In arid regions of the southwestern United States, water is often considered a primary factor limiting distribution and productivity of desert ungulates, including desert bighorn sheep (Ovis canadensis mexicana). Thus, wildlife management agencies and sportsmen's organizations have invested substantial time and resources in the construction and maintenance of water catchments. Although the availability of freestanding water sources is believed to influence many aspects of the ecology of desert bighorn sheep, the efficacy of these water sources has been questioned and has not been examined experimentally. We used a before-after–control-impact study design to determine if removal of water catchments changed diet, characteristics of foraging areas used by female desert bighorn sheep, home-range size, movement rates, distance to catchments, adult mortality, productivity, or juvenile recruitment in 2 mountain ranges on the Cabeza Prieta National Wildlife Refuge, Arizona, USA. During pretreatment (2002–2003), we ensured that water catchments were available to desert bighorn sheep in both mountain ranges; during posttreatment (2004–2005), we drained all water catchments in the treatment range. We measured diet composition, characteristics of foraging areas, 50% and 95% kernel home ranges, movement rates, and distance to water catchments seasonally from 2002 to 2005. We also estimated adult survival, lamb:female, and yearling:female ratios from 2002 to 2005. We predicted that removal of water catchments would result in 1) increased use of foraging areas with more vegetation cover, more thermal cover, and higher succulent abundance; 2) increased consumption of cacti and other succulents; 3) an increase in home-range size, movement rates, and distance to nearest catchment; and 4) a decrease in adult survival, productivity, and juvenile recruitment. Removal of water catchments in the treatment range did not result in predicted changes in diet, foraging area selection, home-range size, movement rates, mortality, productivity, or recruitment. Female desert bighorn sheep did use areas with more thermal cover during the summer after removal of water catchments, but other characteristics of foraging areas used by bighorn sheep and their diet did not change appreciably with removal of water catchments. We did not document changes in home-range area, movement rates, or distance sheep were from water during hotter months; we only documented changes in home-range area, movement rates, and distance to water catchments during winter and autumn. There were 10 desert bighorn sheep mortalities in the treatment range and 8 in the control range; 7 mortalities in each mountain range were during pretreatment. Twelve of the 18 total mortalities occurred during summer. Survival rate was lower during pretreatment than posttreatment in both mountain ranges. We did not document increased mortality or a change in lamb:female or yearling:female ratios after removal of water catchments. Home-range area and movement rates declined with increasing precipitation. Annual survival rates increased with increases in the current year's total precipitation and total precipitation during the previous year; annual survival rates declined with increases in average daily temperature during winter. There was a severe drought during pretreatment and abnormally wet conditions during posttreatment. The increase in precipitation that coincided with removal of water sources improved forage conditions during posttreatment and may have provided adequate water for female desert bighorn sheep. The lack of change in home-range size, movement rates, and distance to the nearest water catchment during hot, dry seasons after removal of water sources suggests that forage conditions played a greater role in determining home-range area and movement rates than did the presence of water catchments. Higher mortality rates during the drought of the pretreatment p