Transformation of Geospatial Modelling of Soil Erosion Susceptibility Using Machine Learning DOI
Muhammad Ramdhan Olii,

Sartan Nento,

Nurhayati Doda

и другие.

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Сен. 18, 2024

Abstract Soil erosion creates substantial environmental and economic challenges, especially in areas vulnerable to land degradation. This study investigates the use of machine learning (ML) techniques—namely Support Vector Machines (SVM) Generalized Linear Models (GLM)—for geospatial modeling soil susceptibility (SES). By leveraging data incorporating a range factors including hydrological, topographical, variables, research aims improve accuracy reliability SES predictions. Results show that SVM model predominantly identifies as having moderate (40.59%) or low (38.50%) susceptibility, whereas GLM allocates higher proportion very (24.55%) (38.59%) susceptibility. Both models exhibit high performance, with achieving accuracies 87.4% 87.2%, respectively, though slightly surpasses AUC (0.939 vs. 0.916). places greater emphasis on hydrological such distance rivers drainage density, while provides more balanced assessment across various variables. demonstrates ML-based can significantly enhance assessments, offering nuanced accurate approach than traditional methods. The findings highlight value adopting innovative, data-driven techniques offer practical insights for management conservation practices.

Язык: Английский

Improving the spatial prediction of topsoil properties in a typical grazing area using multi-season PlanetScope spectral covariates and data mining techniques DOI Creative Commons
Kwanele Phinzi, László Bertalan,

Gashaw Gismu Chakilu

и другие.

Earth Science Informatics, Год журнала: 2025, Номер 18(2)

Опубликована: Фев. 1, 2025

Abstract Understanding the spatial distribution of topsoil properties in grassland ecosystems is essential for improving soil ecosystem services, quality, and erosion resilience. The availability free, high-resolution satellite imagery advanced data mining techniques offers new opportunities efficient property assessment. This study aimed to evaluate potential utility multi-season PlanetScope predict organic carbon (SOC), pH, calcium carbonate (CaCO 3 ). Using random sampling, 121 samples (0–30 cm depth) were collected with an auger across grasslands, bare soil, eroded areas within a typical grazing land use. Three techniques: forest (RF), extreme gradient boosting (XGB), support vector machines (SVM), applied evaluated using 10-fold cross-validation. results indicated that spectral covariates considerably improved accuracy target compared single-season imagery. SVM was most effective algorithm predicting SOC, achieving root mean square error (RMSE) 0.52%, absolute (MAE) 0.24%, R² 0.92. RF best-performing pH (RMSE = 0.22, MAE 0.17, 0.97) CaCO 0.55%, 0.42%, 0.96). While XGB failed capture variability other models generated interpretable maps accurately represented different cover categories. green-red vegetation index (GRVI) critical covariate while elevation topographic wetness (TWI) key predictors , respectively. underscores recommends conducting similar studies diverse geographical settings validate these findings develop more generalizable models.

Язык: Английский

Процитировано

1
Integrating Remote Sensing, GIS, and AI Technologies in Soil Erosion Studies DOI Creative Commons
Salman Selmy, Dmitry E. Kucher, Ali RA Moursy

и другие.

IntechOpen eBooks, Год журнала: 2025, Номер unknown

Опубликована: Март 7, 2025

Soils are one of the most valuable non-renewable natural resources, and conserving them is critical for agricultural development ecological sustainability because they provide numerous ecosystem services. Soil erosion, a complex process caused by forces such as rainfall wind, poses significant challenges to ecosystems, agriculture, infrastructure, water quality, necessitating advanced monitoring modeling techniques. It has become global issue, threatening systems food security result climatic changes human activities. Traditional soil erosion field measurement methods have limitations in spatial temporal coverage. The integration new techniques remote sensing (RS), geographic information (GIS), artificial intelligence (AI) revolutionized our approach understanding managing erosion. RS technologies widely applicable investigations due their high efficiency, time savings, comprehensiveness. In recent years, advancements sensor technology resulted fine spatial-resolution images increased accuracy detection mapping purposes. Satellite imagery provides data on land cover properties, whereas digital elevation models (DEMs) detailed required assess slope flow accumulation, which important factors modeling. GIS enhances analysis integrating multiple datasets, making it easier identify hot spots utilizing like Revised Universal Loss Equation (RUSLE) estimate loss guide management decisions. Furthermore, AI techniques, particularly machine learning (ML) deep (DL), significantly improve predictions analyzing historical extracting relevant features from imagery. These use convolutional neural networks (CNNs) augmentation, well risk factors. Additionally, innovative methods, including biodegradable materials, hydroseeding, autonomous vehicles precision being developed prevent mitigate effectively. Although specific case studies demonstrate successful implementation this integrated framework variety landscapes, ongoing availability model validation must be addressed. Ultimately, collaboration RS, GIS, not only but also paves way effective control strategies, underscoring importance continued research vital area. This chapter addresses basic concerns related application erosion: concepts, acquisition, tools, types, management, visualization, an overview type its role

Язык: Английский

Процитировано

1
Pixel-scale gully erosion susceptibility: Predictive modeling with R using gully inventory consistent with terrain variables DOI Creative Commons
Christian Conoscenti, Grazia Azzara, Aleksey Y. Sheshukov

и другие.

CATENA, Год журнала: 2025, Номер 257, С. 109091 - 109091

Опубликована: Май 20, 2025

Язык: Английский

Процитировано

0
Transformation of Geospatial Modelling of Soil Erosion Susceptibility Using Machine Learning DOI
Muhammad Ramdhan Olii,

Sartan Nento,

Nurhayati Doda

и другие.

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Сен. 18, 2024

Abstract Soil erosion creates substantial environmental and economic challenges, especially in areas vulnerable to land degradation. This study investigates the use of machine learning (ML) techniques—namely Support Vector Machines (SVM) Generalized Linear Models (GLM)—for geospatial modeling soil susceptibility (SES). By leveraging data incorporating a range factors including hydrological, topographical, variables, research aims improve accuracy reliability SES predictions. Results show that SVM model predominantly identifies as having moderate (40.59%) or low (38.50%) susceptibility, whereas GLM allocates higher proportion very (24.55%) (38.59%) susceptibility. Both models exhibit high performance, with achieving accuracies 87.4% 87.2%, respectively, though slightly surpasses AUC (0.939 vs. 0.916). places greater emphasis on hydrological such distance rivers drainage density, while provides more balanced assessment across various variables. demonstrates ML-based can significantly enhance assessments, offering nuanced accurate approach than traditional methods. The findings highlight value adopting innovative, data-driven techniques offer practical insights for management conservation practices.

Язык: Английский

Процитировано

0