Comparison of the Various Methodologies for Estimating Thermoelectric Power Generation Water Withdrawals and Their Effect on Water-Use Trends from 1985-2010 in the United States

The U.S Geological Survey (USGS) has estimated thermoelectric water withdrawals at 5-year intervals since 1950, and consumptive use from 1950 to 1995. Changes in water demand for cooling water, a significant part of the thermoelectric water use, has important implications for water availability to meet future energy demand, especially at the local level. USGS data show total water withdrawals peaked in 1980, declined in 1985, and have remained relatively stable through 2005. Total water use has been dominated by thermoelectric withdrawals since 1965. USGS estimates through 2005 have been primarily based on compiling self-reported data by powerplant operators to State water regulatory agencies and to the Department of Energy-Energy Information Administration (EIA). The reported data from these sources have often been inconsistent because techniques for measuring or estimating the main water flows are not standardized; and, incomplete because reporting thresholds for water withdrawals vary from State-to-State. EIA only requires the reporting of water use from powerplants that are 100 megawatts or more. Some withdrawals have also been estimated with a gallon per kilowatt-hour coefficient and powerplant net electric generation; however, coefficients were mostly based on reported data, and although the coefficients accounted for differences in cooling systems, fuel type, and flue gas desulfurization and other factors, the coefficients are averages and have not accounted for either weather or climatic conditions. The USGS National Water Use Information Program (NWUIP) developed consistent estimates of water withdrawals and water consumption based on linked heat and water budgets for the entire fleet of 1,284 active water-using powerplants for 2010. In 2010, 802 powerplants reported water-use data to EIA. The linked heat and water budget calculates condenser duty for a powerplant, and estimated water withdrawal is a function of condenser duty and change in temperature in the cooling water. Condenser duty is the amount of waste heat delivered to the cooling system through the condenser. The modeled water withdrawal results were expressed as a single value for each powerplant along with a minimum and maximum that bracketed a thermodynamically reasonable range of values. This range also provided a quality assurance check for other self-reported operator or coefficient derived water withdrawal estimates for a powerplant. To varying degrees, the USGS modeled results differed from the self-reported operator values used by most USGS State offices for the 2010 national compilation. Importantly, these two USGS nationally-generated sets of 2010 withdrawal values show a notable shift in the relatively stable thermoelectric water-use trend from 1985 to 2005. To understand the shift in the 2010 thermoelectric withdrawal estimates, the methodologies were analyzed by comparing EIA reported data for 1985, 1990, 1995, 2000, and 2005 to USGS national compilation estimates for the same years. Further, 2010 self-reported water withdrawal data from EIA, the USGS national compilation data, and USGS model results were also compared.