Forecasting formation of a Tropical Cyclone Using Reanalysis Data

TL;DR

This paper presents a deep learning model that accurately forecasts tropical cyclone formation up to 60 hours in advance using high-resolution reanalysis data, outperforming traditional models in speed and accuracy.

Contribution

The study introduces a novel deep learning approach that captures spatial and temporal factors for tropical cyclone formation prediction with high accuracy and efficiency.

Findings

Achieves 86.9% - 92.9% accuracy across six ocean basins for 60-hour lead time.

Predicts cyclone formation within seconds after training.

Requires 5-15 minutes of training depending on data and basin.

Abstract

The tropical cyclone formation process is one of the most complex natural phenomena which is governed by various atmospheric, oceanographic, and geographic factors that varies with time and space. Despite several years of research, accurately predicting tropical cyclone formation remains a challenging task. While the existing numerical models have inherent limitations, the machine learning models fail to capture the spatial and temporal dimensions of the causal factors behind TC formation. In this study, a deep learning model has been proposed that can forecast the formation of a tropical cyclone with a lead time of up to 60 hours with high accuracy. The model uses the high-resolution reanalysis data ERA5 (ECMWF reanalysis 5th generation), and best track data IBTrACS (International Best Track Archive for Climate Stewardship) to forecast tropical cyclone formation in six ocean basins of the world. For 60 hours lead time the models achieve an accuracy in the range of 86.9% - 92.9% across the six ocean basins. The model takes about 5-15 minutes of training time depending on the ocean basin, and the amount of data used and can predict within seconds, thereby making it suitable for real-life usage.