Ground Water it,the extra 40,000 km of water returns to the oceans. primarily via rivers and underground aquifers.A number Approximately 99 percent of all liquid fresh water of factors affect how much of this water is available for is in underground aquifers(Hgure 2).and at least a quarter human use on its journey to the oceans.These factors of the world's population draws its water from these include whethe the pitation falls as rair lies.Estimates of the global wate timing of precipitati relative 8代 erally trea dwate temperature and sunlight,and the regional topography as if they were balanced.In reality.however,this resource For example,in many mountain regions,most precipita is being depleted globally.Ground water typically turns tion falls as snow during winter,and spring snowmelt causes over more slowly than most other water pools.often in peak flows that flood major river sys. tems.In some tropical regions,mon- soons rather than s owmelt create s sona flo In othe pre cipitation thes to recharge ground water or is stored in wetlands.Widespread loss of wetlands and floodplains.however.reduces their ability to absorb these high flows and speeds the runoff of excess nutrie nts and con aminan ts to estua s and othe coastal e onments ore than halt of all wetlands in the U.S have already been drained,dredged,filled,or planted. Available water is not evenly distributed globally Two thirds of all pitation falls in the t latituo due to greater sola radiation and evaporation there.Daily evaporation from the oceans ranges from 0.4 cm at the equator to less than O.I cm at the poles.Typically,tropical regions also have larg noff.Roughly half of the that falls rainforests in thedeser rainfall and high evaporation rates combine to greatly reduce runoff.The Figure 3-Locations of non-renewable groundw Amazon,for example.carries 15 percent of all water returning to the global 1997
4 Issues in Ecology Number 9 Spring 2001 of plant growth and microbial decomposition, and this coupling creates additional feedbacks between vegetation and climate. This second source of recycled water contributes two-thirds of the 70 cm of precipitation that falls over land each year. Taken together, these two sources account for the 110,000 km3 of renewable freshwater available each year for terrestrial, freshwater, and estuarine ecosystems (Figure 2). Because the amount of rain that falls on land is greater than the amount of water that evaporates from it, the extra 40,000 km3 of water returns to the oceans, primarily via rivers and underground aquifers. A number of factors affect how much of this water is available for human use on its journey to the oceans. These factors include whether the precipitation falls as rain or snow, the timing of precipitation relative to patterns of seasonal temperature and sunlight, and the regional topography. For example, in many mountain regions, most precipitation falls as snow during winter, and spring snowmelt causes peak flows that flood major river systems. In some tropical regions, monsoons rather than snowmelt create seasonal flooding. In other regions, excess precipitation percolates into the soil to recharge ground water or is stored in wetlands. Widespread loss of wetlands and floodplains, however, reduces their ability to absorb these high flows and speeds the runoff of excess nutrients and contaminants to estuaries and other coastal environments. More than half of all wetlands in the U. S. have already been drained, dredged, filled, or planted. Available water is not evenly distributed globally. Two thirds of all precipitation falls in the tropics (between 30 degrees N and 30 degree S latitude) due to greater solar radiation and evaporation there. Daily evaporation from the oceans ranges from 0.4 cm at the equator to less than 0.1 cm at the poles. Typically, tropical regions also have larger runoff. Roughly half of the precipitation that falls in rainforests becomes runoff, while in the deserts low rainfall and high evaporation rates combine to greatly reduce runoff. The Amazon, for example, carries 15 percent of all water returning to the global oceans. In contrast, the Colorado River drainage, which is one-tenth the size of the Amazon, has a historic annual runoff 300 times smaller. Similar variation occurs at continental scales. Average runoff in Australia is only 4 cm per year, eight times less than in North America and orders of magnitude less than in tropical South America. As a result of these and many other disparities, freshwater availability varies dramatically worldwide. Ground Water Approximately 99 percent of all liquid fresh water is in underground aquifers (Figure 2), and at least a quarter of the worlds population draws its water from these groundwater supplies. Estimates of the global water cycle generally treat rates of groundwater inflow and outflow as if they were balanced. In reality, however, this resource is being depleted globally. Ground water typically turns over more slowly than most other water pools, often in Figure 3 — Locations of non-renewable groundwater resources (light blue) and the main locations of groundwater mining (dark gray) (Shiklomanov 1997). The inset shows the location of the High Plains (Ogallala) Aquifer
