To quantify the effects of land use and climate change on runoff in the upper Urumqi River watershed,the soil and water assessment tool (SWAT) was calibrated and validated using the correlation index (R~2) , Nash-Sutcliffe efficiency (NSE) , and percent bias ( PBIAS). In addition, the river runoffs under a changing environment were simulated by designing various land use and climate change scenarios. The results showed that values of R~2 and NSE for simulated streamflow during the calibration period were 0.89 and 0.84, respectively. During the validation period, the value of R2 was 0.75 and that of NSE was 0.74. Values of PBIAS were between 18% and 28% both during calibration and validation periods. Simulation results showed that the SWAT model was capable for simulating runoff responses in Urumqi River upstream. River runoff in the study area was more sensitive to climate change than land use change. The annual average runoff decreased by 1.41 m~3/s under the scenario of land use and climate change. Meanwhile, land use change caused an annual streamflow reduction of 0.04 m~3/s, whereas climate change caused streamflow reduction of 1.37 m~3/s, indicating that the impact of climate change on streamflow was more obvious than that of land use change. The results of extreme land use scenario showed that the annual average runoff increased by 0.1 m~3/s in the grassland scenario, and decreased 0.58 m~3/s in the woodland scenario. The climate variability scenario analysis indicated that a 10% increase in annual precipitation could increase the annual streamflow by 3.05 m~3/s, and precipitation decrease of the same percentage could reduce the annual streamflow by 0.93 m~3/s. In addition, a 20% increase in annual precipitation could induce an increase in annual streamflow by 4.02 m~3/s, whereas a 20% decrease in annual precipitation could reduce annual streamflow by 2.25 m~3/s. When the annual mean temperature increased by 1℃ and 2℃, the annual streamflow decreased by 2.71 m~3/s and 3.02 m~3/s, respectively. Therefore, we concluded that the annual streamflow variation had the same change tendency as that of annual precipitation.