Monitoring Soil Salinity in Arid Areas of Northern Xinjiang Using Multi-Source Satellite Data: A Trusted Deep Learning Framework

Land, Journal Year: 2025, Volume and Issue: 14(1), P. 110 - 110

Published: Jan. 8, 2025

Soil salinization affects agricultural productivity and ecosystem health in Xinjiang, especially arid areas. The region’s complex topography limited data emphasize the pressing need for effective, large-scale monitoring technologies. Therefore, 1044 soil samples were collected from farmland northern potential effectiveness of salinity was explored by combining environmental variables with Landsat 8 Sentinel-2. study applied four types feature selection algorithms: Random Forest (RF), Competitive Adaptive Reweighted Sampling (CARS), Uninformative Variable Elimination (UVE), Successive Projections Algorithm (SPA). These are then integrated into various machine learning models—such as Ensemble Tree (ETree), Extreme Gradient Boosting (XGBoost), LightBoost—as well deep models, including Convolutional Neural Networks (CNN), Residual (ResNet), Multilayer Perceptrons (MLP), Kolmogorov–Arnold (KAN), modeling. results suggest that fertilizer use plays a critical role processes. Notably, interpretable model KAN achieved an accuracy 0.75 correctly classifying degree salinity. This highlights integrating multi-source remote sensing technologies, offering pathway to monitoring, thereby providing valuable support management.

Language: Английский

---

Development of Novel Soil Salinity Spectral Index Using Remotely Sensed Data: A Case Study on Balod District, Chhattisgarh, India

Journal of Landscape Ecology, Journal Year: 2025, Volume and Issue: unknown

Published: March 28, 2025

Abstract Soil salinity is a known phenomenon worldwide. It has substantial influence on crop productivity and environmental well-being. Conventional approaches to evaluate soil are laborious expensive, which need efficient such as geospatial. Geospatial have led the development of several indices for estimation. Existing region specific not verified different regions. This study was conducted in Balod district Chhattisgarh, India. Landsat 9 imagery along with field electrical conductivity (EC) were used existing index develop new index. A multi-parameter recorder collect 69 EC samples April May 2024. Sixteen spectral evaluated verify applicability area. The results showed that had weak correlation values. Therefore, we developed by combining Near Infrared surface reflectance, redsurface Shortwave Infrared-1 reflectance bands using linear regression analysis.The classification categorize 78.40 % slightly saline, 16.50 moderately saline 1.46 strongly saline. demonstrates strong between values data an R 2 value 0.83 mean relative error 10 %. provides reliable geospatial approach evaluation sustainable land management techniques improve agricultural semi-arid, arid regions varying properties levels.

Language: Английский

---

Optimization of Multi-Source Remote Sensing Soil Salinity Estimation Based on Different Salinization Degrees

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(7), P. 1315 - 1315

Published: April 7, 2025

The timely and accurate monitoring of regional soil salinity is crucial for the sustainable development land stability ecological environment in arid semi-arid regions. However, due to spatiotemporal heterogeneity properties environmental conditions, improving accuracy salinization remains challenging. This study aimed explore whether partitioned modeling based on degrees during both bare vegetation cover periods can enhance prediction. Specifically, this integrated situ hyperspectral data satellite multispectral using spectral response functions. Subsequently, machine learning methods such as random forest (RF), extreme gradient boosting (XGBoost), support vector (SVM), multiple linear regression (MLR) were employed, combination with sensitive indices, develop a multi-source remote sensing estimation model optimized different (mild or lower vs. moderate higher salinization). performance approach was then compared an overall that does not distinguish between determine optimal strategy. results highlight effectiveness considering enhancing sensitivity indices accuracy. Classifying helps identify variable combinations are more construction content (SSC) models, positively impacting estimation. strategy outperformed stability, R2 values reaching 0.84 0.80 corresponding RMSE 0.1646% 0.1710% periods, respectively. proposes degrees, providing scientific evidence technical precise assessment effective management salinization.

Language: Английский

---

Advancing Multi-Scale Geographic Environmental Monitoring: A Synthesis of Cutting-Edge Research and Scalable Solutions

Land, Journal Year: 2025, Volume and Issue: 14(5), P. 1059 - 1059

Published: May 13, 2025

The geographic environment is a complex concept that encompasses various natural elements of the Earth’s surface and human activities [...]

Language: Английский

---