A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 Hour Precipitation Nowcasting
Multi-source meteorological data can reflect the development process of single meteorological elements from different angles. Making full use of multi-source meteorological data is an effective method to improve the performance of weather nowcasting. For precipitation nowcasting, this paper proposes...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/12/12/1596 |
| _version_ | 1797506672541827072 |
|---|---|
| author | Fuhan Zhang Xiaodong Wang Jiping Guan |
| author_facet | Fuhan Zhang Xiaodong Wang Jiping Guan |
| author_sort | Fuhan Zhang |
| collection | DOAJ |
| description | Multi-source meteorological data can reflect the development process of single meteorological elements from different angles. Making full use of multi-source meteorological data is an effective method to improve the performance of weather nowcasting. For precipitation nowcasting, this paper proposes a novel multi-input multi-output recurrent neural network model based on multimodal fusion and spatiotemporal prediction, named MFSP-Net. It uses precipitation grid data, radar echo data, and reanalysis data as input data and simultaneously realizes 0–4 h precipitation amount nowcasting and precipitation intensity nowcasting. MFSP-Net can perform the spatiotemporal-scale fusion of the three sources of input data while retaining the spatiotemporal information flow of them. The multi-task learning strategy is used to train the network. We conduct experiments on the dataset of Southeast China, and the results show that MFSP-Net comprehensively improves the performance of the nowcasting of precipitation amounts. For precipitation intensity nowcasting, MFSP-Net has obvious advantages in heavy precipitation nowcasting and the middle and late stages of nowcasting. |
| first_indexed | 2024-03-10T04:35:51Z |
| format | Article |
| id | doaj.art-8e5fa40ab54947e8a4660fdcb6ac4d76 |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-03-10T04:35:51Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-8e5fa40ab54947e8a4660fdcb6ac4d762023-11-23T03:45:59ZengMDPI AGAtmosphere2073-44332021-11-011212159610.3390/atmos12121596A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 Hour Precipitation NowcastingFuhan Zhang0Xiaodong Wang1Jiping Guan2School of Computer, National University of Defense Technology, Changsha 410000, ChinaSchool of Computer, National University of Defense Technology, Changsha 410000, ChinaSchool of Meteorology and Oceanography, National University of Defense Technology, Changsha 410000, ChinaMulti-source meteorological data can reflect the development process of single meteorological elements from different angles. Making full use of multi-source meteorological data is an effective method to improve the performance of weather nowcasting. For precipitation nowcasting, this paper proposes a novel multi-input multi-output recurrent neural network model based on multimodal fusion and spatiotemporal prediction, named MFSP-Net. It uses precipitation grid data, radar echo data, and reanalysis data as input data and simultaneously realizes 0–4 h precipitation amount nowcasting and precipitation intensity nowcasting. MFSP-Net can perform the spatiotemporal-scale fusion of the three sources of input data while retaining the spatiotemporal information flow of them. The multi-task learning strategy is used to train the network. We conduct experiments on the dataset of Southeast China, and the results show that MFSP-Net comprehensively improves the performance of the nowcasting of precipitation amounts. For precipitation intensity nowcasting, MFSP-Net has obvious advantages in heavy precipitation nowcasting and the middle and late stages of nowcasting.https://www.mdpi.com/2073-4433/12/12/1596radar echo datareanalysis dataprecipitation amount grid datadeep learningspatiotemporal predictionmultimodal fusion |
| spellingShingle | Fuhan Zhang Xiaodong Wang Jiping Guan A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 Hour Precipitation Nowcasting Atmosphere radar echo data reanalysis data precipitation amount grid data deep learning spatiotemporal prediction multimodal fusion |
| title | A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 Hour Precipitation Nowcasting |
| title_full | A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 Hour Precipitation Nowcasting |
| title_fullStr | A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 Hour Precipitation Nowcasting |
| title_full_unstemmed | A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 Hour Precipitation Nowcasting |
| title_short | A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 Hour Precipitation Nowcasting |
| title_sort | novel multi input multi output recurrent neural network based on multimodal fusion and spatiotemporal prediction for 0 4 hour precipitation nowcasting |
| topic | radar echo data reanalysis data precipitation amount grid data deep learning spatiotemporal prediction multimodal fusion |
| url | https://www.mdpi.com/2073-4433/12/12/1596 |
| work_keys_str_mv | AT fuhanzhang anovelmultiinputmultioutputrecurrentneuralnetworkbasedonmultimodalfusionandspatiotemporalpredictionfor04hourprecipitationnowcasting AT xiaodongwang anovelmultiinputmultioutputrecurrentneuralnetworkbasedonmultimodalfusionandspatiotemporalpredictionfor04hourprecipitationnowcasting AT jipingguan anovelmultiinputmultioutputrecurrentneuralnetworkbasedonmultimodalfusionandspatiotemporalpredictionfor04hourprecipitationnowcasting AT fuhanzhang novelmultiinputmultioutputrecurrentneuralnetworkbasedonmultimodalfusionandspatiotemporalpredictionfor04hourprecipitationnowcasting AT xiaodongwang novelmultiinputmultioutputrecurrentneuralnetworkbasedonmultimodalfusionandspatiotemporalpredictionfor04hourprecipitationnowcasting AT jipingguan novelmultiinputmultioutputrecurrentneuralnetworkbasedonmultimodalfusionandspatiotemporalpredictionfor04hourprecipitationnowcasting |