Establishing a high-precision real-time Precipitable Water Vapor model in China with Global Navigation Satellite System and Fifth-Generation Reanalysis Model data

The real-time high precision of Precipitable Water Vapor (PWV) is widely recognized as crucial for advancing numerical weather prediction and enhancing our understanding of climate change. PWV is usually measured with radiosondes, microwave radiometers, and meteorological satellites. Nevertheless, these instruments have limitations including low temporal or spatial resolutions, high cost and weather dependence. To address these issues, we developed a real-time PWV grid model for China by integrating the ground-based Global Navigation Satellite System (GNSS) network with ERA5 reanalysis products. In this study, we explored an alternative method using ERA5 products to construct an accurate Elevation Normalization Factor Model (ENFM) and T-m models. The culmination of our efforts resulted in the creation of a real-time 1 & DEG; x 1 & DEG; PWV grid model for China. To validate the models of T-m and PWV, we compared them against the data obtained from radiosondes in China throughout 2021. The results indicate that the root-mean-square error (RMSE) value of the deviation between the T-m derived from our new T-m model, excluding meteorological parameters, and the radiosonde-derived T-m, is better than 4.22 K. The T-m values of the new T-m model are improved by 15.31% compared to those estimated based on the Bevis model plus HGPT2 temperature model. The RMSE values of the deviations between the new grid PWV and the radiosonde-derived PWV are less than 3.45 mm. The precision of our new PWV grid model is improved by 37.2% compared to that of the traditional Askne and Nordius model.