Revolutionizing Humidity Profile Retrieval with Multi-Angle Aware Networks for Ground-Based Microwave Radiometers

Accurate retrieval of atmospheric relative humidity (RH) profiles is essential for improving our understanding of atmospheric thermodynamics and climate change. Here, we present a deep learning model called AngleNet, which is designed to retrieve RH profiles by leveraging multi-angle brightness temperature observation from ground-based microwave radiometers, and then producing a total of 7-year (2018-2024) AngleNet relative humidity profiles at hourly temporal resolution respectively in Beijing, Nanjing, and Shanghai. The AngleNet relative humidity profiles are uploaded in CSV format. However, in some cases, AngleNet RH profiles may be unavailable due to field observation experiment limitations or weather restrictions such as cloudy or rainy days.