Issues in Ecology Number 9 Spring 2001 hundreds to tens of thousands of years.although the range face and ground waters for water supply water quality in turnover rates is large.Indeed,a majority of grou nd and aquatic habitats where extraction of s und wate water is not actively tur xceeds recharge rates.the o vater tables the earth's surface at all fosi water. In summer when a high water table is needed to sustair relic of wetter ancient climatic conditions and melting minimum flows in rivers and streams.low groundwater Pleistocene ice sheets that accumulated over tens of levels can decrease low-flow rates.reduce perennial thousands of years.Once used,it cannot readily be stream habitat.increase summer stream temperatures replenished. and impair water quality.trout and salmon species selec The distinction between renewable and non areas of groundwater upwelling in streams to moderate renewable iscritica for water managemen extreme s al tem to keer egg and policy. an three-quarters of un ergroun ange of sur water is non-renewable,meaning it has a replenishment face and ground waters alters the dissolved oxygen and period of centuries or more(Figure 3).The High Plains or nutrient concentrations of streams and dilutes concen Ogallala Aquifer that underlies half a million km2 of the trations of dissolved contaminants such as pesticides and central United States is arguably the largest aquifer in the volatile organic compounds.Because of such links.hu elopment of either ground water or surface wa 20.000%e % tity and quality of the othe The links between surface ar ater the Ogalla ala the primary water source for a fifth of especially important in regions with low rainfall (see Box irrigated U.S.farmland.The extent of irrigated cropland 1,Table I,and Figure 4).Arid and semi-arid regions in the region peaked around 1980 at 5.6 million hectares cover a third of the earth's lands and hold a fifth of the and at pumping rates of about 6 trillion gallons of water a global population.Ground water is the primary source of year.That has since declined somewhat due water for drinking and irrigation in these regions.which gro ndwater depletion ic ch possess m the orld's la est aqu limited re region.However. the average thickness of the charge n aquifers high le to groun declined by more than 5 percent across a fifth of its area water depletion.For example,explo tation of the North in the 1980s alone. em Sahara Basin Aquifer in the I990s was almost twic In contrast,renewable aquifers depend on cur. the rate of replenishment,and many springs associated rent rainfall for refilling and so are vulnerable to changes with this aquifer are drying up.For non-renewable ground- water.For water sources,discussing Edwards Aqu stainabler appropriate rates ction is difficult As with s much, of coal and oil suppl ng water amost any extraction nable portant que has increased four-fold since the tions for society include at what rate groundwater pump exceeds annual recharge rates. Increased water with ing should be allowed,for what purpose,and who if any drawal makes aquifers more susceptible to drought and one will safeguard the needs of future generations in other changes in weather and to contamination from pol the Ogallala Aquifer,for example,the water may be gone lutants and wastes that percolate into the ground wa in as little as a century. Depletion of ground wa can als caus land subs porous sand. gravel or rock HUMAN APPROPRIATION OF FRESHWATER SUPPLY water.The Central Valley of California has lost about 25 Global Renewable Water Supplies km3 of storage in this way,a capacity equal to more than 40 percent of the combined storage capacity of all hu- Growth in global population and water consump man-made reservoirs in the state tion will place additional pr essure on freshwater resources s hav in the cc available several nty the wa im s more sh wa er each tha legally.This view is changing.however.as studies in needed to sustain the world ds population of six billior streams,rivers,reservoirs,wetlands.and estuaries show people(Table 2).However,the distribution of this water the importance of interactions between renewable sur- both geographically and temporally,is not well matched
5 Issues in Ecology Number 9 Spring 2001 hundreds to tens of thousands of years, although the range in turnover rates is large. Indeed, a majority of ground water is not actively turning over or being recharged from the earths surface at all. Instead, it is fossil water, a relic of wetter ancient climatic conditions and melting Pleistocene ice sheets that accumulated over tens of thousands of years. Once used, it cannot readily be replenished. The distinction between renewable and nonrenewable ground water is critical for water management and policy. More than three-quarters of underground water is non-renewable, meaning it has a replenishment period of centuries or more (Figure 3). The High Plains or Ogallala Aquifer that underlies half a million km2 of the central United States is arguably the largest aquifer in the world. The availability of turbine pumps and relatively inexpensive energy has spurred the drilling of about 200,000 wells into the aquifer since the 1940s, making the Ogallala the primary water source for a fifth of irrigated U.S. farmland. The extent of irrigated cropland in the region peaked around 1980 at 5.6 million hectares and at pumping rates of about 6 trillion gallons of water a year. That has since declined somewhat due to groundwater depletion and socioeconomic changes in the region. However, the average thickness of the Ogallala declined by more than 5 percent across a fifth of its area in the 1980s alone. In contrast, renewable aquifers depend on current rainfall for refilling and so are vulnerable to changes in the quantity and quality of recharge