Estimates indicate that after increases in water withdrawals from 1965 to 1980 in the Tennessee River watershed, withdrawals declined from 1980 to 1985 and remained steady from 1985 to 1995. Water withdrawals in the Tennessee River watershed during 2000 averaged about 12,211 million gallons per day (Mgal/d) of freshwater for offstream uses?22 percent more than the 1995 estimate. The 2000 estimate is nearly the same as the estimate for 1980, the highest year of record, with 12,260 Mgal/d. The reuse potential of water from the Tennessee River is high because most of the water withdrawn for offstream use is returned to the river system. Besides water quality, reuse potential reflects the quantity of water available for subsequent uses and is gaged by consumptive use, which is the difference between water withdrawals and return flow. For the Tennessee River watershed, return flow was estimated to be 11,562 Mgal/d, or 95 percent of the water withdrawn during 2000. Total consumptive use accounts for the remaining 5 percent, or 649 Mgal/d. Estimates of water withdrawals by source indicate that during 2000, withdrawals from surface water accounted for 98 percent of the total withdrawals, or 11,996 Mgal/d, 23 percent more than during 1995. Total ground-water withdrawals during 2000 were 215 Mgal/d, or 17 percent less than during 1995. During 2000, thermoelectric power withdrawals were estimated to be 10,276 Mgal/d; industrial, 1,205 Mgal/d; public supply, 662 Mgal/d; and irrigation, 68.9 Mgal/d. Return flows were estimated to be: thermoelectric power, 10,244 Mgal/d; industrial, 942 Mgal/d; and public supply, 377 Mgal/d. Consumptive use was estimated to be: thermoelectric power, 32.2 Mgal/d; industrial, 263 Mgal/d; public supply, 285 Mgal/d; and irrigation, 68.9 Mgal/d. Each category of use affects the reuse potential of the return flows differently. The consumptive use in the river is comparatively small because most of the water withdrawn from the Tennessee River watershed is used for once-through cooling for the thermoelectric power and industrial sectors. Average per capita use for all offstream uses was 2,710 gallons per day per person in 2000, compared to the record high of 3,200 in 1975 and 1980. The intensity of use for the Tennessee River watershed as measured as a function of area was 298,489 gallons per day per square mile in 2000. In 2030, water withdrawals are projected to increase by about 15 percent to 13,990 Mgal/d. By category, water withdrawals are projected to increase as follows: thermoelectric power, 11 percent or 1,152 Mgal/d; industry, 31 percent or 368 Mgal/d; public supply, 35 percent or 232 Mgal/d; and irrigation, 37 percent or 25.2 Mgal/d. Total consumptive use is projected to increase about 51 percent or 334 Mgal/d to 980 Mgal/d. Per capita use in 2030 is calculated to be about 2,370 gallons per day, about 26 percent less than in 1980. Water transfers to the Tennessee-Tombigbee waterway for navigation lockages were estimated as 200 Mgal/d for 2000 and 800 Mgal/d for 2030. Water transfers for hydropower commitments through Barkley Canal averaged 3,361 Mgal/d for 2000 and are estimated to be an average of 4,524 Mgal/d in 2030.