Samudra: An AI Global Ocean Emulator for Climate

Abstract AI emulators for forecasting have emerged as powerful tools that can outperform conventional numerical predictions. The next frontier is to build emulators for long climate simulations with skill across a range of spatiotemporal scales, a particularly important goal for the ocean. Our work...

Full description

Saved in:
Bibliographic Details
Main Authors: Surya Dheeshjith, Adam Subel, Alistair Adcroft, Julius Busecke, Carlos Fernandez‐Granda, Shubham Gupta, Laure Zanna
Format: Article
Language:English
Published: Wiley 2025-05-01
Series:Geophysical Research Letters
Subjects:
Online Access:https://doi.org/10.1029/2024GL114318
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849725257402810368
author Surya Dheeshjith
Adam Subel
Alistair Adcroft
Julius Busecke
Carlos Fernandez‐Granda
Shubham Gupta
Laure Zanna
author_facet Surya Dheeshjith
Adam Subel
Alistair Adcroft
Julius Busecke
Carlos Fernandez‐Granda
Shubham Gupta
Laure Zanna
author_sort Surya Dheeshjith
collection DOAJ
description Abstract AI emulators for forecasting have emerged as powerful tools that can outperform conventional numerical predictions. The next frontier is to build emulators for long climate simulations with skill across a range of spatiotemporal scales, a particularly important goal for the ocean. Our work builds a skillful global emulator of the ocean component of a state‐of‐the‐art climate model. We emulate key ocean variables, sea surface height, horizontal velocities, temperature, and salinity, across their full depth. We use a modified ConvNeXt UNet architecture trained on multi‐depth levels of ocean data. We show that the ocean emulator—Samudra—which exhibits no drift relative to the truth, can reproduce the depth structure of ocean variables and their interannual variability. Samudra is stable for centuries and 150 times faster than the original ocean model. Samudra struggles to capture the correct magnitude of the forcing trends and simultaneously remain stable, requiring further work.
format Article
id doaj-art-2dac4fa24e404ab98e4e1dc5d8deb7e5
institution DOAJ
issn 0094-8276
1944-8007
language English
publishDate 2025-05-01
publisher Wiley
record_format Article
series Geophysical Research Letters
spelling doaj-art-2dac4fa24e404ab98e4e1dc5d8deb7e52025-08-20T03:10:31ZengWileyGeophysical Research Letters0094-82761944-80072025-05-015210n/an/a10.1029/2024GL114318Samudra: An AI Global Ocean Emulator for ClimateSurya Dheeshjith0Adam Subel1Alistair Adcroft2Julius Busecke3Carlos Fernandez‐Granda4Shubham Gupta5Laure Zanna6Courant Institute of Mathematical Sciences New York University New York NY USACourant Institute of Mathematical Sciences New York University New York NY USAProgram in Atmospheric and Oceanic Sciences Princeton University Princeton NJ USALamont Doherty Earth Observatory Columbia University Palisades NY USACourant Institute of Mathematical Sciences New York University New York NY USACourant Institute of Mathematical Sciences New York University New York NY USACourant Institute of Mathematical Sciences New York University New York NY USAAbstract AI emulators for forecasting have emerged as powerful tools that can outperform conventional numerical predictions. The next frontier is to build emulators for long climate simulations with skill across a range of spatiotemporal scales, a particularly important goal for the ocean. Our work builds a skillful global emulator of the ocean component of a state‐of‐the‐art climate model. We emulate key ocean variables, sea surface height, horizontal velocities, temperature, and salinity, across their full depth. We use a modified ConvNeXt UNet architecture trained on multi‐depth levels of ocean data. We show that the ocean emulator—Samudra—which exhibits no drift relative to the truth, can reproduce the depth structure of ocean variables and their interannual variability. Samudra is stable for centuries and 150 times faster than the original ocean model. Samudra struggles to capture the correct magnitude of the forcing trends and simultaneously remain stable, requiring further work.https://doi.org/10.1029/2024GL114318oceansclimate emulatormachine learning
spellingShingle Surya Dheeshjith
Adam Subel
Alistair Adcroft
Julius Busecke
Carlos Fernandez‐Granda
Shubham Gupta
Laure Zanna
Samudra: An AI Global Ocean Emulator for Climate
Geophysical Research Letters
oceans
climate emulator
machine learning
title Samudra: An AI Global Ocean Emulator for Climate
title_full Samudra: An AI Global Ocean Emulator for Climate
title_fullStr Samudra: An AI Global Ocean Emulator for Climate
title_full_unstemmed Samudra: An AI Global Ocean Emulator for Climate
title_short Samudra: An AI Global Ocean Emulator for Climate
title_sort samudra an ai global ocean emulator for climate
topic oceans
climate emulator
machine learning
url https://doi.org/10.1029/2024GL114318
work_keys_str_mv AT suryadheeshjith samudraanaiglobaloceanemulatorforclimate
AT adamsubel samudraanaiglobaloceanemulatorforclimate
AT alistairadcroft samudraanaiglobaloceanemulatorforclimate
AT juliusbusecke samudraanaiglobaloceanemulatorforclimate
AT carlosfernandezgranda samudraanaiglobaloceanemulatorforclimate
AT shubhamgupta samudraanaiglobaloceanemulatorforclimate
AT laurezanna samudraanaiglobaloceanemulatorforclimate