基于组合超快星历GPS数据的实时可降水量反演

Real-time Precipitable Water Vapor Inversion Based on a Combination of Ultra-rapid GPS Ephemeris Data

针对静态模式下GPS水汽反演存在的实时性不强、时间分辨率较低、反演代表性较差等问题,利用组合超快星历GPS数据进行实时动态可降水量反演研究。根据组合超快星历,利用假设检验法识别和剔除预报星历中的粗差卫星,采用动态模糊度分解方法基于单历元实时反演获得可降水量;对比实验结果显示,动态模式偏差值约为1.5 mm,精度约下降40%,在反演允许范围内。该方法在恶劣天气短临预报中有一定实际应用价值。

The static mode of GPS water vapor retrieval has issues of weak realtime, low temporal resolution, poor inversion representation. According to this, we used a combination of ultra-rapid GPS ephemeris data to inverse the realtime dynamic PWV. Based on the combination of ultra-rapid ephemeris, we used hypothesis testing to identify and eliminate the crude satellites from ephemeris, and used dynamic ambiguity decomposition method to obtain the real-time PWV based on a single epoch. Experimental results show that the dynamic mode deviation value is about 1.5 mm, and the precision is decreased by about 40% in the inversion, within the allowable range. This method has some practical value in the short impending severe weather forecasting.