Skip to Content

Dividing the Water

Water may seem to be everywhere, but for a rising portion of the world’s population, there may soon be hardly a drop to drink -or to use for growing food, supporting industries and cities, and preserving life-giving ecosystems.

Deep in the delta of the Colorado River, the Cocopa people have fished and farmed for perhaps 2,000 years. They once harvested a grain they called nipa, a unique salt-loving plant known to botanists as Distichlis palmeri that tastes much like wild rice. Protein was also abundant: they sometimes ate fish three times a day, and they hunted deer, wild boar, ducks, and geese. Known as “people of the river,” the Cocopa had no formal calendar but keyed their lives to the Colorado’s seasonal floods. While no census documented their numbers, historical accounts suggest that about 5,000 Cocopa were living in the delta 400 years ago.

Today the Cocopa culture is at risk of extinction. Their water has been siphoned away from the Colorado to fill swimming pools in Los Angeles, generate electricity to illuminate Las Vegas, and irrigate crops in the deserts of Arizona, California, and Mexico’s Mexicali Valley. Fishing and farming can no longer sustain them. They last harvested nipa in the early 1950s; by then, U.S. dams upstream had largely eliminated the annual floods that had naturally irrigated their staple grain. Now just 40 to 50 Cocopa families remain south of the border. With little means of subsistence or livelihood in the delta countryside, many of the tribal members have migrated to the cities. Anita Alvarez de Williams, a Mexicali-based expert on the Cocopa, worries that by the end of the twentieth century they “may no longer be river people at all.”

It might be tempting to dismiss the Cocopa’s plight as the price of progress. Supporting ever larger populations and higher levels of consumption has always involved taking more and more of nature’s bounty, and those last in line are bound to suffer. But apart from the tragedy of losing yet another culture in a world of dwindling cultural diversity, the fading of the Cocopa people is a harbinger of far more widespread disruption to society at large today.

Indeed, a growing scarcity of freshwater is now an impediment to global future food security, health of aquatic ecosystems, and social and political stability. Each year, millions of tons of grain are grown by depleting groundwater, a clear case of robbing the future to pay for the present. Competition for water is increasing-between cities and farms, between neighboring states and provinces, and between nations-as demands bump up against the limits of a finite supply. And critical ecosystem functions such as flood protection, water purification, habitat maintenance, and the sustenance of fisheries are being destroyed by excessive damming, diversion, and pollution of rivers.

As world population expands by a projected 2.6 billion people over the next 30 years, and as consumption levels spiral upward, water problems are bound to intensify. With the best dam sites already developed and many rivers and groundwater reserves already overtapped, opportunities to solve these problems by exploiting new sources are limited. A fresh approach is needed, one focused on using water more efficiently and allocating it more equitably.

Global Mirage

The photographs of earth taken by astronauts show a strikingly blue planet, seemingly a world of water spinning in space. Yet this impression of water wealth may be as deceptive as a desert mirage. Only about 2.5 percent of all the water on earth is fresh, and two-thirds of that is locked in glaciers and ice caps. The renewable freshwater supply on land-that made available year after year by the solar-powered hydrologic cycle in the form of precipitation-totals some 110,300 cubic kilometers (1 cubic kilometer equals 1 billion cubic meters), a mere 0.008 percent of all the water on earth.

Each year, nearly two-thirds of this renewable supply returns to the atmosphere through evaporation or transpiration, the uptake and release of moisture by plants. This process supplies the water needed for forests, grasslands, rain-fed croplands, and all other nonirrigated vegetation. The remainder, just over one-third of the renewable supply-about 40,700 cubic kilometers per year-is runoff, the flow of freshwater from land to sea through rivers, streams, and underground aquifers. This is the source for all human diversions or withdrawals of water-for irrigated agriculture, industry, and households as well as a variety of “instream” water services, including the dilution of pollutants, navigation, and the generation of hydroelectric power. Rivers also carry nutrients from the land to the seas and in this way help support the highly productive fisheries of coastal bays and estuaries. Thus, by virtue of the hydrologic cycle, the oceans water the continents, and the continents nourish the oceans.

