Study On The Applicability Of Reanalysis Data For Albedo Modeling In The Qinghai-Tibet Plateau

The surface albedo plays a crucial role in controlling the amount of solar radiation absorbed by the Earth’s surface,and thus affects regional and global climate change.As a warning zone and sensitive area for climate change,the temporal and spatial characteristics of surface albedo on the Qinghai-Tibet Plateau are of great significance for climate research and regional environmental management.However,obtaining long-term surface albedo data is challenging due to the limited number of ground observation sites and the complex underlying surface effects.Reanalysis data has become an important source of data for studying the temporal and spatial variations of surface albedo,but its applicability and sources of errors need further investigation.Therefore,in this paper,multiple sets of widely used reanalysis data were used to study the temporal and spatial characteristics of surface albedo on the Qinghai-Tibet Plateau,and the evaluation was conducted based on MODIS albedo data.The applicability of these reanalysis data in different time and space scales was explored,and the important factors influencing surface albedo,such as snow cover and vegetation,were analyzed.The main conclusions are as follows:(1)Reanalysis data and MODIS albedo data show similarities in their overall spatial distribution patterns.The spatial distribution of albedo generally exhibits higher values in the northwest and lower values in the southeast,with some high albedo regions in the central part of the Qinghai-Tibet Plateau,including the Kunlun Mountains,Himalayas,and Nyainqentanglha Mountains.The low albedo regions are mainly concentrated in the Hengduan Mountains and the Three Rivers Source area.In terms of temporal variations of multi-year and seasonal mean surface albedo in the Qinghai-Tibet Plateau,JRA-55 and MERRA-2 tend to overestimate,while MERRA-2 and CERES-EBAF show smaller differences compared to MODIS.CERES-EBAF can well capture the interannual variability trend of MODIS albedo,while ERA5 and JRA-55 can better match the trend of variability.(2)The variability of surface albedo in MERRA-2 and CERES-EBAF is lower than that in MODIS,while JRA-55 and ERA5 show higher variability compared to MODIS.The root mean square error(RMSE)and correlation coefficient between surface albedo from CERESEBAF and MERRA-2 with MODIS are better than those between JRA-55 and ERA5 with MODIS.The assimilation quality of ERA5 albedo is greatly influenced by season,possibly due to errors in snow data during assimilation;while JRA-55 product is less affected by season.In summary,the ranking of assimilation quality of albedo at annual and seasonal scales is as follows: autumn > summer > annual mean > spring > winter.The ranking of assimilation quality of albedo from reanalysis data is as follows:CERES-EBAF > MERRA-2 > JRA-55 > ERA5,with little difference between CERES-EBAF and MERRA-2,indicating similar performance in simulations.(3)The high snow cover area in spring and winter from JRA-55 is consistent with the high surface albedo area,while ERA5 shows higher spatial consistency between snow cover and surface albedo data.In spring and winter,JRA-55 overestimates surface albedo data,but the difference between snow cover data and MODIS observations is small,indicating that snow data may not be the main factor affecting JRA-55 surface albedo data.ERA5 snow cover data shows consistent trends with surface albedo data in both annual and vertical variations,and both are significantly overestimated,indicating that snow data is a major factor affecting ERA5 surface albedo data.There is a high correlation between snow cover and albedo data,indicating consistent trends in trend changes between snow data and albedo data in reanalysis data.Leaf area index(LAI)is an important factor affecting surface albedo variations in MERRA-2 in annual,summer,and autumn periods,while the impact of ERA5 LAI on surface albedo variations is relatively small in summer and autumn.