基于CMIP6气候模式的华北平原极端气温指数评估和预测

Evaluation and Projection of Extreme Temperature Indices in the North China Plain Based on CMIP6 Models

随着全球气候变暖,极端气候事件频发,对自然生态系统、社会经济及人类生产生活带来重要影响。预估未来气候情景下极端气候事件的发生有助于准确评估气候变化的影响程度。该文基于华北平原54个气象站的气候观测资料及CMIP6中20个全球气候模式(GCM)在4个气候情景下的未来气候预测数据,计算9种代表性极端气温指数,利用独立权重均值法(IWM)对多GCMs预测结果进行集合平均,旨在揭示华北平原历史(1971-2010年)和未来(2061-2100年)极端气温指数的时空变化。结果表明:基于IWM方法获得的多模式集合平均值与观测值的均方根误差(RMSE)和相对标准偏差(RSD)均低于算术平均法,能更好地反演历史极端气温指数的变化趋势;预测未来极端高温指数呈显著上升趋势,而极端低温指数将显著降低;极端气温指数(DTR除外)在历史和未来时期均存在明显的空间差异,其中高辐射强迫情景(SSP585)下的极端气温指数变幅最大,空间差异最显著。

With the global climate warming,extreme weather events occur frequently,which have important impacts on natural ecosystems,social economy,and human production and life.Predicting the occurrence of extreme climates under future climate scenarios helps to assess the impact of climate change on agriculture and natural ecosystems.This study was to calculate 9 representative extreme temperature indices using the historical daily climate data of 54 meteorological stations and the projection data from 20 global climate models(GCMs)in the sixth international coupled model comparison program(CMIP6)under 4 typical climate scenarios in the North China Plain.The independent weighted mean method(IWM)was applied to average the projection results of multiple GCMs,aiming to reveal the temporal and spatial changes of the extreme temperature indices in the history(1971-2010)and the future(2061-2100)periods in the study area.The results show that the root mean square error(RMSE)and relative standard deviation(RSD)between the ensemble average result obtained by the IWM model and the observation value were lower than that of the arithmetic average method,which can better reproduce the change trends of the extreme temperature indices in the historical period.In the future,the extreme high temperature index will show a significant upward trend,while the extreme low temperature index will decrease significantly.The extreme temperature indices(except DTR)have obvious spatial differences in both historical and future stages.Among them,the extreme temperature index under the high radiative forcing scenario(SSP585)has the largest amplitude of variation and the most significant spatial difference.