 Though the volume of runoff seems huge, nature’s delivery of this freshwater supply does not correlate well with the distribution of world population. Asia, for example, receives 36 percent of global runoff but is home to 60 percent of the world’s people; South America, on the other hand, supports 6 percent of the population yet has 26 percent of the world’s runoff. The Amazon River alone carries 15 percent of the earth’s runoff but is accessible to only 0.4 percent of the world’s population. Much of the river flow in the tropics and high latitudes is virtually inaccessible to people and economic activity and is likely to remain so for the foreseeable future, since water is difficult and expensive to transport long distances. In fact, 55 rivers in northern North America, Europe, and Asia, with combined annual flows equal to about 5 percent of global runoff, are so remote that no dams have been constructed on them, even for hydropower.

According to a 1996 study conducted by this author and Gretchen Daily and Paul Ehrlich of Stanford University, the total amount of runoff within reach geographically totals some 32,900 cubic kilometers or about 81 percent of the total runoff. But that’s not the end of the story. About three-fourths of this amount is flood water and therefore not accessible on demand when it is needed most. To add to the remaining one-fourth that is accessible, engineers have built large dams and reservoirs, raising the stable supply of water provided by underground aquifers and year-round river flows by about half. This brings the total stable renewable supply to 12,500 cubic kilometers.

Globally, people now use about 35 percent of this accessible supply, or some 4,430 cubic kilometers per year. At least an additional 19 percent is used “instream” to dilute pollution, sustain fisheries, and transport goods. Thus humanity is already appropriating, directly or indirectly, more than half of the water supply that is now accessible. The problem is that water use tripled between 1950 and 1990 as world population soared by some 2.7 billion. Given that the population is projected to climb by nearly the same amount over the next 30 years, this is a troubling prospect. Worldwide demand for water cannot triple again without causing severe shortages for crop irrigation, industrial use, basic household needs, and critical life-supporting ecosystems.

Water Stressed

The scarcity of renewable freshwater not only poses a long-term threat but has already begun taking a toll in many countries, especially where population has grown out of proportion to water resources. The United Nations Food and Agriculture Organization estimates that producing the food needed for a nutritious, low-meat diet requires about 1,600 cubic meters of water per person per year. In humid climates, virtually all of this could be provided directly to the soil by natural rainfall. But in drier regions and in those with distinct wet and dry seasons, a portion of the needed moisture would have to be supplied by irrigation water drawn from rivers, lakes, or aquifers. Estimating conservatively that a third of the 1,600 cubic meters per person would need to be supplied by irrigation, annual water demand for food-above and beyond what direct rainfall provides-would average about 530 cubic meters per person.

Of course, countries have more than just food needs to meet. Estimates by Russian hydrologist Igor Shiklomanov suggest that worldwide household, municipal, and industrial water uses average about 240 cubic meters per capita per year. More widespread use of efficient technologies could reduce this level substantially, but the resulting savings would partially be offset by the more than 1 billion people now lacking minimum household water supplies and by rising affluence, which translates into higher water use. Assuming an average for household, municipal, and industrial uses of 200 cubic meters per capita per year, and adding this to the freshwater required for food production, yields a requirement of some 730 cubic meters per capita per year.

Unfortunately, in many, if not most, countries it is difficult to access and control more than 30 to 50 percent of runoff. Moreover, a portion of runoff must remain in rivers to dilute pollution and satisfy other “instream” needs. Thus the total amount of runoff must be 2 to 3 times higher than the amount required to meet irrigation, industrial, and household water demands, which works out to be approximately 1,700 cubic meters per person per year. Consequently, countries can be considered “water stressed” when the total annual runoff per capita drops below 1,700 cubic meters.

Some water analysts argue that this “water stress” indicator can be misleading. Hillel Shuval, professor of environmental science at Hebrew University, points out, for example, that Israel maintains a highly successful modern economy and high per-capita income even though its renewable water per person is less than a fifth of the water-stress level of 1,700 cubic meters per year. In part, Israel has succeeded so well with its limited supplies by importing much of its grain-which Shuval and others sometimes refer to as “virtual water.”

