Using NCAR Community Atmosphere Model (CAM 3.1), two numerical experiments are designed to explore the possible influence of the soil moisture anomaly on the simulation of extreme climate events. Results show that the simulated extreme climate events are sensitive to the soil moisture anomaly, and the anomaly has significant impacts on the climatology, interannual variability, and interannual variations of extreme climate events. Without the interannual anomaly of soil moisture included, the simulated Tn95p (warm nights), Tx95p (warm days), and HWDI (heat wave duration) significantly decrease in most areas of China, while FD (frost days) increases on the whole. It is also found that the response of the simulated precipitation extremes to the soil moisture anomaly exhibits significant spatial difference. The simulated P95p (the frequency of extreme heavy precipitation) obviously decreases in the Huaihe River basin, and R95p (the intensity of extreme heavy precipitation) also drops in the northeast of China but increases in the Yangtze River valley. Both the simulated R10 (number of days with precipitation greater than 10 mm) and CWD (the consecutive wet days) have experienced an evident drop in most areas of China. In addition, the interannual variability of the temperature extremes has decreased in most areas of China, but the situations of precipitation extremes are much complicated. Both the interannual variability of P95p and R95p decrease in North China and increase over the regions south of the Yangtze River. Throughout the whole country, the interannual variability of R10 and CWD decrease. Results also suggest that the performance of the model in simulating the interannual variations of the extreme climate indices, especially for Tn95p and FD, will become worse when the interannual anomaly of the soil moisture is removed.