Make Thunderbolts Less Frightening -- Predicting Extreme Weather Using Deep Learning

TL;DR

This paper presents a deep learning approach using CNN architectures inspired by UNet++ and ResNet to predict thunderstorms and lightning with high accuracy, aiming to make severe weather forecasts more reliable and less intimidating.

Contribution

It introduces a novel CNN-based method for lightning prediction using satellite images and lightning history, improving detection rates and reducing false alarms.

Findings

Over 94% lightning detection probability within 15 minutes

Reduced false alarm ratio compared to previous methods

Effective binary classification of thunderstorms from satellite data

Abstract

Forecasting severe weather conditions is still a very challenging and computationally expensive task due to the enormous amount of data and the complexity of the underlying physics. Machine learning approaches and especially deep learning have however shown huge improvements in many research areas dealing with large datasets in recent years. In this work, we tackle one specific sub-problem of weather forecasting, namely the prediction of thunderstorms and lightning. We propose the use of a convolutional neural network architecture inspired by UNet++ and ResNet to predict thunderstorms as a binary classification problem based on satellite images and lightnings recorded in the past. We achieve a probability of detection of more than 94% for lightnings within the next 15 minutes while at the same time minimizing the false alarm ratio compared to previous approaches.