Laboratory Experiments to Evaluate Diffusion of 14C into Nevada Test Site Carbonate Aquifer Matrix

Abstract

Determination of groundwater flow velocities at the Nevada Test Site is important since groundwater is the principal transport medium of underground radionuclides. However, 14C-based groundwater velocities in the carbonate aquifers of the Nevada Test Site are several orders of magnitude slower than velocities derived from the Underground Test Area regional numerical model. This discrepancy has been attributed to the loss or retardation of 14C from groundwater into the surrounding aquifer matrix making 14C-based groundwater ages appear much older. Laboratory experiments were used to investigate the retardation of 14C in the carbonate aquifers at the Nevada Test Site. Three sets of experiments were conducted evaluating the diffusion of 14C into the carbonate aquifer matrix, adsorption and/or isotopic exchange onto the pore surfaces of the carbonate matrix, and adsorption and/or isotopic exchange onto the fracture surfaces of the carbonate aquifer. Experimental results a nd published aquifer matrix and fracture porosities from the Lower Carbonate Aquifer were applied to a 14C retardation model. The model produced an extremely wide range of retardation factors because of the wide range of published aquifer matrix and fracture porosities (over three orders of magnitude). Large retardation factors suggest that groundwater with very little measured 14C activity may actually be very young if matrix porosity is large relative to the fracture porosity. Groundwater samples collected from highly fractured aquifers with large effective fracture porosities may have relatively small correction factors, while samples from aquifers with a few widely spaced fractures may have very large correction factors. These retardation factors were then used to calculate groundwater velocities from a proposed flow path at the Nevada Test Site. The upper end of the range of 14C correction factors estimated groundwater velocities that appear to be at least an order of magnitude too high compared to published velocities. The lower end of the range of 14C correction factors falls within the range of reported velocities. From these results, future experimental studies (both laboratory and field scale) to support 14C groundwater age dating should focus on obtaining better estimates of aquifer properties including matrix and fracture porosities.