May 24, 2026J/K move · Enter open · S save
Yuan Yuan, Jesse Rusak, Alexander Merose, Adam Subel+4
Courant Institute School of Mathematics, Computing, and Data Science, New York University, New York, NY, USA · Open Athena AI Foundation, Inc., New York, NY, USA · Program in Atmospheric and Oceanic Sciences, Princeton University, Princeton, NJ, USA
Ocean general circulation models (OGCMs) are essential to climate science but computationally expensive, limiting ensemble size and forcing scenarios. Neural emulators promise orders-of-magnitude speedups, yet existing ocean emulators have not combined fine spatial resolution with multi-year autoregressive rollouts. Samudra, the first autoregressive neural ocean emulator to produce multi-decade global rollouts, is limited to
1∘ resolution and exhibits two long-horizon failure modes:
variance collapse, the loss of temporal variability, and
imprinting artifacts, in which velocity patterns leak into deep-ocean fields. We present Samudra 2, which introduces a wider U-Net backbone with modified ConvNeXt-style blocks and a reduced block-internal expansion factor, together with a dynamic loss that reweights output channels according to their prediction errors, strengthening gradients for slow-evolving deep-ocean fields. At
1∘, Samudra 2 increases upper-ocean global-mean temperature
R2 from 0.56 to 0.87 and reduces deep-ocean temperature error by roughly sevenfold. The same architecture scales to
1/2∘ and
1/4∘ over approximately 8-year autoregressive rollouts, recovering mesoscale eddies and sharp western boundary currents. Running on a single GPU, Samudra 2 enables larger ensembles for sea-level projections, ocean heat uptake, and climate variability studies. All artifacts are publicly available: project page, code, checkpoints, documentation.