Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning Framework
Soil salinization affects agricultural productivity and ecosystem health in Xinjiang, especially in arid areas. The region’s complex topography and limited agricultural data emphasize the pressing need for effective, large-scale monitoring technologies. Therefore, 1044 soil samples were collected fr...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Land |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-445X/14/1/110 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588107300470784 |
|---|---|
| author | Mengli Zhang Xianglong Fan Pan Gao Li Guo Xuanrong Huang Xiuwen Gao Jinpeng Pang Fei Tan |
| author_facet | Mengli Zhang Xianglong Fan Pan Gao Li Guo Xuanrong Huang Xiuwen Gao Jinpeng Pang Fei Tan |
| author_sort | Mengli Zhang |
| collection | DOAJ |
| description | Soil salinization affects agricultural productivity and ecosystem health in Xinjiang, especially in arid areas. The region’s complex topography and limited agricultural data emphasize the pressing need for effective, large-scale monitoring technologies. Therefore, 1044 soil samples were collected from arid farmland in northern Xinjiang, and the potential effectiveness of soil salinity monitoring was explored by combining environmental variables with Landsat 8 and Sentinel-2. The study applied four types of feature selection algorithms: Random Forest (RF), Competitive Adaptive Reweighted Sampling (CARS), Uninformative Variable Elimination (UVE), and Successive Projections Algorithm (SPA). These variables are then integrated into various machine learning models—such as Ensemble Tree (ETree), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and LightBoost—as well as deep learning models, including Convolutional Neural Networks (CNN), Residual Networks (ResNet), Multilayer Perceptrons (MLP), and Kolmogorov–Arnold Networks (KAN), for modeling. The results suggest that fertilizer use plays a critical role in soil salinization processes. Notably, the interpretable model KAN achieved an accuracy of 0.75 in correctly classifying the degree of soil salinity. This study highlights the potential of integrating multi-source remote sensing data with deep learning technologies, offering a pathway to large-scale soil salinity monitoring, and thereby providing valuable support for soil management. |
| format | Article |
| id | doaj-art-08365d643f584c00a2ff7a0a615a525e |
| institution | Kabale University |
| issn | 2073-445X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Land |
| spelling | doaj-art-08365d643f584c00a2ff7a0a615a525e2025-01-24T13:37:56ZengMDPI AGLand2073-445X2025-01-0114111010.3390/land14010110Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning FrameworkMengli Zhang0Xianglong Fan1Pan Gao2Li Guo3Xuanrong Huang4Xiuwen Gao5Jinpeng Pang6Fei Tan7College of Information Science and Technology, Shihezi University, Shihezi 832061, ChinaAgricultural College, Shihezi University, Shihezi 832003, ChinaCollege of Information Science and Technology, Shihezi University, Shihezi 832061, ChinaCollege of Information Science and Technology, Shihezi University, Shihezi 832061, ChinaCollege of Information Science and Technology, Shihezi University, Shihezi 832061, ChinaCollege of Information Science and Technology, Shihezi University, Shihezi 832061, ChinaCollege of Information Science and Technology, Shihezi University, Shihezi 832061, ChinaCollege of Information Science and Technology, Shihezi University, Shihezi 832061, ChinaSoil salinization affects agricultural productivity and ecosystem health in Xinjiang, especially in arid areas. The region’s complex topography and limited agricultural data emphasize the pressing need for effective, large-scale monitoring technologies. Therefore, 1044 soil samples were collected from arid farmland in northern Xinjiang, and the potential effectiveness of soil salinity monitoring was explored by combining environmental variables with Landsat 8 and Sentinel-2. The study applied four types of feature selection algorithms: Random Forest (RF), Competitive Adaptive Reweighted Sampling (CARS), Uninformative Variable Elimination (UVE), and Successive Projections Algorithm (SPA). These variables are then integrated into various machine learning models—such as Ensemble Tree (ETree), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and LightBoost—as well as deep learning models, including Convolutional Neural Networks (CNN), Residual Networks (ResNet), Multilayer Perceptrons (MLP), and Kolmogorov–Arnold Networks (KAN), for modeling. The results suggest that fertilizer use plays a critical role in soil salinization processes. Notably, the interpretable model KAN achieved an accuracy of 0.75 in correctly classifying the degree of soil salinity. This study highlights the potential of integrating multi-source remote sensing data with deep learning technologies, offering a pathway to large-scale soil salinity monitoring, and thereby providing valuable support for soil management.https://www.mdpi.com/2073-445X/14/1/110neural networkmulti-source satellite datainterpretable deep learningGoogle Earth Engine |
| spellingShingle | Mengli Zhang Xianglong Fan Pan Gao Li Guo Xuanrong Huang Xiuwen Gao Jinpeng Pang Fei Tan Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning Framework Land neural network multi-source satellite data interpretable deep learning Google Earth Engine |
| title | Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning Framework |
| title_full | Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning Framework |
| title_fullStr | Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning Framework |
| title_full_unstemmed | Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning Framework |
| title_short | Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning Framework |
| title_sort | monitoring soil salinity in arid areas of northern xinjiang using multi source satellite data a trusted deep learning framework |
| topic | neural network multi-source satellite data interpretable deep learning Google Earth Engine |
| url | https://www.mdpi.com/2073-445X/14/1/110 |
| work_keys_str_mv | AT menglizhang monitoringsoilsalinityinaridareasofnorthernxinjiangusingmultisourcesatellitedataatrusteddeeplearningframework AT xianglongfan monitoringsoilsalinityinaridareasofnorthernxinjiangusingmultisourcesatellitedataatrusteddeeplearningframework AT pangao monitoringsoilsalinityinaridareasofnorthernxinjiangusingmultisourcesatellitedataatrusteddeeplearningframework AT liguo monitoringsoilsalinityinaridareasofnorthernxinjiangusingmultisourcesatellitedataatrusteddeeplearningframework AT xuanronghuang monitoringsoilsalinityinaridareasofnorthernxinjiangusingmultisourcesatellitedataatrusteddeeplearningframework AT xiuwengao monitoringsoilsalinityinaridareasofnorthernxinjiangusingmultisourcesatellitedataatrusteddeeplearningframework AT jinpengpang monitoringsoilsalinityinaridareasofnorthernxinjiangusingmultisourcesatellitedataatrusteddeeplearningframework AT feitan monitoringsoilsalinityinaridareasofnorthernxinjiangusingmultisourcesatellitedataatrusteddeeplearningframework |