Paper ID: 2208.05855

Predicting Tornadoes days ahead with Machine Learning

Davide Alessandro Coccomini, Giuliano Zara

Developing methods to predict disastrous natural phenomena is more important than ever, and tornadoes are among the most dangerous ones in nature. Due to the unpredictability of the weather, counteracting them is not an easy task and today it is mainly carried out by expert meteorologists, who interpret meteorological models. In this paper we propose a system for the early detection of a tornado, validating its effectiveness in a real-world context and exploiting meteorological data collection systems that are already widespread throughout the world. Our system was able to predict tornadoes with a maximum probability of 84% up to five days before the event on a novel dataset of more than 5000 tornadic and non-tornadic events. The dataset and the code to reproduce our results are available at: https://tinyurl.com/3brsfwpk

Submitted: Aug 11, 2022