Global daily actual and snow-free blue-sky land surface albedo climatology from 20-year MODIS products Journal of Geophysical Research: Atmospheres

Land surface albedo plays a critical role in climate, hydrological and biogeochemical modeling, and weather forecasting. It is often assigned in models and satellite retrievals by albedo climatology look-up tables using land cover type and other variables; however, there are considerable differences in albedo simulations among models, which partially result from uncertainty in obsolete albedo climatology. Therefore, this study introduces a new global 500 m daily blue-sky land surface albedo climatology data set under both actual and snow-free surface conditions utilizing 20-year Moderate Resolution Imaging Spectroradiometer products from Google Earth Engine. In situ measurements from 38 long-term-maintained sites were utilized to validate the accuracies of different albedo climatology datasets. The root-mean-square error, bias, and correlation coefficient of the new climatology are 0.031, −0.003, and 0.96, respectively, which are more accurate than the Global Land Surface Satellite, GlobAlbedo, and 16 model datasets. Data intercomparison suggests that ERA5 exhibits better performance than Modern-Era Retrospective Analysis for Research and Applications Version 2 (MERRA2) and 14 Coupled Model Intercomparison Projects Phase 6 models. However, it contains positive biases in the snow-free season, while MERRA2 underestimates the snow albedo. Global albedo variation associated with basic surface plant functional types was also characterized, and snow impact was considered separately. Temporal variability analysis indicates that traditional climatology datasets with coarser temporal resolutions (≥8 days) cannot capture albedo variation over areas with distinct snow seasons, especially in central Eurasia and boreal regions. These results confirm the high reliability and robustness of the new albedo climatology in model assessment, data assimilation, and satellite product retrievals. Grant no. NA19NES4320002