Search

The environmental mobility of newly deposited radionuclides in surface soil is driven by complex biogeochemical relationships, which have significant impacts on transport pathways. The partition coefficient (Kd) is useful for characterizing the soil-solution exchange kinetics and is an important factor for predicting relative amounts of a radionuclide transported to groundwater compared to that remaining on soil surfaces and thus available for transport through erosion processes. Measurements of Kd for 238U are particularly useful because of the extensive use of 238U in military applications and associated testing, such as done at Los Alamos National Laboratory (LANL). Site-specific measurements of Kd for 238U are needed because Kd is highly dependent on local soil conditions and also on the fine soil fraction because 238U concentrates onto smaller soil particles, such as clays and soil organic material, which are most susceptible to wind erosion and contribute to inhalation exposure in off-site populations. We measured Kd for uranium in soils from two neighboring semiarid forest sites at LANL using a U.S. Environmental Protection Agency (EPA)-based protocol for both whole soil and the fine soil fraction (diameters<45 microm). The 7-d Kd values, which are those specified in the EPA protocol, ranged from 276-508 mL g-1 for whole soil and from 615-2249 mL g-1 for the fine soil fraction. Unexpectedly, the 30-d Kd values, measured to test for soil-solution exchange equilibrium, were more than two times the 7-d values. Rates of adsorption of 238U to soil from solution were derived using a 2-component (FAST and SLOW) exponential model. We found significant differences in Kd values among LANL sampling sites, between whole and fine soils, and between 7-d and 30-d Kd measurements. The significant variation in soil-solution exchange kinetics among the soils and soil sizes promotes the use of site-specific data for estimates of environmental transport rates and suggests possible differences in desorption rates from soil to solution (e.g., into groundwater or lung fluid). We also explore potential relationships between wind erosion, soil characteristics, and Kd values. Combined, our results highlight the need for a better mechanistic understanding of soil-solution partitioning kinetics for accurate risk assessment.

This paper describes the structure of the dynamic food-chain model PATHWAY and its utility for estimating radionuclide ingestion after fallout deposition from nuclear testing in Nevada. Model input requirements are described and output examples are provided. The basic output of PATHWAY is the time-integrated radionuclide ingestion by humans per unit fallout deposition (Bq per Bq m-2). Output specific to sex, age, life-style (diet), location (agricultural practice), event (calendar date), and radionuclide may be generated. Uncertainties of model predictions, based on "Monte Carlo" simulations using parameter value distributions, are described. Results of a sensitivity analysis, based on a ranking of partial correlation coefficients, are reviewed to illustrate the relative importance of parameters and associated transport pathways. Output data for 131I and 137Cs in milk are compared with predictions from several well known food-chain models. Preliminary efforts to validate PATHWAY results with real data sets are described.

Uncertainty in predictions from the PATHWAY food-chain model was estimated using Monte Carlo simulation. Uncertainty estimates, measured by the geometric standard deviation (GSD), were obtained for median values of time-integrated concentrations of 131I, 136Cs, and 137Cs in foods and for the corresponding time-integrated intakes resulting from ingestion of all foods. The GSDs associated with a given food for the two short-lived radionuclides, 131I and 136Cs, were not significantly different, but they differed from the GSDs for the longer-lived radionuclide. The GSDs for integrated concentrations of radionuclides in milk varied with the time of year fallout was deposited, but uncertainty for nondairy products was relatively independent of the date of fallout deposition. The estimated GSDs were applied to other radionuclides of interest based on physical half-life and ranged from 1.7 to 2.7 for time-integrated intake across all foods for radionuclides with physical half-lives less than 30 d, from 1.8 to 2.3 for half-lives ranging from 30 to 500 d, and from 1.9 to 2.1 when half-lives were greater than 500 d.