Vulnerability Indexing to Saltwater Intrusion from Models at Two Levels using Artificial Intelligence Multiple Model...

Area of Science:

• Hydrogeology
• Artificial Intelligence
• Environmental Science

Background:

• Unplanned groundwater extraction in coastal areas leads to aquifer depletion and saltwater intrusion (SWI).
• Accurate mapping of SWI vulnerability is crucial for sustainable groundwater resource management.

Purpose of the Study:

• To develop and validate a novel AI-based modeling strategy for mapping vulnerability to saltwater intrusion (SWI) in coastal aquifers.
• To apply the AI model to the Urmia aquifer, a region experiencing significant groundwater decline.

Main Methods:

• A two-level Artificial Intelligence (AI) modeling strategy was formulated, integrating the GALDIT method.
• Level 1 models (Artificial Neural Network, Sugeno Fuzzy Logic, Neuro-Fuzzy) predicted SWI vulnerability, with their outputs feeding into a Level 2 Support Vector Machine model.
• Model performance was evaluated using RMSE and R² metrics, and results were validated with groundwater sample analysis (Piper diagram).

Main Results:

• The AI-based models successfully mapped SWI vulnerability in the Urmia aquifer.
• Level 1 models demonstrated fitness-for-purpose, while the Level 2 Support Vector Machine model showed improved accuracy with reduced scatter.
• Validation confirmed a strong correspondence between the predicted SWI vulnerability index and the actual SWI status.

Conclusions:

• The developed AI-based GALDIT modeling strategy offers a robust tool for assessing and mapping saltwater intrusion vulnerability in coastal aquifers.
• This approach provides valuable insights for effective groundwater resource management in vulnerable coastal regions.
• The study highlights the potential of integrating AI with established hydrogeological methods for environmental risk assessment.