Accurate Shortwave Radiation Simulation with a Two-Layer Aerosol Model in Xinjiang Region

To harness the rich solar energy resources in Xinjiang Region of Northwest China, this study tries to address the issue of lack of downward surface shortwave radiation (DSSR) observations and the need to improve the accuracy of satellite retrieval and numerical simulation of DSSR under varied sky and meteorological conditions. (1) A two-layer aerosol model specific to Xinjiang was developed to capture the vertical distributions of aerosols based on multiple data sources including lidar, GPS sounding, ground meteorological observations, and profiles from the ECMWF reanalysis version 5 (ERA5) data. The results show that the ERA5/PBLH (planetary boundary layer height) and ERA5/ALH (aerosol layer height) could be used to establish the two-layer aerosol model and characterize the vertical distribution of aerosols in Xinjiang Region. (2) Using the Santa Barbara Discrete Atmospheric Radiative Transfer (SBDART) model, a localized inverse model of clear-sky DSSR was established. After parameter adjustment and using the optimal combination of input parameters for DSSR simulation together with the two-layer aerosol model, the model-simulated DSSR (DSSRSBD) under clear-sky conditions improved significantly compared to the initial results, with all fitting indices greatly improved. (3) In addition, the study demonstrated that the impact of the two-layer aerosol model on DSSR was more pronounced under dust conditions than clear-sky conditions. (4) Using the localized clear-sky DSSR inversion model and its required parameters, simulations were also conducted to capture the spatiotemporal distribution of DSSR under clear-sky conditions in Xinjiang from 2017 to 2019. The annual average DSSRSBD under clear-sky conditions in Xinjiang during 2017-2019 was 606.78 W m-2, while DSSR from CERES (DSSRCER) under the same conditions was generally higher (703.95 W m-2). (5) It is found that satellite remote sensing products experienced data loss in high-altitude snow areas, where numerical simulation technology could serve as a valuable complement.