Abstract
The Goddard Earth Observing System (GEOS) General Circulation Model (GCM) is part of the GEOS Data Assimilation System (DAS), which is being developed at the Goddard Data Assimilation Office for the production of climate datasets. This study examines Version 1 of the GEOS CYCM by evaluating the quality of the fields that relate most closely to the GCM physical parameterizations and examines the impact of the GCM climate errors on the climate of the DAS assimilated fields.
The climate characteristics are evaluated using independent satellite and ground-based data for comparison. The GEOS-1 GCM shows reasonably good agreement with available observations in terms of general global distribution and seasonal cycles. The major biases or systematic errors are a tendency toward a dry tropical atmosphere and an inadequate cloud radiative impact in the extratropics. Other systematic errors are a generally wet subtropical atmosphere, slightly excess precipitation over the continents, and excess cloud radiative effects over the Tropics. There is also an underestimation of surface sensible and latent heat fluxes over the area of maximum flux.
The DAS climate characteristics, in general, show better agreement with available observations than the GCM. Four distinct ways that the GCM impacts the DAS have been identified, ranging from a DAS climate with little or no impact from the GCM bias to a DAS climate with a greater bias than the GCM due to a spurious feedback between the GCM and the input data.
