Using the Fourier Neural Operator and Real‐Time GOES‐R Satellite Data for Precipitation Retrievals in the Southern Great Plains

• Published in: Journal of geophysical research. Machine learning and computation Vol. 2; no. 3
• Main Authors: Johncox, Max, Pu, Zhaoxia, Zhe, Shandian
• Format: Journal Article
• Language: English
• Published: 01.09.2025
• ISSN: 2993-5210; 2993-5210
• DOI: 10.1029/2024JH000531

In the U. S. Southern Great Plains (SGP) region, severe weather occurs regularly during the warm season (e.g., June–August), causing extensive property damage and loss of life. Despite advancements in observations and numerical models, estimating precipitation associated with these severe weather events remains challenging, thereby complicating accurate public warnings. Recently, machine learning (ML) models have been employed as a data‐driven approach to quantify precipitation during severe storms. In this study, we evaluate the performance of a ML model which utilizes the Fourier Neural Operator (FNO) for obtaining hourly precipitation retrievals in the SGP region. The FNO‐based model uses water vapor‐absorbing band brightness temperatures, lightning flash counts, and lightning average flash areas from NOAA's latest generation of Geostationary Operational Environmental Satellites (known as GOES‐R) as inputs to produce hourly precipitation retrievals at approximately 4‐km horizontal resolution. The “ground truth” rainfall data are hourly National Centers for Environmental Prediction (NCEP) Stage IV precipitation analysis totals. Results demonstrate that the FNO‐based model effectively generates accurate precipitation retrieval totals, offering improvements over the operational GOES‐R Quantitative Precipitation Estimation in both estimating and detecting precipitation at hourly intervals. Severe weather events are common in the Southern Great Plains (SGP) region during the warm season (e.g., June–August), often bringing heavy rainfall that can cause flash flooding. These floods frequently result in significant property damage and, in some cases, loss of life. To help protect at‐risk communities, it is essential to accurately measure how much rain these storms produce. This research introduces a new machine learning model based on the Fourier Neural Operator (FNO) architecture to estimate hourly precipitation amounts. The model uses cloud imagery and lightning products from the GOES‐16 satellite as inputs, with precipitation amounts from the National Center for Environmental Protection (NCEP) Stage‐IV analysis serving as the truths. The FNO‐based method offers faster and more accurate rainfall estimates compared to the current standard, the GOES‐R Quantitative Precipitation Estimation product. The Fourier Neural Operator (FNO) architecture has been effectively applied for precipitation retrievals using GOES‐R satellite data The FNO‐based model retrieves precipitation at 1‐hr intervals for the months of June to August, demonstrating promising results The FNO‐based model enables the generation of precipitation retrievals in under 1 min after the end of each hour