Scholarlink

Management of the resources of groundwater has become increasingly significant as a consequence of the complex interplay of climatological, geological, and anthropogenic factors, necessitating cutting-edge methods like Geographic Information Systems (GIS) and RS (Remote Sensing). This study integrates methodologies from RS, GIS, Multi-Influence Factor (MIF), and Analytic Hierarchy Process (AHP) to identify groundwater potential zones (GWPZs). Eight thematic layers-soil, slope, rainfall, lineament density (LD), drainage density (DD), geology, geomorphology (GM), and land use and land cover (LULC)-were utilized in this analysis. Each thematic layer was assigned weights based on its relevance to groundwater occurrence. Both MIF and AHP techniques employed quantile categorization to classify GWPZs into five categories: very poor, poor, moderate, good, and very good. The analysis revealed that approximately 42.47% of the study area exhibits good GWPZs, while 45.93% shows very good GWPZs. Groundwater levels were used to validate the maps through the Receiver Operating Characteristic (ROC) curve using machine learning techniques. The MIF and AHP approaches demonstrated accuracies of 77% and 86%, respectively, validating their effectiveness in delineating GWPZs. This study establishes MIF and AHP as reliable models for groundwater potential zoning, with potential applications across other regions of Andhra Pradesh.

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

Evaluation of Groundwater Potential Zones Using Multi-influence Factor (MIF), Analytical Hierarchy Process (AHP) and Machine Learning (ML) Techniques from Uddanam Area, Srikakulam District, Andhra Pradesh, India

  • 章节

  • 发表日期：

    2025年

查看全文
  • Vinod Kumar Yarlanki,
  • Gope Naik Vadithya,
  • Etikala Balaji,
  • Vangala Sunitha,
  • Veeraswamy Golla,
  • B. C. Sundara Raja Reddy

摘要

Management of the resources of groundwater has become increasingly significant as a consequence of the complex interplay of climatological, geological, and anthropogenic factors, necessitating cutting-edge methods like Geographic Information Systems (GIS) and RS (Remote Sensing). This study integrates methodologies from RS, GIS, Multi-Influence Factor (MIF), and Analytic Hierarchy Process (AHP) to identify groundwater potential zones (GWPZs). Eight thematic layers-soil, slope, rainfall, lineament density (LD), drainage density (DD), geology, geomorphology (GM), and land use and land cover (LULC)-were utilized in this analysis. Each thematic layer was assigned weights based on its relevance to groundwater occurrence. Both MIF and AHP techniques employed quantile categorization to classify GWPZs into five categories: very poor, poor, moderate, good, and very good. The analysis revealed that approximately 42.47% of the study area exhibits good GWPZs, while 45.93% shows very good GWPZs. Groundwater levels were used to validate the maps through the Receiver Operating Characteristic (ROC) curve using machine learning techniques. The MIF and AHP approaches demonstrated accuracies of 77% and 86%, respectively, validating their effectiveness in delineating GWPZs. This study establishes MIF and AHP as reliable models for groundwater potential zoning, with potential applications across other regions of Andhra Pradesh.