Variations of precipitation, temperature, and runoff in the Yellow River source region were analyzed with the Mann–Kendall and Spearman rank correlation tests over the past 60 years. Based on the seven climate scenarios from CMIP5 climate models under RCP2.6, RCP4.5, and RCP8.5, responses of hydrological process to climate change were simulated using the Variable Infiltration Capacity (VIC) model. Variation analysis results indicated that recorded temperature presented significant increasing trend. Daily minimum temperature presented higher increasing trend than daily maximum temperature. Annual gross precipitation presented minor increasing and annual runoff presented minor decreasing. The VIC model performed well on simulating monthly discharge at Tangnaihai station, with NSE of 0.91 and 0.93 in calibration and validation periods, respectively. The projected annual mean temperature would rise (with 25th and 75th percentiles) 1.07–1.32 °C, 1.76–2.33 °C, 3.45–4.29 °C, annual precipitation is expected to increase 3.43%–11.77%, 8.05%–17.27%, 12.84%–27.89%, and runoff would moderately increase with high variability of 0.82%–14.26%, −3.41%–19.14%, 1.43%–38.26% relative to the baseline of 1961–1990 under each RCP in the 2080s, respectively. The inhomogeneity of runoff may increase in the future. Many more droughts and floods under climate change may threaten social development in this region in the future.