GPS-coupled contaminant monitors on free-ranging Chernobyl wolves challenge a fundamental assumption in exposure...

Measurements of external contaminant exposures on individual wildlife are rare because of difficulties in using contaminant monitors on free-ranging animals. Most wildlife contaminant exposure data are therefore simulated with computer models. Rarely are empirical exposure data available to verify model simulations, or to test fundamental assumptions inherent in exposure assessments. We used GPS-coupled contaminant monitors to quantify external exposures to individual wolves (Canis lupus) living within the Belarus portion of Chernobyl's 30-km exclusion zone. The study provided data on animal location and contaminant exposure every 35 min for 6 months, resulting in ~6600 individual locations and ¹³⁷Cs external exposure readings per wolf, representing the most robust external exposure data published to date on free ranging animals. The data provided information on variation in external exposure for each animal over time, as well as variation in external exposure among the eight wolves across the landscape of Chernobyl. The exposure data were then used to test a fundamental assumption in screening-level risk assessments, espoused in guidance documents of the U.S. Environmental Protection Agency and U.S. Department of Energy, - Mean contaminant concentrations conservatively estimate individual external exposures. We tested this assumption by comparing our empirical data to a series of simulations using the ERICA modeling tool. We found that modeled simulations of mean external exposure (10.5 mGy y^-1), based on various measures of central tendency, under-predicted mean exposures measured on five of the eight wolves wearing GPS-contaminant monitors (i.e., 12.3, 26.3, 28.0, 28.8 and 35.7 mGy y^-1). If under-prediction of exposure occurs for some animals, then arguably the use of averaged contaminant concentrations to predict external exposure is not as conservative as proposed by current risk assessment guidance. Thus, a risk assessor's interpretation of simulated exposures in a screening-level risk analysis might be misguided if contaminant concentrations are based on measures of central tendency. We offer three suggestions for risk assessors to consider in order to reduce the probability of underestimating exposure in a screening-level risk assessment.