Time-Evolution of Vapor Intrusion Risk from Gasoline-Derived Multiphase and Multicomponent Sources in Soil

Abstract

Abstract

Human health risk assessment of vapor intrusion caused by organic pollutants is commonly based on steady-state predictions of partition and vapor migration in the subsoil. This study develops a pseudo-dynamic, process-based Partition–Diffusion Risk Model (PDRM) using a one-dimensional numerical model for organic mixtures to assess the time evolution of cancer and non-cancer risks, indoor air concentrations, and non-aqueous phase liquid (NAPL) formation. The model has been applied to a low-carbon sandy soil without microbial degradation, which might be a worst-case scenario. Six simulation scenarios combined two source concentrations (1000 and 3000 mg/kg) and three source depths (1, 3, and 5 m) over 30 years. Results show that source depth governs exposure dynamics: shallow contamination poses unacceptable risks rapidly but declines quickly, whereas at greater depths, unacceptable levels appear later and persist throughout the exposure period. NAPL formation may act as a secondary source, sustaining vapor release and extending indoor exposure under high-loading conditions. Multicomponent partitioning induces nonlinear, compound-specific behavior, with the first 3–5 years representing a critical period for rapid risk changes. Conventional models show that neglecting NAPL formation and time variability may lead to an underestimation of cancer risk by up to an order of magnitude. These findings highlight the importance of incorporating depth and time-dependent characterization to reduce uncertainty in vapor intrusion risk assessments.