Aerosol Optical Depth Forecasts for Solar Irradiance Forecasting in the Middle East

One of the largest sources of error for predictions of solar irradiance and solar power in clear-sky conditions is the aerosol optical depth (AOD). Direct normal irradiance (DNI) is particularly sensitive to AOD, while global horizontal irradiance (GHI) is less sensitive but still impacted. Therefore, accurate predictions of the AOD are crucial for accurate predictions of solar irradiance and power, particularly for concentrated solar power (CSP) plants, but also for standard photovoltaic (PV) solar power plants. Aerosol loading can also be significant in arid regions such as the Middle East, where a number of new CSP and PV plants have been or will soon be constructed.

In this study we present a climatological analysis of AOD forecasts from December 2017 through mid-2019 from the Global Forecast System (GFS) model, the NASA Goddard Earth Observing System Model, Version 5 (GEOS-5), and from operational dust optical depth forecasting models run by the Barcelona Supercomputing Center (BSC). AOD predictions from these models are validated against Aerosol Robotic Network (AERONET) observations across the Middle East. It is found that GEOS-5 is the best model on average.

In addition, AOD forecasts from the Weather Research and Forecasting (WRF-Solar®) model with its current internal aerosol module are compared with those obtained using a new explicit dust model, which consists of a simplified version of the GOCART aerosol scheme used in WRF-Chem. In addition to standard error metrics to characterize the overall performance of the models, a multi-site time series analysis is performed to assess how well these models are able to represent significant dust storm events. The impacts of dust storm and high-AOD events on power production at the Shagaya Renewable Energy Park in western Kuwait, which has co-located irradiance and AERONET observations, is also explored.