BASELINE GEOCHEMISTRY OF AN UNDEVELOPED ORE DEPOSIT: LESSONS FROM THE COLES HILL URANIUM DEPOSIT, VIRGINIA

Assessing the baseline environmental geochemistry of an ore deposit prior to mining provides a record that can be used to establish operational monitoring and post-closure reclamation goals. Pre-mining baseline studies are distinct from studies to distinguish natural geochemical background or identify geochemical anomalies in that they include the full geologic history of the area, including ore formation, weathering processes, and anthropogenic influences.

Recently, a baseline geochemical study was conducted on surface water, stream sediments, and soils at the Coles Hill uranium deposit in Pittsylvania County, Virginia. Samples were collected within and distant from areas impacted by the deposit. Results suggest that the Coles Hill deposit creates a statistically significant signature of elevated U, Ba, V, Pb, and radionuclide concentrations near and downstream of the deposit in the media studied. However, the deposit lies on a fault between two geologic provinces, and comparisons between downstream and reference sites within the same geologic province show no significant difference in stream water and sediment element concentrations. Statistically significant temporal differences in stream water concentrations were also found in element concentrations between monthly collections, but variations can be mostly accounted for by differences in discharge, which is negatively correlated with many elements.

During this study, a number of challenges arose, many of which are relevant to other studies of this type. One particular issue is that “baseline” lacks a specific definition, and so there are a number of ways in which results can be presented and interpreted. Another difficulty was that the deposit is located in an area with diverse geology, land use, and climate, all of which can complicate environmental signatures. Selection of appropriate reference or background sites can aid in distinguishing these signals. Another challenge was the prevalence of non-detect concentrations in surface water. Non-standard statistical methods, particularly regression on order statistics, were employed to meaningfully and robustly analyze these data. However, comparisons of statistics calculated using this method show that as the fraction of data below detection increases, results become more uncertain.