四川区域CMIP6模式模拟能力评估

Evaluation of CMIP6 Model Simulation Capability in Sichuan Region

摘要: 为了进一步研究CMIP6模式对四川地区地表温度、降水的模拟能力，本文使用CMIP6气候模式数据集中14个模式温度数据和8个模式降水数据的历史模拟试验结果，与CN05.1格点化观测数据进行了对比分析。通过对时间平均结果的对比分析以及空间相关分析，对CMIP6模式数据关于四川地区的模式降水、模式地表温度在时间平均值和空间分布相关性的模拟能力进行对比分析，结果表明：1) 模式对于降水的模拟值偏高1.3 mm/day，基本都能较好的模拟出四川地区的降水空间分布，MIROC6模式的模拟效果最优，就相关性而言，其次为CanESM5、NESM3模式，就空间分布而言，CESM2、CESM2-WACCM模拟效果也不错。2) 模式对于地表温度的预估值偏低10 K，基本都能较好的模拟出四川地区气温的空间分布，但因为气候模型的水平分辨率相对较粗糙，大多数局部特征难以表征，尤其是在东南盆地以及中西部地区。关于地表温度、预估量和空间分布的结论比较一致，CESM2、CESM2-WACCM模式在两个方面的模拟效果都是最好，其次为CNRM-ESM2、CNRM-CM6模式。

Abstract: This paper used the historical simulation data of CMIP6 model and CN05.1 grid observation data to evaluate the ability of CMIP6 model of simulation of the surface temperature and precipitation in Sichuan province. Through the time-averaged comparison and spatial correlation analysis, the CMIP6 model is compared and analyzed on the simulation ability of the model surface temperature and precipitation in the Sichuan area. The results show that: 1) The time-averaged estimated precipitation is 1.3 mm/day higher, which can basically simulate the spatial distribution of precipitation in Sichuan. The simulation effect of the MIROC6 is the best, followed by the CanESM5 and NESM3. In terms of spatial distribution, CESM2, CESM2-WACCM simulation results are also good. 2) The estimated surface temperature of the model is 10 K lower, which also can basically simulate the spatial distribution of the temperature in Sichuan. However, because the horizontal resolution of the climate model is relatively rough, most local features are difficult to characterize, especially in the Southeast Basin and the Midwest in Sichuan. Regarding the surface temperature, the conclusions of the estimated amount and the spatial distribution are relatively consistent. The simulation effects of the CESM2 and CESM2-WACCM are the best in two aspects. CNRM-ESM2 and CNRM-CM6 are followed.