Title: Investigating Surface Bias Errors in the Weather Research and Forecasting (WRF) Model using a Geographic Information System (GIS)
Abstract: MET) to evaluate the accuracy of WRF forecasts using observations of meteorological variables.The traditional use of MET calculates model performance over the entire model domain.High-resolution modeling requires more focused verification at subdomain levels.A Geographical Information System (GIS) enables consideration of terrain variables in model assessments using a location-based approach.We discuss our GIS approach to verify WRF-ARW with a 1-kilometer horizontal resolution inner domain centered near San Diego, California.We selected 5 case study days from February and March of 2012.The literature indicated that elevation is strongly correlated with meteorological parameters, which became our focus.We found that elevation accounts for a significant portion of the variance in the model error.The study demonstrated that a GIS can analyze spatially distributed point forecast errors over subdomains, to reveal the dependence of forecast errors on elevation.15.