CMIP5模式对西南地区气温的模拟能力评估

Evaluation of Surface Air Temperature in Southwestern China Simulated by the CMIP5 Models

利用1961—2005年西南地区(四川、贵州、云南、重庆)115个站点的地表气温观测资料以及国际耦合模式比较第五阶段(CMIP5)的历史模拟试验数据,从气温增暖强度、年代际变化和突变三个角度,评估了40个全球气候系统模式对西南地区地表气温的模拟能力。结果表明:大部分模式能模拟出近45年来西南地区不同分区年平均气温的显著升高趋势,但仅6个模式能较好的模拟出地表气温增温幅度的海拔依赖性特征。海拔较低的四川盆地、重庆丘陵地区年平均气温在20世纪60年代至80年代后期呈降温趋势,80年代末开始升温,70年代中期到90年代中期是一个相对较冷的时期,10个模式能模拟出这种降温趋势,其中3个模式模拟降温趋势、年代际偏冷时间与观测结果最为接近,模拟效果较好。所有模式均不能模拟出气温的突变特征。总体来说,对西南地区气温变化模拟相对较好的模式有ACCESS1.0、CESM 1-WACCM、CM CC-CM S、GFDL-CM 2.1、GISS-E2-R-CC、M RI-ESM 1、Nor ESM 1-M E,其中,模拟效果最好的模式为ACCESS1.0。

The simulated abilities of 40 models for the surface air temperature in Southwestern China（ SWC） are evaluated from the perspective of the warming amplitude,interdecadal variation and abrupt changes by using the surface air temperature observational data at 115 stations in Southwestern China（ Sichuan,Guizhou,Yunnan and Chongqing） from 1961 to 2005 and the output of the historical simulations from 40 models participating in the Fifth Phase of the Coupled Model Intercomparison Project（ CMIP5）. The results showthat most of the models can reproduce the increasing trend of the annual mean temperature in different areas of SWC in recent 45 years,however,only six models can better capture the features that the increasing amplitude of the surface air temperature is dependent on altitude. Annual mean temperature in lower altitude Sichuan Basin and Chongqing hills presents dropping trend from 1960 s to 1980 s,increasing trend from the late 1980 s,and cool period from the middle of 1970 s to the middle of 1990 s. 10 modes could simulated the cooling trend,but the cooling trend and decadal slants cold time is well simulated in only 3 models. All models failed to simulate the characteristics of the temperature mutations. In general,model ACCESS1. 0、CESM1-WACCM、CMCC-CMS 、GFDL-CM2. 1、GISS-E2-R-CC 、MRI-ESM1、Nor ESM1-ME are relatively well in simulating the change of the surface air temperature in SWC,among them,the best is ACCESS1. 0.