Three-dimensional soil salinity mapping with uncertainty using Bayesian Hierarchical Modelling, Random Forest Regression and remote sensing data

Agricultural Water Management, Год журнала: 2025, Номер 309, С. 109318 - 109318

Опубликована: Янв. 27, 2025

Язык: Английский

---

A Specific Time Lag Regulation of Soil Moisture Across Layers on Soil Salinization in the Northeast Tibetan Plateau Agroecosystem

Agriculture, Год журнала: 2025, Номер 15(1), С. 106 - 106

Опубликована: Янв. 5, 2025

The evaporation of soil water drives the upward movement salt and its accumulation on surface, which ultimately leads to salinization in agroecosystems. With rapid development remote sensing technology, transport can be monitored accurately. Based Landsat 8 satellite imagery ERA5-Land reanalysis datasets, this study explored variation characteristics northeast Tibetan Plateau from 2013 2023, inferred by geostatistical methods like ridge regression, windowed cross correlation, machine learning algorithms. results show that negative correlation effect between deep moisture (100–289 cm) is stronger. Moreover, also have a time lag compared with instant responses, meaning caused may require longer times. As potential driving factors, an increase organic carbon runoff beneficial for alleviating while abundant promotes humidification. This has elucidated specific regulation within different profiles, understanding ecological balance

Язык: Английский

---

Estimating the groundwater table threshold for mitigating soil salinization in the Songnen Plain of China

Journal of Hydrology Regional Studies, Год журнала: 2025, Номер 59, С. 102326 - 102326

Опубликована: Март 25, 2025

Язык: Английский

---

Monitoring of Soil Salinization and Analysis of Driving Factors in the Oasis Zone of South Xinjiang

Land, Год журнала: 2025, Номер 14(4), С. 803 - 803

Опубликована: Апрель 8, 2025

Soil salinization significantly jeopardizes agricultural productivity and ecological stability in southern Xinjiang’s oasis regions, highlighting the urgent need to examine its spatial–temporal trends driving mechanisms for improved resource management. Utilizing soil salinity measurements collected 2010 2023, current research applied multiple environmental variables processed via Google Earth Engine (GEE) platform evaluate predictive capability of four machine learning algorithms—random forest (RF), Gradient Boosting Decision Tree (GBDT), Classification Regression (CART), Support Vector Machine (SVM)—for accurate large-scale mapping. Subsequently, a piecewise structural equation model (piecewiseSEM) was employed quantitatively analyze factors salinization. Correlation analysis revealed seven critical variables—Red, NDSI, kNDVI, SDI, ET, elevation, SM—as most influential among 41 assessed their impact on salinity. The performance evaluation ranked models as follows: RF > GBDT SVM CART, with achieving highest accuracy (R2 = 0.756, RMSE 2.265 g·kg−1, MAE 1.468 g·kg−1). Between severity region exhibited slight overall decrease; however, extent this reduction relatively modest. proportion moderately severely salinized areas declined, accompanied by reduced spatial variability, whereas mildly soils increased markedly. These findings imply that primarily experiences internal redistribution within surface layers, limited downward leaching. Evapotranspiration (ET) moisture (SM) were identified dominant drivers affecting dynamics during both periods, influence SM becoming more pronounced over time. This trend highlights conditions natural human-induced irrigation practices have emerged primary regulator levels. study provide novel methodologies data support monitoring prevention arid regions.

Язык: Английский

---

Mapping Soil Cadmium Content Using Multi-Spectral Satellite Images and Multiple-Residual-Stacking Model: Incorporating Information from Homologous Pollution and Spectrally Active Materials

Journal of Hazardous Materials, Год журнала: 2024, Номер 485, С. 136755 - 136755

Опубликована: Дек. 6, 2024

Язык: Английский

---