Study Of Temperature Changes And Applicability Of Reanalysis Data In The Tianshan Mountains

With the increase of global warming and extreme weather events,it is necessary to analyze and forecast regional climate change.Because of its special geographical location and complex topography,Tianshan mountain has an extremely important influence on the climate change and ecological environment in the northwest region of China.Tianshan is also a natural barrier in the climate and ecological environment of the central and western regions of China.Tianshan mountain is composed of mountains,mountain basins and valleys,and mountains and plains.The mountain natural system is complete,and it has good typicality and representativeness.This research mainly includes the following two aspects:(1)In this study,14 representative meteorological stations were selected according to integrated physicogeographical regionalization of Tianshan Mountains.The linear trend analysis.Mann-Kendall test and R/S method were used to analyze the characteristics of daily temperature changes and trends in 4 natural zones and 14 natural regions of Tianshan Mountains from 1960-2013.(2)According to daily temperature data from 32 meteorological stations in Tianshan Mountainous,the root mean square error(RMSE).absolute error(MBE)and probability density function(PDF)were selected to verify the reliability of ERA-Interim and ERA-20CM(10 sets of ensemble forecast data)from European medium-term forecast center(ECMWF)based on measures-oriented and distributions-oriented evaluation methods.The following results are obtained through the above study:(1)The temperature was increasing since 1990s,especially significant since 21th century with a mutation in the late of 1990s.The temperature trend rates were different in 4 nature zones,which decreased in the order of Ili valley,north slope of Tianshan Mountains.south slope of Tianshan Mountains,and Bayanbulak.The daily minimum temperature increased greater than the daily average and maximum temperatures.In terms of seasons.the warmer winter was mainly due to the increasing daily minimum temperature.The temperature would continue to increase in Ili valley,south slope of Tianshan Mountains and Bayanbulak in the future according to Hurst index.In particular,the daily minimum temperature may be more significant.The increasing minimum temperature has a significant contribution to the warming trend of daily average temperature.The warming trend is possible to slow down or decrease in north slope of Tianshan Mountains in autumn and winter.(2)ERA-Interim and ERA-20CM have a good credibility in Tianshan Mountainous.However,the performance differences in regional are significant.From the comparison of sites aspect,RMSE and MBE from ERA-I are superior to the ERA-20CM on average daily temperature,daily maximum temperature,daily minimum temperature.According to the PDF skill score,the performances of ERA-I in the daily maximum temperature and daily minimum temperature are inferior to the ERA-20CM.There is a tiny difference between the ten ensemble sets of ERA-20CM.The regional differences of reanalysis are more significant than that of reanalysis data.From the seasonal variation,reanalysis performed the best in autumn,the worst in winter,the middle level in spring and autumn.In general,ERA-I is better than ERA-20CM.The regression analysis shows that the elevation difference between ECMWF grid point and ground station is an important error source,especially for the daily maximum temperature.