Saved in:
Bibliographic Details
Main Authors: Chen, Huaguan, Han, Wei, Sun, Haofei, Lin, Ning, Song, Xingtao, Yang, Yunfan, Tian, Jie, Liu, Yang, Wen, Ji-Rong, Zhang, Xiaoye, Shen, Xueshun, Sun, Hao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.04659
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908784605003776
author Chen, Huaguan
Han, Wei
Sun, Haofei
Lin, Ning
Song, Xingtao
Yang, Yunfan
Tian, Jie
Liu, Yang
Wen, Ji-Rong
Zhang, Xiaoye
Shen, Xueshun
Sun, Hao
author_facet Chen, Huaguan
Han, Wei
Sun, Haofei
Lin, Ning
Song, Xingtao
Yang, Yunfan
Tian, Jie
Liu, Yang
Wen, Ji-Rong
Zhang, Xiaoye
Shen, Xueshun
Sun, Hao
contents Reliable nowcasting of extreme precipitation remains difficult because convective systems are strongly nonlinear, multiscale, and nonstationary in 3D. Radar is the backbone of nowcasting, yet existing methods struggle to predict extremes: physics-based extrapolation cannot capture growth and decay, deterministic learning tends to oversmooth and underestimate peaks, and purely generative models often lack physical consistency. Hybrid schemes help but are mostly limited to 2D composite reflectivity, collapsing the atmosphere into one layer and discarding vertical structure critical for height-dependent dynamics. We introduce Nowcast3D, a gray-box, fully 3D framework that works directly on volumetric radar reflectivity. The end-to-end model couples physically constrained neural operators (advection, local diffusion, and microphysics) with a conditional diffusion model to generate ensemble forecasts with quantified uncertainty. Trained on provincial-scale 3D volumes over a $10.24^\circ \times 10.24^\circ$ region and fine-tuned on a $2.56^\circ \times 2.56^\circ$ city region ($0.01^\circ \approx 1$ km), Nowcast3D provides near-real-time forecasts up to 3 h and outperforms competitive baselines in cross-region and temporal out-of-sample tests. It can also infer wind fields without labeled supervision, supporting physically plausible transport. In a nationwide blind evaluation by 160 meteorologists, Nowcast3D ranked first and was preferred in 57% of post-hoc assessments, surpassing the leading baseline (27%). These results highlight its reliability and operational value for extreme precipitation nowcasting.
format Preprint
id arxiv_https___arxiv_org_abs_2511_04659
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Nowcast3D: Reliable precipitation nowcasting via gray-box learning
Chen, Huaguan
Han, Wei
Sun, Haofei
Lin, Ning
Song, Xingtao
Yang, Yunfan
Tian, Jie
Liu, Yang
Wen, Ji-Rong
Zhang, Xiaoye
Shen, Xueshun
Sun, Hao
Machine Learning
Atmospheric and Oceanic Physics
Reliable nowcasting of extreme precipitation remains difficult because convective systems are strongly nonlinear, multiscale, and nonstationary in 3D. Radar is the backbone of nowcasting, yet existing methods struggle to predict extremes: physics-based extrapolation cannot capture growth and decay, deterministic learning tends to oversmooth and underestimate peaks, and purely generative models often lack physical consistency. Hybrid schemes help but are mostly limited to 2D composite reflectivity, collapsing the atmosphere into one layer and discarding vertical structure critical for height-dependent dynamics. We introduce Nowcast3D, a gray-box, fully 3D framework that works directly on volumetric radar reflectivity. The end-to-end model couples physically constrained neural operators (advection, local diffusion, and microphysics) with a conditional diffusion model to generate ensemble forecasts with quantified uncertainty. Trained on provincial-scale 3D volumes over a $10.24^\circ \times 10.24^\circ$ region and fine-tuned on a $2.56^\circ \times 2.56^\circ$ city region ($0.01^\circ \approx 1$ km), Nowcast3D provides near-real-time forecasts up to 3 h and outperforms competitive baselines in cross-region and temporal out-of-sample tests. It can also infer wind fields without labeled supervision, supporting physically plausible transport. In a nationwide blind evaluation by 160 meteorologists, Nowcast3D ranked first and was preferred in 57% of post-hoc assessments, surpassing the leading baseline (27%). These results highlight its reliability and operational value for extreme precipitation nowcasting.
title Nowcast3D: Reliable precipitation nowcasting via gray-box learning
topic Machine Learning
Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2511.04659