Uncertainty versus variability in Monte Carlo simulations of human exposure through food pathways

An important issue in both the risk characterization and subsequent risk management of contaminated soil is how precisely we can characterize the distribution among individuals of potential doses associated with chemical contaminants in soil and whether this level of precision favors the use of population distributions of exposure over the use of single scenario representations. For lipophilic contaminants, such as dioxins, furans, polychlorinated biphenyls, pesticides, and for metals such as lead and mercury, exposures through food have been demonstrated to be dominant contributors to total dose within non-occupationally exposed populations. However, overall uncertainties in estimating potential doses through food chains are much larger than uncertainties associated with other exposure pathways. A general model is described here for estimating the ratio of potential dose to contaminant concentration in soil for homegrown foods contaminated by lipophilic, nonionic organic chemicals. This model includes parameters describing homegrown food consumption rates, exposure duration, biotransfer factors, and partition factors. For the parameters needed in this model, the mean and variance are often the only moments of the parameter distribution available. Parameters are divided into three categories, uncertain parameters, variable parameters, and mixed uncertain/variable parameters. Using soils contaminated by hexachlorbenzene (HCB) and benzo(a)pyrene (BaP) as cases studies, a stepwise Monte Carlo analysis is used to develop a histogram that apportions variance in the outcome (ratio of potential dose by food pathways to soil concentration) to variance in each of the three input categories. The results represent potential doses in households consuming homegrown foods