May 9, 2014
Journal Article

Modeling spatial patterns in soil arsenic to estimate natural baseline concentrations

Abstract

ABSTRACT Arsenic in soil is an important public health concern. Toxicity guidelines and models based on laboratory studies (i.e., U.S. EPA’s Integrated Risk Information System) should consider natural soil As concentrations to avoid unnecessary remediation burdens on society. We used soil and stream sediment samples from the USGS National Geochemical Survey database to assess the spatial distribution of natural As in a 1.16E+5 km2 area. Samples were collected at 348 soil and 144 stream locations, providing approximately one sample for every 290 km2. Sample sites were selected to minimize the potential influence of anthropogenic inputs. Samples were processed using acid digestion of whole samples (concentrated HCl and ascorbic acid) and concentrations were measured using hydride-generation atomic absorption spectrometry. Soil As ranged from 2.0 to 45.6 mg kg-1. Geostatistical techniques were used to model and map the spatial variability of As. The mean and variance at unsampled locations were estimated using sequential Gaussian simulation. Five areas of elevated concentration (> the median of 10 mg kg-1) were identified and the relationships to geologic parent materials, glacial sedimentation patterns, and soil conditions interpreted. Our results showed As concentrations >10 mg kg-1 were common, and >20 mg kg-1 were not unusual for the central and west central portions of Ohio (USA). In contrast, concentrations

Revised: May 19, 2014 | Published: May 9, 2014

Citation

Venteris E.R., N.T. Basta, J.M. Bigham, and R. Rea. 2014. Modeling spatial patterns in soil arsenic to estimate natural baseline concentrations. Journal of Environmental Quality 43, no. 3:936-946. PNNL-SA-89577. doi:10.2134/jeq2013.11.0459