water. For example, groundwater pumping of the Edwards Aquifer, which supplies much of central Texas with drinking water, has increased four-fold since the 1930s and at times now exceeds annual recharge rates. Increased water withdrawal makes aquifers more susceptible to drought and other changes in weather and to contamination from pollutants and wastes that percolate into the ground water. Depletion of ground water can also cause land subsidence and compaction of the porous sand, gravel, or rock of the aquifer, permanently reducing its capacity to store water. The Central Valley of California has lost about 25 km3 of storage in this way, a capacity equal to more than 40 percent of the combined storage capacity of all human-made reservoirs in the state. Renewable ground water and surface waters have commonly been viewed separately, both scientifically and legally. This view is changing, however, as studies in streams, rivers, reservoirs, wetlands, and estuaries show the importance of interactions between renewable surface and ground waters for water supply, water quality, and aquatic habitats. Where extraction of ground water exceeds recharge rates, the result is lower water tables. In summer, when a high water table is needed to sustain minimum flows in rivers and streams, low groundwater levels can decrease low-flow rates, reduce perennial stream habitat, increase summer stream temperatures, and impair water quality. Trout and salmon species select areas of groundwater upwelling in streams to moderate extreme seasonal temperatures and to keep their eggs from overheating or freezing. Dynamic exchange of surface and ground waters alters the dissolved oxygen and nutrient concentrations of streams and dilutes concentrations of dissolved contaminants such as pesticides and volatile organic compounds. Because of such links, human development of either ground water or surface water often affects the quantity and quality of the other. The links between surface and ground waters are especially important in regions with low rainfall (see Box 1, Table 1, and Figure 4). Arid and semi-arid regions cover a third of the earths lands and hold a fifth of the global population. Ground water is the primary source of water for drinking and irrigation in these regions, which possess many of the worlds largest aquifers. Limited recharge makes such aquifers highly susceptible to groundwater depletion. For example, exploitation of the Northern Sahara Basin Aquifer in the 1990s was almost twice the rate of replenishment, and many springs associated with this aquifer are drying up. For non-renewable groundwater sources, discussing sustainable or appropriate rates of extraction is difficult. As with deposits of coal and oil, almost any extraction is non-sustainable. Important questions for society include at what rate groundwater pumping should be allowed, for what purpose, and who if anyone will safeguard the needs of future generations. In the Ogallala Aquifer, for example, the water may be gone in as little as a century. HUMAN APPROPRIATION OF FRESHWATER SUPPLY Global Renewable Water Supplies Growth in global population and water consumption will place additional pressure on freshwater resources in the coming century. Currently, the water cycle makes available several times more fresh water each year than is needed to sustain the worlds population of six billion people (Table 2). However, the distribution of this water, both geographically and temporally, is not well matched
Issues in Ecology Number 9 Spring 2001 Box I:A Case Study-the Middle Rio Grande Perhaps nowhere are man impacts on river and floodplain ecosystems greater than in arid and semi-arid regionso the world.The Middle Rio Grande Basin of central New Mexico is a rapidly growing area that holds more than half of the state's population.The desire to balance water needs there has led to development of a careful water budget for the basin (Table I).highlighting annual variability,measurement uncertainty,and conflicting water demands for the region.The goal of the water budget is to help design a sustainable water policy management has already greatly altered this floodplain ecosystem (Figure 4).Dams and constructed river channels zones, f I a mosaic of cottonwoo wet meado .once hosted with signific ant cotton wood establishment occurred in 1942,and cottonwoods are declining in most areas.Half of the wetlands in the drainage were lost in just 50 years.Invasion by nonnative deep-rooted trees such as saltcedar and Russian-olive has dramatically altered riparian forest composition.Without changes in water management.exotic species will likely dominate riparian zones within half a century The water budget of the Middle Rio Grande reflects recent changes in hydrology,riparian ecology,and ground epletions in the basin or dep etion The largest loss is open-water evaporation,comprising one-third of the total.This loss is large compared to pre-dam values-direct evaporation from Elephant Butte Reservoir alone ranges from 50 to 280 million cubic meters(m)per year depending on reservoir size and climate.The second largest depletion is riparian plant transpiration(135 to 340 million m/y). There is considerable uncertainty in this estimate because of the unknown effects of fluctuating river discharge on Table I Sources and average annual water depletion from Water Source 1972 to 1997 for the Middle Rio Grande reach(the 64.000 km- dra between Otowi Gage north of Santa Fe and El Average Otowi Flow 1360 ephant But m Flow records