Surface temperature extremes produced by huge machine learning hindcasts of summer 2023

TL;DR

This study uses a large machine learning ensemble to simulate summer 2023 temperature extremes, revealing new insights into the frequency and severity of heatwaves globally.

Contribution

It introduces a massive ML-based ensemble approach that surpasses traditional methods in predicting extreme temperature events and generating dangerous heatwave scenarios.

Findings

ML ensemble produced heatwave scenarios exceeding traditional models.

Two-thirds of global land areas showed no unusual extremes.

One-third exhibited extreme events outside traditional prediction envelopes.

Abstract

The summer of 2023 was the second hottest on record, with numerous extreme heatwaves across the globe. Using the Spherical Fourier Neural Operator machine learning (ML) weather model, we generated a massive ensemble of 7,424 weather scenarios simulating summer temperature extremes. The ML ensemble produced extreme heatwave scenarios exceeding temperatures from reanalysis and numerical weather prediction ensembles. Our results show that the ML model's extreme surface temperatures were not unusual for approximately two-thirds of the global land area. However, for the other one-third, ML-generated extreme events were well outside the prediction envelope from extrapolating smaller ensembles with extreme value theory. Furthermore, the ML ensemble readily generates storyline simulations of humid heat extremes, which yield more dangerous categories of public safety alerts than can be simulated from smaller ensembles. This research highlights the potential of huge ensemble simulations to improve understanding and prediction of both humid and dry temperature extremes.