You are here

Past and future water use in the United States

January, 1999 to December, 2012


The Forest and Rangeland Renewable Resources Planning Act of 1974 requires the Forest Service to periodically assess anticipated resource supply and demand conditions of the nation's renewable resources. This project focuses on fresh water demand.

The likelihood of future water shortages depends on how water supply compares with demands for water use. Comparison of supply and demand within a probabilistic framework yields an estimate of the probability of shortage and thus a measure of the vulnerability of the water supply system. This comparison was performed for current conditions and for several possible future conditions reflecting alternative socio-economic scenarios and climatic projections. Examining alternative futures provides a measure of the extent to which serious future risks of water shortage must be anticipated.


Water supply was quantified by first estimating freshwater input as precipitation minus evapotranspiration for each point in a grid covering the study area. These water inputs were then allocated to major river basins and made available to meet basic in-stream flow requirements, satisfy off-stream demands including those from downstream basins or those reached by trans-basin diversions, and add to reservoir storage. Off-stream demands were estimated as threshold quantities of desired water use based on extending past trends in water use under the assumption that water supply would be no more constraining to future water withdrawals than in the recent past.

Change from current conditions (2010) in mean water yield (cm/yr) with the A1B-CGCM climate scenario future for 2020, 2040, 2060, and 2080.
Change from current conditions (2010) in mean water yield (cm/yr) with the A1B-CGCM climate scenario future for 2020, 2040, 2060, and 2080.
Vulnerability was defined as the probability of shortage, that is, of off-stream demand exceeding supply. Demand and supply were modeled on an annual basis for 98 river basins covering the coterminous United States called Assessment Sub-Regions (ASRs). Current levels of inter-ASR diversion were accounted for, as were existing reservoir storage capacity and basic in-stream flow needs. Only renewable sources of supply were considered; thus, lowering of groundwater tables was not considered a source of supply.

Modeling water supply and demand in this way does not provide a forecast of future shortage levels. Rather, it provides a projection of the degree to which water shortages would occur in the absence of adaptation measures to either increase supply or decrease demand.

Key Findings

  • On a per capita basis, aggregate water withdrawal in the United States has been dropping since at least 1985. This reduction has occurred largely because of changes in the irrigation, thermoelectric, and industrial water use sectors.

  • Despite the reductions in per-capita water withdrawal, total U.S. withdrawal rose from 1985 to 2000, largely in response to population growth of roughly 2.7 million persons per year. However, the most recent data show a drop in total withdrawals, attributable to large reductions in irrigation and industrial withdrawals plus a slowing of the increase in domestic and public withdrawals.

  • Future climate change will increase water use for agricultural irrigation and landscape maintenance in response to rising plant water requirements, and at thermoelectric plants to accommodate rising electricity demands for space cooling. Including these effects, per-capita withdrawals are projected to drop only moderately for the next few decades and then level off as the effects of climate change become greater, and total withdrawals are projected to rise nearly continuously into the future.

  • Although precipitation is projected to increase in much of the United States with future climate change, in most locations that additional precipitation will merely accommodate rising evapotranspiration demand in response to temperature increases. For the United States as a whole, projected declines in precipitation are substantial, exceeding 30% of current levels by 2080 for some scenarios examined.

  • Vulnerability to water shortages is greatest in arid and semiarid areas of the U.S.—including the Southwest, parts of California, and the central and southern Great Plains—where current conditions are already precarious. In some cases, important reservoirs are left with little or no water. Although the detailed results differ depending on which scenario is simulated and which climate model is used, the general finding of increasing and substantial vulnerability in the larger Southwest holds true in all cases.

  • The gradually increasing future vulnerability results from the effect of increasing population on water demand, and of climate change on both water supply and water demand. In about one-half of the ASRs where vulnerability is projected to increase, decreases in water yield, and thus in water supply, have a greater effect on vulnerability than do increases in water demand, whereas in the other ASRs the reverse is true.

  • The projected levels of vulnerability in some ASRs are clearly untenable, indicating that adaptation will be essential. Adaptation options that are likely to be considered include groundwater mining (while supplies last), reductions in in-stream flows, water transfers, water conservation beyond the levels assumed here, alterations of reservoir operating rules and other water management agreements, population shifts, and, in selected locations, increases in water storage and diversion capacity.

Related Publications

Brown, Thomas C.; Foti, Romano; Ramirez, Jorge A. 2013. Projected freshwater withdrawals in the United States under a changing climate. Water Resources Research. 49: 1259-1276.

Foti, Romano; Ramirez, Jorge A.; Brown, Thomas C. 2012. Vulnerability of U.S. water supply to shortage: a technical document supporting the Forest Service 2010 RPA Assessment. Gen. Tech. Rep. RMRS-GTR-295. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 147 p.

Brown, T.C. 2000. Projecting U.S. freshwater withdrawals. Journal of Water Resources Research 36(3):769-780. 

Brown, Thomas C. 1999. Past and future freshwater use in the United States. Gen. Tech. Rep. RMRS-GTR-39. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, 47 p.

Principal Investigators:
Romano Foti - Princeton University
Jorge A. Ramirez - Colorado State University