Characterizing the surface radiation budget over the Tibetan Plateau with ground-measured, reanalysis, and remote sensing data sets: 1. Methodology

摘要

The surface radiation budget (SRB) over the Tibetan Plateau (TP) greatly influences local climate, climate extremes (e.g., drought, flood) in China, and the East Asian monsoon. However, current estimates of SRB from models and satellite data are subject to large errors, and ground-measured data sets within this region are rather limited over the TP. Our objective is to determine the SRB over the TP by integrating information from three sources: (1) four ground-measured data sets from AsiaFlux, ChinaFLUX, GAME/Tibet, and CAMP/Tibet; (2) four reanalysis data sets from Climate Forecast System Reanalysis, Modern-Era Retrospective Analysis for Research and Applications, ERA-Interim, and Japanese 25-year Reanalysis; and (3) two remote sensing data sets, Global Energy and Water Cycle Experiment Surface Radiation Budget and International Satellite Cloud Climatology Project FD. This study, the first of a two-paper series, presents the methodology. Individual radiation components of reanalysis and remote sensing data set were first validated using ground-measured data from 1997 to 2007; then, a linear regression method was applied to generate the fused data from July 1983 to December 2007. The cross-validation results indicate that the monthly mean root-mean-square errors (RMSEs) of fused downward shortwave irradiance and albedo are 15.1 W m~(-2) and 0.05, respectively; the RMSEs of the downward and upward longwave fluxes are 13.3 and 8.4 W m~(-2), respectively; and the RMSE of all-wave net radiation is as low as 18.9 W m~(-2). Compared to nine sites with long-term observation of downward shortwave irradiance, the fused data represent the decadal variations with higher correlation than using individual products, suggesting the potential for application of the fused data sets in climatic and environmental research. Key Points The surface radiation budget over the Tibetan Plateau is validatedA multiple linear regression method is applied to integrate multiple datasetsThe fused data insolation agreed well with that of other stations