Accurate Rainfall Prediction Using GNSS PWV Based on Pre-Trained Transformer Model

With an increase in the intensity and frequency of extreme rainfall events, there is a pressing need for accurate rainfall nowcasting applications. In recent years, precipitable water vapor (PWV) data obtained from GNSS observations have been widely used in rainfall prediction. Unlike previous studi...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenjie Yin, Chen Zhou, Yuan Tian, Hui Qiu, Wei Zhang, Hua Chen, Pan Liu, Qile Zhao, Jian Kong, Yibin Yao
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/12/2023
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849704816563978240
author Wenjie Yin
Chen Zhou
Yuan Tian
Hui Qiu
Wei Zhang
Hua Chen
Pan Liu
Qile Zhao
Jian Kong
Yibin Yao
author_facet Wenjie Yin
Chen Zhou
Yuan Tian
Hui Qiu
Wei Zhang
Hua Chen
Pan Liu
Qile Zhao
Jian Kong
Yibin Yao
author_sort Wenjie Yin
collection DOAJ
description With an increase in the intensity and frequency of extreme rainfall events, there is a pressing need for accurate rainfall nowcasting applications. In recent years, precipitable water vapor (PWV) data obtained from GNSS observations have been widely used in rainfall prediction. Unlike previous studies mainly focusing on rainfall occurrences, this study proposes a transformer-based model for hourly rainfall prediction, integrating the GNSS PWV and ERA5 meteorological data. The proposed model employs the ProbSparse self-attention to efficiently capture long-range dependencies in time series data, crucial for correlating historical PWV variations with rainfall events. Additionally, the adoption of the DILATE loss function better captures the structural and timing aspects of rainfall prediction. Furthermore, traditional rainfall prediction models are typically trained on datasets specific to one region, which limits their generalization ability due to regional meteorological differences and the scarcity of data in certain areas. Therefore, we adopt a pre-training and fine-tuning strategy using global datasets to mitigate data scarcity in newly deployed GNSS stations, enhancing model adaptability to local conditions. The evaluation results demonstrate satisfactory performance over other methods, with the fine-tuned model achieving an MSE = 3.954, DTW = 0.232, and TDI = 0.101. This approach shows great potential for real-time rainfall nowcasting in a local area, especially with limited data.
format Article
id doaj-art-2134bfafea3244bfad7b4ff1c87e8943
institution DOAJ
issn 2072-4292
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-2134bfafea3244bfad7b4ff1c87e89432025-08-20T03:16:39ZengMDPI AGRemote Sensing2072-42922025-06-011712202310.3390/rs17122023Accurate Rainfall Prediction Using GNSS PWV Based on Pre-Trained Transformer ModelWenjie Yin0Chen Zhou1Yuan Tian2Hui Qiu3Wei Zhang4Hua Chen5Pan Liu6Qile Zhao7Jian Kong8Yibin Yao9School of Earth and Space Science Technology, Wuhan University, Wuhan 430072, ChinaSchool of Earth and Space Science Technology, Wuhan University, Wuhan 430072, ChinaSchool of Earth and Space Science Technology, Wuhan University, Wuhan 430072, ChinaSchool of Earth and Space Science Technology, Wuhan University, Wuhan 430072, ChinaDepartment of Forecasting and Networking, China Meteorological Administration, Beijing 100081, ChinaSchool of Water Resources and Hydropower Engineering, Wuhan University, Wuhan 430062, ChinaSchool of Water Resources and Hydropower Engineering, Wuhan University, Wuhan 430062, ChinaGNSS Research Center, Wuhan University, Wuhan 430062, ChinaSchool of Geodesy and Geomatics, Wuhan University, Wuhan 430062, ChinaSchool of Geodesy and Geomatics, Wuhan University, Wuhan 430062, ChinaWith an increase in the intensity and frequency of extreme rainfall events, there is a pressing need for accurate rainfall nowcasting applications. In recent years, precipitable water vapor (PWV) data obtained from GNSS observations have been widely used in rainfall prediction. Unlike previous studies mainly focusing on rainfall occurrences, this study proposes a transformer-based model for hourly rainfall prediction, integrating the GNSS PWV and ERA5 meteorological data. The proposed model employs the ProbSparse self-attention to efficiently capture long-range dependencies in time series data, crucial for correlating historical PWV variations with rainfall events. Additionally, the adoption of the DILATE loss function better captures the structural and timing aspects of rainfall prediction. Furthermore, traditional rainfall prediction models are typically trained on datasets specific to one region, which limits their generalization ability due to regional meteorological differences and the scarcity of data in certain areas. Therefore, we adopt a pre-training and fine-tuning strategy using global datasets to mitigate data scarcity in newly deployed GNSS stations, enhancing model adaptability to local conditions. The evaluation results demonstrate satisfactory performance over other methods, with the fine-tuned model achieving an MSE = 3.954, DTW = 0.232, and TDI = 0.101. This approach shows great potential for real-time rainfall nowcasting in a local area, especially with limited data.https://www.mdpi.com/2072-4292/17/12/2023rainfall predictiondeep learningtransfer learningGlobal Navigation Satellite System (GNSS)precipitable water vapor (PWV)
spellingShingle Wenjie Yin
Chen Zhou
Yuan Tian
Hui Qiu
Wei Zhang
Hua Chen
Pan Liu
Qile Zhao
Jian Kong
Yibin Yao
Accurate Rainfall Prediction Using GNSS PWV Based on Pre-Trained Transformer Model
Remote Sensing
rainfall prediction
deep learning
transfer learning
Global Navigation Satellite System (GNSS)
precipitable water vapor (PWV)
title Accurate Rainfall Prediction Using GNSS PWV Based on Pre-Trained Transformer Model
title_full Accurate Rainfall Prediction Using GNSS PWV Based on Pre-Trained Transformer Model
title_fullStr Accurate Rainfall Prediction Using GNSS PWV Based on Pre-Trained Transformer Model
title_full_unstemmed Accurate Rainfall Prediction Using GNSS PWV Based on Pre-Trained Transformer Model
title_short Accurate Rainfall Prediction Using GNSS PWV Based on Pre-Trained Transformer Model
title_sort accurate rainfall prediction using gnss pwv based on pre trained transformer model
topic rainfall prediction
deep learning
transfer learning
Global Navigation Satellite System (GNSS)
precipitable water vapor (PWV)
url https://www.mdpi.com/2072-4292/17/12/2023
work_keys_str_mv AT wenjieyin accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT chenzhou accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT yuantian accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT huiqiu accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT weizhang accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT huachen accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT panliu accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT qilezhao accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT jiankong accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel
AT yibinyao accuraterainfallpredictionusinggnsspwvbasedonpretrainedtransformermodel