Scholarlink

To develop a comprehensive watershed management plan, it is essential to have a deep understanding of the geographic features and geomorphological conditions within the watershed. This research concentrates on assessing the priority of the Panam watershed concerning soil erosion. It achieves this by utilizing morphometric characteristics and three Topo-hydrological factors, employing Remote Sensing (RS) and Geographical Information Systems (GIS) techniques. The application of a 30-m resolution Shuttle Radar Topography Mission Digital Elevation Model (SRTM DEM) allows for the delineation of sub-watersheds and numerous parameters. Our prioritization approach combines Weighted Sum Analysis (WSA) and Principal Component Analysis (PCA) into a comprehensive methodology. The two machine learning models were employed, specifically the multilayer perceptron (MLP) and the support vector machine (SVM), to predict Compound Factor (CF) values. The performance evaluation of these machine learning models relies on three fundamental metrics: mean absolute percentage error (MAPE), Pearson correlation coefficient (PCC), and root mean square error (RMSE). It is worth noting that the combination of WSA and MLP produces the highest achievable PCC value. Given the outstanding performance of WSA, CF values are calculated and applied to delineate watershed boundaries. The proposed technique introduced in this study provides a reliable approach for forecasting CF values and offers a valuable means of defining suitable watershed boundaries. As a result, this methodology aids in facilitating well-informed decision-making processes and the effective implementation of conservation measures related to soil and water management.

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

Integrating Morphometric Analysis, Prioritization Strategies, and Machine Learning for Enhanced Watershed Management

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Akil V. Memon,
  • Nirav V. Shah,
  • Dharam N. Thakkar,
  • Yogesh S. Patel

摘要

To develop a comprehensive watershed management plan, it is essential to have a deep understanding of the geographic features and geomorphological conditions within the watershed. This research concentrates on assessing the priority of the Panam watershed concerning soil erosion. It achieves this by utilizing morphometric characteristics and three Topo-hydrological factors, employing Remote Sensing (RS) and Geographical Information Systems (GIS) techniques. The application of a 30-m resolution Shuttle Radar Topography Mission Digital Elevation Model (SRTM DEM) allows for the delineation of sub-watersheds and numerous parameters. Our prioritization approach combines Weighted Sum Analysis (WSA) and Principal Component Analysis (PCA) into a comprehensive methodology. The two machine learning models were employed, specifically the multilayer perceptron (MLP) and the support vector machine (SVM), to predict Compound Factor (CF) values. The performance evaluation of these machine learning models relies on three fundamental metrics: mean absolute percentage error (MAPE), Pearson correlation coefficient (PCC), and root mean square error (RMSE). It is worth noting that the combination of WSA and MLP produces the highest achievable PCC value. Given the outstanding performance of WSA, CF values are calculated and applied to delineate watershed boundaries. The proposed technique introduced in this study provides a reliable approach for forecasting CF values and offers a valuable means of defining suitable watershed boundaries. As a result, this methodology aids in facilitating well-informed decision-making processes and the effective implementation of conservation measures related to soil and water management.