Online Test of a Neural Network Deep Convection Parameterization in ARP-GEM1

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Balogh, Blanka, Saint-Martin, David, Geoffroy, Olivier
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912093175808000
author Balogh, Blanka
Saint-Martin, David
Geoffroy, Olivier
author_facet Balogh, Blanka
Saint-Martin, David
Geoffroy, Olivier
contents In this study, we present the integration of a neural network-based parameterization into the global atmospheric model ARP-GEM1, leveraging the Python interface of the OASIS coupler. This approach facilitates the exchange of fields between the Fortran-based ARP-GEM1 model and a Python component responsible for neural network inference. As a proof-of-concept experiment, we trained a neural network to emulate the deep convection parameterization of ARP-GEM1. Using the flexible Fortran/Python interface, we have successfully replaced ARP-GEM1's deep convection scheme with a neural network emulator. To assess the performance of the neural network deep convection scheme, we have run a 5-years ARP-GEM1 simulation using the neural network emulator. The evaluation of averaged fields showed good agreement with output from an ARP-GEM1 simulation using the physics-based deep convection scheme. The Python component was deployed on a separate partition from the general circulation model, using GPUs to increase inference speed of the neural network.
format Preprint
id arxiv_https___arxiv_org_abs_2410_21920
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Online Test of a Neural Network Deep Convection Parameterization in ARP-GEM1
Balogh, Blanka
Saint-Martin, David
Geoffroy, Olivier
Atmospheric and Oceanic Physics
Machine Learning
In this study, we present the integration of a neural network-based parameterization into the global atmospheric model ARP-GEM1, leveraging the Python interface of the OASIS coupler. This approach facilitates the exchange of fields between the Fortran-based ARP-GEM1 model and a Python component responsible for neural network inference. As a proof-of-concept experiment, we trained a neural network to emulate the deep convection parameterization of ARP-GEM1. Using the flexible Fortran/Python interface, we have successfully replaced ARP-GEM1's deep convection scheme with a neural network emulator. To assess the performance of the neural network deep convection scheme, we have run a 5-years ARP-GEM1 simulation using the neural network emulator. The evaluation of averaged fields showed good agreement with output from an ARP-GEM1 simulation using the physics-based deep convection scheme. The Python component was deployed on a separate partition from the general circulation model, using GPUs to increase inference speed of the neural network.
title Online Test of a Neural Network Deep Convection Parameterization in ARP-GEM1
topic Atmospheric and Oceanic Physics
Machine Learning
url https://arxiv.org/abs/2410.21920