Optimization and Construction of Jinan Regional Tm Model Based on LSTM and Analysis of Its Influence on the Accuracy of GNSS Inversion PWV
Water vapor constitutes a vital component of atmospheric precipitation, serving as the fundamental material basis for weather phenomena such as rainfall, and is a significant factor contributing to extreme weather events. The Weighted Mean Temperature (Tm) is a crucial factor in the calculation of P...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-02-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/16/3/278 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850204083130990592 |
|---|---|
| author | Shukai Wang Qiuying Guo Guihong Hua Yingjun Sun Wengang Sang Zhengyu Wang |
| author_facet | Shukai Wang Qiuying Guo Guihong Hua Yingjun Sun Wengang Sang Zhengyu Wang |
| author_sort | Shukai Wang |
| collection | DOAJ |
| description | Water vapor constitutes a vital component of atmospheric precipitation, serving as the fundamental material basis for weather phenomena such as rainfall, and is a significant factor contributing to extreme weather events. The Weighted Mean Temperature (Tm) is a crucial factor in the calculation of Precipitable Water Vapor (PWV) in the atmosphere, directly impacting the quality of GNSS-PWV inversion. The Tm<sub>N</sub>, Tm<sub>L1,</sub> and Tm<sub>L2</sub> models were constructed through regression analysis and LSTM based on data from the Zhangqiu Radiosonde Station in the Jinan region from 2020 to 2022, as well as ERA5 data. The six Tm models (Tm<sub>N</sub>, Tm<sub>L1</sub>, Tm<sub>L2</sub>, Bevis, GTm, and GPT3) were analyzed by comparing them with the Tm value from the Radiosonde station in 2023. Compared with the Bevis, GTm, and GPT3 models, the accuracy of Tm<sub>N</sub> was improved by 24%, 19%, and 45%, Tm<sub>L1</sub> was improved by 20%, 16%, and 42%, and Tm<sub>L2</sub> was increased by 34%, 29%, and 52%. The influence of the above six Tm models on GNSS-PWV accuracy was analyzed using both theoretical and experimental methods. It was demonstrated that the impact of Tm<sub>L1</sub> and Tm<sub>L2</sub> on the accuracy of the PWV solution is significantly enhanced in comparison with the other Tm models. The Tm<sub>L1</sub> and Tm<sub>L2</sub> models developed in this study offer enhanced accuracy for Tm data utilized in GNSS PWV inversion within the Jinan region. |
| format | Article |
| id | doaj-art-a075a3ed18274c7d86ce8d58f3ab67ae |
| institution | OA Journals |
| issn | 2073-4433 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj-art-a075a3ed18274c7d86ce8d58f3ab67ae2025-08-20T02:11:21ZengMDPI AGAtmosphere2073-44332025-02-0116327810.3390/atmos16030278Optimization and Construction of Jinan Regional Tm Model Based on LSTM and Analysis of Its Influence on the Accuracy of GNSS Inversion PWVShukai Wang0Qiuying Guo1Guihong Hua2Yingjun Sun3Wengang Sang4Zhengyu Wang5College of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250101, ChinaCollege of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250101, ChinaCollege of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250101, ChinaCollege of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250101, ChinaCollege of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250101, ChinaCollege of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250101, ChinaWater vapor constitutes a vital component of atmospheric precipitation, serving as the fundamental material basis for weather phenomena such as rainfall, and is a significant factor contributing to extreme weather events. The Weighted Mean Temperature (Tm) is a crucial factor in the calculation of Precipitable Water Vapor (PWV) in the atmosphere, directly impacting the quality of GNSS-PWV inversion. The Tm<sub>N</sub>, Tm<sub>L1,</sub> and Tm<sub>L2</sub> models were constructed through regression analysis and LSTM based on data from the Zhangqiu Radiosonde Station in the Jinan region from 2020 to 2022, as well as ERA5 data. The six Tm models (Tm<sub>N</sub>, Tm<sub>L1</sub>, Tm<sub>L2</sub>, Bevis, GTm, and GPT3) were analyzed by comparing them with the Tm value from the Radiosonde station in 2023. Compared with the Bevis, GTm, and GPT3 models, the accuracy of Tm<sub>N</sub> was improved by 24%, 19%, and 45%, Tm<sub>L1</sub> was improved by 20%, 16%, and 42%, and Tm<sub>L2</sub> was increased by 34%, 29%, and 52%. The influence of the above six Tm models on GNSS-PWV accuracy was analyzed using both theoretical and experimental methods. It was demonstrated that the impact of Tm<sub>L1</sub> and Tm<sub>L2</sub> on the accuracy of the PWV solution is significantly enhanced in comparison with the other Tm models. The Tm<sub>L1</sub> and Tm<sub>L2</sub> models developed in this study offer enhanced accuracy for Tm data utilized in GNSS PWV inversion within the Jinan region.https://www.mdpi.com/2073-4433/16/3/278perceptible water vapor (PWV)GNSS water vapor inversionatmospheric weighted mean temperature (Tm)long short-term memory network (LSTM) |
| spellingShingle | Shukai Wang Qiuying Guo Guihong Hua Yingjun Sun Wengang Sang Zhengyu Wang Optimization and Construction of Jinan Regional Tm Model Based on LSTM and Analysis of Its Influence on the Accuracy of GNSS Inversion PWV Atmosphere perceptible water vapor (PWV) GNSS water vapor inversion atmospheric weighted mean temperature (Tm) long short-term memory network (LSTM) |
| title | Optimization and Construction of Jinan Regional Tm Model Based on LSTM and Analysis of Its Influence on the Accuracy of GNSS Inversion PWV |
| title_full | Optimization and Construction of Jinan Regional Tm Model Based on LSTM and Analysis of Its Influence on the Accuracy of GNSS Inversion PWV |
| title_fullStr | Optimization and Construction of Jinan Regional Tm Model Based on LSTM and Analysis of Its Influence on the Accuracy of GNSS Inversion PWV |
| title_full_unstemmed | Optimization and Construction of Jinan Regional Tm Model Based on LSTM and Analysis of Its Influence on the Accuracy of GNSS Inversion PWV |
| title_short | Optimization and Construction of Jinan Regional Tm Model Based on LSTM and Analysis of Its Influence on the Accuracy of GNSS Inversion PWV |
| title_sort | optimization and construction of jinan regional tm model based on lstm and analysis of its influence on the accuracy of gnss inversion pwv |
| topic | perceptible water vapor (PWV) GNSS water vapor inversion atmospheric weighted mean temperature (Tm) long short-term memory network (LSTM) |
| url | https://www.mdpi.com/2073-4433/16/3/278 |
| work_keys_str_mv | AT shukaiwang optimizationandconstructionofjinanregionaltmmodelbasedonlstmandanalysisofitsinfluenceontheaccuracyofgnssinversionpwv AT qiuyingguo optimizationandconstructionofjinanregionaltmmodelbasedonlstmandanalysisofitsinfluenceontheaccuracyofgnssinversionpwv AT guihonghua optimizationandconstructionofjinanregionaltmmodelbasedonlstmandanalysisofitsinfluenceontheaccuracyofgnssinversionpwv AT yingjunsun optimizationandconstructionofjinanregionaltmmodelbasedonlstmandanalysisofitsinfluenceontheaccuracyofgnssinversionpwv AT wengangsang optimizationandconstructionofjinanregionaltmmodelbasedonlstmandanalysisofitsinfluenceontheaccuracyofgnssinversionpwv AT zhengyuwang optimizationandconstructionofjinanregionaltmmodelbasedonlstmandanalysisofitsinfluenceontheaccuracyofgnssinversionpwv |