基于CMIP5资料的西南地区2020—2050年气温多模式集合预估

Multimodel Superensemble Prediction of Air Temperature in Southwestern China During 2020-2050 Based on CMIP5 Data

利用1961—2005年西南地区2 m气温的观测资料及同期CMIP5的11个全球气候系统模式的历史模拟数据,对比分析模式模拟、统计降尺度方法模拟、多模式集合模拟、统计降尺度和多模式集合相结合方法模拟的西南地区及不同分区气温误差。结果表明统计降尺度方法和多模式集合方法都能有效降低模拟误差,多模式集合的模拟误差相对较小。选取多模式集合方法预估RCP4. 5中等偏低辐射强迫情景下2020—2050年西南地区2 m气温的变化,发现2020—2050年西南地区年平均及四季气温都呈显著上升趋势,冬季气温增幅相对较高,夏季相对较低;气温增幅较高的区域主要位于102°E以西,较低的区域位于四川西南部和云南西北部交界处。

Based on the observed data of2m air temperature in southwestern China from1961-2005and the corresponding period data simulated by11global climate system models of CMIP5,the2m air temperature biases in different regions in southwestern China from the models simulation and that obtained by the statistical downscaling,multi-mode ensemble simulation,as well as the joint combination of the both methods were analyzed.The results show that both statistical downscaling and multi-mode ensemble simulation method could effectively reduce simulated errors,and the RMS(root-mean-square)error from the latter was relatively small.Based on the multi-mode ensemble simulated results under the RCP4.5scenario,the annual and seasonal averaged2m air temperature in southwestern China showed obviously increasing trend during2020-2050,which was relatively larger in winter and smaller in summer.The increasing amplitude of2m air temperature was higher in the west of the102°E in southwestern China,and relatively lower in the junction of southwestern Sichuan and northwestern Yunnan.