Monthly radiation data of ground stations in mainland China based on Ångström-Prescott formula ( 2003-2022 )

Solar radiation is a necessary parameter for calculating reference crop water requirement and crop growth model calibration, and it is also one of the characterization indexes of surface energy cycle. Obtaining long-term and accurate solar radiation values can provide an important theoretical basis for the application of solar energy industry and water resources allocation. Limited by factors such as operation and maintenance costs, there are few ground radiation observation stations in China that can publicly release data, and the measured data are scarce, which is difficult to meet the application needs of large-scale meteorology, agriculture and solar energy industries. Based on the correction coefficient of the existing Ångström-Prescott formula, combined with some sunshine hours observation data, this study constructed a monthly solar radiation data set of more than 2000 meteorological stations in mainland China from 2003 to 2022. By comparing with the observed data of the existing radiation stations, the correlation coefficient (R) of the estimated results is 0.806, the root mean square error (RMSE) is 3.653MJ/m2, the mean deviation (MBE) is-2.443 MJ/m2, and the coefficient of determination (R2) reaches 0.777, indicating that the estimated value and the observed value fit well and have strong reliability. Although there are still some systematic deviations in the estimated data, it has obvious advantages in terms of data continuity, spatial integrity and applicability compared with the traditional method of relying only on radiation observation sites. This data set can provide important data support and theoretical basis for solar energy resource assessment, agricultural water demand model driving, regional energy balance analysis and other fields.