Estimation of Soil Organic Carbon Content in Coastal Wetlands with Measured VIS-NIR Spectroscopy Using Optimized Support Vector Machines and Random Forests
Coastal wetland soil organic carbon (CW-SOC) is crucial for both “blue carbon” and carbon sequestration. It is of great significance to understand the content of soil organic carbon (SOC) in soil resource management. A total of 133 soil samples were evaluated using an indoor spectral curve and were...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/17/4372 |
| _version_ | 1827665067452137472 |
|---|---|
| author | Jingru Song Junhai Gao Yongbin Zhang Fuping Li Weidong Man Mingyue Liu Jinhua Wang Mengqian Li Hao Zheng Xiaowu Yang Chunjing Li |
| author_facet | Jingru Song Junhai Gao Yongbin Zhang Fuping Li Weidong Man Mingyue Liu Jinhua Wang Mengqian Li Hao Zheng Xiaowu Yang Chunjing Li |
| author_sort | Jingru Song |
| collection | DOAJ |
| description | Coastal wetland soil organic carbon (CW-SOC) is crucial for both “blue carbon” and carbon sequestration. It is of great significance to understand the content of soil organic carbon (SOC) in soil resource management. A total of 133 soil samples were evaluated using an indoor spectral curve and were categorized into silty soil and sandy soil. The prediction model of CW-SOC was established using optimized support vector machine regression (OSVR) and optimized random forest regression (ORFR). The Leave-One-Out Cross-Validation (LOO-CV) method was used to verify the model, and the performance of the two prediction models, as well as the models’ stability and uncertainty, was examined. The results show that (1) The SOC content of different coastal wetlands is significantly different, and the SOC content of silty soils is about 1.8 times that of sandy soils. Moreover, the characteristic wavelengths associated with SOC in silty soils are mainly concentrated in the spectral range of 500–1000 nm and 1900–2400 nm, while the spectral range of sandy soils is concentrated in the spectral range of 600–1400 nm and 1700–2400 nm. (2) The organic carbon prediction model of silty soil based on the OSVR method under the first-order differential of reflectance (R′) is the best, with the <i>Adjusted-R<sup>2</sup></i> value as high as 0.78, the <i>RPD</i> value is much greater than 2.0 and 5.07, and the <i>RMSE</i> value as low as 0.07. (3) The performance of the OSVR model is about 15~30% higher than that of the support vector machine regression (SVR) model, and the performance of the ORFR model is about 3~5% higher than that of the random forest regression (RFR) model. OSVR and ORFR are better methods of accurately predicting the CW-SOC content and provide data support for the carbon cycle, soil conservation, plant growth, and environmental protection of coastal wetlands. |
| first_indexed | 2024-03-10T01:18:20Z |
| format | Article |
| id | doaj.art-add8f2286998482384c78eb7bd676b47 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T01:18:20Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-add8f2286998482384c78eb7bd676b472023-11-23T14:05:32ZengMDPI AGRemote Sensing2072-42922022-09-011417437210.3390/rs14174372Estimation of Soil Organic Carbon Content in Coastal Wetlands with Measured VIS-NIR Spectroscopy Using Optimized Support Vector Machines and Random ForestsJingru Song0Junhai Gao1Yongbin Zhang2Fuping Li3Weidong Man4Mingyue Liu5Jinhua Wang6Mengqian Li7Hao Zheng8Xiaowu Yang9Chunjing Li10College of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaTangshan Branch, CCTEG Ecological Environment Technology Co., Ltd., Tangshan 063012, ChinaCollege of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaCollege of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaCollege of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaCollege of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaCollege of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaCollege of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaCollege of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaCollege of Mining Engineering, North China University of Science and Technology, Tangshan 063210, ChinaCollege of Geography and Ocean Sciences, Yanbian University, Yanji 133000, ChinaCoastal wetland soil organic carbon (CW-SOC) is crucial for both “blue carbon” and carbon sequestration. It is of great significance to understand the content of soil organic carbon (SOC) in soil resource