Anomaly-Aware Tropical Cyclone Track Prediction Using Multi-Scale Generative Adversarial Networks  

Authors / CoAuthors

Huang, H | Deng, D | Hu, L. | Sun, N

Abstract

<div>Tropical cyclones (TCs) frequently encompass multiple hazards, including extreme winds, intense rainfall, storm surges, flooding, lightning, and tornadoes. Accurate methods for forecasting TC tracks are essential to mitigate the loss of life and property associated with these hazards. Despite significant advancements, accurately forecasting the paths of TCs remains a challenge, particularly when they interact with complex land features, weaken into remnants after landfall, or are influenced by abnormal satellite observations. To address these challenges, we propose a generative adversarial network (GAN) model with a multi-scale architecture that processes input data at four distinct resolution levels. The model is designed to handle diverse inputs, including satellite cloud imagery, vorticity, wind speed, and geopotential height, and features an advanced center detection algorithm to ensure precise TC center identification. Our model demonstrates robustness during testing, accurately predicting TC paths over both ocean and land, while also identifying weak TC remnants. Compared to other deep learning approaches, our method achieves superior detection accuracy, with an average error of 41.0 km for all landfalling TCs in Australia from 2015 to 2020. Notably, for five TCs with abnormal satellite observations, our model maintains high accuracy with a prediction error of 35.2 km, a scenario often overlooked by other approaches. <b>Citation:</b> Huang, H.; Deng, D.; Hu, L.; Sun, N. Anomaly-Aware Tropical Cyclone Track Prediction Using Multi-Scale Generative Adversarial Networks. Remote Sens. 2025, 17, 583. https://doi.org/10.3390/rs17040583

Product Type

document

eCat Id

150153

Contact for the resource

Keywords

• tropical cyclone

• track prediction

• deep learning

• satellite image

theme.ANZRC Fields of Research.rdf

• Meteorology

• generative adversarial network - GAN

• Published_External

Publication Date

2025-02-20T21:07:35

Creation Date

2025-02-02T18:00:00

Security Constraints

Australian Government Security Classification System    

Classification - unclassified

Legal Constraints

Creative Commons Attribution 4.0 International Licence    

Use - license otherRestrictions

Status

completed

Purpose

Journal paper submitted to Remote Sensing

Maintenance Information

asNeeded

Topic Category

geoscientificInformation

Series Information

AGSO Remote Sensing Volume 17 Issue 4

Lineage

Journal paper submitted to Remote Sensing

Parent Information

Extents

[-54.75, -9.2402, 112.92, 159.11]

Reference System

Spatial Resolution

Service Information

Associations

Downloads and Links

Link to Journal  

Source Information