As one of the major satellite-estimated surface radiation budget (SRB) dataset, the Clouds and the Earth's Radiant Energy System-Monthly Gridded Radiative Fluxes and Clouds (CERES-FSW) data have been widely used in various applications. Although CERES/ARM Validation Experiment (CAVE) has performed the validation work of the input data of CERES-FSW, accuracy of the CERES-FSW data itself is still unknown. So there is a need for a validation of this radiation dataset, especially at some regions with special climate or terrain conditions. The Tibetan Plateau, due to its protruding nature, has an impact on energy and water cycles on both regional and global scales, and it is a perfect research area for comprehensive awareness of data quality and applicability of estimation algorithms. Accordingly, the Plateau has been the subject of climate research for several decades. In this paper, we validates the four major components SRB of the CERES-FSW dataset with ground measurement at the Tibetan Plateau. The four components include: downward shortwave radiation (SWDN), upward shortwave radiation (SWUP), downward longwave radiation (LWDN) and upward longwave radiation (LWUP). It is found that CERES-FSW data have comparatively high accuracy in LWDN and LWUP, while there is slight underestimation at most sites in these two components. The mean bias is -27.3W/m(2) in LWDN and -7.0W/m(2) in LWUP, the mean root mean square errors (RMSE) is 40.0W/m(2) in LWDN and 40.4W/m(2) in LWUP. Mean coefficient of determination (R-2) of the two components are both over 0.80. But the accuracy of SW is not as good as that of LW. SWDN of CERES-FSW data are overestimated at most sites with lower R-2, large bias and RMSE. SWDN has a mean bias of 41.7 W/m(2), mean RMSE of 174.0 W/m(2) and mean R-2 of 0.39. And for SWUP, CERES-FSW data have low accuracy with very small R-2 and large RMSE. SWUP has a mean bias of 9.7W/m(2) and mean RMSE of 95.2 W/m(2), but the mean R-2 is only 0.08. Therefore, special attention should be given when using CERES-FSW data at Tibetan Plateau.