management. A total of 133 soil samples were evaluated using an indoor spectral curve and were categorized into silty soil and sandy soil. The prediction model of CW-SOC was established using optimized support vector machine regression (OSVR) and optimized random forest regression (ORFR). The Leave-One-Out Cross-Validation (LOO-CV) method was used to verify the model, and the performance of the two prediction models, as well as the models’ stability and uncertainty, was examined. The results show that (1) The SOC content of different coastal wetlands is significantly different, and the SOC content of silty soils is about 1.8 times that of sandy soils. Moreover, the characteristic wavelengths associated with SOC in silty soils are mainly concentrated in the spectral range of 500–1000 nm and 1900–2400 nm, while the spectral range of sandy soils is concentrated in the spectral range of 600–1400 nm and 1700–2400 nm. (2) The organic carbon prediction model of silty soil based on the OSVR method under the first-order differential of reflectance (R′) is the best, with the <i>Adjusted-R<sup>2</sup></i> value as high as 0.78, the <i>RPD</i> value is much greater than 2.0 and 5.07, and the <i>RMSE</i> value as low as 0.07. (3) The performance of the OSVR model is about 15~30% higher than that of the support vector machine regression (SVR) model, and the performance of the ORFR model is about 3~5% higher than that of the random forest regression (RFR) model. OSVR and ORFR are better methods of accurately predicting the CW-SOC content and provide data support for the carbon cycle, soil conservation, plant growth, and environmental protection of coastal wetlands.https://www.mdpi.com/2072-4292/14/17/4372optimized support vector machine regressionoptimized random forest regressioncoastal wetland soil organic carbonspectrumprediction model |
| spellingShingle | Jingru Song Junhai Gao Yongbin Zhang Fuping Li Weidong Man Mingyue Liu Jinhua Wang Mengqian Li Hao Zheng Xiaowu Yang Chunjing Li Estimation of Soil Organic Carbon Content in Coastal Wetlands with Measured VIS-NIR Spectroscopy Using Optimized Support Vector Machines and Random Forests Remote Sensing optimized support vector machine regression optimized random forest regression coastal wetland soil organic carbon spectrum prediction model |
| title | Estimation of Soil Organic Carbon Content in Coastal Wetlands with Measured VIS-NIR Spectroscopy Using Optimized Support Vector Machines and Random Forests |
| title_full | Estimation of Soil Organic Carbon Content in Coastal Wetlands with Measured VIS-NIR Spectroscopy Using Optimized Support Vector Machines and Random Forests |
| title_fullStr | Estimation of Soil Organic Carbon Content in Coastal Wetlands with Measured VIS-NIR Spectroscopy Using Optimized Support Vector Machines and Random Forests |
| title_full_unstemmed | Estimation of Soil Organic Carbon Content in Coastal Wetlands with Measured VIS-NIR Spectroscopy Using Optimized Support Vector Machines and Random Forests |
| title_short | Estimation of Soil Organic Carbon Content in Coastal Wetlands with Measured VIS-NIR Spectroscopy Using Optimized Support Vector Machines and Random Forests |
| title_sort | estimation of soil organic carbon content in coastal wetlands with measured vis nir spectroscopy using optimized support vector machines and random forests |
| topic | optimized support vector machine regression optimized random forest regression coastal wetland soil organic carbon spectrum prediction model |
| url | https://www.mdpi.com/2072-4292/14/17/4372 |
| work_keys_str_mv | AT jingrusong estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT junhaigao estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT yongbinzhang estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT fupingli estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT weidongman estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT mingyueliu estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT jinhuawang estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT mengqianli estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT haozheng estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT xiaowuyang estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests AT chunjingli estimationofsoilorganiccarboncontentincoastalwetlandswithmeasuredvisnirspectroscopyusingoptimizedsupportvectormachinesandrandomforests |