at the Otowi gage,the in San Juan-Chama Diversion 70 flow poin the Middle Rio Grande reach.are more than a entur ter supplemented from the San Juan Water Use Depletion 9/2 and inci sed Oto (106 m/vr) a89 pen-werevaporatio 165 on (ground water) Net aquifer recharg 85 ng prog y severe drought years 10 to human needs the laree river flows of the amazon other regions,spring floodwaters from snowmelt are and Zaire-Congo basins and the tier of undeveloped riv. captured in reservoirs for later use in tropical regions ers in the northern tundra and taiga regions of Eurasia a substantial share of annual runoff occurs during mon and north am soon flooding.In Asia,fore xample,80 Perc occurs between May and October. Although this flood gether.these remote rivers account for nearly one-fifth water provides a variety of ecological services,including of total global runoff. sustaining wetlands,it is not a practical supply for Approximately half of the global renewable water irrigation.industry,and household uses that need supply runs rapidly toward the sea in floods (Table 2). water to be delivered in controlled quantities at specific In managed river systems of North America and many times
6 Issues in Ecology Number 9 Spring 2001 Box 1: A Case Study the Middle Rio Grande Increasing water demands create potential conflicts between human needs and those of native ecosystems. Perhaps nowhere are human impacts on river and floodplain ecosystems greater than in arid and semi-arid regions of the world. The Middle Rio Grande Basin of central New Mexico is a rapidly growing area that holds more than half of the states population. The desire to balance water needs there has led to development of a careful water budget for the basin (Table 1), highlighting annual variability, measurement uncertainty, and conflicting water demands for the region. The goal of the water budget is to help design a sustainable water policy. Water management has already greatly altered this floodplain ecosystem (Figure 4). Dams and constructed river channels prevent spring floods. Riparian zones, now limited by a system of levees, once hosted a mosaic of cottonwood and willow woodlands, wet meadows, marshes, and ponds. The last major floods with significant cottonwood establishment occurred in 1942, and cottonwoods are declining in most areas. Half of the wetlands in the drainage were lost in just 50 years. Invasion by nonnative deep-rooted trees such as saltcedar and Russian-olive has dramatically altered riparian forest composition. Without changes in water management, exotic species will likely dominate riparian zones within half a century. The water budget of the Middle Rio Grande reflects recent changes in hydrology, riparian ecology, and groundwater pumping. Estimating all major water depletions in the basin is critical for managing its water. Major depletions include urban uses, irrigation, plant transpiration, open-water evaporation, and aquifer recharge. The largest loss is open-water evaporation, comprising one-third of the total. This loss is large compared to pre-dam values direct evaporation from Elephant Butte Reservoir alone ranges from 50 to 280 million cubic meters (m3 ) per year depending on reservoir size and climate. The second largest depletion is riparian plant transpiration (135 to 340 million m3/y). There is considerable uncertainty in this estimate because of the unknown effects of fluctuating river discharge on Table 1 Sources and average annual water depletion from 1972 to 1997 for the Middle Rio Grande reach (the 64,000 km2 drainage between Otowi Gage north of Santa Fe and Elephant Butte Dam). Flow records at the Otowi gage, the inflow point for the Middle Rio Grande reach, are more than a century old. Water supplemented from the San Juan-Chama diversion project began in 1972 and increased Otowi flow by 70 million m3 /y (average flow without this water was about 1400 million m3/y). Major municipal water systems in the basin currently pump ground water at a rate of 85 million m3 /y. Maximum allowable depletion for the reach is 500 million m3 /y when adjusted annual flow exceeds 1900 million m3 /y, decreasing progressively to 58 million m3 /y in severe drought years (inflows of 120 million m3 /y at Otowi Gage). Water Source Supply (106 m3 /yr) Average Otowi Flow 1360 San Juan-Chama Diversion 70 Water Use Depletion (106 m3/yr) Open-water evaporation 270 Riparian plant transpiration 220 Irrigated agriculture 165 Urban consumption (ground water) 85 Net aquifer recharge 85 to human needs. The large river flows of the Amazon and Zaire-Congo basins and the tier of undeveloped rivers in the northern tundra and taiga regions of Eurasia and North America are largely inaccessible for human uses and will likely remain so for the foreseeable future. Together, these remote rivers account for nearly one-fifth of total global runoff. Approximately half of the global renewable water supply runs rapidly toward the sea in floods (Table 2). In managed river systems of North America and many other regions, spring floodwaters from snowmelt are captured in reservoirs for later use. In tropical regions, a substantial share of annual runoff occurs during monsoon flooding. In Asia, for example, 80 percent of runoff occurs between May and October. Although this floodwater provides a variety of ecological services, including sustaining wetlands, it is not a practical supply for irrigation, industry, and household uses that need water to be delivered in controlled quantities at specific times.
