6.3    Improvement of the WRF Model for Solar Resource Assessment    and Forecast Under Clear Skies

Ruiz-Arias, J.A., University of JaŽn, Spain, Dudhia, J., NCAR, Gueymard, C., Pozo-V‡zquez, D., University of JaŽn, Spain

Numerical Weather Prediction (NWP) models constitute a very promising tool for solar resource assessment and forecasting. However, they still need further development to respond to the specific requirements of the solar industry. Very usually, NWP do not provide the direct component (DNI) of the solar irradiance, which is required for concentrating solar energy applications. Additionally, aerosol effect has most often been oversimplified. However, current satellite imagery is providing a rich source of broad measurements, which opens new avenues of research.

 In this contribution, we present preliminary results of a new parameterization of the aerosol optical depth (AOD) introduced in the shortwave GSFC scheme to yield DNI from the total AOD at 550 nm. The performance of the model has been tested based on NOAAŐs SURFRAD radiometric stations. The methodology reduces the MBE and RMSE in global horizontal irradiance (GHI) and DNI by more than 50% on clear-sky days. On August 21st, 2009 at the Boulder (Colorado) BSRN station, for instance, the MBE was reduced from 7% to 2% and the RMSE from 9% to 4% for GHI. For DNI, the improvement was 17% to 3%, and 24% to 7%, respectively.