Comparing deterministic and probabilistic risk assessments for sites contaminated by polycyclic aromatic hydrocarbons (PAHs)

Abstract

Deterministic risk assessments provide a single point estimate of risk at a site of concern, while probabilistic risk assessment methods generate a range of values from probability distribution functions (PDFs). This study compares deterministic and probabilistic risk assessment methods for two different sites using uncertainty analysis and evaluates the use of toxic equivalency factors (TEFs) for polycyclic aromatic hydrocarbons (PAHs) in each method. The use of TEFs allowed more of the PAHs to be included which resulted in higher risk estimates for both adults and children at both study sites. Exposure factor variability generated greater uncertainty in risk estimates than sample heterogeneity or toxicity estimates. TEFs allowed for a more complete representation of the toxicity contributions from PAH species present at the study sites. Exposure factor distributions together with TEF estimations for toxicity of PAH species provide an assessment that can reduce and account for uncertainty. Probabilistic risk estimates provide a sensible improvement to deterministic risk estimates by generating a range of risk values, along with an average estimate and the degree of conservatism of those estimates.