Indeed, with the production of each ton of grain requiring some 1,000 tons of water, importing grain becomes a key strategy for balancing water budgets. Such a strategy would seem to make economic and environmental sense for countries short of water, since they can get much higher value from their limited supplies by devoting them to commercial and industrial enterprises and using the resulting income to purchase food through international markets. The Middle East, for instance, which is the most concentrated region of water scarcity in the world, imports 30 percent of its grain. As long as surplus food is produced elsewhere, nations with surpluses are willing to trade, and the countries in need can afford to pay for the imports, it would seem that water-short countries can have food security without needing to be food self-sufficient.

This tidy logic is shaken, however, by the growing number of people living in countries where water availability is a constraint to food self-sufficiency, and by widespread signs of unsustainable water use in key food-producing regions. As of 1995, a total of 44 countries with a combined population of 733 million people had annual renewable water supplies per person below 1,700 cubic meters. Just over half of these people live in Africa or the Middle East, where the populations of many countries are projected to double within 30 years. Water-short Algeria, Egypt, Libya, Morocco, and Tunisia are each already importing more than a third of their grain. With their collective population projected to grow by 87 million people over the next 30 years, these countries’ dependence on grain imports is bound to increase. Indeed, this is a likely scenario for much of Africa: given current population projections, more than 1.1 billion Africans will be living in water-stressed countries by 2025-three-quarters of the continent’s projected population.

Portions of many large countries, including China, India, and the United States, would also qualify as water stressed if breakdowns of water supplies and population were available by region. Even using national statistics, China-with 7 percent of global runoff but 21 percent of world population-will narrowly miss the 1,700 cubic meter per capita mark in 2030; India, the world’s second most populous country, will join the list by then.

Water for Food

Many physical signs of unsustainable water use authenticate the numerical indicator of water stress. Perhaps most important, evidence suggests that the amount of freshwater that can sustainably be supplied to farmers is nearing its limit. Groundwater overpumping and aquifer depletion are now occurring in many of the world’s most important crop-producing regions-including the western United States and large portions of India, as well as parts of north China, where water tables are dropping 1 meter a year. This signals not only that limits to groundwater use have been exceeded in many areas, but also that a portion of the world’s food supply is produced through unsustainable water use.

Like groundwater, many of the planet’s major rivers are suffering from overexploitation. In Asia, where the majority of world population growth and additional food needs will be centered in the years ahead, many rivers are completely tapped out during the drier part of the year, when irrigation is essential. These include most rivers in India-among them the mighty Ganges, a principal water source for densely populated and rapidly growing South Asia, and China’s Yellow River, whose lower reaches ran dry for an average of 70 days a year in each of the last 10 years and for 122 days in 1995. Demand for water is exceeding the Yellow River’s capacity to supply it.

Crop production may be even harder hit in these and other areas as population growth and urbanization push up water demands. Worldwide, the number of urban dwellers is expected to double to 5 billion by 2025. With political power and money concentrated in the cities, and with insufficient water to meet all demands, governments will face strong pressures to shift water out of agriculture even as food demands are rising.
In fact, the reallocation of water from farms to cities is well under way in both industrial and developing countries. In California, for instance, a 1957 water plan projected that 8 million hectares of irrigated land would ultimately be developed statewide, yet the state’s irrigated area peaked in 1981 at 3.9 million hectares, less than half this amount. Net irrigated area fell by more than 121,000 hectares during the 1980s. California officials project an additional net decline of nearly 162,000 hectares between 1990 and 2020, with most of the loss owing to urbanization as the population expands from 30 million to a projected 49 million.

In China, water supplies are being siphoned away from farmlands surrounding Beijing to meet that city’s rising domestic, industrial, and tourist demands. The capital’s water use now exceeds the capacity of its two main reservoirs, and farmers in the agricultural belt that rings the city have been cut off from traditional sources of irrigation water. With some 300 Chinese cities now experiencing water shortages, this shift is bound to accelerate.

Similarly, growing demand in the megacities of Southeast Asia-including Bangkok, Manila, and Jakarta-is already partially met by overpumping groundwater. With limited new sources to tap, pressures to shift water out of agriculture will mount in these regions as well.

Unfortunately, no one has tallied the potential effect on future food production of the progressive shift of water from agriculture to cities combined with groundwater overpumping, aquifer depletion, and the other forms of unsustainable water use. Without such assessments, countries have no clear idea how secure their agricultural foundations are, no ability to accurately predict their future food import requirements, and no sense of how or when to prepare for the economic and social disruption that may ensue as farmers lose their water.

