Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification Approach
Global navigation satellite system-reflectometry (GNSS-R) can retrieve Earth's surface parameters, such as soil moisture (SM) using the reflected signals from GNSS constellations with advantages of noncontact, all-weather, real-time, and continuity, particularly the space-borne cyclone GN...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9419719/ |
| _version_ | 1819002677959327744 |
|---|---|
| author | Yan Jia Shuanggen Jin Haolin Chen Qingyun Yan Patrizia Savi Yan Jin Yuan Yuan |
| author_facet | Yan Jia Shuanggen Jin Haolin Chen Qingyun Yan Patrizia Savi Yan Jin Yuan Yuan |
| author_sort | Yan Jia |
| collection | DOAJ |
| description | Global navigation satellite system-reflectometry (GNSS-R) can retrieve Earth's surface parameters, such as soil moisture (SM) using the reflected signals from GNSS constellations with advantages of noncontact, all-weather, real-time, and continuity, particularly the space-borne cyclone GNSS (CYGNSS) mission. However, the accuracy and efficiency of SM estimation from CYGNSS still need to improve. In this article, the global SM is estimated using machine learning (ML) regression aided by a preclassification strategy. The total observations are classified by land types and corresponding subsets are built for constructing ML regression submodels. Ten-fold cross-validation technique is adopted. The overall performance of SM estimation with/without preclassification is compared, and the results show that the SM estimations using different ML algorithms all have substantial improvement with the preclassification strategy. Then, the optimal XGBoost predicted model with root-mean-square error (RMSE) of 0.052 cm<sup>3</sup>/cm<sup>3</sup> is adopted. In addition, the satisfactory daily and seasonal SM prediction outcomes with an overall correlation coefficient value of 0.86 and an RMSE value of 0.056 cm<sup>3</sup>/cm<sup>3</sup> are achieved at a global scale, respectively. Furthermore, the extensive temporal and spatial variations of CYGNSS SM predictions are evaluated. It shows that the reflectivity plays a main role among the predictors in SM estimation, and the next is vegetation. In some extremely dry places, the roughness may become more important. The value of SM is positively correlated with RMSE and also another limit condition that will constrain the variation of predictors, thus affecting correlation coefficient R and RMSE. Also, we compare both SMAP and CYGNSS SM predictions against <italic>in situ</italic> SM measurements from 301 stations. Similar low-median unbiased RMSEs are obtained, and the daily averaged CYGNSS-based SM against the <italic>in situ</italic> networks is 0.049 cm<sup>3</sup>/cm<sup>3</sup>. The presented approach succeeds in providing SM estimation at a global scale with employing the least ancillary data with superior results and this article reveals the spatio-temporal heterogeneity for SM estimation using CYGNSS data. |
| first_indexed | 2024-12-20T23:08:55Z |
| format | Article |
| id | doaj.art-2e642fcfcf954df1b4ebd31b03e81dcd |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-20T23:08:55Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-2e642fcfcf954df1b4ebd31b03e81dcd2022-12-21T19:23:47ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01144879489310.1109/JSTARS.2021.30764709419719Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification ApproachYan Jia0https://orcid.org/0000-0002-8282-8105Shuanggen Jin1Haolin Chen2Qingyun Yan3https://orcid.org/0000-0001-6693-957XPatrizia Savi4https://orcid.org/0000-0001-9585-310XYan Jin5https://orcid.org/0000-0002-6978-163XYuan Yuan6https://orcid.org/0000-0003-1860-3275Department of Surveying and Geoinformatics, Nanjing University of Posts and Telecommunications, Nanjing, ChinaSchool of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science and Technology, Nanjing, ChinaDepartment of Surveying and Geoinformatics, Nanjing University of Posts and Telecommunications, Nanjing, ChinaSchool of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science and Technology, Nanjing, ChinaDepartment of Electronic and Telecommunication, Politecnico di Torino, Torino, ItalyDepartment of Surveying and Geoinformatics, Nanjing University of Posts and Telecommunications, Nanjing, ChinaDepartment of Surveying and Geoinformatics, Nanjing University