Enregistré dans:
Détails bibliographiques
Auteurs principaux: Brinkerhoff, Douglas, Fischer, Elizabeth
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2604.25172
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913067435032576
author Brinkerhoff, Douglas
Fischer, Elizabeth
author_facet Brinkerhoff, Douglas
Fischer, Elizabeth
contents Precipitation in complex terrain is governed by orographic processes operating at scales of a few kilometers, yet climate models typically run at resolutions of 50--100~km where this topographic detail is absent. Dynamical downscaling with high-resolution regional models such as WRF can resolve these processes, but the computational cost -- months of wall-clock time per scenario -- precludes the large ensembles needed for uncertainty quantification. We present WxFlow, a conditional generative model based on flow matching that learns to map coarse-resolution climate model output and high-resolution topography to calibrated probabilistic ensembles of fine-scale precipitation fields. Applied to 4~km WRF simulations of maximum 3-day snowfall over southeast Alaska, WxFlow achieves 87.8\% improvement in spectral fidelity and dramatically lower Continuous Ranked Probability Scores relative to conventional lapse-rate-corrected bicubic downscaling, while generating 50-member ensembles in seconds on a laptop. Ensemble spread is spatially coherent and governed by topography, reflecting physically plausible uncertainty structure. All code is available at https://github.com/glide-ism/wrf-flow.
format Preprint
id arxiv_https___arxiv_org_abs_2604_25172
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Conditional Flow Matching for Probabilistic Downscaling of Maximum 3-day Snowfall in Alaska
Brinkerhoff, Douglas
Fischer, Elizabeth
Computational Physics
Machine Learning
Atmospheric and Oceanic Physics
Precipitation in complex terrain is governed by orographic processes operating at scales of a few kilometers, yet climate models typically run at resolutions of 50--100~km where this topographic detail is absent. Dynamical downscaling with high-resolution regional models such as WRF can resolve these processes, but the computational cost -- months of wall-clock time per scenario -- precludes the large ensembles needed for uncertainty quantification. We present WxFlow, a conditional generative model based on flow matching that learns to map coarse-resolution climate model output and high-resolution topography to calibrated probabilistic ensembles of fine-scale precipitation fields. Applied to 4~km WRF simulations of maximum 3-day snowfall over southeast Alaska, WxFlow achieves 87.8\% improvement in spectral fidelity and dramatically lower Continuous Ranked Probability Scores relative to conventional lapse-rate-corrected bicubic downscaling, while generating 50-member ensembles in seconds on a laptop. Ensemble spread is spatially coherent and governed by topography, reflecting physically plausible uncertainty structure. All code is available at https://github.com/glide-ism/wrf-flow.
title Conditional Flow Matching for Probabilistic Downscaling of Maximum 3-day Snowfall in Alaska
topic Computational Physics
Machine Learning
Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2604.25172