Weather Radar Echo Extrapolation Method Based on Deep Learning
In order to forecast some high intensity and rapidly changing phenomena, such as thunderstorms, heavy rain, and hail within 2 h, and reduce the influence brought by destructive weathers, this paper proposes a weather radar echo extrapolation method based on deep learning. The proposed method include...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/13/5/815 |
| _version_ | 1797501589014970368 |
|---|---|
| author | Fugui Zhang Can Lai Wanjun Chen |
| author_facet | Fugui Zhang Can Lai Wanjun Chen |
| author_sort | Fugui Zhang |
| collection | DOAJ |
| description | In order to forecast some high intensity and rapidly changing phenomena, such as thunderstorms, heavy rain, and hail within 2 h, and reduce the influence brought by destructive weathers, this paper proposes a weather radar echo extrapolation method based on deep learning. The proposed method includes the design and combination of the data preprocessing, convolutional long short-term memory (Conv-LSTM) neuron and encoder–decoder model. We collect eleven thousand weather radar echo data in high spatiotemporal resolution, these data are then preprocessed before they enter the neural network for training to improve the data’s quality and make the training better. Next, the neuron integrates the structure and the advantages of convolutional neural network (CNN) and long short-term memory (LSTM), called Conv-LSTM, is applied to solve the problem that the full-connection LSTM (FC-LSTM) cannot extract the spatial information of input data. This operation replaced the full-connection structure in the input-to-state and state-to-state parts so that the Conv-LSTM can extract the information from other dimensions. Meanwhile, the encoder–decoder model is adopted due to the size difference of the input and output data to combine with the Conv-LSTM neuron. In the neural network training, mean square error (<i>MSE</i>) loss function weighted according to the rate of rainfall is added. Finally, the matrix “point-to-point” test method, including the probability of detection (<i>POD</i>), critical success index (<i>CSI</i>), false alarm ratio (<i>FAR</i>) and spatial test method contiguous rain areas (CRA), is used to examine the radar echo extrapolation’s results. Under the threshold of 30 dBZ, at the time of 1 h, we achieved 0.60 (<i>POD</i>), 0.42 (<i>CSI</i>) and 0.51 (<i>FAR</i>), compared with 0.42, 0.28 and 0.58 for the CTREC algorithm, and 0.30, 0.24 and 0.71 for the TITAN algorithm. Meanwhile, at the time of 1 h, we achieved 1.35 (total MSE ) compared with 3.26 for the CTREC algorithm and 3.05 for the TITAN algorithm. The results demonstrate that the radar echo extrapolation method based on deep learning is obviously more accurate and stable than traditional radar echo extrapolation methods in near weather forecasting. |
| first_indexed | 2024-03-10T03:20:36Z |
| format | Article |
| id | doaj.art-9c1b381032964dbdb5adcb140a0f8a97 |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-03-10T03:20:36Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-9c1b381032964dbdb5adcb140a0f8a972023-11-23T10:03:13ZengMDPI AGAtmosphere2073-44332022-05-0113581510.3390/atmos13050815Weather Radar Echo Extrapolation Method Based on Deep LearningFugui Zhang0Can Lai1Wanjun Chen2School of Electronic Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaSchool of Electronic Engineering, Chengdu University of Information Technology, Chengdu 610225, ChinaSchool of Electronic Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, ChinaIn order to forecast some high intensity and rapidly changing phenomena, such as thunderstorms, heavy rain, and hail within 2 h, and reduce the influence brought by destructive weathers, this paper proposes a weather radar echo extrapolation method based on deep learning. The proposed method includes the design and combination of the data preprocessing, convolutional long short-term memory (Conv-LSTM) neuron and encoder–decoder model. We collect eleven thousand weather radar echo data in high spatiotemporal resolution, these data are then preprocessed before they enter the neural network for training to improve the data’s quality and make the training better. Next, the neuron integrates the structure and the advantages of convolutional neural network (CNN) and long short-term memory (LSTM), called Conv-LSTM, is applied to solve the problem that the full-connection LSTM (FC-LSTM) cannot extract the spatial information of input data. This operation replaced the full-connection structure in the input-to-state and state-to-state parts so that the Conv-LSTM can extract the information from other dimensions. Meanwhile, the encoder–decoder model is adopted due to the size difference of the input and output data to combine with the Conv-LSTM neuron. In the neural network training, mean square error (<i>MSE</i>) loss function weighted according to the rate of rainfall is added. Finally, the matrix “point-to-point” test method, including the probability of detection (<i>POD</i>), critical success index (<i>CSI</i>), false alarm ratio (<i>FAR</i>) and spatial test method contiguous rain areas (CRA), is used to examine the radar echo extrapolation’s results. Under the threshold of 30 dBZ, at the time of 1 h, we achieved 0.60 (<i>POD</i>), 0.42 (<i>CSI</i>) and 0.51 (<i>FAR</i>), compared with 0.42, 0.28 and 0.58 for the CTREC algorithm, and 0.30, 0.24 and 0.71 for the TITAN algorithm. Meanwhile, at the time of 1 h, we achieved 1.35 (total MSE ) compared with 3.26 for the CTREC algorithm and 3.05 for the TITAN algorithm. The results demonstrate that the radar echo extrapolation method based on deep learning is obviously more accurate and stable than traditional radar echo extrapolation methods in near weather forecasting.https://www.mdpi.com/2073-4433/13/5/815weather radar echo extrapolationdeep learningConv-LSTMencoder–decoder modelloss function |
| spellingShingle | Fugui Zhang Can Lai Wanjun Chen Weather Radar Echo Extrapolation Method Based on Deep Learning Atmosphere weather radar echo extrapolation deep learning Conv-LSTM encoder–decoder model loss function |
| title | Weather Radar Echo Extrapolation Method Based on Deep Learning |
| title_full | Weather Radar Echo Extrapolation Method Based on Deep Learning |
| title_fullStr | Weather Radar Echo Extrapolation Method Based on Deep Learning |
| title_full_unstemmed | Weather Radar Echo Extrapolation Method Based on Deep Learning |
| title_short | Weather Radar Echo Extrapolation Method Based on Deep Learning |
| title_sort | weather radar echo extrapolation method based on deep learning |
| topic | weather radar echo extrapolation deep learning Conv-LSTM encoder–decoder model loss function |
| url | https://www.mdpi.com/2073-4433/13/5/815 |
| work_keys_str_mv | AT fuguizhang weatherradarechoextrapolationmethodbasedondeeplearning AT canlai weatherradarechoextrapolationmethodbasedondeeplearning AT wanjunchen weatherradarechoextrapolationmethodbasedondeeplearning |