WRF-Solar: Upgrading the WRF representation of the aerosol-cloud-radiation feedbacks in support of solar energy forecasting

Abstract:

WRF-Solar is an upgraded version of the Weather Research and Forecasting (WRF) model aimed at improving solar power forecasting that provides a better representation of the aerosol-cloud-radiation feedbacks. Model developments include efficient numerical approaches to support operational forecasting and focus on particular feedbacks of the aerosol-cloud-radiation system:

• Aerosol-radiation feedbacks: A new parameterization of the aerosol direct effect was implemented to improve the representation of the aerosol variability.
• Cloud-aerosol feedbacks: The microphysics parameterization was upgraded to include water- and ice-nucleation aerosols.
• Cloud-radiation feedbacks: A shallow cumulus parameterization was implemented to connect sub-grid clouds to the radiation scheme. In addition, the microphysics parameterization provides the cloud droplet radius and ice crystal size to the radiation parameterizations to fully represent the first and second aerosol indirect effect.
• Initialization of the cloud field from infrared radiances recorded by satellites.

The different components have been interconnected to provide a complete representation of the aerosol-cloud-radiation system and its feedbacks. In addition, new developments were introduced to output the diffuse and direct normal irradiance (DNI) at temporal resolutions only limited by the time step of the model.

This presentation will provide an overview of the model physics packages upgraded for solar energy applications together with an assessment of different upgraded components. This includes the clear sky assessment wherein improvements of up to 58%, 76%, and 83% are found in global horizontal irradiance, DNI, and diffuse irradiance, respectively, compared to a standard version of the WRF model. The benefits of including a representation of the effects of unresolved clouds in the solar irradiance that largely reduce a positive bias in the model (~50W/m2). Finally, we will discuss an ongoing evaluation of the improvements resulting from the activation of the aerosol indirect effect.