World's Fresh Water Supply

The Water Resources of Earth

Over 70% of our Earth's surface is covered by water ( we should really call our planet "Ocean" instead of "Earth"). Although water is seemingly abundant, the real issue is the amount of fresh water available. 

  • 97.5% of all water on Earth is salt water, leaving only 2.5% as fresh water 
  • Nearly 70% of that fresh water is frozen in the icecaps of Antarctica and Greenland; most of the remainder is present as soil moisture, or lies in deep underground aquifers as groundwater not accessible to human use. 
  • Only ~1% of the world's fresh water is accessible for direct human uses. This is the water found in lakes, rivers, reservoirs and those underground sources that are shallow enough to be tapped at an affordable cost. Only this amount is regularly renewed by rain and snowfall, and is therefore available on a sustainable basis.

Water as a Resource

Since antiquity, irrigation, drainage, and impoundment have been the three types of water control having a major impact on landscapes and water flows. Since the dawn of irrigated agriculture at least 5000 years ago, controlling water to grow crops has been the primary motivation for human alteration of freshwater supplies. Today, principal demands for fresh water are for irrigation, household and municipal water use, and industrial uses. Most supplies come from surface runoff, although mining of "fossil water" from underground aquifers is an important source in some areas. The pattern of water withdrawal over the past 300 years shows the dramatic increases in this century.

A timeline of human water use:

  • 12,000 yrs. ago: hunter-gatherers continually return to fertile river valleys 
  • 7,000 yrs. ago: water shortages spur humans to invent irrigation 
  • 1,100 yrs ago: collapse of Mayan civilization due to drought 
  • Mid 1800's: fecal contamination of surface water causes severe health problems (typhoid, cholera) in some major North American cities, notably Chicago
  • 1858: "Year of the Great Stink" in London, due to sewage and wastes in Thames 
  • Late 1800s-early 1900: Dams became popular as a water management tool 
  • 1900s: The green revolution strengthens human dependency on irrigation for agriculture 
  • World War II: water quality impacted by industrial and agricultural chemicals 
  • 1972: Clean Water Act passed; humans recognize need to protect water
Figure 1: The water usage of different regions of the world per capita in cubic meters.

Consumptive and Non-Consumptive Water Use

Consumptive water use refers to water that is not returned to streams after use. For the most part, this is water that enters the atmospheric pool of water via evaporation (from reservoirs in arid areas) and from plant transpiration (especially from "thirsty" crops such as cotton and alfalfa). Irrigated agriculture is responsible for most consumptive water use, and decreases surface run-off. An extreme example is the Colorado River, which has most of its water diverted to irrigated agriculture, so that in a normal year, no water at all reaches the river’s mouth. 

Agriculture is responsible for 87 % of the total water used globally. In Asia it accounts for 86% of total annual water withdrawal, compared with 49% in North and Central America and 38% in Europe. Rice growing, in particular, is a heavy consumer of water: it takes some 5000 liters of water to produce 1 kg of rice. Compared with other crops, rice production is less efficient in the way it uses water. Wheat, for example, consumes 4000 m3/ha, while rice consumes 7650 m3/ha.

A great deal of water use is non-consumptive, which means that the water is returned to surface runoff. Usually that water is contaminated however, whether used for agriculture, domestic consumption, or industry. The WHO estimates that more than 5 million people die each year from diseases caused by unsafe drinking water, and lack of sanitation and water for hygiene. This has economic effects as well: an outbreak of cholera in Latin America killed hundreds of people, and cost hundreds of millions of dollars. 

Some believe that fresh water will be a critical limiting resource for many regions in the near future. About one-third of the world's population lives in countries that are experiencing water stress. In Asia, where water has always been regarded as an abundant resource, per capita availability declined by 40-60% between 1955 and 1990. Projections suggest that most Asian countries will have severe water problems by the year 2025. Most of Africa historically has been water-poor. 

What's the problem?:

  • The population is growing rapidly, putting more pressure on our water supply (demand is increasing)
  • The amount of water is effectively reduced by pollution and contamination (supply is decreasing)
  • What does the future hold? We can best explore this question by looking carefully at the world's water resources. 
Figure 2: This picture depicts the global hydrological cycle adapted from Gleick. Flows are approximate estimates and are in cubic kilometers per year.

Human Appropriation of Renewable Fresh Water

The hydrological cycle:

  • The water cycle on Earth is essentially a closed system – we always have the same amount of water. 
  • The only parts of this cycle appropriated by humans is water held as surface water or shallow aquifers. Let us try to quantify present use.

