Universal Diffusion-Based Probabilistic Downscaling

Roberto Molinaro, Niall Siegenheim, Henry Martin, Mark Frey, Niels Poulsen, Philipp Seitz, Marvin Vincent Gabler

Keywords: weather forecasting, statistical downscaling, diffusion-based models, model-agnostic post-processing

TL;DR: We present a universal zero-shot diffusion-based downscaling model that improves both point accuracy and probabilistic skill of weather forecasts across a variety of operational forecasting models at European scale.

Abstract:

We introduce a universal diffusion-based downscaling framework that lifts
deterministic low-resolution weather forecasts into probabilistic
high-resolution predictions without any model-specific fine-tuning.
A single conditional diffusion model is trained on paired coarse-resolution
inputs ($\sim$25\,km resolution) and high-resolution regional reanalysis targets ($\sim$5\,km resolution),
and is applied in a fully zero-shot manner to deterministic forecasts from
heterogeneous upstream weather models. Focusing on near-surface variables, we evaluate probabilistic forecasts against
independent in situ station observations over lead times up to 90\,h.
Across a diverse set of AI-based and numerical weather prediction (NWP) systems,
the ensemble mean of the downscaled forecasts consistently improves upon each
model's own raw deterministic forecast, and substantially larger gains are
observed in probabilistic skill as measured by CRPS.
These results demonstrate that diffusion-based downscaling provides a scalable,
model-agnostic probabilistic interface for enhancing spatial resolution and
uncertainty representation in operational weather forecasting pipelines.

Journal Opt In: Yes, I want to participate in the IOP focus collection submission

Journal Corresponding Email: [e-mail]

Submission Number: 47