| The global climate is undergoing significant changes characterized primarily by warming in the past century,and it will continue for some time to come.Global warming had led to changes in weather and climate extremes around the world.On this background,apparent changes of features of extreme cold events in China were observed.Previous studies have mainly analyzed the changes of characteristics of extreme cold events after the mid-20 th century,and mostly adopted the equalweighted multi-model ensemble method in the future projection.Based on a series of climate extremes indices describing the characteristics of extreme cold events,the ability of reanalysis data and the newest model data to reproduce the changes of extreme cold events in China since the mid-20 th century was evaluated.Finally,this article analyzed the characteristics of the changes in the last century and projected the future changes in 2021-2100 of extreme cold events in China based on the above datasets.In projection,a weighted multi-model ensemble method was used.Results are as follows.(1)Consisting with global warming,the intensity,frequency,and duration of extreme cold events in China had tended to decrease since the early 20 th century,and the number and duration of cold waves defined by the Excess Cold Factor showed the trends of reduction.Meanwhile,the decrease in Heating Degree Day reflected the decline of heating energy demand in China.Trends of the characteristics of extreme cold events were more pronounced after 1951,and the variation of some indices significantly exceeded the century-scale changes,which indicated a rapid weakening and reduction of cold events after the mid-20 th century.It’s worth noting that the two sets of centennial scale reanalysis data can be used to study the changes of regional extreme cold events in China,but the uncertainties are large in areas with insufficient spatio-temporal coverage of site observations,the deficiencies caused by spatial coverage and quality issues need to be carefully considered.(2)The global climate models in CMIP6 could basically reproduce the distribution of mean climate states and trends of extreme cold indices in China,such as north-south distribution characteristics of the climate states and north-south gradient distribution of the trends,colder climate states and more significant trends in the Qinghai-Tibet Plateau than other regions at the same latitude.However,the cold bias which was greater in the high-altitude area remained in most regions in China as in previous phases of CMIP.The multi-model ensemble means performed better than most of the individual models,and they could capture the characteristics of long-term trend of extreme cold events in China since the mid-20 th century.Nevertheless,there were quantitative differences between the model-simulated and observed trends.By comparison,the climate models had a considerable capacity for simulating characteristics of temporal and spatial variation of regional extreme cold events in China since the mid-20 th century,but there were still some limitations in portraying some details.The simulated results were different for different multi-model ensembles.(3)Under the medium to high emission scenarios,the intensity,frequency and the number of days of extreme cold events in China continue to decrease over time in2021-2100,and the change is more significant under the high emission scenario.Under the low emission scenario,there were no significant changes in indices after2081.The weighted results based on model performance and independence show smaller warming and narrower ranges of uncertainty,the reductions of the uncertainty ranges mainly occur at the warm end.Absolute difference between the weighted and unweighted projections increases with time and the rise of the greenhouse emission in scenarios.The effect of observational constraint projection is not obvious in projecting the percentile threshold indices.Spatially,the Qinghai-Tibet Plateau region remains to be the high value area for the changes of indices. |
PDF Full Text Request