A framework for spatial correlations between industrial pollution sources and groundwater vulnerabilities based on...

Area of Science:

• Environmental Science
• Hydrogeology
• Data Science

Background:

• Groundwater pollution risk control is hindered by poor understanding of spatial relationships between pollution sources and aquifer vulnerability.
• Effective management requires detailed spatial correlation analysis.

Purpose of the Study:

• To develop and apply a novel data-driven framework for revealing spatial correlations between industrial pollution sources and groundwater vulnerabilities.
• To enhance groundwater pollution risk assessment and control strategies in Guangdong Province, China.

Main Methods:

• A hybrid model combining genetic algorithm (GA) and back propagation neural network (BPNN) to objectively weight indicators for the DRASTICL model, creating a groundwater vulnerability map.
• Kernel density estimation (KDE) for smoothing industrial pollution source distribution.
• Bivariate local Moran's I (BLMI) to generate a spatial clustering map of pollution sources and groundwater vulnerabilities.

Main Results:

• The GA-BPNN-DRASTICL model successfully reduced subjectivity in indicator weighting for groundwater vulnerability mapping.
• KDE provided a smoothed representation of industrial pollution source distribution.
• BLMI effectively visualized spatial clustering, highlighting areas with high industrial pollution and groundwater vulnerability.

Conclusions:

• The integrated data-driven framework provides a robust method for assessing groundwater pollution risks.
• The findings offer crucial insights for targeted groundwater pollution control measures in different regions.
• This approach can significantly improve the efficacy of groundwater resource management and protection.