Spatial Super-Resolution Reconstruction of Remote Sensing Precipitation Products Using Generative Adversarial Network by Multi-Source Fused Precipitation Data Features

Objectives: High-resolution precipitation data is one of the important basic data for studying water and energy cycles at regional scales, but the spatial resolution of existing precipitation products cannot yet meet the needs of regional refinement studies. Methods: This paper makes full use of the high spatial resolution (5 km) feature of the fused precipitation data (CMPA) from the National Intelligent Grid Live Analysis product to construct a two sets of superresolution reconstruction model based on Generative Adversarial Networks, which is applied to the spatial downscaling of the IMERG (The Integrated Multi-satellite Retrievals for GPM) daily precipitation data and ERA5 (The fifth generation ECMWF atmospheric reanalysis) daily precipitation reanalysis data, respectively. The final high-resolution IMERG daily precipitation product (0.05°) and ERA5 daily precipitation product (0.125°) were obtained and the accuracy of both downscaled data was evaluated using measured meteorological station data. Results: The results show that (1) The super-resolution reconstruction model based on CMPA data can be used for spatial downscaling of other precipitation products. The reconstructed IMERG products and ERA5 products show different degrees of improvement in all accuracy indicators. (2) The model is able to effectively retain the essential data characteristics of the CMPA data. The reconstructed IMERG product and ERA5 product are more closely matched to the CMPA data in terms of CC, POD and Bias, with higher overall accuracy. (3) The effect of model reconstruction is influenced by the spatial resolution and accuracy of the original data, as well as the correlation between the original data and the CMPA data. The higher the spatial resolution and accuracy of the raw data, the better the correlation between the raw data and the training data set, and the better the downscaling effect. As a result, the application on the IMERG product is significantly better than the EAR5 product. (4) The GAN model constructed in this paper outperforms the MF and RF models in terms of applicability to IMERG daily precipitation data, and its detailed reconstruction is better. Compared with the original IMERG data, the accuracy statistics of the GAN reconstructed daily precipitation in the three time dimensions of year, season and month are mostly improved, which indicates better model stability. Conclusions: Therefore, the GAN model constructed in this paper is able to provide a more refined precipitation distribution and is of some research value.