Prediction of severe thunderstorm events with ensemble deep learning and radar data

TL;DR

This paper presents a deep learning approach using radar data to predict severe thunderstorms, demonstrating improved warning capabilities through a novel skill score evaluation method.

Contribution

It introduces a deep neural network model for thunderstorm nowcasting that leverages radar videos and a new skill score for performance assessment.

Findings

Effective early warning of severe thunderstorms achieved

Deep learning model outperforms traditional methods

Validated on regional radar data in Italy

Abstract

The problem of nowcasting extreme weather events can be addressed by applying either numerical methods for the solution of dynamic model equations or data-driven artificial intelligence algorithms. Within this latter framework, the present paper illustrates how a deep learning method, exploiting videos of radar reflectivity frames as input, can be used to realize a warning machine able to sound timely alarms of possible severe thunderstorm events. From a technical viewpoint, the computational core of this approach is the use of a value-weighted skill score for both transforming the probabilistic outcomes of the deep neural network into binary classification and assessing the forecasting performances. The warning machine has been validated against weather radar data recorded in the Liguria region, in Italy,