The Variation Characteristics And Spatial Differences Of Extremely High Temperature Days Over China During The Recent 53 Years

China is one of the countries with the most extreme weather or climate events and disasters in the world.Extremely high temperature days lead to more and more serious disaster risk under the Global Warming.It is helpful to develop the corresponding countermeasures of disaster prevention and mitigation based on the local conditions,which by analyzing different regions and various influences about extremely high temperature days.It will provide some reference bases for agricultural production and people's travel and so on.Based on the daily maximum temperature data from 511 meteorological stations in China during 1961?2013,the variation characteristics and spatial differences of extremely high temperature days were studied by ways of Percentiles,linear trend,standard deviation,Mann-Kendall test and Wavelet analysis with t test.(1)This paper calculates the trend rate of national average and each station's extremely high temperature days to analyze the characteristics and regional differences of trend change within liner trend and significant level test.Extremely high temperature days identify a significant increasing;the spatial characteristics reflect a closely connection with the location of land and sea and topography.(2)Using the annual and each station's spatial standard deviation and two adjacent periods'variability change to analyze the characteristics and regional differences of variability and variability change.As time goes by,regional differences of extremely high temperature days are becoming more and more obvious;regional characteristics are sensitive to the latitudinal zonality and topography.(3)Judging from the Mann-Kendall test to analyze national average and each station's mutation,then differentiate the initial and significant mutant periods to reflect the spatial characteristics dynamically.Nearly each station experiences a mutation,the initial mutant periods concentrate in 2000s,1960s and 1990s;1960s and 1990s display a significant decreasing and increasing period respectively;stations' distribution connects with the topography and climate,the later mutations occur,the more concentrate to the inland areas with or without initial and significant mutant periods.(4)This paper identifies the cycle period oscillations with wavelet analysis,then combines(5)with the t-test to judge the cycle period mutations and analyze the spatial and temporal distribution characteristics.The cycle period oscillations show obvious regularity,and the more close to the "dividing line",the lower the cycle period oscillations;the significant decreasing and increasing mutation periods respectively distribute in the earlier 1960s and the late 2000s?2010s,the late 1990s?2000s;the superposed phenomenon of different mutant periods exists in different regions in different degrees;spatial differences are closely related to the location of land and sea and topography.