A dynamic downscaling simulation experiment based on the WRF (Weather Research and Forecasting) model was conducted to examine the performance of the model in simulating the climate in China. The following results were found. MIROC (Model for Interdisciplinary Research on Climate) and the WRF model can reproduce the observed spatial patterns of surface air temperature well. The WRF model provides more detailed descriptions of the temperature, with both high temperatures in the Sichuan Basin and low temperatures in North China simulated well. The annual and seasonal spatial correlation coefficient between the MIROC simulated and observed precipitation ranges from 0.79 to 0.83, indicating that MIROC can simulate precipitation well. Overall, the WRF model can reproduce the observed precipitation distribution more accurately than the MIROC. Both the MIROC and the WRF models simulate precipitation better in South and East China than in North and West China. The WRF model can remove the artificial precipitation center over the eastern edge of the Tibetan Plateau simulated by the MIROC. Both models do a poor job of simulating the interannual variations in annual mean temperature and total precipitation, although the WRF model performs better.