Available renewable fresh water:

  • Fossil ground water can be tapped but is non-replenishable. 
  • Terrestrial replenishable fw supply (RFWS land). 
  • RFWSland = ppte on land. 
  • Pland = evapotranspiration from the land (ETland) and runoff to sea (R). 
  • Estimates of annual runoff range from 33,500 to 47, 000 km3 (Postel uses 40,000 km3). 
Because it is difficult to separate evaporation from transpiration, they are combined as evapotranspiration. Evapotranspiration represents the water supply for all non-irrigated vegetation, both natural and crops. 

Runoff is the source for all human diversions or withdrawals for irrigation, industry, municipal uses, navigation, dilution, hydropower, and maintenance of aquatic life including fisheries. 

ET Estimates Appropriated for Human Dominated Land Uses.
Land Type
NPP co-opted 
(x 10^9 metric tons)
ET co-opted* 
(km ^3)
Cultivated land 
5,500 **
Grazing land 
Forest land 
Human occupied areas
(lawns, parks, etc.)
Total appropriated
Table 1: A total of 26.2% of terrestrial ET is appropriated (18,200 cubic km/69,600 cubic km). * Assumes 2 g of biomass produced for each liter of water evapotranspired. ** Adjusts for share of ET requirement me through irrigation.
Human appropriation of evapotranspiration

Vitousek et al. (1986) estimated the human co-option of terrestrial NPP at 40.6 billion metric tons, or more than 30% of terrestrial NPP. This includes cropland, grazing land, and trees harvested for fuelwood and timber. You can review Net Primary Production (NPP) from a lecture in Global Change I. 

The volume of ET required to produce a unit of biomass =total terrestrial NPP (132 billion metric tons) divided by terrestrial ET (70,00 km3) = 1.9 kg of biomass per ton of ET. 

The final estimate of appropriated ET is downward corrected for irrigation (approx. 16% of world’s cropland is irrigated ) and a rough estimate of irrigation of lawns, parks, and human-occupied areas. 

Some 18,200 km3 (26%) of total terrestrial ET is appropriated for human use (see table 1). The remaining 74% must meet the needs of remaining terrestrial ecosystems. 

Continental Breakdown of Share of Global Runoff and Population
Total river runoff
Share of global river runoff
Share of global population
Europe  3,240  8.0  13.0
Asia  14,550  35.8  60.5
Africa  4,320  10.6  12.5
N & C
6,200  15.2  8.0
S America  10,420  25.6  5.5
Australia& Oceania  1,970  4.8  0.5
Totals  40,700  100.0  100.0
Table 2

Human appropriation of runoff

    Distribution of global runoff is highly uneven and corresponds poorly to the distribution of the world population (see table 2). Asia has 69% of world population but 36% of global runoff. South America has 5% of world population, 25% of runoff. 

    Much of runoff is inaccessible. Amazon River accounts for 15% of runoff and is currently accessible to 25 million people (0.4% of world’s pop). Estimate it to be 95% inaccessible. Zaire may be 50% inaccessible. The mostly untapped northern rivers have an average annual flow of 1815 km3/yr, consider 95% to be inaccessible. 

    Together, this amounts to 7774 km3 or 19% of total annual runoff, leaving 32,900 km3 geographically accessible. Does not correct for many northern rivers with large flows relative to their pop sizes. 

    Temporal availability: about 27% of global runoff (11,100 km3) is renewable ground water and base river flow. Remainder is flood water and harder to capture (table 3). Present storage capacity of large dams totals 5500 km3, of which 3500 km3 is used to regulate river runoff. Adding together base flow and surface runoff controlled by dams gives total stable flow. Correct for spatially inaccessible flows yields and estimate of available runoff (AR) as 12,500 km3/yr. 

Estimates of inaccessible runoff of selected remote rivers. 
River basin or region 
Remote flow
(km cubed/year) 
Amazon (95% of total flow)  5,387 
Zaire-Congo (50% of total)  5,387 
Remote undammed northern
rivers (95% of totals)
  • North America 
  • 979 
  • Eurasia 
  • 746 
    Total inaccessible remote 
    Table 3

    Water Use and Consumption Estimates on a Global Scale, 1990
    (km cubed/year) 
    (km cubed/year)
    Agriculture*  2,880  1870
    Industry  975  90
    Municipalities  300  50
    Reservoir losses#  275  275
    Subtotal  4,430  2,285
    Instream flow needs  275  275
    Total as a percent
    of AR (12,500 km cubed) 
    54%  18%
    Table 4: *Assumes average applied water use of 12,000 (m cubed/ha) and consumption equal to ~65% of withdrawls. #Assumes evaporation loss equal to 5% of gross reservoir storage capacity.

    What fraction of AR is used by humanity?