of Posts and Telecommunications, Nanjing, ChinaGlobal navigation satellite system-reflectometry (GNSS-R) can retrieve Earth's surface parameters, such as soil moisture (SM) using the reflected signals from GNSS constellations with advantages of noncontact, all-weather, real-time, and continuity, particularly the space-borne cyclone GNSS (CYGNSS) mission. However, the accuracy and efficiency of SM estimation from CYGNSS still need to improve. In this article, the global SM is estimated using machine learning (ML) regression aided by a preclassification strategy. The total observations are classified by land types and corresponding subsets are built for constructing ML regression submodels. Ten-fold cross-validation technique is adopted. The overall performance of SM estimation with/without preclassification is compared, and the results show that the SM estimations using different ML algorithms all have substantial improvement with the preclassification strategy. Then, the optimal XGBoost predicted model with root-mean-square error (RMSE) of 0.052 cm<sup>3</sup>/cm<sup>3</sup> is adopted. In addition, the satisfactory daily and seasonal SM prediction outcomes with an overall correlation coefficient value of 0.86 and an RMSE value of 0.056 cm<sup>3</sup>/cm<sup>3</sup> are achieved at a global scale, respectively. Furthermore, the extensive temporal and spatial variations of CYGNSS SM predictions are evaluated. It shows that the reflectivity plays a main role among the predictors in SM estimation, and the next is vegetation. In some extremely dry places, the roughness may become more important. The value of SM is positively correlated with RMSE and also another limit condition that will constrain the variation of predictors, thus affecting correlation coefficient R and RMSE. Also, we compare both SMAP and CYGNSS SM predictions against <italic>in situ</italic> SM measurements from 301 stations. Similar low-median unbiased RMSEs are obtained, and the daily averaged CYGNSS-based SM against the <italic>in situ</italic> networks is 0.049 cm<sup>3</sup>/cm<sup>3</sup>. The presented approach succeeds in providing SM estimation at a global scale with employing the least ancillary data with superior results and this article reveals the spatio-temporal heterogeneity for SM estimation using CYGNSS data.https://ieeexplore.ieee.org/document/9419719/CYGNSSGNSS-Reflectometrypreclassifica- tionSMAPsoil moistureXGBoost |
| spellingShingle | Yan Jia Shuanggen Jin Haolin Chen Qingyun Yan Patrizia Savi Yan Jin Yuan Yuan Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification Approach IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing CYGNSS GNSS-Reflectometry preclassifica- tion SMAP soil moisture XGBoost |
| title | Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification Approach |
| title_full | Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification Approach |
| title_fullStr | Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification Approach |
| title_full_unstemmed | Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification Approach |
| title_short | Temporal-Spatial Soil Moisture Estimation from CYGNSS Using Machine Learning Regression With a Preclassification Approach |
| title_sort | temporal spatial soil moisture estimation from cygnss using machine learning regression with a preclassification approach |
| topic | CYGNSS GNSS-Reflectometry preclassifica- tion SMAP soil moisture XGBoost |
| url | https://ieeexplore.ieee.org/document/9419719/ |
| work_keys_str_mv | AT yanjia temporalspatialsoilmoistureestimationfromcygnssusingmachinelearningregressionwithapreclassificationapproach AT shuanggenjin temporalspatialsoilmoistureestimationfromcygnssusingmachinelearningregressionwithapreclassificationapproach AT haolinchen temporalspatialsoilmoistureestimationfromcygnssusingmachinelearningregressionwithapreclassificationapproach AT qingyunyan temporalspatialsoilmoistureestimationfromcygnssusingmachinelearningregressionwithapreclassificationapproach AT patriziasavi temporalspatialsoilmoistureestimationfromcygnssusingmachinelearningregressionwithapreclassificationapproach AT yanjin temporalspatialsoilmoistureestimationfromcygnssusingmachinelearningregressionwithapreclassificationapproach AT yuanyuan temporalspatialsoilmoistureestimationfromcygnssusingmachinelearningregressionwithapreclassificationapproach |