Evaluation of GHI from NASA MERRA-2 and ECMWF ERA5-Land Reanalysis Against In-Situ Observations in Al-Ain, UAE

Solar reanalysis datasets are a valuable resource for estimating solar energy potential in situations where in-situ solar measurements are not available. A detailed assessment with short-term in-situ data is crucial to evaluate and understand the extent of errors. This study evaluates Global Horizontal Irradiance (GHI) reanalysis data from NASA's MERRA-2 and ECMWF's ERA5-Land, using high-quality in-situ observations collected in Al-Ain, UAE. The evaluation spans from January 1 to December 31, 2022, and employs statistical metrics such as Relative Mean Bias Error (rMBE), Relative Mean Absolute Error (rMAE), Relative Root Mean Square Error (rRMSE), and the Coefficient of Determination (R2). The analysis reveals that for Al-Ain, the annual ranges for rMBE, rRMSE, and R2 are 2.9-6.2%, 5.0-8.1%, and 0.941-0.980 for ERA5-Land, and 8.7-13.6%, 12.6-15.9%, and 0.919-0.959 for MERRA-2. Findings indicate that MERRA-2 and ERA5-Land yield comparable accuracy under clear sky conditions, whereas ERA5-Land consistently outperformed MERRA-2 across all error metrics for both all-sky and clear-sky conditions, demonstrating lower bias and greater accuracy..