Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sundar, Rahul, Hu, Yucong, Parashar, Nishant, Blanchard, Antoine, Dodov, Boyko
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2412.13627
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915195216986112
author Sundar, Rahul
Hu, Yucong
Parashar, Nishant
Blanchard, Antoine
Dodov, Boyko
author_facet Sundar, Rahul
Hu, Yucong
Parashar, Nishant
Blanchard, Antoine
Dodov, Boyko
contents Deterministic regression-based downscaling models for climate variables often suffer from spectral bias, which can be mitigated by generative models like diffusion models. To enable efficient and reliable simulation of extreme weather events, it is crucial to achieve rapid turnaround, dynamical consistency, and accurate spatio-temporal spectral recovery. We propose an efficient correction diffusion model, TAUDiff, that combines a deterministic spatio-temporal model for mean field downscaling with a smaller generative diffusion model for recovering the fine-scale stochastic features. We demonstrate the efficacy of this approach on downscaling atmospheric wind velocity fields obtained from coarse GCM simulations. We then extend TAUDiff for computationally efficient kilometer-scale downscaling of atmospheric wind velocity fields. Owing to low inference times, our approach can ensure quicker simulation of extreme events necessary for estimating associated risks and economic losses.
format Preprint
id arxiv_https___arxiv_org_abs_2412_13627
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models
Sundar, Rahul
Hu, Yucong
Parashar, Nishant
Blanchard, Antoine
Dodov, Boyko
Machine Learning
Deterministic regression-based downscaling models for climate variables often suffer from spectral bias, which can be mitigated by generative models like diffusion models. To enable efficient and reliable simulation of extreme weather events, it is crucial to achieve rapid turnaround, dynamical consistency, and accurate spatio-temporal spectral recovery. We propose an efficient correction diffusion model, TAUDiff, that combines a deterministic spatio-temporal model for mean field downscaling with a smaller generative diffusion model for recovering the fine-scale stochastic features. We demonstrate the efficacy of this approach on downscaling atmospheric wind velocity fields obtained from coarse GCM simulations. We then extend TAUDiff for computationally efficient kilometer-scale downscaling of atmospheric wind velocity fields. Owing to low inference times, our approach can ensure quicker simulation of extreme events necessary for estimating associated risks and economic losses.
title TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models
topic Machine Learning
url https://arxiv.org/abs/2412.13627