MSRGAN: A Multi-Scale Residual GAN for High-Resolution Precipitation Downscaling

To address the challenge of insufficient spatial resolution in remote sensing precipitation data, this paper proposes a novel Multi-Scale Residual Generative Adversarial Network (MSRGAN) for reconstructing high-resolution precipitation images. The model integrates multi-source meteorological informa...

Full description

Saved in:
Bibliographic Details
Main Authors: Yida Liu, Zhuang Li, Guangzhen Cao, Qiong Wang, Yizhe Li, Zhenyu Lu
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/13/2281
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850115996147253248
author Yida Liu
Zhuang Li
Guangzhen Cao
Qiong Wang
Yizhe Li
Zhenyu Lu
author_facet Yida Liu
Zhuang Li
Guangzhen Cao
Qiong Wang
Yizhe Li
Zhenyu Lu
author_sort Yida Liu
collection DOAJ
description To address the challenge of insufficient spatial resolution in remote sensing precipitation data, this paper proposes a novel Multi-Scale Residual Generative Adversarial Network (MSRGAN) for reconstructing high-resolution precipitation images. The model integrates multi-source meteorological information and topographic priors, and it employs a Deep Multi-Scale Perception Module (DeepInception), a Multi-Scale Feature Modulation Module (MSFM), and a Spatial-Channel Attention Network (SCAN) to achieve high-fidelity restoration of complex precipitation structures. Experiments conducted using Weather Research and Forecasting (WRF) simulation data over the continental United States demonstrate that MSRGAN outperforms traditional interpolation methods and state-of-the-art deep learning models across various metrics, including Critical Success Index (CSI), Heidke Skill Score (HSS), False Alarm Rate (FAR), and Jensen–Shannon divergence. Notably, it exhibits significant advantages in detecting heavy precipitation events. Ablation studies further validate the effectiveness of each module. The results indicate that MSRGAN not only improves the accuracy of precipitation downscaling but also preserves spatial structural consistency and physical plausibility, offering a novel technological approach for urban flood warning, weather forecasting, and regional hydrological modeling.
format Article
id doaj-art-484b3ef451ef46af8ede7b7c96a41e41
institution OA Journals
issn 2072-4292
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-484b3ef451ef46af8ede7b7c96a41e412025-08-20T02:36:27ZengMDPI AGRemote Sensing2072-42922025-07-011713228110.3390/rs17132281MSRGAN: A Multi-Scale Residual GAN for High-Resolution Precipitation DownscalingYida Liu0Zhuang Li1Guangzhen Cao2Qiong Wang3Yizhe Li4Zhenyu Lu5School of Mathematics and Statistics, Huazhong University of Science and Technology, Wuhan 430074, ChinaSchool of Electronics and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaKey Laboratory of Radiometric Calibration and Validation for Environmental Satellites, National Satellite Meteorological Center (National Center for Space Weather), China Meteorological Administration, Beijing 100081, ChinaSchool of Mathematics and Statistics, Huazhong University of Science and Technology, Wuhan 430074, ChinaSchool of Artificial Intelligence, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaSchool of Electronics and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaTo address the challenge of insufficient spatial resolution in remote sensing precipitation data, this paper proposes a novel Multi-Scale Residual Generative Adversarial Network (MSRGAN) for reconstructing high-resolution precipitation images. The model integrates multi-source meteorological information and topographic priors, and it employs a Deep Multi-Scale Perception Module (DeepInception), a Multi-Scale Feature Modulation Module (MSFM), and a Spatial-Channel Attention Network (SCAN) to achieve high-fidelity restoration of complex precipitation structures. Experiments conducted using Weather Research and Forecasting (WRF) simulation data over the continental United States demonstrate that MSRGAN outperforms traditional interpolation methods and state-of-the-art deep learning models across various metrics, including Critical Success Index (CSI), Heidke Skill Score (HSS), False Alarm Rate (FAR), and Jensen–Shannon divergence. Notably, it exhibits significant advantages in detecting heavy precipitation events. Ablation studies further validate the effectiveness of each module. The results indicate that MSRGAN not only improves the accuracy of precipitation downscaling but also preserves spatial structural consistency and physical plausibility, offering a novel technological approach for urban flood warning, weather forecasting, and regional hydrological modeling.https://www.mdpi.com/2072-4292/17/13/2281precipitation downscalinggenerative adversarial networkmulti-scale featureshigh-resolution reconstructiondeep learningextreme precipitation detection
spellingShingle Yida Liu
Zhuang Li
Guangzhen Cao
Qiong Wang
Yizhe Li
Zhenyu Lu
MSRGAN: A Multi-Scale Residual GAN for High-Resolution Precipitation Downscaling
Remote Sensing
precipitation downscaling
generative adversarial network
multi-scale features
high-resolution reconstruction
deep learning
extreme precipitation detection
title MSRGAN: A Multi-Scale Residual GAN for High-Resolution Precipitation Downscaling
title_full MSRGAN: A Multi-Scale Residual GAN for High-Resolution Precipitation Downscaling
title_fullStr MSRGAN: A Multi-Scale Residual GAN for High-Resolution Precipitation Downscaling
title_full_unstemmed MSRGAN: A Multi-Scale Residual GAN for High-Resolution Precipitation Downscaling
title_short MSRGAN: A Multi-Scale Residual GAN for High-Resolution Precipitation Downscaling
title_sort msrgan a multi scale residual gan for high resolution precipitation downscaling
topic precipitation downscaling
generative adversarial network
multi-scale features
high-resolution reconstruction
deep learning
extreme precipitation detection
url https://www.mdpi.com/2072-4292/17/13/2281
work_keys_str_mv AT yidaliu msrganamultiscaleresidualganforhighresolutionprecipitationdownscaling
AT zhuangli msrganamultiscaleresidualganforhighresolutionprecipitationdownscaling
AT guangzhencao msrganamultiscaleresidualganforhighresolutionprecipitationdownscaling
AT qiongwang msrganamultiscaleresidualganforhighresolutionprecipitationdownscaling
AT yizheli msrganamultiscaleresidualganforhighresolutionprecipitationdownscaling
AT zhenyulu msrganamultiscaleresidualganforhighresolutionprecipitationdownscaling