Scholarlink

<p>Surface water scarcity is a critical concern in arid and semi-arid regions, particularly in Rajasthan, India, where over 60% of the area faces chronic water stress. This study assesses 38 years (1984–2021) of surface water dynamics across 49 districts using high-resolution satellite data to identify hydrological disparities and inform district-level planning. The research addresses a significant gap in quantifying long-term changes in both seasonal and permanent surface water availability, normalized by population to estimate per capita access. We used the JRC Global Surface Water Mapping Layer derived from Landsat imagery to extract indicators such as water occurrence, absolute change, per capita seasonal and permanent water availability. A clustering approach using KMeans and Silhouette Score analysis was implemented to group hydrologically similar districts. In addition, we developed a surface water depletion index and increment index to identify districts with the greatest water loss and gain, respectively. A log-scaled seasonal-to-permanent per capita water ratio was used to assess hydrological imbalance. Results reveal stark spatial variation in surface water distribution. Chittorgarh and Tonk (Cluster 3) showed the highest availability indices (&gt; 0.6), supported by perennial reservoirs and rainwater harvesting structures. In contrast, districts like Sri Ganganagar and Hanumangarh exhibited surface water depletion indices of 0.85 and 0.72. Conversely, Tonk and Jhalawar showed increment indices of 1.05 and 0.95, reflecting successful watershed interventions. Per capita permanent water availability ranged from &lt; 50&#xa0;m² in urban districts like Jaipur to &gt; 1200&#xa0;m² in low-density districts such as Jaisalmer. This district-level analysis provides a first-of-its-kind water vulnerability mapping for Rajasthan, enabling cluster-specific planning and targeted policy interventions. The findings support national initiatives such as Jal Shakti Abhiyan and contribute to the Sustainable Development Goals, particularly SDG 6 (Clean Water and Sanitation) and SDG 13 (Climate Action).</p> Graphical Abstract <p>The graphical abstract presents a comprehensive analysis combining machine learning, spatial mapping, and cluster-based comparisons to assess district-level water availability in Rajasthan. On the left, machine learning models are compared using the silhouette score to determine clustering effectiveness. KMeans is identified as the best-performing method, with the optimal number of clusters being five. The middle section shows district-wise maps of permanent and seasonal water availability per capita, highlighting spatial disparities. The right section presents boxplots that compare permanent and seasonal water per capita across the five clusters. These visualizations reveal clear differences in water availability patterns among regions. This combined approach is important because it not only identifies natural groupings in data but also links them to geographic and quantitative insights. Such analysis supports targeted water management strategies, helping policymakers prioritize interventions in areas with limited water access and design region-specific resource allocation plans under current and future climatic conditions. This analysis directly supports Sustainable Development Goal (SDG) 6: Clean Water and Sanitation by identifying spatial disparities in water availability, aiding in efficient and equitable water resource management. It also contributes to SDG 13: Climate Action by using data-driven clustering to understand seasonal variations and water stress under changing climatic conditions. Together, these insights promote informed decision-making for climate-resilient and inclusive water planning.</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

Clustering-Based Assessment of Long-Term Surface Water Scarcity and Per Capita Vulnerability in Arid India

  • 期刊论文

  • 发表日期：

    2025年10月9日

查看全文
  • Vikas Poonia,
  • Ankita Mukherjee,
  • Arun Dev Singh,
  • Somil Swarnkar

摘要

Surface water scarcity is a critical concern in arid and semi-arid regions, particularly in Rajasthan, India, where over 60% of the area faces chronic water stress. This study assesses 38 years (1984–2021) of surface water dynamics across 49 districts using high-resolution satellite data to identify hydrological disparities and inform district-level planning. The research addresses a significant gap in quantifying long-term changes in both seasonal and permanent surface water availability, normalized by population to estimate per capita access. We used the JRC Global Surface Water Mapping Layer derived from Landsat imagery to extract indicators such as water occurrence, absolute change, per capita seasonal and permanent water availability. A clustering approach using KMeans and Silhouette Score analysis was implemented to group hydrologically similar districts. In addition, we developed a surface water depletion index and increment index to identify districts with the greatest water loss and gain, respectively. A log-scaled seasonal-to-permanent per capita water ratio was used to assess hydrological imbalance. Results reveal stark spatial variation in surface water distribution. Chittorgarh and Tonk (Cluster 3) showed the highest availability indices (> 0.6), supported by perennial reservoirs and rainwater harvesting structures. In contrast, districts like Sri Ganganagar and Hanumangarh exhibited surface water depletion indices of 0.85 and 0.72. Conversely, Tonk and Jhalawar showed increment indices of 1.05 and 0.95, reflecting successful watershed interventions. Per capita permanent water availability ranged from < 50 m² in urban districts like Jaipur to > 1200 m² in low-density districts such as Jaisalmer. This district-level analysis provides a first-of-its-kind water vulnerability mapping for Rajasthan, enabling cluster-specific planning and targeted policy interventions. The findings support national initiatives such as Jal Shakti Abhiyan and contribute to the Sustainable Development Goals, particularly SDG 6 (Clean Water and Sanitation) and SDG 13 (Climate Action).

Graphical Abstract

The graphical abstract presents a comprehensive analysis combining machine learning, spatial mapping, and cluster-based comparisons to assess district-level water availability in Rajasthan. On the left, machine learning models are compared using the silhouette score to determine clustering effectiveness. KMeans is identified as the best-performing method, with the optimal number of clusters being five. The middle section shows district-wise maps of permanent and seasonal water availability per capita, highlighting spatial disparities. The right section presents boxplots that compare permanent and seasonal water per capita across the five clusters. These visualizations reveal clear differences in water availability patterns among regions. This combined approach is important because it not only identifies natural groupings in data but also links them to geographic and quantitative insights. Such analysis supports targeted water management strategies, helping policymakers prioritize interventions in areas with limited water access and design region-specific resource allocation plans under current and future climatic conditions. This analysis directly supports Sustainable Development Goal (SDG) 6: Clean Water and Sanitation by identifying spatial disparities in water availability, aiding in efficient and equitable water resource management. It also contributes to SDG 13: Climate Action by using data-driven clustering to understand seasonal variations and water stress under changing climatic conditions. Together, these insights promote informed decision-making for climate-resilient and inclusive water planning.