Issues in Ecology Number g Spring 2001 transpiration and differences between native and non-native plants in transpiration rates.Irrigated agriculture in the Middle Rio Grande accounts for an estimated 20 percent of annual average depletions,with cropping patterns. weather,and water availability contributing to annual variations.Urban consumption and net aquifer recharge are similar and account for 20-25 percent of the remaining depletion in the Middle Rio Grande rtially offset by water from the San Juan-Chama Project,inflows from tribu taries within the basin,and mur icipal v ated for an average water year.Municipal use of San Juan-Chama water.sustained drought.and continued population growth will increase pressure on surface water resources.No new water will likely be available in the near future,so water conservation must play a dominant role. A careful water budget such as the one described here is essential in designing sustainable water policy.For the Middle Rio Grande,accurate long-term measurements of surface flows,evapotranspiration.net aquifer recharge and goudervr orionsic peie oodpn emi-arid regions have similar needs for fundamental data and careful water planning. Figure 4-Contrasting riparian vegetation in the Middle Rio Grande reach south of Albuque ue NM.a native cottor od-dominated site (A) a and a otic saltcedar-dominated site (B)o he National Wildlife Refuge construction and river channeling.has greatly altered this floodplain cosystem.The last maior floods with significant cottonwood establishment were in 194)Invasions by evotic deen.rooted nlants such as saltcedar.pictured here.and Russian-olive have dramatically altered riparian forest nposition. Without changes in water 09ssns puc auU Daunid Jep3ics Thus.there are two categories of accessible run water is chronically overpumped.Data for China,India off available to meet human water needs:(1)renewable North Africa.Saudi Arabia,and the United States indi ground water and base river flow,and (2)floodwater that cate that groundwater depletion in key basins totals at is captured and stored in reservoirs. least 160 kmper year.Groundwater depletion is par Base river flows and renewable ground water ac ticularly serious in India and some water percent of global each year long as the rate of water ceed replenishment by rainfall,these sources can serve ea9eareewocPsamto as a sustainable supply.Unfortunately.in many places bal recharge rate does not mean that groundwater use in including many important agricultural regions.ground a specific region is sustainable.What matters is how water
7 Issues in Ecology Number 9 Spring 2001 transpiration and differences between native and non-native plants in transpiration rates. Irrigated agriculture in the Middle Rio Grande accounts for an estimated 20 percent of annual average depletions, with cropping patterns, weather, and water availability contributing to annual variations. Urban consumption and net aquifer recharge are similar and account for 20-25 percent of the remaining depletion in the Middle Rio Grande. Average annual depletions are partially offset by water from the San Juan-Chama Project, inflows from tributaries within the basin, and municipal wastewater discharge. Nonetheless, water depletions are already fully appropriated for an average water year. Municipal use of San Juan-Chama water, sustained drought, and continued population growth will increase pressure on surface water resources. No new water will likely be available in the near future, so water conservation must play a dominant role. A careful water budget such as the one described here is essential in designing sustainable water policy. For the Middle Rio Grande, accurate long-term measurements of surface flows, evapotranspiration, net aquifer recharge, and groundwater levels are necessary. Reservoir operations, exotic species control, land use planning, and agricultural and urban water conservation will all play an important role in a sustainable water future for the region. Other arid and semi-arid regions of the world, where balancing diverse water demands will be a formidable and important challenge, have similar needs for fundamental data and careful water planning. Figure 4 Contrasting riparian vegetation in the Middle Rio Grande reach south of Albuquerque, NM: a native cottonwood-dominated site (A) near Los Lunas and an exotic saltcedar-dominated site (B) on the Sevilleta National Wildlife Refuge. Water management, especially dam construction and river channeling, has greatly altered this floodplain ecosystem. The last major floods with significant cottonwood establishment were in 1942. Invasions by exotic deep-rooted plants such as saltcedar, pictured here, and Russian-olive have dramatically altered riparian forest composition. Without changes in water management, exotic species will likely dominate riparian zones in the Middle Rio Grande basin