Enhancing WRF-Solar to Provide Probabilistic Cloud Optimized Day-Ahead Forecasts

WRF-Solar is an enhanced numerical weather prediction (NWP) model specifically designed to provide solar irradiance forecasts tailored for solar energy applications. The enhancements focused on improving the cloud-aerosol-radiation representation. New efforts funded by the U.S department of energy (DOE) are underway to further develop the model in order to provide probabilistic forecasts specifically tailored for solar energy applications. With this aim, WRF modules responsible for the cloud representation are being examined to develop adjoint models. The adjoint models will be used to identify variables and parameters with the largest impact on irradiance forecasts. Subsequently, the WRF-Solar model will be developed to introduce perturbations on these variables and parameters in order to produce the probabilistic forecasts. This presentation will describe these new efforts which will convert WRF-Solar into the first standalone NWP framework to provide probabilistic forecast specifically tailored for solar energy applications.