Improving the Completion of Weather Radar Missing Data with Deep Learning
Weather radars commonly suffer from the data-missing problem that limits their data quality and applications. Traditional methods for the completion of weather radar missing data, which are based on radar physics and statistics, have shown defects in various aspects. Several deep learning (DL) model...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/18/4568 |
| _version_ | 1797577191752466432 |
|---|---|
| author | Aofan Gong Haonan Chen Guangheng Ni |
| author_facet | Aofan Gong Haonan Chen Guangheng Ni |
| author_sort | Aofan Gong |
| collection | DOAJ |
| description | Weather radars commonly suffer from the data-missing problem that limits their data quality and applications. Traditional methods for the completion of weather radar missing data, which are based on radar physics and statistics, have shown defects in various aspects. Several deep learning (DL) models have been designed and applied to weather radar completion tasks but have been limited by low accuracy. This study proposes a dilated and self-attentional UNet (DSA-UNet) model to improve the completion of weather radar missing data. The model is trained and evaluated on a radar dataset built with random sector masking from the Yizhuang radar observations during the warm seasons from 2017 to 2019, which is further analyzed with two cases from the dataset. The performance of the DSA-UNet model is compared to two traditional statistical methods and a DL model. The evaluation methods consist of three quantitative metrics and three diagrams. The results show that the DL models can produce less biased and more accurate radar reflectivity values for data-missing areas than traditional statistical methods. Compared to the other DL model, the DSA-UNet model can not only produce a completion closer to the observation, especially for extreme values, but also improve the detection and reconstruction of local-scale radar echo patterns. Our study provides an effective solution for improving the completion of weather radar missing data, which is indispensable in radar quantitative applications. |
| first_indexed | 2024-03-10T22:05:37Z |
| format | Article |
| id | doaj.art-a0cf2b5e4bd5464682c7d046b164d7cb |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T22:05:37Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-a0cf2b5e4bd5464682c7d046b164d7cb2023-11-19T12:49:33ZengMDPI AGRemote Sensing2072-42922023-09-011518456810.3390/rs15184568Improving the Completion of Weather Radar Missing Data with Deep LearningAofan Gong0Haonan Chen1Guangheng Ni2State Key Laboratory of Hydroscience and Engineering, Department of Hydraulic Engineering, Tsinghua University, Beijing 100084, ChinaElectrical and Computer Engineering, Colorado State University, Fort Collins, CO 80523, USAState Key Laboratory of Hydroscience and Engineering, Department of Hydraulic Engineering, Tsinghua University, Beijing 100084, ChinaWeather radars commonly suffer from the data-missing problem that limits their data quality and applications. Traditional methods for the completion of weather radar missing data, which are based on radar physics and statistics, have shown defects in various aspects. Several deep learning (DL) models have been designed and applied to weather radar completion tasks but have been limited by low accuracy. This study proposes a dilated and self-attentional UNet (DSA-UNet) model to improve the completion of weather radar missing data. The model is trained and evaluated on a radar dataset built with random sector masking from the Yizhuang radar observations during the warm seasons from 2017 to 2019, which is further analyzed with two cases from the dataset. The performance of the DSA-UNet model is compared to two traditional statistical methods and a DL model. The evaluation methods consist of three quantitative metrics and three diagrams. The results show that the DL models can produce less biased and more accurate radar reflectivity values for data-missing areas than traditional statistical methods. Compared to the other DL model, the DSA-UNet model can not only produce a completion closer to the observation, especially for extreme values, but also improve the detection and reconstruction of local-scale radar echo patterns. Our study provides an effective solution for improving the completion of weather radar missing data, which is indispensable in radar quantitative applications.https://www.mdpi.com/2072-4292/15/18/4568weather radarmissing datadata completiondeep learning |
| spellingShingle | Aofan Gong Haonan Chen Guangheng Ni Improving the Completion of Weather Radar Missing Data with Deep Learning Remote Sensing weather radar missing data data completion deep learning |
| title | Improving the Completion of Weather Radar Missing Data with Deep Learning |
| title_full | Improving the Completion of Weather Radar Missing Data with Deep Learning |
| title_fullStr | Improving the Completion of Weather Radar Missing Data with Deep Learning |
| title_full_unstemmed | Improving the Completion of Weather Radar Missing Data with Deep Learning |
| title_short | Improving the Completion of Weather Radar Missing Data with Deep Learning |
| title_sort | improving the completion of weather radar missing data with deep learning |
| topic | weather radar missing data data completion deep learning |
| url | https://www.mdpi.com/2072-4292/15/18/4568 |
| work_keys_str_mv | AT aofangong improvingthecompletionofweatherradarmissingdatawithdeeplearning AT haonanchen improvingthecompletionofweatherradarmissingdatawithdeeplearning AT guanghengni improvingthecompletionofweatherradarmissingdatawithdeeplearning |