Abstract: The uncertainty of the influence of climate change on the North China's winter wheat yield is estimated by using the ensemble climate projections of CMIP5 and the probability of increase or reduction of the wheat yield in main production areas is analyzed. We combined 54 runs of projections from 15 global climate models of CMIP5 under different greenhouse gas emission scenarios in 2006-2030. Meanwhile, the CERES-Wheat was employed to stimulate the North China's winter wheat yield in the future. The results indicate that the projection of precipitation and solar radiation in future climate by the climate models has the largest uncertainty. Take the three representative points as an example: although in some years the yield will increases slightly, the fluctuation of winter wheat yield from year to year can be significant. An increased risk of lower yield is inevitable. And the probabilistic distributions of winter wheat yield in Middle and Eastern China during 2011-2030 over 2000, 4000, 6000, 8000, and 10000 kg/hm² are elaborated.

Key words: model assessment of CMIP5, North China, probabilistic distribution, uncertainty, winter wheat