Observation,Simulation And Future Prediction Of Evaporation In China

Evaporation is an important link between the atmosphere and hydrosphere,and one of the main links in the hydrological cycle,maintaining the heat balance and water balance at the surface.Evapotranspiration has important applications in the estimation of surface evaporation,the water demand of crops and the maintenance of water balance.Therefore,the study of evapotranspiration is of great importance to reveal the change pattern of hydrological cycle and to gain insight into climate change.In this thesis,based on the monthly pan evaporation（PE）of 573 meteorological stations in China from 1960 to 2021 collected and compiled by the National Meteorological Information Center,we used two homogenization methods,MASH（multiple analysis of series for homogenization）and Climatol to test and revise the inhomogeneity,and quantitatively assessed the uncertainty of the results of the two homogenization methods in terms of the number of non-homogeneous stations,the number of detected breakpoints,and the magnitude of revisions.We reveal the spatial and temporal evolution characteristics of annual and seasonal PE in China in the past 60 years using the equal-weighted averaged monthly PE series dataset after homogenization.And on the basis of which,the potential evaporation is calculated by K_pan,and the simulation effect of CMIP6 potential evaporation is evaluated qualitatively.Finally,we perform future prediction and uncertainty analysis for potential evaporation in the Chinese region.The conclusions are as follows:（1）Both MASH and Climatol are good at detecting and capturing breakpoints associated with relocation and manual-to-automated observations,etc.However,there are differences in the number of breakpoints,drift values and revision magnitude.（2）After the homogenization,most stations in the country showed a decreasing trend in spring,summer and annual PE from 1960 to 2021,accounting for 81.7%,80.8%and80.3%,respectively,while 57.1%and 60.4%of stations showed an increasing trend in winter and autumn.In winter,there is an upward trend in northeast,eastern Qinghai-Tibet Plateau,south and southwest China（except Yunnan）;in spring,most of the stations show a downward trend except for the coastal areas of east China,southern Shaanxi,northern Sichuan and Chongqing,and western Hubei;in summer,most of the stations show a downward trend especially in the eastern areas south of north China;in autumn,there is a downward trend except for northwestern northeast,northwestern Xinjiang,and central Inner Mongolia extending to the eastern part of Qinghai-Tibet Plateau In autumn,except for northwestern northeast,northwestern Xinjiang,and central Inner Mongolia extending to the eastern part of the Tibetan Plateau,other areas are on the rise;annual PE in northwestern Xinjiang,central Inner Mongolia,Shandong,southern Hebei and Henan,central Yunnan and other areas are on the decline.（3）After the homogenization,the national average PE showed a slight upward trend of 0.27 mm/10a and 1.10 mm/10a in winter and autumn,while the spring,summer,and annual PE showed significant downward trends of-8.38 mm/10a,-9.83 mm/10a,and-16.83 mm/10a,respectively.The ranges of seasonal and annual PE trends all tend to be smaller,showing that the spatial consistency of large-scale trends tends to be better.In spring,autumn,winter and annual,PE abrupt changes exist in 1977,2018 and 2020,2020and 1972,respectively,while no abrupt changes exist in summer.（4）Spatially,ET0_CMIP6 generally showed a greater trend than ET0pan across seasons and years.Temporally,the difference between the two is smaller in spring,summer,and year than in winter and autumn.SSP245 and SSP585 scenarios show a continuous increase in ET0 over time for each season as well as for the whole year,with SSP585 increasing more than SSP245 in the near and distant future,both peaking at the end of the 21st century,and increasing most in summer,followed by autumn and spring,and least in winter.SSP245 scenario has less uncertainty than the SSP585 scenario.The uncertainty of the winter prediction is the smallest;the uncertainty of the CNRM-ESM2-1 and EC-Earth3-Veg-LR predictions of potential evaporation is the largest.