摘要
本文利用中国660个站点逐日地面温度资料,评估了参与政府间气候变化专门委员会第五次报告(IPCC AR5)的9个全球气候模式(Global Climate Models,GCMs)及多模式集合(Multi-Model Ensemble,MME)对中国地区气温的模拟精度。结果表明:9个IPCC AR5全球气候模式和MME模拟的中国地区1996-2005年日平均气温与气象站点观测值的相关系数都大于0.86,表明相关性较好;气候模式模拟的中国东南部地区1996-2005年日平均气温的模拟精度较高,模拟值的偏差、平均相对误差、平均绝对误差和均方根误差都比较小;而西部地区的模拟效果较差,模拟精度较低。综合考虑模式模拟值与站点观测值的相关系数、偏差、平均相对误差、平均绝对误差和均方根误差发现,MME在中国地区的气温模拟精度优于大部分单个模式。
This article evaluates the precision of the temperature simulated by nine IPCC AR5(the Fifth Assessment Report of the Intergovernmental Panel on Climate Change) GCMs(Global Climate Models) and the multimodel ensemble(MME), based on the observed temperature of 660 stations in China from 1996 to 2005. The results show that the correlation coefficients between the average daily temperature simulated by GCMs and station observations in China during 1996-2005 were very high, all above 0.86. The precision of the simulated average daily temperature in the southeast by the 10 models was higher than that in the west, judged by the lower Biases, mean relative errors(MREs), mean absolute errors(MAEs), and root mean square errors(RMSEs) in the southeast as compared to those in the west. The precision of the simulated temperature by IPSL- CM5A- LR,MRI-CGCM3, and Nor ESM1-M was poorer than that of the others—specially, the Biases, MREs, and RMSEs of the simulation result by IPSL-CM5A-LR, the Biases and RMSEs of the simulation result by MRI-CGCM3, and MREs and RMSEs of the simulation result by Nor ESM1-M, were larger. Taken into account the Biases, MREs,MAEs, and RMSEs, the simulation precision of MME was the highest.
出处
《地理科学进展》
CSCD
北大核心
2015年第10期1229-1240,共12页
Progress in Geography
基金
国家自然科学基金项目(41471463)