within the next half century. Thus, there are two categories of accessible runoff available to meet human water needs: (1) renewable ground water and base river flow, and (2) floodwater that is captured and stored in reservoirs. Base river flows and renewable ground water account for about 27 percent of global runoff each year. As long as the rate of water withdrawals does not exceed replenishment by rainfall, these sources can serve as a sustainable supply. Unfortunately, in many places, including many important agricultural regions, ground water is chronically overpumped. Data for China, India, North Africa, Saudi Arabia, and the United States indicate that groundwater depletion in key basins totals at least 160 km3 per year. Groundwater depletion is particularly serious in India, and some water experts have warned that as much as one-fourth of Indias grain harvest could be jeopardized by overpumping. The fact that global groundwater extractions remain well below the global recharge rate does not mean that groundwater use in a specific region is sustainable. What matters is how water
Issues in Ecology Number 9 Spring 2001 isusedand managed n there many regions of the world whe s.In the strips supply. Turning floodwater into an accessible supply gen- gation have caused the lake to shrink more than three erally requires dams and reservoirs to capture.store.and quarters in volume and fifteen meters in depth over the control the water.Worldwide.there are approximately past four decades.The shoreline of the Aral Sea has 40.000 large dams more than 15 meters (m)high and retreated 120 km in places,and a commercial fishery 3 smaller dams that once landed 45.000 tonnes a year and employed estimated 6.600 the 60,000 water each year.Considerably less water than this is de decined.Salinity tripled from 9 tondth livered to farms,industries. water that remains is now and cities,however,because saltier than the oceans. dams and reservoirs are also For purposes of wa used to generate electricity. Total Global Runoff 40.700 control floods,and enha Remote Flow n river navigat Amazon Basin Finally,after sub Zaire-Congo Basin 5删 1,740 tracting remote rivers from cluding evaporative losses base flows and discounting 2040 from reservoirs)total water reservoir capacity allocated 12.500 4 430 km3 a vear and 52) to functions other than wa A 880 the man use is abou Mun tr e remaining water in a ba km per year,or 3I percent alities 300 275 sin or channel by increasing of total annual runoff. Total Global Withdrawals concentration of major ions 4.430 nutrients.or contaminants Human Water Use 2.350 otal Human Appropriation 6.780 As the example of the Aral Sea showed.this effect car People use fres Table 2 Global runoff.withdraw water for many purposes and hun ion of f ater supply(km/yr ele for future use There are three broad cat In addition to water re egories of extractive uses nclude 95%of 959%0 moved from natural sys for which people withdraw n and furasian river flow tems.human enterprises de water from its natural chan half of off The est of ins in its natural chan crops indus 18%(or 2285 km/yr)of acc nels. These m activities,and res although humans use.directly or indirectly.6.780 km/v include dilution of pollut dential life.In many cases. or 54%of accessible runoff Water that is withdrawn bu ants.recreation,navigation not consumed is not always returned to the same river o water can be used more maintenance of healthy es than once after it is with lake from which it was taken.From Postel et al.(1996) tuaries.sustenance of fish drawn.Water that is used based on additional data in Czaya (19).L'Vovich et al (1990).and Shiklomanov (1997). eries.and protection of biodiversity instream uses of water v ample may be usec by region and season,it i again,although it sometimes requires further treatment. difficult to estimate their global total.If pollution dilution In contrast,about half the water diverted for irrigation is is taken as a rough global proxy,however,instream uses lost through evapotranspiration and is unavailable for fur may total 2.350 km a year,a conservative estimate that ther use. does not incorporate all instream uses
