Comparison of point-source pollutant loadings to soil and groundwater for 72 chemical substances

Abstract

Fate and transport of 72 chemicals in soil and groundwater were assessed by using a multiphase compositional model (CompFlow Bio) because some of the chemicals are non-aqueous phase liquids or solids in the original form. One metric ton of chemicals were assumed to leak in a stylized facility. Scenarios of both surface spills and subsurface leaks were considered. Simulation results showed that the fate and transport of chemicals above the water table affected the fate and transport of chemicals below the water table, and vice versa. Surface spill scenarios caused much less concentrations than subsurface leak scenarios because leaching amounts into the subsurface environment were small (at most 6% of the 1 t spill for methylamine). Then, simulation results were applied to assess point-source pollutant loadings to soil and groundwater above and below the water table, respectively, by multiplying concentrations, impact areas, and durations. These three components correspond to the intensity of contamination, mobility, and persistency in the assessment of pollutant loading, respectively. Assessment results showed that the pollutant loadings in soil and groundwater were linearly related (r = 0.64). The pollutant loadings were negatively related with zero-order and first-order decay rates in both soil (r = - 0.5 and - 0.6, respectively) and groundwater (- 1.0 and - 0.8, respectively). In addition, this study scientifically defended that the soil partitioning coefficient (K) significantly affected the pollutant loadings in soil (r = 0.6) and the maximum masses in groundwater (r = - 0.9). However, K was not a representative factor for chemical transportability unlike the expectation in chemical ranking systems of soil and groundwater pollutants. The pollutant loadings estimated using a physics-based hydrogeological model provided a more rational ranking for exposure assessment, compared to the summation of persistency and transportability scores in the chemical ranking systems. In the surface spill scenario, the pollutant loadings were zeros for all chemicals, except methylamine to soil whose pollutant loading was smaller than that in the subsurface leak scenario by 4 orders of magnitude. The maximum mass and the average mass multiplied by duration in soil greatly depended on leaching fluxes (r = 1.0 and 0.9, respectively), while the effect of leaching fluxes diminished below the water table. The contribution of this work is that a physics-based numerical model was used to quantitatively compare the subsurface pollutant loading in a chemical accident for 72 chemical substances, which can scientifically defend a simpler and more qualitative assessment of pollutant loadings. Besides, this study assessed pollutant loadings to soil (unsaturated zone) and groundwater (saturated zone) all together and discussed their interactions.