Supply-Side Options

Preventing water scarcity from undermining food security, ecological life-support systems, and social stability will not be easy. In much of the world, expanding the water supply for one user now means taking it away from another. New groundwater wells may expand supplies in some regions, but groundwater use will need to be reduced to the level of recharge in others. New dams and river diversions will rarely offer sustainable solutions, because in most cases they entail drawing more water from freshwater systems that are already overtaxed. In fact, the construction of new dams has slowed markedly over the last couple of decades as the public, governments, and financial backers have begun to pay more attention to their high economic, social, and environmental costs. Whereas nearly 1,000 large dams began operation each year from the 1950s through the mid-1970s, the number dropped to about 260 annually during the early 1990s. Even if conditions become more favorable to dam construction, it seems unlikely that new reservoirs built over the next 30 years will increase accessible runoff by more than 10 percent while population is projected to expand by 45 percent during that period.

Another option, desalination, is often held up as the ultimate solution to the world’s water problems, since the oceans hold more than 97 percent of the earth’s water. As early as 1961, President John F. Kennedy noted that if humanity could find an inexpensive way to obtain freshwater from the seas, the achievement “would really dwarf any other scientific accomplishment.”

Some 35 years later, desalination is a proven technology experiencing solid growth. As of December 1995, a total of 11,066 desalting units had been installed or contracted for worldwide, with a collective capacity of 7.4 billion cubic meters per year.

Despite considerable growth, however, desalination still plays a minor part in the global supply picture, accounting for less than 0.2 percent of world water use. Removing salt from water either by heating it and condensing the steam (distillation) or by filtering it through a membrane (reverse osmosis) is highly energy intensive. And although costs have come down to $1.00-$1.60 per cubic meter, desalination remains one of the most expensive supply options. The wealthy countries of Saudi Arabia, United Arab Emirates, and Kuwait-which together encompass only 0.4 percent of world population-accounted for 46 percent of the world’s 1993 desalting capacity. These countries are essentially turning oil into water, and they are among the few that can afford to do so. For the foreseeable future, seawater desalination will likely continue to be a lifeline technology for water-scarce, energy-rich countries as well as island nations with no other options. But desalination capacity would have to expand 30-fold to supply even 5 percent of current world water use. As such, the option will likely remain a minor contributor to total water supplies worldwide.

Other options, such as towing icebergs, transporting water by tanker, or shipping it in large bags, may increase drinking water supplies in some specific water-scarce areas, but like desalination, are expensive and not likely to make much of a dent in the global supply picture over the next 30 years.

Cutting Demand

Measures to reduce demand for water through conservation, recycling, and higher efficiency are typically more economical than efforts to gain new supplies of freshwater. Costing between 5 and 50 per cubic meter of water, nearly the entire spectrum of conservation options-including leak repair, the adoption of more efficient technologies, and water recycling-cost less than the development of new water sources and much less than desalination.

Unfortunately, large subsidies to water users continue to discourage investments in efficiency and convey the false message that water is abundant and can be wasted-even as rivers are drying up and aquifers are being depleted. Farmers in water-short Tunisia pay 5 per cubic meter for irrigation water-one-seventh the cost of supplying it. Jordanian farmers pay less than 3 per cubic meter, a small fraction of the water’s full cost. And federal subsidies to irrigators in the western United States total at least $20 billion, representing 86 percent of total construction costs of installing the systems, according to Richard Wahl, formerly an economist at the U.S. Department of Interior. Although poverty alleviation and other social goals may justify some degree of irrigation subsidy, especially for poor farmers, the levels of subsidization that exists today is an invitation to waste water.

Experience in the Broadview water district in California, where farmers irrigate 4,000 hectares of melons, tomatoes, cotton, wheat, and alfalfa, reveals the gains an intermediate policy can yield. In the late 1980s, when the district was faced with the need to reduce polluting drainage to the San Joaquin River, it established a tiered water pricing structure. The district determined the average volume of water used over the 1986-88 period and applied a base rate of $16 per acre-foot (1.3 per cubic meter) to 90 percent of this amount. Any water used above that level was charged at a rate 2.5 times higher. In 1991, only 7 out of the 47 fields in the district used any water charged at the higher level: the higher price encouraged farmers to switch crops and irrigate more efficiently, thus cutting the average amount of water applied to the district’s farms by 19 percent.

