Understanding the factors influencing soil erosion using data science and remote sensing
Understanding the factors influencing soil erosion using data science and remote sensing
One of the significant challenges facing modern Algerian society is the siltation of dams, a strategic issue for the sustainable water supply, primarily caused by soil erosion. This study investigates the factors influencing soil erosion by leveraging geographical information systems (GIS), data science, and machine learning techniques. By analyzing extensive datasets on climate, topography, and land use, the research identifies key predictors of erosion. Interpretable machine learning models, including Random Forest and Logistic Regression algorithms, are employed to model erosion patterns. The findings reveal that slope aspect, NDVI (Normalized Difference Vegetation Index), and aspect are the most significant factors influencing erosion. Additionally, a counterintuitive result shows that areas with lower precipitation areas are more affected by erosion, likely due to less cohesive soils and sparse vegetation. These insights are intended to enhance soil conservation strategies and promote sustainable land management practices, offering valuable guidance for mitigating the impact of soil erosion on water resources.
Djoumana Sakhri; Zakaria Matougui; A. Medjnoun; Ramdane Bahar
18th African Regional Conference on Soil Mechanics and Geotechnical Engineering (ARCSMGE2024)
Environmental geotechnics
Contact Information
- Secretary General - Dr A M McNamara
- Email: [email protected]
- Phone: +44 20 7040 8154
- City, University of London
- Northampton Square
- London EC1V 0HB. United Kingdom