Scholarlink

<p>The Mahanadi Catchment requires improved water quality to support its extensive cropped areas and to meet drinking and other non-agricultural water demands of the surrounding population. The densely populated deltaic regions are particularly vulnerable to pollution from sewage discharge, industrial effluents, and agricultural runoff containing fertilizers. In this context, the present task was undertaken to assess the overall status of water quality, providing a basis for effective, sustainable, and long-term water resource management.&#xa0;The present study provided a methodical evaluation for the pollution level of water and its effect on local people, residing near the Mahanadi River Basin, Odisha. For this assessment, twenty significant surface water (SW) regulating variables from nineteen sampling sites were developed, during a pre-monsoon period of 2022–2025, gathered annual average, by combining different methods:&#xa0;weighted arithmetic (WA)-water quality index (WQI), Statistical analysis namely, Correlation Matrix, Cluster Analysis (CA), Discriminant Analysis (DA), Absolute Principal Component Score Multiple Linear Regression (APCS-MLR) model, and decision-making techniques namely, Integrated Determination of Objective Criteria Weights (IDOCRIW), and Fuzzy (F) Analytic Hierarchy Process (AHP). In terms of physicochemical characteristics, the pH values at all sampling stations indicate a slightly alkaline nature. The study further reveals that the area is dominated by parameters such as TKN, which exhibits elevated concentrations across all locations, exceeding the World Health Organization (WHO) permissible limits for drinking water.&#xa0;By using Geographical Information System (GIS), the findings of the prospective surface water zone model using WA WQI, were categorized into four groups: excellent, good, poor, and very poor. Approximately, 15.8% of the research area was recognized as having high potential, 68.4% as having good potential, and 10.5% and 5.3% as having poor and very poor potential, respectively.&#xa0;Later, surface water prospect maps were examined and confirmed using the multivariate data from numerous production locations in the studied region. Based on the CA classification, it distinguishes into three groups, based on the similar patterns and its pollution level, and therefore, requires monitoring at cluster III, for ecosystem sustainability. The DA often, culminated in the retrieval of two components, that accounts for about 100% and 97.92%, using 20 and 10 variables, in standard and stepwise mode, respectively. Again, APCS-MLR demonstrates expected process-based and covert correlations among the main dissolved ions in the river as well as factors that generally control the water chemistry. It identified five factors that accumulated for 50.21%-98.69% of the cumulative variance,&#xa0;and the urban areas, rural regions, enterprises, the environment, chemicals, downstream regions, and automobiles were the primary causes. The U-WQI resulted five different types of water, which were identified as: excellent (36.84%), good (10.53%), poor (31.68%), very poor (15.79%) and unsuitable (5.26%). From the F-AHP WQI diagram, there were five distinct kinds of water detected: out of the total 20 samples collected from the study area, around 15 samples (78.95%) were classified under the category of good-medium class. As a result, the largest proportion at the polluted sites, was most clearly explained by sources from urban and downstream areas. In this context, pollution sources were identified using the various techniques discussed above, indicating contributions from both natural processes—such as geogenic influences (weathering of parent rocks), evaporation, ion exchange, mineral dissolution, and subsequent mineral weathering—as well as anthropogenic activities, including urbanization and industrial operations. It is therefore concluded that the combined application of statistical procedures and indexing strategies provides a robust and effective framework for developing a surface water quality map for the selected river in Odisha. As a result,&#xa0;the findings provide critical insights for optimizing hydrological modelling, improving water resource management, and supporting sustainable agricultural practices in the chosen catchment and its similar environments.</p> Graphical abstract <p></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

Reliable water quality classification assessment and evaluating the influences of hydrochemistry variations using explainable multi-criteria and statistical models: implications for management strategies

  • 期刊论文

  • 发表日期：

    2026年1月16日

查看全文
  • Abhijeet Das

摘要

The Mahanadi Catchment requires improved water quality to support its extensive cropped areas and to meet drinking and other non-agricultural water demands of the surrounding population. The densely populated deltaic regions are particularly vulnerable to pollution from sewage discharge, industrial effluents, and agricultural runoff containing fertilizers. In this context, the present task was undertaken to assess the overall status of water quality, providing a basis for effective, sustainable, and long-term water resource management. The present study provided a methodical evaluation for the pollution level of water and its effect on local people, residing near the Mahanadi River Basin, Odisha. For this assessment, twenty significant surface water (SW) regulating variables from nineteen sampling sites were developed, during a pre-monsoon period of 2022–2025, gathered annual average, by combining different methods: weighted arithmetic (WA)-water quality index (WQI), Statistical analysis namely, Correlation Matrix, Cluster Analysis (CA), Discriminant Analysis (DA), Absolute Principal Component Score Multiple Linear Regression (APCS-MLR) model, and decision-making techniques namely, Integrated Determination of Objective Criteria Weights (IDOCRIW), and Fuzzy (F) Analytic Hierarchy Process (AHP). In terms of physicochemical characteristics, the pH values at all sampling stations indicate a slightly alkaline nature. The study further reveals that the area is dominated by parameters such as TKN, which exhibits elevated concentrations across all locations, exceeding the World Health Organization (WHO) permissible limits for drinking water. By using Geographical Information System (GIS), the findings of the prospective surface water zone model using WA WQI, were categorized into four groups: excellent, good, poor, and very poor. Approximately, 15.8% of the research area was recognized as having high potential, 68.4% as having good potential, and 10.5% and 5.3% as having poor and very poor potential, respectively. Later, surface water prospect maps were examined and confirmed using the multivariate data from numerous production locations in the studied region. Based on the CA classification, it distinguishes into three groups, based on the similar patterns and its pollution level, and therefore, requires monitoring at cluster III, for ecosystem sustainability. The DA often, culminated in the retrieval of two components, that accounts for about 100% and 97.92%, using 20 and 10 variables, in standard and stepwise mode, respectively. Again, APCS-MLR demonstrates expected process-based and covert correlations among the main dissolved ions in the river as well as factors that generally control the water chemistry. It identified five factors that accumulated for 50.21%-98.69% of the cumulative variance, and the urban areas, rural regions, enterprises, the environment, chemicals, downstream regions, and automobiles were the primary causes. The U-WQI resulted five different types of water, which were identified as: excellent (36.84%), good (10.53%), poor (31.68%), very poor (15.79%) and unsuitable (5.26%). From the F-AHP WQI diagram, there were five distinct kinds of water detected: out of the total 20 samples collected from the study area, around 15 samples (78.95%) were classified under the category of good-medium class. As a result, the largest proportion at the polluted sites, was most clearly explained by sources from urban and downstream areas. In this context, pollution sources were identified using the various techniques discussed above, indicating contributions from both natural processes—such as geogenic influences (weathering of parent rocks), evaporation, ion exchange, mineral dissolution, and subsequent mineral weathering—as well as anthropogenic activities, including urbanization and industrial operations. It is therefore concluded that the combined application of statistical procedures and indexing strategies provides a robust and effective framework for developing a surface water quality map for the selected river in Odisha. As a result, the findings provide critical insights for optimizing hydrological modelling, improving water resource management, and supporting sustainable agricultural practices in the chosen catchment and its similar environments.

Graphical abstract