ABSTRACT

With the rapid progress of industrialization, the groundwater contamination has become a tricky environmental issue. In the past, the groundwater contamination issues usually focus on the soluble pollutants. Recently, the insoluble pollutants have attracted the attention of the public. The insoluble aqueous pollutants are also called non-aqueous phase liquids (NAPLs). When the pollutant density is over 1g/cm3 it is dense non-aqueous phase liquids (DNAPLs). Part of the DNAPLs are toxic and pose a great threat to the ecological environment and human health. In this paper, we propose a human health risk assessment method based on the numerical simulation process and parameter uncertainty analysis. Based on a sandbox experiment, the DNAPL migration process is simulated by TOUGH program. In addition, the modeling uncertainty is considered explicitly in this study. The unreliable model parameters would cause inaccurate results. In order to reduce parameter uncertainty, we calibrate the unknown model parameter in Bayesian theory. The Markov chain Monte Carlo simulation is applied to reduce the uncertainty of parameters. The human health risk is illustrated in the distribution of a WTO metric – Maximum Concentration Level. Compared with the results assessed by the specific parameters, the distribution of risk could provide more flexible and reliable information.