MCSPF-Net: A Precipitation Forecasting Method Using Multi-Channel Cloud Observations of FY-4A Satellite by 3D Convolution Neural Network
Accurate precipitation forecasting plays an important role in disaster prevention and mitigation. Currently, precipitation forecasting mainly depends on numerical weather prediction and radar observation. However, ground-based radar observation has limited coverage and is easily influenced by the en...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/18/4536 |
| _version_ | 1797577266306220032 |
|---|---|
| author | Yuhang Jiang Feng Gao Shaoqing Zhang Wei Cheng Chang Liu Shudong Wang |
| author_facet | Yuhang Jiang Feng Gao Shaoqing Zhang Wei Cheng Chang Liu Shudong Wang |
| author_sort | Yuhang Jiang |
| collection | DOAJ |
| description | Accurate precipitation forecasting plays an important role in disaster prevention and mitigation. Currently, precipitation forecasting mainly depends on numerical weather prediction and radar observation. However, ground-based radar observation has limited coverage and is easily influenced by the environment, resulting in the limited coverage of precipitation forecasts. The infrared observations of geosynchronous earth orbit (GEO) satellites have been widely used in precipitation estimation due to their extensive coverage, continuous monitoring, and independence from environmental influences. In this study, we propose a multi-channel satellite precipitation forecasting network (MCSPF-Net) based on 3D convolutional neural networks. The network uses real-time multi-channel satellite observations as input to forecast precipitation for the future 4 h (30-min intervals), utilizing the observation characteristics of GEO satellites for wide coverage precipitation forecasting. The experimental results showed that the precipitation forecasting results of MCSPF-Net have a high correlation with the Global Precipitation Measurement product. When evaluated using rain gauges, the forecasting results of MCSPF-Net exhibited higher critical success index (0.25 vs. 0.21) and correlation coefficients (0.33 vs. 0.23) and a lower mean square error (0.36 vs. 0.93) compared to the numerical weather prediction model. Therefore, the multi-channel satellite observation-driven MCSPF-Net proves to be an effective approach for predicting near future precipitation. |
| first_indexed | 2024-03-10T22:05:46Z |
| format | Article |
| id | doaj.art-c79ee3529ad446e2a01f47ed08a73dec |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T22:05:46Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-c79ee3529ad446e2a01f47ed08a73dec2023-11-19T12:49:05ZengMDPI AGRemote Sensing2072-42922023-09-011518453610.3390/rs15184536MCSPF-Net: A Precipitation Forecasting Method Using Multi-Channel Cloud Observations of FY-4A Satellite by 3D Convolution Neural NetworkYuhang Jiang0Feng Gao1Shaoqing Zhang2Wei Cheng3Chang Liu4Shudong Wang5College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin 150001, ChinaCollege of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin 150001, ChinaCollege of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin 150001, ChinaBeijing Institute of Applied Meteorology, Beijing 100029, ChinaCollege of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin 150001, ChinaPublic Meteorological Service Center, China Meterological Administration, Beijing 100081, ChinaAccurate precipitation forecasting plays an important role in disaster prevention and mitigation. Currently, precipitation forecasting mainly depends on numerical weather prediction and radar observation. However, ground-based radar observation has limited coverage and is easily influenced by the environment, resulting in the limited coverage of precipitation forecasts. The infrared observations of geosynchronous earth orbit (GEO) satellites have been widely used in precipitation estimation due to their extensive coverage, continuous monitoring, and independence from environmental influences. In this study, we propose a multi-channel satellite precipitation forecasting network (MCSPF-Net) based on 3D convolutional neural networks. The network uses real-time multi-channel satellite observations as input to forecast precipitation for the future 4 h (30-min intervals), utilizing the observation characteristics of GEO satellites for wide coverage precipitation forecasting. The experimental results showed that the precipitation forecasting results of MCSPF-Net have a high correlation with the Global Precipitation Measurement product. When evaluated using rain gauges, the forecasting results of MCSPF-Net exhibited higher critical success index (0.25 vs. 0.21) and correlation coefficients (0.33 vs. 0.23) and a lower mean square error (0.36 vs. 0.93) compared to the numerical weather prediction model. Therefore, the multi-channel satellite observation-driven MCSPF-Net proves to be an effective approach for predicting near future precipitation.https://www.mdpi.com/2072-4292/15/18/4536precipitation forecastingneural networksatelliteAGRI |
| spellingShingle | Yuhang Jiang Feng Gao Shaoqing Zhang Wei Cheng Chang Liu Shudong Wang MCSPF-Net: A Precipitation Forecasting Method Using Multi-Channel Cloud Observations of FY-4A Satellite by 3D Convolution Neural Network Remote Sensing precipitation forecasting neural network satellite AGRI |
| title | MCSPF-Net: A Precipitation Forecasting Method Using Multi-Channel Cloud Observations of FY-4A Satellite by 3D Convolution Neural Network |
| title_full | MCSPF-Net: A Precipitation Forecasting Method Using Multi-Channel Cloud Observations of FY-4A Satellite by 3D Convolution Neural Network |
| title_fullStr | MCSPF-Net: A Precipitation Forecasting Method Using Multi-Channel Cloud Observations of FY-4A Satellite by 3D Convolution Neural Network |
| title_full_unstemmed | MCSPF-Net: A Precipitation Forecasting Method Using Multi-Channel Cloud Observations of FY-4A Satellite by 3D Convolution Neural Network |
| title_short | MCSPF-Net: A Precipitation Forecasting Method Using Multi-Channel Cloud Observations of FY-4A Satellite by 3D Convolution Neural Network |
| title_sort | mcspf net a precipitation forecasting method using multi channel cloud observations of fy 4a satellite by 3d convolution neural network |
| topic | precipitation forecasting neural network satellite AGRI |
| url | https://www.mdpi.com/2072-4292/15/18/4536 |
| work_keys_str_mv | AT yuhangjiang mcspfnetaprecipitationforecastingmethodusingmultichannelcloudobservationsoffy4asatelliteby3dconvolutionneuralnetwork AT fenggao mcspfnetaprecipitationforecastingmethodusingmultichannelcloudobservationsoffy4asatelliteby3dconvolutionneuralnetwork AT shaoqingzhang mcspfnetaprecipitationforecastingmethodusingmultichannelcloudobservationsoffy4asatelliteby3dconvolutionneuralnetwork AT weicheng mcspfnetaprecipitationforecastingmethodusingmultichannelcloudobservationsoffy4asatelliteby3dconvolutionneuralnetwork AT changliu mcspfnetaprecipitationforecastingmethodusingmultichannelcloudobservationsoffy4asatelliteby3dconvolutionneuralnetwork AT shudongwang mcspfnetaprecipitationforecastingmethodusingmultichannelcloudobservationsoffy4asatelliteby3dconvolutionneuralnetwork |