Model variations plus uncertainties in the future climate change scenarios create a big challenge for such evaluation. In this work, we developed the statistical models for six different regions in China, using the historical yield data between 1981 and 2010 from the National Bureau of Statistics combined with meteorological station observations and analyzed the impact of climate variation (i.e., temperature and precipitation changes) on the grain yields into the 2030s, based on28 ensemble climate predictions from six state-of-the-art Coupled Model Inter comparison Project Phase 5 (CMIP5) model outputs. Our results indicate that the four crops (i.e., rice, maize, wheat, and soybean) respond similarly to the climate variation in different regions of China, with the sensitivity to warming increasing from north to south and from inner land to coast regions. In addition, the yields of all the four crops in East and Central-South China are also positively correlated with precipitation change.