Determination of surface albedo from satellites

The dynamical features of the climate system result from the interaction of the atmosphere with the surface. The hope for improving climate prediction, on seasonal and interannual time scales, is based on the premise that slowly varying boundary conditions force well defined predictable patterns of general circulation.It is now recognized that in models of surface climate over land, surface properties should not be regarded as constants and that it is important to specify a realistic value of the surface albedo¹ in order to correctly estimate the amount of solar radiation absorbed at the surface. The albedo is also considered by some as an internal dependent parameter of climate, since it time integrates the effects of changes in more variable quantities, such as rainfall.The global nature of the links between forcing and response imply global monitoring of the parameters which control the transfer and feedback of the energy at the surface/atmosphere interface. Satellites can play an important role in resolving the difficult problem of properly representing the average surface albedo over large areas. However, satellites measure only the earth-atmosphere reflectance in narrow spectral intervals, narrow solid angles, and in most cases at a fixed local time. To derive from these observations the effective surface albedo, one has to compute the total reflected planetary flux, integrated over the whole solar spectrum, over all the viewing angles, and over time. A transformation from top of the atmosphere to the surface is also required.In this presentation the implications of atmospheric corrections, bidirectional reflectance, and transformations from narrow spectral bands to the total solar spectrum for albedo derivations will be addressed. Also, the various attempts to derive surface reflectivity and surface albedo from satellites will be reviewed.