Skip to content

Earth observation embeddings are effective sub-grid descriptors for probabilistic weather downscaling

Aug 2026 · arXiv.org · Vol abs/2608.12271 · 3 citations · 56 references
Computer Science Physics

TL;DR

This work augments a convolutional conditional neural process (ConvCNP) that downscales coarse ERA5 reanalysis fields at ~25 km resolution with a learned local surface descriptor, obtained by compressing a patch of TESSERA embeddings at 10 m resolution, and shows the first evidence that long-timescale Earth observation embeddings can support short-timescale weather downscaling where subgrid departures are systematically structured by persistent surface properties.

Abstract

Global weather reanalyses and forecasts resolve the evolving atmospheric state on coarse grids, but site-specific applications require predictions at arbitrary locations where near-surface conditions also depend on unresolved terrain and land-surface properties. Existing probabilistic downscalers address this gap using hand-crafted topographic and surface descriptors. We ask instead whether Earth observation foundation models can provide transferable subgrid surface representations for probabilistic weather downscaling. We augment a convolutional conditional neural process (ConvCNP) that downscales coarse ERA5 reanalysis fields at ~25 km resolution with a learned local surface descriptor, obtained by compressing a patch of TESSERA embeddings at 10 m resolution. Although these embeddings summarize annual surface conditions, they improve downscaling by encoding persistent surface properties that capture a location's departure from the coarse-grid atmospheric state. Across five climatically diverse regions, the embedding improves point and probabilistic skill at stations held out in both space and time, overall improving CRPS skill by 11.5% for 2 m temperature and 6.2% for 10 m wind speed relative to a topography-only ConvCNP baseline. A hand-crafted descriptor incorporating richer surface information than topography alone captures comparable persistent subgrid signal but yields far smaller predictive gains than the learned embedding representation. These improvements persist when forecasts from the Aurora AI model replace ERA5 reanalysis fields and when predicting at newly deployed weather station networks. To our knowledge, this is the first evidence that long-timescale Earth observation embeddings can support short-timescale weather downscaling where subgrid departures are systematically structured by persistent surface properties.

View source

Similar papers

#machine learning Review Nov 2026

Fusing Earth observation embeddings and spaceborne terrain with meteorological forcing for flood peak depth prediction from high-water marks

Flood peak depth on normally dry land governs damage and risk to life, but is observed only sparsely, by post-event high-water mark (HWM) surveys, and depth derived from satellite flood extent is produced after the event and inherits the vertical error of the digital elevation model (DEM) it is computed from. Deep-lear...

Wen Zhou, Shao-Wen Wang, Upmanu Lall et al. · 0 citations
#artificial intelligence Preprint Sep 2026

Efficient Analog-Initialized Latent Transport for Spatially Coherent Probabilistic Downscaling from Global to Kilometer Scales

Probabilistic downscaling must represent kilometer-scale structure left unresolved by a deterministic regional prediction. We introduce an analog-initialized latent transport method that uses historical regional residuals as a meteorologically informed empirical source. A frozen graph neural network predicts the determ...

Ophélia Miralles · 0 citations
#machine learning Preprint Sep 2026

Planetary Feature Fields are Scalable Earth Representations

Satellite observations, precomputed embeddings, and map products describe the same evolving Earth, yet are stored as independent, petabyte-scale data products. Their continued growth calls for compact representations of multiple products while preserving spatial and temporal detail. We introduce Planetary Feature Field...

Arjun Rao, Sebastian Loeschcke, A. Fuller et al. · 0 citations
Preprint Oct 2026

Observation Assimilation Transformer: Probabilistic Weather Analysis from Sparse Station Observations

We present the Observation Assimilation Transformer (OAT), a geometry-aware neural process that corrects and refines a prior numerical weather prediction (NWP) field into a high-resolution posterior field by assimilating sparse observations. OAT is an encoder-decoder transformer that considers a set of observed locatio...

Christian Donner, Michele Cattaneo, M. Samarin et al. · 0 citations
Open access Oct 2026

Generative and deterministic deep learning models comparison for fine-scale precipitation retrievals from infrared brightness temperature

Accurate precipitation estimation at fine spatial scales is critical for hydrology, agriculture, and climate studies. Infrared brightness temperatures from geostationary satellites offer excellent temporal coverage over continental-scale domains. However, because these measurements primarily characterize cloud-top pr...

Matthieu Meignin, C. Mallet, N. Viltard · 1 citation
Preprint Nov 2026

Forecasting Earth Surface Change in Embedding Space

Repeated satellite imagery permits understanding of how our planet changes over time, letting us predict its future and plan mitigating actions. Most existing studies that rely on a time series of satellite observations forecast the spectral values at each pixel for a future target date. However, clouds, shadows, and s...

Jing-Tao Li, S. Jaffer, A. Madhavapeddy et al. · 0 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.