Search papers, labs, and topics across Lattice.
This study investigates the use of Earth observation embeddings as sub-grid descriptors for enhancing probabilistic weather downscaling, addressing the limitations of traditional hand-crafted topographic descriptors. By integrating a convolutional conditional neural process with TESSERA embeddings, the authors achieve significant improvements in downscaling accuracy for 2 m temperature and 10 m wind speed across diverse climatic regions. The findings reveal an 11.5% improvement in CRPS skill for temperature and 6.2% for wind speed, demonstrating the potential of these embeddings to capture persistent surface properties that influence local weather conditions.
Earth observation embeddings can boost weather downscaling accuracy by over 11% by effectively encoding persistent surface properties.
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 descriptors. We ask instead whether Earth observation foundation models can provide transferable sub-grid surface representations for probabilistic weather downscaling. We augment a convolutional conditional neural process 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 summarise surface conditions over annual timescales, they improve downscaling of instantaneous 2 m temperature and 10 m wind speed 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. We further analyse how its contribution differs by variable, finding that topography explains more of temperature's sub-grid structure, while TESSERA provides additional surface information for wind speed. These improvements persist when the coarse input is changed from ERA5 to forecasts from the Aurora AI forecasting model, and when predicting at newly deployed stations with no regional history. To our knowledge, this is the first evidence that long-timescale Earth-observation embeddings can support short-timescale weather downscaling where sub-grid departures are systematically structured by persistent surface properties.
