Scholarlink

<p>In coastal aquifers, saltwater intrusion is becoming more severe due to excessive groundwater extraction and environmental pressures. This study aims to assess how well subsurface dams mitigate saltwater intrusion using SEAWAT code under the Henry problem framework. Simulations were performed to explore the effect of abstraction rate, well depth, well location, subsurface dam height, and subsurface dam location on the intrusion process. Results show that increasing the abstraction rate from 1 × 10<sup>− 6</sup> to 5 × 10<sup>− 6</sup> m<sup>3</sup>/sec aggravated the intrusion, reducing the dam’s effectiveness. The maximum concentration of abstracted water was recorded to be 10,576.6&#xa0;mg/L at the highest abstraction rate of 5 × 10<sup>− 6</sup> m<sup>3</sup>/sec. In contrast, the lowest concentration of 0.1&#xa0;mg/L was achieved under the optimal conditions of subsurface dam location and height and the lowest abstraction rate. Greater dam heights and optimal placement also limited saltwater intrusion, improving mitigation. Then, EXtreme Gradient Boosting (XGBoost) were employed and optimized using Bayesian Optimization (BO) to predict saltwater intrusion. The model achieved high predictive accuracy with R<sup>2</sup> of 0.999 and 0.992 for the training and testing data, respectively. SHAP analysis identified the abstraction rate and dam height as the most influential predictors of the saltwater intrusion. The study illustrates the effectiveness of combining numerical simulations and machine learning to address coastal groundwater challenges. The findings provide actionable insights for improving the design of subsurface dams and offer sustainable solutions to protect freshwater resources in regions vulnerable to saltwater intrusion.</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

Assessing the effectiveness of subsurface dam as a mitigation for saltwater intrusion using SEAWAT and XGBOOST models

  • 期刊论文

  • 发表日期：

    2026年6月23日

查看全文
  • Mohamed Galal Eltarabily,
  • Romysaa Elasbah,
  • Mohamed Elkiki,
  • Ahmed Shalby,
  • Sobhy R. Emara,
  • Tarek Selim

摘要

In coastal aquifers, saltwater intrusion is becoming more severe due to excessive groundwater extraction and environmental pressures. This study aims to assess how well subsurface dams mitigate saltwater intrusion using SEAWAT code under the Henry problem framework. Simulations were performed to explore the effect of abstraction rate, well depth, well location, subsurface dam height, and subsurface dam location on the intrusion process. Results show that increasing the abstraction rate from 1 × 10− 6 to 5 × 10− 6 m3/sec aggravated the intrusion, reducing the dam’s effectiveness. The maximum concentration of abstracted water was recorded to be 10,576.6 mg/L at the highest abstraction rate of 5 × 10− 6 m3/sec. In contrast, the lowest concentration of 0.1 mg/L was achieved under the optimal conditions of subsurface dam location and height and the lowest abstraction rate. Greater dam heights and optimal placement also limited saltwater intrusion, improving mitigation. Then, EXtreme Gradient Boosting (XGBoost) were employed and optimized using Bayesian Optimization (BO) to predict saltwater intrusion. The model achieved high predictive accuracy with R2 of 0.999 and 0.992 for the training and testing data, respectively. SHAP analysis identified the abstraction rate and dam height as the most influential predictors of the saltwater intrusion. The study illustrates the effectiveness of combining numerical simulations and machine learning to address coastal groundwater challenges. The findings provide actionable insights for improving the design of subsurface dams and offer sustainable solutions to protect freshwater resources in regions vulnerable to saltwater intrusion.