Groundwater pollution from valley type landfills is concerning, and natural attenuation by contaminants is increasingly relied upon. However, the reliability of natural attenuation in such complex sites has been called into question due to incomplete understanding of their attenuation mechanisms. Therefore, we conducted field investigations, monitoring analyses, mathematical statistics, and machine learning techniques to elucidate the natural attenuation mechanisms of pollutants within bedrock fissures at a prototypical valley type landfill located in the east Yanshan Mountains, China. Our results indicate that 50% of the monitored indicators showed extreme pollution in bedrock fissure aquifers, due to seepage from the valley type landfill site. Ammonia nitrogen, arsenic, cadmium, lead, iron, manganese, and mercury were among the contaminants that could pose serious risks to human health. Pollutant concentrations in bedrock fissure aquifers were lower during the rainy season compared to the dry season as the aquifer was rapidly recharged by strong rainfall runoff. The initial concentration of bedrock fissure water generally increased during the flow through the landfill. However, significant natural attenuation of total dissolved solids, oxygen consumption, ammonia, cadmium, and lead occurred after passing through the landfill (p<0.05), with attenuation coefficients of 0.0041 m-1, 2.56×E-5m-2, 4.18×E-5m-2、0.0015 m-0.99, and 6.83×E-33m-12.49, respectively. The driving mechanisms for natural attenuation include physical migration, leaching, microbiological degradation, and adsorption, primarily occurring within 600-650 m downstream of the landfill boundary. This study makes fundamental contribution to the understanding of the migration and natural attenuation process of leachate pollutants in bedrock fissure aquifer, which will provide a scientific basis for implementation of natural attenuation strategies in complex site remediation. Future research should examine more precise evidence of natural attenuation feasibility in complex sites in conjunction with monitoring networks.
Read full abstract