DuoCast: Duo-Probabilistic Diffusion for Precipitation Nowcasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wen, Penghui, He, Mengwei, Filippi, Patrick, Zhao, Na, Zhang, Feng, Bishop, Thomas Francis, Wang, Zhiyong, Hu, Kun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918336070156288
author Wen, Penghui
He, Mengwei
Filippi, Patrick
Zhao, Na
Zhang, Feng
Bishop, Thomas Francis
Wang, Zhiyong
Hu, Kun
author_facet Wen, Penghui
He, Mengwei
Filippi, Patrick
Zhao, Na
Zhang, Feng
Bishop, Thomas Francis
Wang, Zhiyong
Hu, Kun
contents Accurate short-term precipitation forecasting is critical for weather-sensitive decision-making in agriculture, transportation, and disaster response. Existing deep learning approaches often struggle to balance global structural consistency with local detail preservation, especially under complex meteorological conditions. We propose DuoCast, a dual-diffusion framework that decomposes precipitation forecasting into low- and high-frequency components modeled in orthogonal latent subspaces. We theoretically prove that this frequency decomposition reduces prediction error compared to conventional single branch U-Net diffusion models. In DuoCast, the low-frequency model captures large-scale trends via convolutional encoders conditioned on weather front dynamics, while the high-frequency model refines fine-scale variability using a self-attention-based architecture. Experiments on four benchmark radar datasets show that DuoCast consistently outperforms state-of-the-art baselines, achieving superior accuracy in both spatial detail and temporal evolution.
format Preprint
id arxiv_https___arxiv_org_abs_2412_01091
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DuoCast: Duo-Probabilistic Diffusion for Precipitation Nowcasting
Wen, Penghui
He, Mengwei
Filippi, Patrick
Zhao, Na
Zhang, Feng
Bishop, Thomas Francis
Wang, Zhiyong
Hu, Kun
Computer Vision and Pattern Recognition
Accurate short-term precipitation forecasting is critical for weather-sensitive decision-making in agriculture, transportation, and disaster response. Existing deep learning approaches often struggle to balance global structural consistency with local detail preservation, especially under complex meteorological conditions. We propose DuoCast, a dual-diffusion framework that decomposes precipitation forecasting into low- and high-frequency components modeled in orthogonal latent subspaces. We theoretically prove that this frequency decomposition reduces prediction error compared to conventional single branch U-Net diffusion models. In DuoCast, the low-frequency model captures large-scale trends via convolutional encoders conditioned on weather front dynamics, while the high-frequency model refines fine-scale variability using a self-attention-based architecture. Experiments on four benchmark radar datasets show that DuoCast consistently outperforms state-of-the-art baselines, achieving superior accuracy in both spatial detail and temporal evolution.
title DuoCast: Duo-Probabilistic Diffusion for Precipitation Nowcasting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.01091