Editorial: EVA 2019 data competition on spatio-temporal prediction of Red Sea surface temperature extremes

TL;DR

This paper discusses the EVA 2019 data competition focused on predicting Red Sea surface temperature extremes, highlighting the challenges of modeling non-stationary spatio-temporal data and the performance of various predictive methods.

Contribution

It introduces a novel benchmark dataset and evaluation framework for spatio-temporal extreme value prediction, and compares different modeling approaches used by participating teams.

Findings

Various modeling approaches were evaluated for predicting temperature extremes.

The competition revealed strengths and limitations of current methods.

Future challenges include improving prediction accuracy for unobserved locations.

Abstract

Large, non-stationary spatio-temporal data are ubiquitous in modern statistical applications, and the modeling of spatio-temporal extremes is crucial for assessing risks in environmental sciences among others. While the modeling of extremes is challenging in itself, the prediction of rare events at unobserved spatial locations and time points is even more difficult. In this editorial, we describe the data competition that was organized for the 11th international conference on Extreme-Value Analysis (EVA 2019), for which several teams modeled and predicted Red Sea surface temperature extremes over space and time. After introducing the dataset and the goal of the competition, we disclose the final ranking of the teams, and we finally discuss some interesting outcomes and future challenges.