Assessment And Prediction Of Snow Cover In The Northern Hemisphere Under Global Climate Change

The fifth report of the IPCC indicates that global temperatures are rising and that this is leading to a steady melting of the global snowpack.As an important component of the climate system,snow has a significant impact on the radiative balance of the earth’s surface and the water cycle.Moreover,98%of the global snowpack is found in the Northern Hemisphere.Therefore,based on NOAA snow cover fraction data,CRU temperature and precipitation data,this study uses trend analysis and correlation analysis to analyze the spatial and temporal distribution of spring(March-May)snow cover fraction and its correlation with climate factors(temperature and precipitation)in the Northern Hemisphere and its 13 sub-regions during the 1982-2014.In addition,the ability of the CMIP6 models to simulate the spatial distribution,trends and intra-annual variability of spring snow cover fraction in the Northern Hemisphere and its 13 sub-regions for 1982-2014 was assessed using Taylor Skill Scores(TSS),Relative Bias(RB),root mean square errors(RMSE)and Comprehensive rating index(CRI)with NOAA as a reference.Then the top three CMIP6 models in terms of comprehensive simulation ability were also selected to predict the spatial and temporal characteristics of future snow cover fraction under three different emission scenarios(SSP1-2.6 low emission scenario,SSP2-4.5 medium emission scenario and SSP5-8.5 high emission scenario).The main conclusions were obtained as follows.(1)The spatial distribution of spring snow cover fraction in the Northern Hemisphere from 1982-2014 is characterized by an increase with increasing latitude(except for high altitude areas such as the Tibetan Plateau and eastern Asia).The interannual variability of spring snow cover fraction in the Northern Hemisphere fluctuates around 40%,with a trend of-0.97%/10 a.Among the sub-regions,the CAS region has the largest trend in snow cover fraction(-3.53%/10a);the CEU region has the smallest trend(0.06%/10a).The multiyear average snow cover fraction for March,April and May in the Northern Hemisphere was 57.45%,41.52% and 21.24%,respectively.Among the subregions,the ALA region had the greatest snow cover fraction(99.85%,97.0%and 59.99%,respectively);the MED region had the least snow cover fraction(5.91%,1.46% and 0.26%,respectively).(2)Spring snow cover fraction in the Northern Hemisphere from 1982-2014 was significantly negatively correlated with temperature with a correlation coefficient of-0.76 and with precipitation with a non-significant negative correlation coefficient of-0.29.All 13 sub-regions also showed a significant negative correlation between spring snow cover fraction and temperature,with the CAS region showing the highest correlation with temperature(R-0.85)and the MED region showing the lowest correlation with temperature(R-0.54).Snow cover fraction was positively correlated with precipitation in the ALA,CGI,CNA,ENA,CEU,WAS,CAS,TIB and EAS regions,while was negatively correlated with precipitation in the WNA,NEU,MED and NAS regions.(3)The top three CMIP6 models in terms of their comprehensive ability to simulate spring snow cover fraction across the Northern Hemisphere are Nor ESM2-MM,CESM2,CESM2-WACCM,CESM2-WACCM-FV2 and Nor ESM2-LM.CMIP6 MME is second only to Nor ESM2-MM in terms of its comprehensive ability to simulate snow cover across the Northern Hemisphere.There are differences in the comprehensive ability of the CMIP6 models to simulate spring snow cover fraction in different sub-regions,with MPI-ESM-1-2-HAM,MPI-ESM1-2-LR and MPI-ESM1-2-HR being less comprehensive in simulating snow cover fraction in most sub-regions(except the TIB and EAS regions).The comprehensive ability of CMIP6 MME to simulate snow cover fraction in most sub-regions(except the ENA and CAS areas)is better than most individual models.(4)As emissions intensity increases,future reductions in spring snow cover fraction in the Northern Hemisphere become more pronounced and the areas showing reductions in snow cover fraction expand.At the end of the 21 st century(2068-2100),snow cover fraction decreases by more than 30% on the Tibetan Plateau and the western margin of North America.Under the SSP1-2.6,SSP2-4.5 and SSP5-8.5 scenarios,spring snow cover fraction in the Northern Hemisphere and its sub-regions(except the MED region)changes relatively similarly until 2040,with significant differences in change after 2040.The trends in future snow cover fraction in the Northern Hemisphere from 2015-2100 under the three different emission scenarios are-0.23%/10 a,-0.54%/10 a and-1.48%/10 a,respectively.In addition,most sub-regions show decreasing trends in spring snow cover under the different emission scenarios,while the NEU and CEU regions show increasing trends in future spring snow cover under the SSP1-2.6 scenario with trends of 0.38%/10 a and 0.14%/10 a,respectively.