Reconstruction of 3D DPR Observations Using GMI Radiances

Three-dimensional global precipitation observation is crucial for understanding climate and weather dynamics. While spaceborne precipitation radars provide precise but limited observations, passive microwave imagers are available much more frequently. In this study, we propose a deep learning approach to reconstruct active radar observations using passive microwave radiances. We introduce the Hybrid Deep Neural Network (HDNN) model, which utilizes reflectivity profiles from the Dual-frequency Precipitation Radar (DPR) onboard the Global Precipitation Measurement (GPM) Core Observatory Satellite as the "target" and combines radiances from the GPM Microwave Imager (GMI) with supplementary reanalysis data to serve as the "features." Results underscore the HDNN's exemplary performance, with a root mean square error below 4 dBZ across all altitude levels, and a consistent accuracy across different precipitation types. Its efficacy is further illustrated when applied to typhoon cases of Haishen and Khanun, emerging as a superior tool for capturing 3D structures of expansive precipitation systems. This research explores a novel way to improve our ability to observe precipitation from space. The study focuses on two instruments aboard a satellite: one that provides detailed 3D images of precipitation (DPR), and another (GMI) that covers a wider area but with less vertical resolution. The challenge is to combine the strengths of both instruments to get a more complete picture of precipitation. To address this, a HDNN was proposed to understand and interpret the data from both instruments. The detailed reflectivity profile from DPR was used as the label data, while the data from GMI with wider coverage along with additional information are used as the input. Evaluations based on validation set show that the model performed well, giving accurate and consistent measurements across different types of precipitation and at all altitudes. It also proved effective when applied to tropical cyclone cases, like Typhoons Haishen and Khanun, showcasing its potential for enhancing our understanding of such weather phenomena. Proposed deep learning method expands radar coverage from 245 to 885 km using passive microwave imager The model shows high precision across varied precipitation types and in real-world applications on Typhoon Haishen (2020) and Typhoon Khanun (2023)