On Solar Energy Disposition:A Perspective from Observation and Modeling

Abstract

Solar energy disposition (SED) concerns the amount of solar radiation reflected to space, absorbed in the atmosphere, and absorbed at the surface. The state of knowledge on SED is examined by comparing eight datasets from surface and satellite observation and modeling by general circulation models. The discrepancies among these contemporary estimates of SED are so large that wisdom on conventional SED is wanting. Thanks to satellite observations, the earth's radiation budget (ERB) at the top of the atmosphere is reasonably well known. Current GCMs manage to reproduce a reasonable global and annual mean ERB, but often fail to simulate the variations in ERB associated with certain cloud regimes such as tropical convection and storm tracks. In comparison to ERB, knowledge of the surface radiation budget (SRB) and the atmospheric radiation budget (ARB) is still rather poor, owing to the inherent problems in both in situ observations and remote sensing. The major shortcoming of in situ observations lies in insufficient sampling, while the remote sensing techniques suffer from lack of information on some variables affecting the radiative transfer process, and dependence, directly or indirectly, on radiative transfer models. Nevertheless, satellite-based SRB products agree fairly well overall with ground-based observations. GCM-simulated SRBs and ARBs are not only subject to large regional uncertainties associated with clouds, but also to systematic errors of the order of 25 W m?2, due possibly to the neglect of aerosol and/or inaccurate computation of water vapor absorption. Analyses of various datasets suggest that the SED based on ERBE satellite data appears to be more reliable, indicating 30% reflection to space, 24% absorption in the atmosphere, and 46% absorption at the surface.