Predicting Tropical Cyclone Intensity Using a Convolutional Neural Network and 20 Years of IMERG Satellite Rainfall Data
Name
wefo-WAF-D-24-0196.1.pdf
Description
Published version
Size
21.35 MB
Format
Adobe PDF
Checksum (MD5)
c518f6709f7577962b58e1b0697d869b
Author(s)
You, Sunny
Zhu, Ping
Guzman, Oscar
Jiang, Haiyan
Date Issued
October 15, 2025
Journal
Weather and Forecasting
Publisher
American Meteorological Society
Citation
You, S., P. Zhu, O. Guzman, and H. Jiang, 2025: Predicting Tropical Cyclone Intensity Using a Convolutional Neural Network and 20 Years of IMERG Satellite Rainfall Data. Wea. Forecasting, 40, 2317–2331.
Version
Final published version
Abstract
The prediction of tropical cyclone (TC) intensity change remains one of the greatest challenges for forecasters. The Statistical Hurricane Intensity Prediction Scheme (SHIPS) is one of the most accurate models used in operational centers. The current version of the SHIPS uses predictors including climatology and persistence, environmental conditions, and infrared satellite information. One critical piece of information that is largely missing from the SHIPS is the rainfall and structural features of TCs. In this study, a novel Hurricane Convolutional Neural Network (HCNN) model is proposed to predict future TC intensity by using satellite rainfall images and existing SHIPS predictors. A 20-yr (2000–19) satellite rainfall dataset is obtained from the NASA Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (GPM) mission (IMERG) product for TCs from the Atlantic basin. The HCNN model is tested for three different radii of the IMERG data from the TC center and 200 km is selected. The model is trained using satellite images for TCs from 2000 to 2017 and tested using TCs from 2018 to 2019. Relative to a multiple linear regression model with SHIPS predictors trained using the same training sample and tested using the same test sample as used for the HCNN, the HCNN model with satellite rainfall input significantly improves forecasts by 9%–13%, 8%–18%, and 5%–9% for all TCs, major hurricanes, and intensifying TCs, respectively, at 6–24-h forecast intervals. Further experiments show that the HCNN can better utilize rainfall structural information than a multilinear regression model integrating SHIPS and rainfall predictors.
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1175/WAF-D-24-0196.1