Because agriculture accounts for two-thirds of water use worldwide, even small-percentage reductions can free up substantial quantities of water for cities, ecosystems, and additional food production. Farmers in northwest Texas, for example, who have had to cope with falling water tables from depletion of the Ogallala aquifer-an underground water reserve in the region that gets extremely limited recharge from rainfall-have reduced their water use by 20 to 25 percent by adopting more efficient sprinkler technologies, special valves to ensure even water distribution, and other water-saving practices.

Likewise, results from a variety of countries show that farmers who have switched from furrow (trench) systems or sprinkler irrigation to drip systems, which deliver water nearer to the roots of crops, have cut their water use by 30 to 60 percent. Crop yields often increase at the same time because plants are effectively “spoon-fed” the optimal amount of water (and often fertilizer) when they need it. Drip systems, which cost in the range of $1,200 to $2,500 per hectare, tend to be too expensive for most poor farmers and for use on low-value row crops, but research is under way to make them more affordable. Colorado-based International Development Enterprises has developed a drip system that costs just $50 per half acre ($123 per half hectare), 10 to 20 percent of the cost of traditional drip systems. The keys to keeping costs down are simple materials and portability: instead of each row of crops getting its own drip line, a single line is rotated by farmers among ten rows.

Along with encouraging irrigation efficiency improvements, more appropriate water pricing would also promote the treatment and reuse of urban wastewater for irrigation, which is typically more expensive than most conservation and efficiency measures but often less expensive than developing new water sources. Wastewater contains nitrogen and phosphorus, which can be pollutants when released to lakes and rivers but are nutrients when applied to farmland. Moreover, unlike many other water sources, treated wastewater will be both an expanding and fairly reliable supply, since urban water use will likely double by 2025. Many large cities located along coastlines dump their wastewater, treated or untreated, into the ocean, rendering it unavailable for any other purpose and harming coastal marine life. As long as the wastewater stream is free of heavy metals and harmful chemicals, and disease-causing microorganisms are controlled, it can become a vital new supply for irrigating crops.

Adapting to Dryness

Raising the water productivity of the global crop base is also critical. Actual strategies used will vary by crop, climate, and the type of water-control system, but the basic aim will necessarily be the same in each: to optimize the timing and amount of moisture in the root zone and to enhance the crops’ ability to use that moisture productively.

Through plant breeding, for example, biologists can hasten the process of plant adaptation to dryness. Studies have shown that if no other factors are limiting plant growth total production is proportional to the amount of water a plant transpires. Larger or deeper root systems that allow plants to take in more moisture can thus increase yield. New genetic techniques are making it possible to screen crop varieties for water-efficiency traits. And developing varieties with shorter growing seasons or the ability to grow in cooler periods, when evaporation and transpiration are lower, could also help improve crops’ water-use efficiency.

The International Rice Research Institute in the Philippines, for one, is focusing on developing more efficient irrigation operations, technologies that reduce water consumption, and changes in the rice plant itself to improve water-use efficiency. Breeders have already shortened the maturation time for irrigated rice from 150 days to 110 days, for example, a major water-saving achievement.

Matching crops to varying water quality can also enhance supplies for irrigated agriculture. In the western Negev of Israel, for example, farmers successfully grow cotton using highly salty water from a local saline aquifer. The Israelis have also found that certain crops-such as tomatoes grown for canning or pastes-may actually benefit from somewhat salty irrigation water. The varying salt tolerances of crops raise the possibility of multiple reuse of irrigation water. In California, for instance, moderately salty drainage water from a crop of average salt tolerance is used to irrigate more highly tolerant cotton. In turn, the drainage from the cotton fields, which is even saltier, is used to irrigate salt-loving crops, a number of which scientists have made considerable progress toward commercializing. For example, when a variety of Salicornia, a seed-bearing plant, was irrigated with seawater in a coastal desert near Mexico’s Sea of Cortez, its yield was equal to or greater than freshwater oilseed crops such as soybean and sunflower.

