Spatial-temporal Variations Of Snow In The Tibetan Plateau And Their Influences On Summer Rainfall Over China

The Tibetan Plateau(TP)is a plateau with the highest elevation and the most complex terrain in the world.With an average elevation exceeding 4,000 meters,TP is termed as “the Third Pole” of the earth.The snow cover over the TP is one of key physical factors that influence the climate in China as well as in other regions in Asia.In this paper,by using the Empirical Orthogonal Function(EOF)decomposition,composite analysis,and other statistical methods,we first examine the interannual and interdecadal variations of winter and spring snow(in terms of the number of days with snow cover(NDSC)and snow depth(SD))in the entire TP,the eastern TP(ETP),and the western TP(WTP),respectively.Then we identify the typical years with anomalous snow cover in the ETP and WTP,respectively,and investigate the features of snow cover anomalies and the associated atmospheric circulation anomalies.Finally we investigate the impacts of snow cover anomalies in TP on the summer rainfall over China.The results show that significant difference exists in the temporal variations of snow cover for the different regions of TP.The results also show apparent correlation between snow cover variations in TP and summer rainfall in China.In particular,the impacts of snow cover anomalies in ETP on the summer rainfall in China are significantly different from those of snow cover anomalies in WTP.Therefore,it is necessary to take into account the different impacts exerted on summer rainfall by snow cover anomalies in different regions of TP when snow cover in TP is used as a factor for short-term predictions of summer climate in China.The main conclusions were summarized as follows:(1)In the past 53 years(1961-2013),the temporal trends of winter and spring snow averaged over the entire TP are general similar.The interdecadal variations of SD exhibit the feature of “less-rich-less-rich” than normal,and those of NDSC shows the feature of “less-rich-less” than normal.For the temporal trends of snow cover in two regions(ETP and WTP),snow increased significantly in the 1960s-1970 s and reduced in the 1980s-1990 s for both the two regions.However,the trends were obviously different since the late 1990 s,when the snow cover reduced significantly in the ETP whereas it significantly increased in the WTP except for the spring NDSC that shows no apparent trends.Corresponding to the significant increase in global surface temperatures since 1980 s,these results indicate that the snow in ETP is more sensitive to global climate change than in WTP.(2)Based on the EOF decomposition of NDSC and SD,only two EOF modespass the significance test for winter NDSC and three EOF modes pass the significance test for other variables(spring NDSC,winter SD,and spring SD).Further analysis shows that the first mode explains relatively low fraction(less than 25%)of total variations for all four quantities.Moreover,as there are larger spatial variations for SD than for NDSC especially for some local regions,the first mode of SD explains mode smaller fraction of variance than that of NDSC.For the spatial patterns of EOF modes,the first mode shows uniform distribution(i.e.,either positive or negative)for spring SD,spring NDSC,and winter NDSC,and the second mode shows anti-phase variations in the ETP and WTP.In addition,the first mode of winter SD exhibits anti-phase variations in the ETP and WTP.For the temporal coefficients corresponding to each EOF mode,the first mode shows obvious decadal variations,which contributes significantly to the decadal variations of snow in TP in the past 50 years.(3)Atmospheric circulation associated with snow cover anomalies in ETP and WTP exhibits obvious difference.In the ETP,the atmospheric circulation in the upper geopotential height is significantly lower than normal in this region when there is abnormally more snow than normal,while it is higher than normal when there is abnormally less snow than normal.A lower(higher)geopotential height is generally favorable for(adverse to)snowfall in the ETP.Similar case is found for the WTP.(4)Snow cover anomalies in the ETP and WTP influence the atmospheric circulation,and thus the summer precipitation in China.There are differences in the summer precipitation change in response to the snow cover anomalies in the ETP and WTP,respectively.With more snow than normal in the ETP,summer precipitation is more than normal in Northwest China,North China,the Yangtze River basin,and the southern part of East China,while summer precipitation is less than normal in the northern part of East China,Huaihe river basin,South China,and Southwest China.With less snow than normal in the ETP,summer precipitation is more than normal in North China,Central China,the Huaihe river basin,and Southwest China,while it is less than normal in the Northwest and East China.For the WTP,with more snow in this region,summer precipitation is more than normal in Northwest China,Northeast China,the Huaihe river basin,and Southwest China,while it is less than normal in North China,the Jianghuai river basin,South China,and the eastern part of Southwest China.In contrast,with less snow in the WTP,summer precipitation are more than normal in the Huanghuai river basin,East China,South China,and Southwest China,while it is less than normal in the Northwest China,the middle and lower reaches of the Yangtze river,and the Huaihe river basin.Because the snow cover anomalies in the ETP and WTP induce different(even opposite)impacts on the summer precipitation in the different regions over China,it is not appropriate to consider the plateau as a whole when studying the impacts of snow cover anomalies on the summer precipitation in China.Instead,it may be necessary to divide TP into ETP and WTP to derive more accurate signals from the snow cover anomalies,and use them for the seasonal prediction of summer precipitation in the different regions over China.