Impact of GPS radio occultation assimilation on the 18–21 July 2021 heavy rainfall event in Henan

Abstract

The Gridpoint Statistical Interpolation (GSI) assimilation system and Weather Research & Forecasting (WRF) mesoscale regional model were used to investigate the sensitivity of assimilating Global Positioning System radio occultation (GPS-RO) data for a “21·7” heavy rainfall event in Henan Province. The Ensemble Forecast Sensitivity to Observation (EFSO) algorithm was used to assess the influence of the GPS-RO assimilation relative to conventional heavy rainfall observations in Henan. Sensitivity experiments revealed that the 6-h accumulated precipitation simulated by the three experiments surpassed 200 mm, which was similar to the automated station observations, and the rainfall center location and intensity simulated by the assimilated GPS-RO experiment (EXPR2) were also similar to the observations. Following GPS-RO data assimilation, the upper-level west border continental high pressure was improved, which increased the pressure gradient on the east and west sides of the heavy rainfall, resulting in enhanced dynamic lifting and water vapor flow from the lower level to the whole Henan downpour center. The entire convective convergence rising height increased, which had an optimum influence on the dynamic and thermal conditions of heavy rainfall. The EFSO estimates revealed that most observations improved the model predictions, with the conventional surface pressure and U-wind observations making a positive standardized contribution of approximately 0.05 and 0.075, respectively, but T(Temperature) and Q(Specific humidity) making a negative standardized contribution of approximately −0.025. The GNSS GPS-RO greatly improved the prediction results by decreasing the error by 0.125, and the National Satellite Meteorological Center (NSMC) FY-3D contributed the most to the model.