Applying an artificial intelligence model using multidimensional spatial-temporal data to predict arsenic contamination of groundwater.

Special attention has been paid in recent years to the social and environmental implications of municipal solid waste. There are potential social and environmental issues associated with storing any waste in landfill. Soil, groundwater, and surface water contamination are issues related to landfill and the migration of contaminants by leaching to adjacent areas. The current research attempts to simulate the penetration of landfill contaminants through groundwater leaching, using PART modeling approaches. Multiple social factors, taken from five years of data from 20 counties and cities in Taiwan were used as predictors of groundwater arsenic content. The results indicate that the proposed model attains a precision rate of 87%. The findings suggest that county and city demographics and the volume of waste removed are two major factors in determining the arsenic content of groundwater. The proposed predictive model can be adopted by agencies as an effective early warning tool for groundwater arsenic levels. [ABSTRACT FROM AUTHOR]