Precipitation measurement based on satellite data and machine learning

Abstract

Satellite infrared (IR) data, with high temporal resolution and wide coverages, have been commonly used in precipitation measurement. However, existing IR-based precipitation retrieval algorithms suffer from various problems such as overestimation in dry regions, poor performance in extreme rainfall events, and reliance on an empirical cloud-top brightness-rain rate relationship. To solve these problems, a deep learning model using a spherical convolutional neural network was constructed to properly represent the Earth's spherical surface. With data inputted directly from IR band 3, 4, and 6 of the operational Geostationary Operational Environmental Satellite (GOES), the new model of Precipitation Estimation based on IR data with Spherical Convolutional Neural Network (PEISCNN) was first trained, tested and validated. Compared to the commonly used IR-based precipitation product PERSIANN CCS (the Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Network, Cloud Classification System), PEISCNN showed significant improvement in the metrics of POD, CSI, RMSE and CC, especially in the dry region and for extreme rainfall events. The PEISCNN model may provide a promising way to produce an improved IR-based precipitation product to benefit a wide range of hydrological applications.