A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial Intelligence
Retrieval of integrated water vapor content (WVC) from remote sensing data is often ill-posed because of insufficient observational information. There are many factors that cause WVC changes, which yield instability in the accuracy of many traditional algorithms. To overcome this problem, we develop...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-08-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/15/17/4250 |
_version_ | 1797581897417621504 |
---|---|
author | Ruyu Mei Kebiao Mao Jiancheng Shi Jeffrey Nielson Sayed M. Bateni Fei Meng Guoming Du |
author_facet | Ruyu Mei Kebiao Mao Jiancheng Shi Jeffrey Nielson Sayed M. Bateni Fei Meng Guoming Du |
author_sort | Ruyu Mei |
collection | DOAJ |
description | Retrieval of integrated water vapor content (WVC) from remote sensing data is often ill-posed because of insufficient observational information. There are many factors that cause WVC changes, which yield instability in the accuracy of many traditional algorithms. To overcome this problem, we developed a novel fully-coupled paradigm for the robust retrieval of WVC from thermal infrared remote sensing data. Through the derivation of the physical radiative transfer equation, we determined two conditions that need to be satisfied for the deep learning retrieval paradigm of WVC. The first condition is that the input parameters and output parameters of the deep learning need to be able to build a complete set of solvable equations in theory. The second condition is that, if there is a strong relationship between input parameters and output parameters, it can be directly retrieved. If it is a weak relationship, we need to use prior knowledge to improve the portability and accuracy of the algorithm. The training and test data of deep learning is composed of representative solutions of physical methods and solutions of statistical methods. The representative solutions of the physical methods were obtained from the physical forward model, and the statistical solutions were obtained from multi-source data which can compensate for the defect that the physical model cannot simulate mixed pixels. MODIS L1B data was used for case analysis of paradigm retrieval, and the analysis indicated that four thermal infrared bands were usually needed as the input parameters of deep learning and the integrated water vapor content as the output parameter. When land surface temperature and emissivity were taken as prior knowledge, the root-mean-square error (RMSE) of the retrieved WVC was 0.07 g/cm<sup>2</sup>. The optimal accuracy RMSE was 0.27 g/cm<sup>2</sup>. When there was a strong correlation between input parameters and output parameters, i.e., if there were two bands that were very sensitive to WVC in the band combination, high-precision retrieval could also be achieved without prior knowledge. All the analyses show that the paradigm of deep learning coupling physics and statistics can accurately retrieve WVC, which is a significant improvement on the traditional method and solves the problem of lack of physical interpretation of deep learning. |
first_indexed | 2024-03-10T23:14:10Z |
format | Article |
id | doaj.art-36b899ba077749a5a9afa6192490e8b6 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-10T23:14:10Z |
publishDate | 2023-08-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-36b899ba077749a5a9afa6192490e8b62023-11-19T08:46:34ZengMDPI AGRemote Sensing2072-42922023-08-011517425010.3390/rs15174250A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial IntelligenceRuyu Mei0Kebiao Mao1Jiancheng Shi2Jeffrey Nielson3Sayed M. Bateni4Fei Meng5Guoming Du6School of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250100, ChinaInstitute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, ChinaNational Space Science Center, Chinese Academy of Sciences, Beijing 100190, ChinaDepartment of Watershed Sciences, Utah State University, Logan, UT 84322, USADepartment of Civil and Environmental Engineering and Water Resources Research Center, University of Hawaii at Manoa, Honolulu, HI 96822, USASchool of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan 250100, ChinaSchool of Public Administration and Law, Northeast Agricultural University, Harbin 150006, ChinaRetrieval of integrated water vapor content (WVC) from remote sensing data is often ill-posed because of insufficient observational information. There are many factors that cause WVC changes, which yield instability in the accuracy of many traditional algorithms. To overcome this problem, we developed a novel fully-coupled paradigm for the robust retrieval of WVC from thermal infrared remote sensing data. Through the derivation of the physical radiative transfer equation, we determined two conditions that need to be satisfied for the deep learning retrieval paradigm of WVC. The first condition is that the input parameters and output parameters of the deep learning need to be able to build a complete set of solvable equations in theory. The second condition is that, if there is a strong relationship between input parameters and output parameters, it can be directly retrieved. If it is a weak relationship, we need to use prior knowledge to improve the portability and accuracy of the algorithm. The training and test data of deep learning is composed of representative solutions of physical methods and solutions of statistical methods. The representative solutions of the physical methods were obtained from the physical forward model, and the statistical solutions were obtained from multi-source data which can compensate for the defect that the physical model cannot simulate mixed pixels. MODIS L1B data was used for case analysis of paradigm retrieval, and the analysis indicated that four thermal infrared bands were usually needed as the input parameters of deep learning and the integrated water vapor content as the output parameter. When land surface temperature and emissivity were taken as prior knowledge, the root-mean-square error (RMSE) of the retrieved WVC was 0.07 g/cm<sup>2</sup>. The optimal accuracy RMSE was 0.27 g/cm<sup>2</sup>. When there was a strong correlation between input parameters and output parameters, i.e., if there were two bands that were very sensitive to WVC in the band combination, high-precision retrieval could also be achieved without prior knowledge. All the analyses show that the paradigm of deep learning coupling physics and statistics can accurately retrieve WVC, which is a significant improvement on the traditional method and solves the problem of lack of physical interpretation of deep learning.https://www.mdpi.com/2072-4292/15/17/4250integrated water vapor contentradiative transferdeep learningprior knowledge |
spellingShingle | Ruyu Mei Kebiao Mao Jiancheng Shi Jeffrey Nielson Sayed M. Bateni Fei Meng Guoming Du A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial Intelligence Remote Sensing integrated water vapor content radiative transfer deep learning prior knowledge |
title | A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial Intelligence |
title_full | A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial Intelligence |
title_fullStr | A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial Intelligence |
title_full_unstemmed | A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial Intelligence |
title_short | A Novel Physics-Statistical Coupled Paradigm for Retrieving Integrated Water Vapor Content Based on Artificial Intelligence |
title_sort | novel physics statistical coupled paradigm for retrieving integrated water vapor content based on artificial intelligence |
topic | integrated water vapor content radiative transfer deep learning prior knowledge |
url | https://www.mdpi.com/2072-4292/15/17/4250 |
work_keys_str_mv | AT ruyumei anovelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT kebiaomao anovelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT jianchengshi anovelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT jeffreynielson anovelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT sayedmbateni anovelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT feimeng anovelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT guomingdu anovelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT ruyumei novelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT kebiaomao novelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT jianchengshi novelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT jeffreynielson novelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT sayedmbateni novelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT feimeng novelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence AT guomingdu novelphysicsstatisticalcoupledparadigmforretrievingintegratedwatervaporcontentbasedonartificialintelligence |