AI-boosted rare event sampling to characterize extreme weather

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lancelin, Amaury, Wikner, Alex, Dubus, Laurent, Priol, Clément Le, Abbot, Dorian S., Bouchet, Freddy, Hassanzadeh, Pedram, Weare, Jonathan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914268060844032
author Lancelin, Amaury
Wikner, Alex
Dubus, Laurent
Priol, Clément Le
Abbot, Dorian S.
Bouchet, Freddy
Hassanzadeh, Pedram
Weare, Jonathan
author_facet Lancelin, Amaury
Wikner, Alex
Dubus, Laurent
Priol, Clément Le
Abbot, Dorian S.
Bouchet, Freddy
Hassanzadeh, Pedram
Weare, Jonathan
contents Weather extremes pose major societal risks, especially in a changing climate, but due to their rarity, they are difficult to study using limited observations or complex climate models. We introduce AI+RES, a framework coupling fast AI weather forecasts with a high-fidelity physics model using a rare-event algorithm to efficiently characterize extremes. This approach enables the study of the statistics and physics of very rare events, such as once per millennium heatwaves at two orders-of-magnitude lower computational cost. AI+RES can be applied broadly across climate science and other fields concerned with rare events.
format Preprint
id arxiv_https___arxiv_org_abs_2510_27066
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AI-boosted rare event sampling to characterize extreme weather
Lancelin, Amaury
Wikner, Alex
Dubus, Laurent
Priol, Clément Le
Abbot, Dorian S.
Bouchet, Freddy
Hassanzadeh, Pedram
Weare, Jonathan
Atmospheric and Oceanic Physics
Computation
Machine Learning
Weather extremes pose major societal risks, especially in a changing climate, but due to their rarity, they are difficult to study using limited observations or complex climate models. We introduce AI+RES, a framework coupling fast AI weather forecasts with a high-fidelity physics model using a rare-event algorithm to efficiently characterize extremes. This approach enables the study of the statistics and physics of very rare events, such as once per millennium heatwaves at two orders-of-magnitude lower computational cost. AI+RES can be applied broadly across climate science and other fields concerned with rare events.
title AI-boosted rare event sampling to characterize extreme weather
topic Atmospheric and Oceanic Physics
Computation
Machine Learning
url https://arxiv.org/abs/2510.27066