Like more realistic water pricing, water marketing can create incentives both to encourage efficiency and reuse, as well as to allocate water more productively. Instead of looking to a new dam or river diversion to get additional water, cities and farmers can purchase supplies from others who are willing to sell, trade, or lease them water or water rights. The Metropolitan Water District of Los Angeles, for example, is investing in conservation measures in southern California’s Imperial Irrigation District in exchange for the water those investments will save. The annual cost of the conserved water is estimated at about 10 per cubic meter, far lower than the water district’s best new-supply option. In Chile, where water policy encourages marketing, water companies that serve expanding cities frequently buy small portions of water rights from farmers, most of whom have gained surpluses through efficiency improvements.

The setting of efficiency standards has also proven to be an effective policy tool for stretching supplies. U.S. legislation passed in late 1992 requires manufacturers of toilets, faucets, and showerheads to meet specified efficiency standards as of January 1994. U.S. residential water use for these three fixtures is expected to drop by more than 35 percent as the more efficient models replace existing stock over the next 30 years.

A number of other governments, including Mexico and the Canadian province of Ontario, have also adopted standards for household plumbing fixtures. The National Community Water Conservation Program in Cairo is working with the Egyptian government to introduce water conservation standards into the plumbing code. Although efficiency standards have so far mainly been applied to household fixtures, they offer potential for water savings in agriculture, industry, and other municipal uses as well.

Further Steps

Here and there, promising efforts inspire hope that the consequences of water shortage can at least be delayed. Yet so far, concerted national and international efforts to bring all the pieces of a sustainable water strategy together are few. One notable exception, however, may be South Africa. In early 1996, the Minister of Water Affairs and Forestry laid out principles for a fundamental overhaul of the nation’s water law and management. Among the top priorities are providing each South African with access to at least 25 liters of water a day to meet the minimum need for drinking water and sanitation, pricing water at levels that reflect its value, encouraging water marketing, mandating that water suppliers adopt conservation measures, allocating water to the environment to prevent the loss of ecosystem functions, and reserving water for countries downstream in order to promote regional cooperation and integration.

Although these principles are promising, turning them into actual laws, policies, and actions will not be easy because it will involve dismantling decades of apartheid-era water legislation. Moreover, the country is still pursuing the socially and environmentally destructive Lesotho Highlands Water Development Project, an $8 billion dam-and-diversion scheme aimed at supplying the Johannesburg region with water from the tiny mountain kingdom of Lesotho. Nonetheless, the nation’s new water plan, which could be taken up by the parliament in early 1997, may emerge as one of the stronger national water strategies thus far.

Beyond the adoption of similar strategies in other nations, there is also an urgent need at the international level to assess and monitor the availability of water for food production. A basic blueprint for satisfying human and ecological needs as well as using and allocating water more efficiently will not guarantee agriculture the water supplies needed to meet the world’s future food demands. For example, many of the policies and strategies to promote more sustainable water use-such as raising water prices and expanding water markets-will likely shift resources away from agriculture toward higher valued uses.

The time may not be far off when a global grain bank will be needed to guard against food shortfalls induced by water shortages. Particularly in Africa, Asia, and the Middle East, water deficits will widen markedly in the coming decades. Together these regions are projected to grow by nearly 2.3 billion people by 2025, accounting for 87 percent of projected population growth over the next 30 years. Many African and Asian countries are unlikely to have the financial resources to balance their water books by purchasing surplus grain on the open market.

Finally, reining in demand for water offers the best hope of preventing scarcity from leading to more hunger, poverty, widespread ecological decline, and social instability. Living within the limits of nature’s water supply will require reduced consumption among the more wealthy social groups and smaller family size among all groups.

Keep Reading

Most Popular

DeepMind’s cofounder: Generative AI is just a phase. What’s next is interactive AI.

“This is a profound moment in the history of technology,” says Mustafa Suleyman.

What to know about this autumn’s covid vaccines

New variants will pose a challenge, but early signs suggest the shots will still boost antibody responses.

Human-plus-AI solutions mitigate security threats

With the right human oversight, emerging technologies like artificial intelligence can help keep business and customer data secure

Next slide, please: A brief history of the corporate presentation

From million-dollar slide shows to Steve Jobs’s introduction of the iPhone, a bit of show business never hurt plain old business.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.