8 Issues in Ecology Number 9 Spring 2001 is used and managed in particular basins, and there are many regions of the world where current demand outstrips supply. Turning floodwater into an accessible supply generally requires dams and reservoirs to capture, store, and control the water. Worldwide, there are approximately 40,000 large dams more than 15 meters (m) high and twenty times as many smaller dams. Collectively, the worlds reservoirs can hold an estimated 6,600 km3 of water each year. Considerably less water than this is delivered to farms, industries, and cities, however, because dams and reservoirs are also used to generate electricity, control floods, and enhance river navigation. Finally, after subtracting remote rivers from base flows and discounting reservoir capacity allocated to functions other than water supply, the total accessible runoff available for human use is about 12,500 km3 per year, or 31 percent of total annual runoff. Human Water Use People use fresh water for many purposes. There are three broad categories of extractive uses for which people withdraw water from its natural channel or basin: irrigation of crops, industrial and commercial activities, and residential life. In many cases, water can be used more than once after it is withdrawn. Water that is used but not physically consumed to wash dishes, for example may be used again, although it sometimes requires further treatment. In contrast, about half the water diverted for irrigation is lost through evapotranspiration and is unavailable for further use. Excessive rates of consumptive water use can have extreme effects on local and regional ecosystems. In the Aral Sea Basin, for example, large river diversions for irrigation have caused the lake to shrink more than three quarters in volume and fifteen meters in depth over the past four decades. The shoreline of the Aral Sea has retreated 120 km in places, and a commercial fishery that once landed 45,000 tonnes a year and employed 60,000 people has disappeared. Water quality has also declined. Salinity tripled from 1960 to 1990, and the water that remains is now saltier than the oceans. For purposes of water management, the difference between use and consumption is important. Global withdrawals of water (including evaporative losses from reservoirs) total 4,430 km3 a year, and 52 percent of that is consumed. Water use or withdrawal also modifies the quality of the remaining water in a basin or channel by increasing concentration of major ions, nutrients, or contaminants. As the example of the Aral Sea showed, this effect can limit the suitability of water for future use. In addition to water removed from natural systems, human enterprises depend heavily on water that remains in its natural channels. These instream uses include dilution of pollutants, recreation, navigation, maintenance of healthy estuaries, sustenance of fisheries, and protection of biodiversity. Because instream uses of water vary by region and season, it is difficult to estimate their global total. If pollution dilution is taken as a rough global proxy, however, instream uses may total 2,350 km3 a year, a conservative estimate that does not incorporate all instream uses. Table 2 Global runoff, withdrawals, and human appropriation of freshwater supply (km3 /yr). Remote flow refers to river runoff that is geographically inaccessible, estimated to include 95% of runoff in the Amazon basin, 95% of remote northern North American and Eurasian river flows, and half of the Zaire-Congo basin runoff. The runoff estimates also include renewable ground water. An estimated 18% (or 2285 km3/yr) of accessible runoff is consumed, although humans use, directly or indirectly, 6,780 km3 /yr or 54% of accessible runoff. Water that is withdrawn but not consumed is not always returned to the same river or lake from which it was taken. From Postel et al. (1996), based on additional data in Czaya (1981), LVovich et al. (1990), and Shiklomanov (1997). Total Global Runoff 40,700 Remote Flow Amazon Basin 5,400 Zaire-Congo Basin 660 Remote northern rivers 1,740 Total Remote Flow 7,800 Uncaptured Floodwater 20,400 Accessible Runoff 12,500 Global Water Withdrawals Agriculture 2,880 Industry 975 Municipalities 300 Reservoir Losses 275 Total Global Withdrawals 4,430 Instream Uses 2,350 Total Human Appropriation 6,780
Issues in Ecology Number 9 Spring 2001 Projected Future Changes in AET Proiected future changes in PRcp ange from change from 1961-1990 baseline to 2061-2090 scenario climate 20020406080166 Figure 5 of future cha transpiration(AET BIOME-RGO precipitation (PRCP)generat by ar system mc this atmos h stem m resp r AET projections 89 we. Re th cted for the Sou VEMAP IL Simula n Group Univ.of Montana) Combining this instream use figu removing a dam outweighed the economic benefits of awals puts the al at 6,780 km mns urrppriauns离 operating it As a result of these and other trends.accessible cent of the accessible freshwater runoff of the planet. runoff is unlikely to increase by more than 5-10 percent Global water demands continue to rise with in- over the next 30 years.During the same period,the creases in human population and consumption.Increases earth's population is projected to grow by approximately in accessible runoff.however.can only be accomplished 35 percent.The demands on freshwater systems wil struction of new dams or desaination of seawater continue to grow throughout the coming century atio globa THE WATER CYCLE AND CLIMATE CHANGE ments.it is likely to remain a minor part of global supply for the foreseeable future.Dams continue to bring more A scientific consensus