Saved in:
Bibliographic Details
Main Authors: Harder, Paula, Lessig, Christian, Chantry, Matthew, Pelletier, Francis, Rolnick, David
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.01400
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908683839995904
author Harder, Paula
Lessig, Christian
Chantry, Matthew
Pelletier, Francis
Rolnick, David
author_facet Harder, Paula
Lessig, Christian
Chantry, Matthew
Pelletier, Francis
Rolnick, David
contents Deep learning offers promising capabilities for the statistical downscaling of climate and weather forecasts, with generative approaches showing particular success in capturing fine-scale precipitation patterns. However, most existing models are region-specific, and their ability to generalize to unseen geographic areas remains largely unexplored. In this study, we evaluate the generalization performance of generative downscaling models across diverse regions. Using a global framework, we employ ERA5 reanalysis data as predictors and IMERG precipitation estimates at $0.1^\circ$ resolution as targets. A hierarchical location-based data split enables a systematic assessment of model performance across 15 regions around the world.
format Preprint
id arxiv_https___arxiv_org_abs_2512_01400
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On Global Applicability and Location Transferability of Generative Deep Learning Models for Precipitation Downscaling
Harder, Paula
Lessig, Christian
Chantry, Matthew
Pelletier, Francis
Rolnick, David
Machine Learning
Deep learning offers promising capabilities for the statistical downscaling of climate and weather forecasts, with generative approaches showing particular success in capturing fine-scale precipitation patterns. However, most existing models are region-specific, and their ability to generalize to unseen geographic areas remains largely unexplored. In this study, we evaluate the generalization performance of generative downscaling models across diverse regions. Using a global framework, we employ ERA5 reanalysis data as predictors and IMERG precipitation estimates at $0.1^\circ$ resolution as targets. A hierarchical location-based data split enables a systematic assessment of model performance across 15 regions around the world.
title On Global Applicability and Location Transferability of Generative Deep Learning Models for Precipitation Downscaling
topic Machine Learning
url https://arxiv.org/abs/2512.01400