    • Withdrawals: agricultural withdrawals = average water application rate (12,000 m3/ha) x world irrigated area (240 x 10^6 ha in 1990) = 2880 km3. Assuming 65% is consumed, 1870 km3
    • Industrial water use is estimated at 975 km3 and roughly 9% (90km3) is consumed. Remainder is discharged back into environment, often polluted. 
    • Municipal use is estimated at 300 km3 per year, of which 50 km3 (17%) is consumed. 
    • Evaporation from reservoirs is estimated to average 5% of gross storage capacity of reservoirs (5500 km3) or 275 km3/yr. 
    • Instream flow needs are estimated from pollution dilution, assuming that this suffices to meet instream needs. A common dilution term is 28.3 liters per second per 1000 population. Using the 1990 population yields a dilution requirement of 4700 km3. If half of water received adequate treatment, the dilution requirement is reduced to 2350 km3/hr. 
    • Combining these estimates (see table 4) indicate that humans appropriate 54% of AR. Human use of ET (18,200 km3) plus runoff (6780 km3) constitutes 30% of total accessible RFWS and 23% of unadjusted RFWS. 

    How much can AR be increased?

      Principal options are to capture and store more flood runoff or desalinate sea water. Latter is too energy-intensive for near future. 

      Worldwide, new dams (> 15 m ht) were constructed at rate of 885 per year during 1950-80, present rate is 500/yr, and future rate is estimated at 350/yr. Over next 30 years, assuming size of reservoirs is unchanged, new construction adds 1200 km3 to accessible supply, and raises total AR in 2025 to 13,700 km3/hr. Assuming average per capita water demand stays unchanged, but adjusting the pollution dilution for additional population, the total human appropriation in 2025 would be 9830 km3/yr, or 70% of estimated AR (compared to current 54%). Clearly we are approaching the limit of available fresh water supply. 

    What are the Solutions?

    Improvements in the efficiency of water use (ex: irrigation systems often perform poorly, wasting as much as 60 percent of the total water pumped before it reaches the intended crop). 

    Efficient management and modern technology can stretch even scarce water supplies much further. Israel, for example, supports its population, its growing industrial base, and intensive irrigation with less than 500 cubic meters per person per year. 

    Water is often wasted because it is underpriced. Direct and indirect subsidies (especially for agricultural use) are still common in both developed and developing countries. Removing such subsidies and letting water prices rise can provide incentives for conservation and for the investments needed to spread more efficient technologies. 

    Water Sustainability, Water Security

    The six billion people of Planet Earth use nearly 30% of the world’s total accessible renewal supply of water.  By 2025, that value may reach 70%.  Yet billions of people lack basic water services, and millions die each year from water-related diseases.  Water is a basis of international conflict.  What is involved in achieving water sustainability and water security?  The following lists some of the criteria that should help us chart our direction.

    1. Basic human needs for water should be fully acknowledged as a top international priority.
    2. Water-related diseases, including Guinea worm, diarrhea, onchocerciasis, malaria and typhoid should be brought under control.
    3. Agricultural water should be efficiently used and allocated.
    4. Basic ecosystem water needs should be identified and met.
    5. Serious water-related conflicts should be resolved through formal negotiations.

    Water conservation through better planning, management, and technologies offers great promise.   Fig. 4 shows per capita water withdrawals in the U.S. from 1900 to 1995.  Per capita water withdrawals began to decline in 1985, despite continued population growth.  More efficient agricultural and industrial water use accounts for this trend. 

    Figure 3. Trends in freshwater withdrawals, and population, for the United States, 1950-1995.  Note that per capita water use peaked in 1985, and has since declined.  Source:  USGS Circ. 1200)

     Figure 4 shows the projected global water withdrawals for the year 2000.  Note that estimates made 10, 20 or 30 years ago substantially over-estimated year 2000 withdrawals.  Less water demand actually materialized, reflecting the considerable improvements in water use over this time period.  Pricing water to its real cost will achieve further gains.  Both graphs provide a basis for cautious optimism. 
    Figure 4.  Actual water global withdrawals, 1960-200, and past forecasts of year 2000 water withdrawal.  Past estimates have been high, because they failed to incorporate improved efficiencies in industrial and agricultural water use.  Source:  P. Gleick, The World’s Water, 1998 Island Press).
    Figure 5. Global Water Scarcity Map. Source: World Resources Institute (WRI).


    • Postel, S.L., G.C. Daily and P.R. Ehrlich.  1996.  Human appropriation of renewable fresh water.  Science 271:785
    • Gleick, P.  2000.  The World's water.  Island Press.
    • Vitousek, P.M., P.R. Ehrlich, A.H. Ehrlich and P.A. Matson.  1986.  Human appropriation of the products of photosynthesis.  BioScience 36:368-373.