The sensitivity of a land surface parameterization scheme to the choice of remotely-sensed landcover data sets

The characteristics of satellite-derived landcover data for climate models vary depending on sensor properties and processing options. To better understand the first order effects of differences in landcover data on a land surface parameterization scheme (VBATS), stand-alone model runs were performed for two adjacent 2.8° by 2.8° GCM gridcells in Wyoming using landcover from two satellite-derived maps (AVHRR, TM) and a global landcover data set commonly used in GCMs. Substantial differences in prescribed landcover were found between the three datasets. Despite these differences, the VBATS simulated surface fluxes were similar in the eastern gridcell for the two satellite data sets. In the western gridcell, the partitioning of net radiation into sensible and latent heat fluxes was affected by the relative proportions of wet cover types (i.e. inland water and irrigated crop) prescribed by the two satellite data sets. This emphasizes the importance of accurately estimating the proportion of wet cover types within a GCM gridcell in arid regions. Spatial aggregation of the satellite data sets reduced the number of cover types used to represent each GCM gridcell. In the western gridcell, a reduction in the number of cover types from 11 to 2 resulted in differences in annual averages of sensible and latent heat fluxes of about 10%. Other simulations involving these data sets suggest that these differences could be reduced if the wet cover types are accounted for. In this respect, fine spatial resolution information is required for some cover types whereas coarser resolution may be adequate for other types. Landcover classifications for land surface modeling need to be based more on model sensitivities than on traditional vegetation-type schemes