[Health Risk Assessment for an Arsenic-contaminated Site Based on Monte Carlo Simulation and Parameters Optimization].

Risk assessment is a critical part of risk management for contaminated sites. However， in the specific management practice of As-contaminated sites， it is difficult to obtain realistic health risks for contaminated sites based on the total amount of pollutants and determined values of the model， thus preventing the control requirements of later remediation to be met. An increasing number of studies have recently been conducting risk assessments by considering bioavailability， modification parameters， and combined probabilistic models. To improve the accuracy of risk assessment results， taking a large As-contaminated site as a case， 432 sampling sites were set up and collected at different depths to analyze the level and distribution characteristics of As pollution， and probabilistic risk assessment was conducted with the modification of model parameters through literature research and Monte Carlo simulation. Then， the impact of traditional methods and probabilistic methods on health risk assessment was explored in comparison. The results indicated that （As） in the top soil of the study area ranged from 2.70-97.0 mg·kg， with a spatial variation coefficient of 0.61 and weaker spatial continuity. The carcinogenic risk and hazard index obtained by the traditional risk assessment method were 2.12E-4 and 8.36， respectively， which obviously overestimated the actual risk level and were not conductive to the refined management of As-contaminated sites. Combined with modification of model parameters and probabilistic risk assessment， the non-carcinogenic risk for adults and children was found to be at an acceptable level， and the carcinogenic risk was reduced by nearly an order of magnitude compared to that in the conventional method. Considering the relative biological effectiveness （RBA） of As， the 95% quantile of the total carcinogenic risk was 1.24E-5， a reduction of up to 36.41% compared to the uncorrected corresponding risk value of 1.95E-5. The carcinogenic risk of soil As for adults and children in the study area exceeded acceptable risk levels 1E-6， with oral ingestion of soil being the primary route of exposure. In addition， the results of the sensitivity analysis of the parameters showed that As concentration， daily oral ingestion rate of soils， and exposure duration of children had relatively larger effects for health risks. This work will provide a methodological and theoretical basis for achieving accurate risk assessment of As-contaminated sites and provide concepts for refined risk management.