2 Hydrogeology, Water Use, and Simulation of Flow in the High Plains Aquifer in Northwestern Oklahoma, Southeastern Colorado, Southwestern Kansas, Northeastern New Mexico, and Northwestern Texas and hydraulic conductivity. Simulated discharge to streams is sensitive to recharge but is insensitive to hydraulic conductivity. Recharge and hydraulic conductivity are closely related with respect to water levels but are not related with respect to discharge to streams. The sensitivity of the development-period model to changes in specific yield and recharge due to dryland cultivation was tested. The model appears more sensitive to specific yield, but the two inputs were not varied over the same range in percent change. The calibrated development-period model was used to simulate water-level changes from 1998 to 2020 using mean 1996-97 pumpage. The largest simulated water-level changes in Oklahoma occur in Texas County where water levels are simulated to decline 25 to 50 additional feet over a large area. Water levels also are simulated to decline 10 to 25 additional feet in two large areas of Cimarron County, two areas in Beaver County, and one area in Ellis County, all in Oklahoma. Water levels are simulated to decline more than 100 additional feet in several areas in Kansas and were simulated to decline 50 to 100 additional feet in several areas in Texas. Purpose and Scope This report describes the hydrology of the High Plains aquifer in Oklahoma and adjacent areas, estimates historical water use, and describes the construction, calibration, and use of a flow model of the aquifer. This report emphasizes the Oklahoma portion of the aquifer but includes parts of adjacent Colorado, Kansas, New Mexico, and Texas. year and major droughts occurred in the 1930s and 1950s (fig. 3). The 1980s tended to be wetter than average and the 1990s had closer to average precipitation. Much of the precipitation occurs as a result of summer thunderstorms, resulting in highly variable rainfall amounts.