Transferable Dual-Stream Representations for Mesoscale-Preserving Sea Surface Temperature Downscaling

• URL: http://arxiv.org/abs/2608.04230v1
• Date: Tue, 04 Aug 2026 21:11:04 GMT
• Title: Transferable Dual-Stream Representations for Mesoscale-Preserving Sea Surface Temperature Downscaling
• Abstract summary: EddyFlow is a representation learning framework for kilometer-scale sea surface temperature downscaling.<n>It balances predictive accuracy, scale-dependent structure, and regional generalization.
• Score: 0.28819200874187073
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deep learning models for scientific spatio-temporal downscaling often minimize reconstruction error while failing to preserve physically meaningful multi-scale structure. For sea surface temperature prediction, this can yield outputs that are numerically plausible yet overly smooth, missing mesoscale variability critical to regional ocean dynamics. Existing methods often focus on pixel-wise objectives or single-context conditioning, which limits their ability to preserve spectral fidelity and generalize across regions. To address this, we propose EddyFlow, a representation learning framework for kilometer-scale sea surface temperature downscaling that balances predictive accuracy, scale-dependent structure, and regional generalization. EddyFlow is trained on the Gulf of St.~Lawrence and evaluated in zero-shot and few-shot settings on the Bay of Fundy and the Gulf of Mexico. EddyFlow demonstrates that physics-informed representation learning reduces zero-shot RMSE by 21%, achieves up to 85.6% skill relative to persistence on unseen domains, and maintains near-ideal spectral fidelity with a PSD ratio of $\approx 1.00$.

Related papers

• Tropospheric temperature and humidity profile retrieval from Meteosat Flexible Combined Imager based on deep learning [41.113886294153346]
  The Meteosat Third Generation (MTG) Flexible Combined Imager (FCI) offers new opportunities for tropospheric temperature and humidity profiling.<n>We develop a spatially aware deep learning framework to retrieve all-sky tropospheric and humidity profiles from FCI without forecast profiles as input.
  arXiv  Detail & Related papers  (2026-08-26T12:16:38Z)
• OceanLight: Efficient Global Ocean Forecasting via Geometry-Adaptive Unstructured Mesh Representation [8.050783873604173]
  OceanLight is an efficient global ocean forecasting framework combining unstructured mesh tokenization with a graph neural network (GNN) backbone.<n>OceanLight achieves pointwise forecast accuracy and kinetic energy spectral fidelity exceeding both operational numerical analyses and state-of-the-art AI-based models.<n>Our unstructured mesh representation establishes a generalizable paradigm for scalable data-driven oceanography.
  arXiv  Detail & Related papers  (2026-08-17T04:03:30Z)
• Predicting Pseudo-nitzschia harmful algal blooms along the Portuguese Coast using satellite-derived predictors [69.38367341895632]
  Pseudo-nitzschia diatoms pose recurrent risks to coastal ecosystems and harvesting along the Portuguese Atlantic coast.<n>We develop and evaluate a machine-learning framework to predict harmful algal bloom occurrence using exclusively satellite-derived predictors.
  arXiv  Detail & Related papers  (2026-07-08T18:14:24Z)
• Physics-Constrained Adaptive Flow Matching for Climate Downscaling [20.06043414924976]
  We present Physics-Constrained Adaptive Flow Matching (PC-AFM) as a generative downscaling model.<n>We train the model on Central Europe climate data, evaluate it on a 10-time downscaling task, and test the generalization on two held-out climate regions.<n>Results indicate that physical consistency is a practical requirement for deploying generative downscaling models in real-world applications.
  arXiv  Detail & Related papers  (2026-04-03T21:09:13Z)
• High-resolution probabilistic estimation of three-dimensional regional ocean dynamics from sparse surface observations [0.0]
  Ocean interior regulates Earth's climate but remains sparsely observed due to limited in situ measurements.<n>We present a depth-aware generative framework for reconstructing high-resolution three-dimensional ocean states from extremely sparse surface data.
  arXiv  Detail & Related papers  (2026-04-03T08:10:55Z)
• Analytical Logit Scaling for High-Resolution Sea Ice Topology Retrieval from Weakly Labeled SAR Imagery [1.6984114279983211]
  High-resolution sea ice mapping is crucial for Arctic navigation and climate monitoring.<n> operational ice charts provide only coarse, region-level polygons (weak labels)<n>We propose a weakly supervised deep learning pipeline that fuses Sentinel-1 SAR and AMSR-2 radiometry data.
  arXiv  Detail & Related papers  (2026-03-13T20:21:33Z)
• CASCADE: Cross-scale Advective Super-resolution with Climate Assimilation and Downscaling Evolution [51.56484100374058]
  CASCADE reconstructs fine structure by iteratively advecting coarse information along learned, flow-conditioned velocity warping fields.<n> evaluated on SEVIR radar data for 4x super-resolution of severe convective storms.<n> produces temporally coherent, physically consistent, and mass-conserving reconstructions well suited to advection-dominated extremes in atmospheric and oceanic applications.
  arXiv  Detail & Related papers  (2026-02-18T14:41:54Z)
• Leveraging an Atmospheric Foundational Model for Subregional Sea Surface Temperature Forecasting [0.0]
  We adapt a deep learning model to predict sea temperature (SST) in the Canary Upwelling System.<n>By fine-tuning this model with high-resolution oceanographic reanalysis data, we demonstrate its ability to capture complex patterns.<n>The model successfully reproduces large-scale SST structures but faces challenges in capturing finer details in coastal regions.
  arXiv  Detail & Related papers  (2025-10-29T14:30:12Z)
• Physics-constrained generative machine learning-based high-resolution downscaling of Greenland's surface mass balance and surface temperature [44.99833362998488]
  High-resolution projections of the Greenland ice sheet's surface mass balance (SMB) and surface temperature are essential for understanding future sea-level rise.<n>Here, we introduce a novel physics-constrained generative modeling framework based on a consistency model (CM) to downscale low-resolution SMB and surface temperature fields.<n>Our approach delivers realistic, high-resolution climate forcing for ice-sheet simulations with fast inference.
  arXiv  Detail & Related papers  (2025-07-30T08:43:48Z)
• ReconMOST: Multi-Layer Sea Temperature Reconstruction with Observations-Guided Diffusion [48.540756751934836]
  ReconMOST is a data-driven guided diffusion model framework for multi-layer sea temperature reconstruction.<n>Our method extends ML-based SST reconstruction to a global, multi-layer setting, handling over 92.5% missing data.
  arXiv  Detail & Related papers  (2025-06-12T06:27:22Z)
• Residual Corrective Diffusion Modeling for Km-scale Atmospheric Downscaling [58.456404022536425]
  State of the art for physical hazard prediction from weather and climate requires expensive km-scale numerical simulations driven by coarser resolution global inputs. Here, a generative diffusion architecture is explored for downscaling such global inputs to km-scale, as a cost-effective machine learning alternative. The model is trained to predict 2km data from a regional weather model over Taiwan, conditioned on a 25km global reanalysis.
  arXiv  Detail & Related papers  (2023-09-24T19:57:22Z)

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.