Scholarlink

<p>This study presents a novel, integrated framework that combines high-resolution morphometric analysis, Principal Component Analysis (PCA), and supervised Land Use/Land Cover (LULC) classification to identify erosion-prone sub-watersheds in the Macta Basin, northwestern Algeria (14,400&#xa0;km<sup>2</sup>). Morphometric parameters were derived from a 30&#xa0;m resolution Shuttle Radar Topography Mission (SRTM) Digital Elevation Model, preprocessed to ensure hydrological accuracy. Twenty-four initial metrics were reduced to ten dominant components using PCA, capturing nearly 100% of the basin’s geomorphological variability. LULC mapping was performed using Sentinel-2 imagery with a supervised Maximum Likelihood Classifier, achieving an overall accuracy of 68.48% and a Kappa coefficient of 0.638, ensuring reliability in semi-arid regions. The prioritization results indicate that high-risk sub-watersheds are concentrated in the central and eastern basin, with the Sefioun sub-catchment consistently identified as the most critical unit. These areas are characterized by steep slopes, dense drainage networks, and extensive agricultural and bare land cover. The framework provides a spatially explicit, reproducible tool for targeted soil conservation and watershed management in semi-arid Mediterranean environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Soil conservation prioritization analysis of Mediterranean Basins: an algerian case study

  • 期刊论文

  • 发表日期：

    2026年5月4日

查看全文
  • Khadidja Semari,
  • Khaled Korichi

摘要

This study presents a novel, integrated framework that combines high-resolution morphometric analysis, Principal Component Analysis (PCA), and supervised Land Use/Land Cover (LULC) classification to identify erosion-prone sub-watersheds in the Macta Basin, northwestern Algeria (14,400 km2). Morphometric parameters were derived from a 30 m resolution Shuttle Radar Topography Mission (SRTM) Digital Elevation Model, preprocessed to ensure hydrological accuracy. Twenty-four initial metrics were reduced to ten dominant components using PCA, capturing nearly 100% of the basin’s geomorphological variability. LULC mapping was performed using Sentinel-2 imagery with a supervised Maximum Likelihood Classifier, achieving an overall accuracy of 68.48% and a Kappa coefficient of 0.638, ensuring reliability in semi-arid regions. The prioritization results indicate that high-risk sub-watersheds are concentrated in the central and eastern basin, with the Sefioun sub-catchment consistently identified as the most critical unit. These areas are characterized by steep slopes, dense drainage networks, and extensive agricultural and bare land cover. The framework provides a spatially explicit, reproducible tool for targeted soil conservation and watershed management in semi-arid Mediterranean environments.