Samudra 2: Scaling Ocean Emulators across Resolutions

• URL: http://arxiv.org/abs/2606.02610v1
• Date: Sun, 24 May 2026 20:17:51 GMT
• Title: Samudra 2: Scaling Ocean Emulators across Resolutions
• Abstract summary: Ocean general circulation models (OGCMs) are essential to climate science but computationally expensive, limiting ensemble size and forcing scenarios.<n>We present Samudra 2, which introduces a wider U-Net backbone with modified ConvNeXt-style blocks and a reduced block-internal expansion factor.<n>At $1circ$, Samudra 2 increases upper-ocean global-mean temperature $R2$ from 0.56 to 0.87 and reduces deep-ocean temperature error by roughly sevenfold.
• Score: 38.716997864890736
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 $1^\circ$ resolution and exhibits two long-horizon failure modes: \emph{variance collapse}, the loss of temporal variability, and \emph{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 $1^\circ$, Samudra 2 increases upper-ocean global-mean temperature $R^2$ from 0.56 to 0.87 and reduces deep-ocean temperature error by roughly sevenfold. The same architecture scales to $1/2^\circ$ and $1/4^\circ$ 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. We provide code, documentation, and benchmark resources at https://openathena.ai/Ocean_Emulator/.

Related papers

• Neptune: An AI model for Global Ocean Subseasonal Prediction [1.6574028140524586]
  Sub-seasonal-to-seasonal (S2S) forecasting is societally critical, supporting decision-making in sectors ranging from water and agricultural management to disaster risk reduction, energy planning, and insurance.<n>Traditional physics-based Ocean General Models (CMs) are computationally expensive and difficult to develop and improve because of the code complexity.<n>Neptune is an end-to-end data-driven framework for global ocean and sea-ice emulation tailored for S2S timescales, up to 60 days.
  arXiv  Detail & Related papers  (2026-09-08T11:42:25Z)
• minWM: A Full-Stack Open-Source Framework for Real-Time Interactive Video World Models [51.45338589543413]
  minWM is a full-stack open-source framework for building real-time interactive video world models.<n> minWM converts existing bidirectional T2V/TI2V video foundation models into camera-controllable few-step autoregressive world models.
  arXiv  Detail & Related papers  (2026-05-28T17:27:03Z)
• Causal Forcing++: Scalable Few-Step Autoregressive Diffusion Distillation for Real-Time Interactive Video Generation [50.55853275866995]
  Real-time interactive video generation requires low-latency, streaming, and controllable rollout.<n>In this paper, we study a more aggressive setting: frame-wise autoregression with only 1--2 sampling steps.<n>We propose textbfCausal Forcing++, a principled and scalable pipeline.
  arXiv  Detail & Related papers  (2026-05-14T17:46:36Z)
• AxiomOcean: Forecasting the Three-Dimensional Structure of the Upper Ocean [40.39306206833454]
  AxiomOcean is a global AI ocean forecasting model that represents vertical hierarchy and cross-layer dependence within the water column.<n>In 10-day forecasts, AxiomOcean outperforms an advanced AI comparison model across variables and lead times.<n>Results show that explicitly preserving upper-ocean three-dimensional structure can improve both forecast accuracy and physical fidelity in AI ocean prediction.
  arXiv  Detail & Related papers  (2026-05-11T12:27:04Z)
• OceanPile: A Large-Scale Multimodal Ocean Corpus for Foundation Models [57.19693589926157]
  OceanPile is a large-scale multimodal corpus designed for ocean foundation models.<n>It comprises three key components: OceanCorpus, OceanInstruction, and OceanBenchmark.<n>All datasets are publicly released to advance the field of marine artificial intelligence.
  arXiv  Detail & Related papers  (2026-04-25T14:53:37Z)
• WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching [50.43356949257532]
  We propose textbfWorldCache, a caching framework tailored to diffusion world models.<n>We show that WorldCache delivers up to textbf3.7$times$ end-to-end speedups while maintaining textbf98% rollout quality.
  arXiv  Detail & Related papers  (2026-03-06T14:43:47Z)
• Simultaneous emulation and downscaling with physically-consistent deep learning-based regional ocean emulators [2.369898950737752]
  We propose an AI-based ocean emulation and downscaling framework focusing on the high-resolution regional ocean over Gulf of Mexico.<n>The framework shows both short-term skills as well as accurate long-term statistics in terms of mean and variability.
  arXiv  Detail & Related papers  (2025-01-09T08:28:31Z)
• Samudra: An AI Global Ocean Emulator for Climate [27.24070831177446]
  We build a global emulator of the ocean component of a state-of-the-art climate model.<n>We use a modified ConvNeXt UNet architecture trained on multi-depth levels of ocean data.<n>We show that the ocean emulator - Samudra - can reproduce the depth structure of ocean variables and their interannual variability.
  arXiv  Detail & Related papers  (2024-12-05T01:25:34Z)
• Accelerate Coastal Ocean Circulation Model with AI Surrogate [10.800252698519634]
  Oceanographers simulate ocean current circulation along the coasts to develop early warning systems that save lives and prevent loss and damage to property from coastal hazards.<n>Traditionally, such simulations are conducted using coastal ocean circulation models such as the Regional Ocean Modeling System (ROMS)<n>Recent advances in deep learning and GPU architecture have enabled the development of faster AI (neural network) surrogates.<n>This paper introduces an AI surrogate based on a 4D Swin Transformer to simulate coastal tidal wave propagation in an estuary for both hindcast and forecast (up to 12 days)
  arXiv  Detail & Related papers  (2024-10-19T02:49:30Z)
• Kick Back & Relax++: Scaling Beyond Ground-Truth Depth with SlowTV & CribsTV [50.616892315086574]
  This paper proposes two novel datasets: SlowTV and CribsTV. These are large-scale datasets curated from publicly available YouTube videos, containing a total of 2M training frames. We leverage these datasets to tackle the challenging task of zero-shot generalization.
  arXiv  Detail & Related papers  (2024-03-03T17:29:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.