A Comparison of deterministic and probabilistic approaches for assessing risks from contaminated aquifers: an Italian case study

Abstract

In this article we consider the methods of deterministic and probabilistic risk analysis regarding the presence of chemical contaminants in soil, water and air, with a broader meaning than usual for the latter, as we extended the probabilistic treatment to the parameters that influence the transport to a greater extent, in particular hydraulic conductivity and partition coefficient. These parameters, to which only one value is assigned, are considered here as random variables. The objective of the study reported herein was to demonstrate that application of the probabilistic method of risk assessment is preferable to the use of the deterministic method. Both methods yield contaminant removal levels that will reduce adverse effects on human health and the environment, but results from the deterministic method are typically more conservative than necessary, and are thus more costly to achieve. In addition, we found it essential to consider the importance of random variables (the parameters influencing the flow and the transport), such as the hydraulic conductivity and the partition coefficient, when assessing health risks. Both methodologies of health risk analysis, deterministic and probabilistic, were applied to a site in southern Italy, contaminated by heavy metals. The results obtained confirm the purposes of this study.

Keywords: Monte Carlo simulation; Risk analysis; deterministic approach; probabilistic approach; soil contamination.