Key points are not available for this paper at this time.
Global, daily, visible and infrared radiance measurements from the NOAA-5 Scanning Radiometer (SR) are analyzed for the months of January, April, July and October 1977 to infer cloud and surface radiative properties and their effects on the Earth and surface radiation budgets. The analysis makes use of several additional datasets to help isolate the cloud contributions. The cloud properties inferred from satellite data are found to be about as accurate as the validation datasets available from other sources, but significant improvements are needed for better diagnosis of cloud-radiative feedbacks. Consequently, advances in cloud retrievals from satellite data will be like “new measurement” that have no independent validation. Reconstruction of regional, monthly mean Earth radiation budgets (ERB) from cloud, atmosphere and surface data is also nearly as accurate as can be checked with summaries of other, more direct measurements, improvements require detailed intercomparisons at smaller space/time scales. Currently, there is no global dataset with which to validate the reconstructed surface radiation budget (SRB). Comparison of monthly, regional mean quantities with those simulated by the GISS climate GCM shows that the differences are only a little larger than the uncertainties in the results; thus, better data and more detailed comparisons will be needed to improve the GCMs. Despite important limitations in these results, several fundamental conclusions about the role of clouds in the radiation balance of Earth are apparent. 1) The magnitude of cloud property variations and their effects on radiation increase strongly with decreasing space/time scales, going from global, annual means to regional, monthly means; 2) Cloud properties are systematically different between land and ocean: oceans have larger cloud cover with somewhat larger optical thicknesses and lower cloud top altitudes; 3) Although cloud variations appear to be the primary cause of regional radiation budget variability at 5–30 day time scales, the effects of their seasonal variations at larger spatial scales are less important than the changes associated with changes in solar declination and atmospheric/surface temperatures, 4) The largest seasonal Variations in radiation occur in the 30°–60° latitude band in each hemisphere, 5) Cloud variations tend to enhance regional and seasonal radiation variations at lower latitudes and mute them at higher latitudes; however, they also affect the average latitudinal gradients of heating/cooling; 6) Although clouds have a net cooling effect in the global, annual mean radiation balance at both the top of the atmosphere and the surface, their net effect on regional, seasonal balances is much more varied; 7) Conclusions (5) and (6), together, are equivalent to saying that the relation between cloud properties and their effect on ERB and SRB depend crucially on the regional and seasonal circumstances of the clouds; 8) Regional cloud and surface seasonal change amplitudes and phase exhibit a wide variety of values; moreover, the correlations between surface temperature and cloud properties vary greatly; 9) There appears to be no simple relation between global mean surface temperature, global mean cloud properties and their global mean effects on ERB and SRB, implying that cloud radiative effects on the seasonal temperature cycle must be described as multiple feedbacks, and 10) The complexity of the cloud radiative effects and the data accuracy required to diagnose cloud-radiative feedbacks indicate the challenge of this problem.
Rossow et al. (Thu,) studied this question.