Sensitivity of the Earth's radiation budget to interannual variations in cloud amount

Abstract

Using two pairs of coincident long-term satellite derived cloud and earth radiation budget data sets (Nimbus-7 ERB/Nimbus-7 Cloud Climatology and ERBE Scanner/ISCCP-C2), estimates are made of the sensitivity of the top of the atmosphere radiation budget to interannual variations in the total cloud amount. Both sets of analyses indicate that the largest net warming due to interannual cloud cover changes occurs over desert regions, while the largest net cooling occurs in areas of persistent marine stratiform cloud. There is generally a large amount of cancellation between the large shortwave cooling and longwave warming effects in tropical convection regions. However, the Nimbus-7 analysis identifies an area of net warming in the tropical eastern Pacific Ocean which is shown to be associated with the 1982–83 ENSO event. In the zonal mean the Nimbus-7 data sets indicate that interannual cloud cover changes lead to a net warming at low latitudes and net cooling polewards of 25° in both hemispheres. In contrast, the analysis of the ERBE and ISCCP data sets indicates net cooling everywhere except for the Northern Hemisphere equatorwards of 20 °N. For the spatial average between 60 °N and 60 °S the ratio of the shortwave and longwave effects is 0.94 in the Nimbus-7 analysis (i.e. clouds cause a small net warming) and 1.21 in the ERBE-ISCCP analysis (i.e. a net cooling). Given their improved spatial and temporal sampling the analysis using the ERBE and ISCCP data sets should be the more reliable. However, the large differences between the two analyses still raises some issues concerning the confidence with which the sign of the effect of clouds on the radiation budget at these time scales is currently known.