射频干扰校正对星载微波资料反演降水的影响

针对全球降水测量（Global Precipitation Measurement，GPM）卫星上搭载的微波成像仪（GPM Microwave Imager，GMI），建立了陆面亮温数据的射频干扰（radio frequency interference，RFI）校正算法，同时提出了适用于GMI的综合指数雨强反演算法，即极化订正温度及散射指数（polarization corrected temperature-scattering index，PCT-SI）综合指数法，估算了2021年第6号强台风“烟花”登陆后，江苏及周边地区的雨强，与同时期双频降水测量雷达（Dual-frequency Precipitation Radar，DPR）的近地面雨强产品进行对比分析，并比较了RFI校正前、后的近地面雨强反演精度。进而用强台风“梅花”和一次江淮气旋检验了该雨强反演算法的精度以及RFI校正算法对提高反演精度的有效性。结果表明，GMI的高频观测与降水强度大小密切相关，利用低频组合拟合89.00 GHz通道亮温获得大气散射指数，散射指数越大，雨强越大；RFI信号对微波资料反演近地面雨强的影响显著，所提出的校正方法可以有效地校正陆面上受RFI污染的GMI观测数据。研究结果有助于揭示微波传感器多通道亮温资料和降水强度之间的相关性，提高微波反演陆面降水的精度，并为陆面降水反演和预报提供参考。

Impacts of radio frequency interference correction on retrieval of precipitation from spaceborne microwave data

For the Global Precipitation Measurement (GPM) Microwave Imager (GMI), the radio frequency interference (RFI) correction algorithm of the land-surface brightness temperatures is proposed, and the comprehensive index method of polarization corrected temperature-scattering index (PCT-SI) suitable for the GMI instrument is proposed. The rainfall intensity in Jiangsu and its surrounding areas after the landing of Typhoon In-fa (2106) is retrieved, and the retrieval data is compared with the detection data of Dual-frequency Precipitation Radar (DPR) in the same period. And also, the retrieval accuracy of near-surface rainfall intensity before and after RFI correction is compared. Furthermore, the accuracy of the rainfall intensity retrieval algorithm and the effectiveness of the RFI correction algorithm in improving the retrieval accuracy are validated by application in Typhoon Muifa and a Jianghuai cyclone process. The results show that the high-frequency observation of GMI is closely related to the precipitation rate. The atmospheric scattering index is obtained by fitting the brightness temperature of 89.00 GHz channel with the low-frequency combination; the larger the scattering index, the greater the rainfall intensity. The RFI signal has a significant impact on the retrieval of near-surface rainfall intensity from microwave measurements, and the proposed correction method can effectively correct the GMI observation data contaminated by RFI on the land surface. The results are helpful to reveal the correlation between multi-channel brightness temperature data of microwave sensors and precipitation intensity, improve the accuracy of microwave retrieval of land-surface precipitation, and provide reference for retrieval and prediction of land-surface precipitation.