A High Resolution Simulation Of Present Climate And Extreme Events Over East Asia By RegCM4.4

Atmospheric-ocean coupled general circulation models(AOGCMs) are the primary tools used for climate change projections. However, due to their coarse resolution, AOGCMs tend to exhibit a poor performance in reproducing the present-day monsoon climate of the region, and dynamical downscaling by high resolution regional climate models(RCMs) are needed. Under the frame of Coordinated Regional climate Downscaling Experiment(CORDEX), twenty years simulation of regional climate over CORDEX(phase II)-EA(East Asia) region has been conducted at a grid spacing of 25 km by the latest version of the Abdus Salam International Centre for Theoretical Physics(ICTP) regional climate model RegCM4.4, for the period 1990-2010. The model is driven by the re-analysis data of ERA-interim.It is important to do some performance tests for the latest regional climate model RegCM4.4 with a lot of improvement. The sensitivity experiments for different combinations of physical parameterization schemes are carried out. The results show that the Emanuel convective precipitation scheme is suitable for the simulation in China by the regional climate model RegCM4.4 with the Community Land Model(CLM) version 3.5.The default land cover data over China of CLM may lead to the bias of the model simulation because of its low resolution and errorness in describing the land surface propertities over the region. Based on the 1:1000000 Vegetation Atlas and 1:6000000 Vegetation Regionalization Map of China, we developed a high resolution land cover data(VEG) with the CLM classification, which can be used in CLM and RegCM4.4. A simulation is used to study the impact of land cover data on surface air temperature, precipitation, and surface energy budget. The results show that the simulation with the VEG data shows a general good performance in describing surface air temperature and precipitation in winter. Most of these improvements are located in South China, which can reduce the dry and cold biases.The contemporary climate is simulated over CORDEX-EA from 1990-2010 by the RegCM4.4 based on the parameters established in the sensitive test. The simulated seasonal mean precipitation and surface temperature in December-January-February(DJF) and June-July-August(JJA) are compared against observation to assess the model performances over the region. The results show that the simulated result of the temperature is close to the observation, especially in the spatial distribution and amount of the temperature. Although the traditional warm-bias in the in high latitudes in the cold season still exists, the revised model has the good capability to reduce the cold bias of surface temperature of most region of China, especially in the Qinghai-Tibet Plateau. The simulation performance of precipitation is not as good as that of temperature. Compared to DJF, the precipitation in JJA is better simulated by RegCM. It can well capture the precipitation of Southeast China. But the overestimation of precipitation in the west and northern China still exists. The bias is higher in DJF and small high in JJA. There is virtual precipitation in the area with a large terrain gradient, especially in the southeast region of the Qinghai—Tibet Plateau. Both the variability and trends of annual mean temperature and precipitation are not as good as that of the mean climate.The extreme events indices related to temperature, such as the extreme maximum temperature(TXx), the extreme minimum temperature(TNn), daily maximum temperature greater than 35°C(T35D), daily maximum temperature less than-10°C(TM35D), the days with a heat index which includes the humidity factor is also greater than 35°C(HI35D), the days with a comfortable index which includes the humidity and wind speed is between 59 and 70(CID), were performed well by RegCM4.4. The overestimates of TXx, T35 D were in eastern China, while the underestimates of TNn, TM10 D, HI35 D, CID were in some area of China. The simulated extreme events indices related to precipitation intensity-RR1(1~10 mm/d), RR10(10~20 mm/d), RR20(≥20 mm/d) according to the daily rainfall were analyzed. Results show that the distributions of the indices are well simulated by the model, except that the overestimates in northern China.