Scaling Issues in the Direct Use of Satellite Data in Mesoscale Model Land-Surface Parameterizations

South Dakota School Of Mines And Technology, Rapid City SD

Investigators

Abstract

This project addresses the accurate representation of the surface in mesoscale atmospheric prediction models. The research methodology is to couple a subgrid land-surface parameterization to a proven mesoscale atmospheric model. The subgrid parameterization is specifically designed to estimate surface properties (e.g., albedo; heat and moisture fluxes) from satellite data. In order to assess the effectiveness of this subgrid scheme, results of the coupled atmospheric-subgrid model will be compared to in-situ observations and remotely-derived estimates of surface energy fluxes. Model results will be compared to gridded analyses of atmospheric properties. In addition, the importance of scale effects on the accuracy of the model system will be investigated. These scale effects include (a) sensitivity of model output to the resolution of the subgrid; (b) the method used to distribute atmospheric data to the subgrid and how calculated subgrid fluxes are transferred back to the atmospheric model's grid; (c) the sensitivity of model output to the orientation of the atmospheric model grid and the surface subgrid; and (d) the determination of a physically reasonable ratio between the atmospheric model and subgrid resolutions. This project will benefit the mesoscale modeling community by providing an improved method for introducing subgrid variability into mesoscale models which should lead to improved numerical weather prediction. This study also will benefit the remote sensing community since many of the scale issues to be investigated are of importance in the remote estimation of surface parameters. Improvements in satellite derived surface parameterizations will provide flux data at scales required to validate the surface component of various numerical modeling schemes.