Sparse Local Implicit Image Function for sub-km Weather Downscaling
Yago del Valle Inclan ⋅ Enrique Arriaga-Varela ⋅ Dmitry Lyamzin ⋅ Pablo Cervantes ⋅ Tiago Ramalho
2025 Poster
in
Workshop: Tackling Climate Change with Machine Learning
in
Workshop: Tackling Climate Change with Machine Learning
Abstract
We introduce SpLIIF to generate implicit neural representations and enable arbitrary downscaling of weather variables. We train a model from sparse weather stations and topography over Japan and evaluate in- and out-of-distribution accuracy predicting temperature and wind, comparing it to both an interpolationbaseline and CorrDiff. We find the model to be up to 50% better than both CorrDiff and the baseline at downscaling temperature, and around 10-20% better for wind.
Video
Chat is not available.
Successful Page Load