Evaluation of the reanalysis surface solar radiation from NCEP, ECMWF, NASA, and JMA using surface observations for Balochistan, Pakistan

Abstract

Solar radiation from reanalysis datasets can be used to estimate solar energy potential in the absence of long-term surface measured data. Comprehensive validation of reanalysis data against short-term measurements of solar radiation is necessary to find uncertainties. The evaluation of the performance of seven reanalysis datasets (ERA-5, MERRA-2, ERA-Interim, JRA-55, NCEP-NCAR, NCEP-DOE, and CFSR) for Balochistan against high-quality surface measured data was performed using statistical analysis. The relative mean bias error (rMBE), relative mean absolute error (rMAE), and relative root mean square error of reanalysis solar radiation range from -4 to 7%, 8 to 14%, and 16 to 21%, respectively, using sub-daily data of six datasets except for NCEP-NCAR. Seasonal statistical analysis was performed to study seasonal variation of performance of datasets, which reveals the significant variation of errors in different seasons. Analysis of monthly mean data shows that rMBE and rMAE are less than 3 and 5% for ERA-5, respectively. The performance of the dataset changes by changing the temporal resolution of the data used for comparison. The monthly mean values of data show better performance, but the accuracy of data is compromised. The solar radiation data of ERA-5 can be used for preliminary solar resource assessment and power estimation.