Arsenic (As) and lead (Pb) are two contaminants of concern associated with urban gardening. In Puerto Rico, data currently is limited on As and Pb levels in urban garden soils, soil metal (loid) bioaccessibility, and uptake of As and Pb in soil by edible plants grown in the region. This study examined total and bioaccessible soil As and Pb concentrations and accumulation in 10 commonly grown garden plants collected from three urban community gardens in Puerto Rico. Bioavailability values were predicted using bioaccessibility data to compare site-specific bioavailability estimates to commonly used default exposure assumptions. Total and bioaccessible As levels in study soils ranged from 2 to 55 mg/kg and 1 to 18 mg/kg, respectively. Total and bioaccessible Pb levels ranged from 19 to 172 mg/kg and 17 to 97 mg/kg, respectively. Measured bioaccessibility values corresponded to 19% to 42% bioaccessible As and 61% to 100% bioaccessible Pb when expressed as a percent of total As and Pb respectively. Predicted relative percent bioavailability of soil As and Pb based on measured bioaccessibility values ranged from 18% to 36% and 51% to 85% for As and Pb respectively. Transfer factors (TFs) measuring uptake of As in plants from soil ranged from 0 to 0.073 in the edible flesh (fruit or vegetable) of plant tissues analyzed and 0.073 to 0.444 in edible leaves. Pb TFs ranged from 0.002 to 0.012 in flesh and 0.023 to 0.204 in leaves. Consistent with TF values, leaves accumulated higher concentrations of As and Pb than the flesh, with the highest tissue concentrations observed in the culantro leaf (3.2 mg/kg dw of As and 8.9 mg/kg dw of Pb). Leaves showed a general but not statistically-significant (α = 0.05) trend of increased As and Pb concentration with increased soil levels, while no trend was observed for flesh tissues. These findings provide critical data that can improve accuracy and reduce uncertainty when conducting site-specific risk determination of potential As and Pb exposure while gardening or consuming garden produce in the understudied region of Puerto Rico.

Interest in improved understanding of relationships among soil properties and arsenic (As) bioaccessibility has motivated the use of regression models for As bioaccessibility prediction. However, limits in the numbers and types of soils included in previous studies restrict the usefulness of these models beyond the range of soil conditions evaluated, as evidenced by reduced predictive performance when applied to new data. In response, hierarchical models that consider variability in relationships among soil properties and As bioaccessibility across geographic locations and contaminant sources were developed to predict As bioaccessibility in 139 soils on both a mass fraction (mg/kg) and % basis. The hierarchical approach improved the estimation of As bioaccessibility in studied soils. In addition, the number of soil elements identified as statistically significant explanatory variables increased when compared to previous investigations. Specifically, total soil Fe, P, Ca, Co, and V were significant explanatory variables in both models, while total As, Cd, Cu, Ni, and Zn were also significant in the mass fraction model and Mg was significant in the % model. This developed hierarchical approach provides a novel tool to (1) explore relationships between soil properties and As bioaccessibility across a broad range of soil types and As contaminant sources encountered in the environment and (2) identify areas of future mechanistic research to better understand the complexity of interactions between soil properties and As bioaccessibility.

The U.S. EPA's in vitro bioaccessibility (IVBA) method 9200.1-86 defines a validated analytical procedure for the determination of lead bioaccessibility in contaminated soils. The method requires the use of a custom-fabricated extraction device that uses a heated water bath for sample incubation. In an effort to improve ease of use, increase sample throughput, and reduce equipment acquisition and maintenance costs, an alternative low-cost, commercially available extraction device capable of sample incubation via heated air and end-over-end rotation was evaluated. An intra-laboratory study was conducted to compare lead bioaccessibility values derived using the two extraction devices. IVBA values were not statistically different (α = 0.05) between the two extraction devices for any of the soils (n = 6) evaluated in this study, with an average difference in mean lead IVBA of 0.8% (s.d. = 0.5%). The commercially available extraction device was able to generate accurate lead IVBA data as compared to the U.S. EPA's expected value for a National Institute of Standards and Technology standard reference material soil. The relative percent differences between high and low IVBA values for each soil, a measure of instrument precision, were also not statistically different (α = 0.05) between the two extraction devices. The statistical agreement of lead IVBA values observed using the two extraction devices supports the use of a low-cost, commercially available extraction device as a reliable alternative to a custom-fabricated device as required by EPA method 9200.1-86.