FASST: A Compact Stateless Atmosphere Model for Long-Term Ocean Coupling

Abstract

Coupled climate simulations are computationally expensive, largely because of the atmosphere. We introduce FASST, a coupled climate model that retains the prognostic POP2 ocean while replacing the atmosphere and sea ice with lightweight U-Net emulators. The atmospheric emulator is diagnostic, predicting surface fluxes from the evolving ocean-ice state every $6$ simulated hours, while the sea-ice emulator advances the ice state monthly. For $140$ years of unforced integration, FASST maintains a stable global-mean climate without SST restoring, with post-transient global-mean SST of 291.30 ± 0.22K compared with 291.25 ± 0.10K in CESM$2$ piControl. FASST also captures interannual Niño-$3.4$ variability, although with weaker amplitude than CESM (0.62K versus 1.02K). These results demonstrate that a stateless learned atmosphere can sustain stable, free-running coupled climate simulations over century-scale integrations.

Phanerozoic atmosphere and ocean conditions estimated by a continuous climate and multi-box ocean model
Alexander James Krause, Mingyu Zhao, Andrew Merdith +1
EarthArXiv15 Sep 2026PreprintCC BY-NC 4.0Full textGeology & geochemistryEarthArXiv
SamudrACE: Fast and Accurate Coupled Climate Modeling With 3D Ocean and Atmosphere Emulators
James Duncan, Elynn Wu, Surya Dheeshjith +15
Geophysical Research Letters30 May 2026Peer-reviewedCC BY31 citedClimate scienceOpenAlex
Holocene millennial climate variability influenced by tropical–Antarctic ocean–atmosphere interactions
Peng Zhou, Zhengguo Shi, Zhisheng An +3
Proceedings of the National Academy of Sciences of the United States of America27 Oct 2026Peer-reviewedCC BY-NC-ND 4.0Full textAtmospheric science & meteorologyPMC
The role of salinity in Snowball Earth: A simplified coupled atmosphere--ocean--ice model
Sounav Sengupta, Francisco E. Spaulding-Astudillo, Dorian S. Abbot +2
arXiv1 Nov 2026PreprintCC0Full textAtmospheric science & meteorologyarXiv
cSINGV v1.0: development and evaluation of a 1.5 km coupled atmosphere–ocean–wave modelling system for the Western Maritime Continent
Rajesh Kumar, Claudio Sanchez, Juan Manuel Castillo +9
Geoscientific Model Development5 Nov 2026Peer-reviewedCC BY 4.0General earth scienceCrossref

Cite

BibTeX RIS

BibTeX

@misc{aggarwal2026fasst,
  title = {{FASST: A Compact Stateless Atmosphere Model for Long-Term Ocean Coupling}},
  author = {Pratham Aggarwal and Duncan Watson-Parris and William E. Chapman},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.36297},
  eprint = {2610.36297},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.36297},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - FASST: A Compact Stateless Atmosphere Model for Long-Term Ocean Coupling
AU  - Pratham Aggarwal
AU  - Duncan Watson-Parris
AU  - William E. Chapman
PY  - 2026
DA  - 2026/10/28
PB  - arXiv
DO  - 10.48550/arxiv.2610.36297
UR  - https://arxiv.org/abs/2610.36297
AB  - Coupled climate simulations are computationally expensive, largely because of the atmosphere. We introduce FASST, a coupled climate model that retains the prognostic POP2 ocean while replacing the atmosphere and sea ice with lightweight U-Net emulators. The atmospheric emulator is diagnostic, predicting surface fluxes from the evolving ocean-ice state every $6$ simulated hours, while the sea-ice emulator advances the ice state monthly. For $140$ years of unforced integration, FASST maintains a stable global-mean climate without SST restoring, with post-transient global-mean SST of 291.30 ± 0.22K compared with 291.25 ± 0.10K in CESM$2$ piControl. FASST also captures interannual Niño-$3.4$ variability, although with weaker amplitude than CESM (0.62K versus 1.02K). These results demonstrate that a stateless learned atmosphere can sustain stable, free-running coupled climate simulations over century-scale integrations.
ER  -

Abstract

Related

Cite