now exists that the con water under human control,but the pace of construction tinuing buildup of human-generated greenhouse gases in has slowed.In developed countries,many of the best sites the atmosphere is warming the earth.The last decade of have alre ady heen Risin the twe centu as the ecord.an ahit tdestruction. pal warm g of the sity.and displacement of hum past50years had no in the past thousand are making further dam construction increasingly dif years.As the earth continues to warm in the coming ficult.About 260 new large dams now come on line world century.a general intensification of the water cycle is wide each year compared with 1.000 a year between the expected to occur.In a warmer climate,greater volumes 1950s and 1970s.Moreover,at least 180 dams in the of water will evaporate from plants,soils and water bod United States were removed in the past decade based on ies globally.lofting more vapor into the atn f afety. ntal impact,ando rain out and in turr sing ru off and making hydr cence. The destruction of the Edwards Dam on Maine' logic extremes such as floods and droughts more com Kennebec River in 1999 marked the first time that fed- mon and more intense.Some decreases in snow and ice eral regulators ruled that the environmental benefits of cover have already been observed.Changes in the tem- 9
9 Issues in Ecology Number 9 Spring 2001 Figure 5 A projection of future changes in actual evapotranspiration (AET) and precipitation (PRCP) generated by an ecosystem model (BIOME-BGC) using a future climate scenario to the year 2100 derived from a global climate model. In this scenario, atmospheric carbon dioxide (CO2) increased approximately 0.5%/yr, and the ecosystem model responded with changes in leaf area index (a measure of plant productivity) based on changes in CO2 , climate, water, and nitrogen availability. In general, these projections suggest higher rainfall and increased plant growth in the arid West, leading to higher AET. Reduced rainfall and the resulting effects of drought on vegetation are the primary causes of lower evapotranspiration projected for the Southeast. For additional information, see Box 2 (Results from VEMAP II, courtesy of P. Thornton, Numerical Terradynamic Simulation Group, Univ. of Montana). Combining this instream use figure with estimated global withdrawals puts the total at 6,780 km3 a year. That means humans currently are appropriating 54 percent of the accessible freshwater runoff of the planet. Global water demands continue to rise with increases in human population and consumption. Increases in accessible runoff, however, can only be accomplished by construction of new dams or desalination of seawater. Today, desalination accounts for less than 0.2 percent of global water use and, because of its high energy requirements, it is likely to remain a minor part of global supply for the foreseeable future. Dams continue to bring more water under human control, but the pace of construction has slowed. In developed countries, many of the best sites have already been used. Rising economic, environmental, and social costs including habitat destruction, loss of biodiversity, and displacement of human communities are making further dam construction increasingly difficult. About 260 new large dams now come on line worldwide each year compared with 1,000 a year between the 1950s and 1970s. Moreover, at least 180 dams in the United States were removed in the past decade based on evaluations of safety, environmental impact, and obsolescence. The destruction of the Edwards Dam on Maines Kennebec River in 1999 marked the first time that federal regulators ruled that the environmental benefits of removing a dam outweighed the economic benefits of operating it. As a result of these and other trends, accessible runoff is unlikely to increase by more than 5-10 percent over the next 30 years. During the same period, the earths population is projected to grow by approximately 35 percent. The demands on freshwater systems will continue to grow throughout the coming century. THE WATER CYCLE AND CLIMATE CHANGE A scientific consensus now exists that the continuing buildup of human-generated greenhouse gases in the atmosphere is warming the earth. The last decade of the twentieth century was the warmest on record, and paleoclimate records indicate that the warming of the past 50 years had no counterpart in the past thousand years. As the earth continues to warm in the coming century, a general intensification of the water cycle is expected to occur. In a warmer climate, greater volumes of water will evaporate from plants, soils and water bodies globally, lofting more vapor into the atmosphere to rain out and in turn, increasing runoff and making hydrologic extremes such as floods and droughts more common and more intense. Some decreases in snow and ice cover have already been observed. Changes in the tem-