Session F2: Atmospheric Effects on GNSS

Research on Prediction of Heavy Rainfall Based on BDS-2/3
Longfei Lv, Hang Guo, School of Information Engineering, Nanchang University; Min Yu, College of Computer Software, Jiangxi Normal University; Jian Xiong, School of Advanced Manufacturing, Nanchang University; Qun Tian, Guangzhou Institute of Tropical and Marine Meteorology; Ting Ni, School of Information Engineering, Nanchang University; Sai Du, Meteorological Bureau of Chenghai; Wenjing Kong, School of Information Engineering, Nanchang University

Peer Reviewed

With the rapid development of Global Navigation Satellite System (GNSS), the application of GNSS atmospheric inversion technology in the meteorological field has received widespread attention. It has the characteristics of all-weather continuous observation, wide coverage, high accuracy and resolution. Firstly, five different observation stations in Hong Kong were selected for the calculation and analysis of the Precision of Precipitable Water Vapor (PWV) inversion using Global Positioning System (GPS) and BeiDou Navigation Satellite System (BDS) observation data, the results show that the RMSE of PWV retrieved by GPS, BDS2 and BDS-2/3 are 2.36 mm,4.17 mm and 3.02 mm. Secondly, By calculating the PWV values of five different observation stations in Hong Kong during the 2020 rainy season and analyzing the characteristics of PWV changes during heavy rainfall, the results show that when the PWV value starts to increase from a lower value and the rate of PWV change continues to increase to 1.08 mm/h, heavy rainfall may occur within 12 hours. The PWV sequence and other related parameters obtained from the BDS-2/3 data inversion of five Hong Kong CORS stations were used as prediction factors for the heavy rainfall threshold model. After obtaining the optimal threshold, the heavy rainfall was predicted. The results showed that the CSI (Critical Success Index) of the six-factor threshold prediction model was 82.5%, POD (Probability of Detection) was 92.1%, and PAR (False Alarm Rate) was 11.2%, which was 21.6% higher than the traditional three-factor model in terms of comprehensive evaluation indicators. This provides valuable references for the application of PWV in short-term heavy rainfall forecasting, disaster prevention, and mitigation in related meteorological services. It also further demonstrates the enormous potential of BDS-2/3-based atmospheric precipitable water vapor inversion technology in short-term heavy rainfall forecasting.