Using dual-domain advective-transport simulation to reconcile multiple-tracer ages and estimate dual-porosity transport parameters

Dual-domain transport is an alternative conceptual and mathematical paradigm to advection-dispersion for describing the movement of dissolved constituents in groundwater. Here we test the use of a dual-domain algorithm combined with advective pathline tracking to help reconcile environmental tracer concentrations measured in springs within the Shenandoah Valley, USA. The approach also allows for the estimation of the three dual-domain parameters: mobile porosity, immobile porosity, and a domain exchange rate constant. Concentrations of CFC-113, SF6, H-3, and He-3 were measured at 28 springs emanating from carbonate rocks. The different tracers give three different mean composite piston-flow ages for all the springs that vary from 5 to 18 years. Here we compare four algorithms that interpret the tracer concentrations in terms of groundwater age: piston flow, old-fraction mixing, advective-flow path modeling, and dual-domain modeling. Whereas the second two algorithms made slight improvements over piston flow at reconciling the disparate piston-flow age estimates, the dual-domain algorithm gave a very marked improvement. Optimal values for the three transport parameters were also obtained, although the immobile porosity value was not well constrained. Parameter correlation and sensitivities were calculated to help quantify the uncertainty. Although some correlation exists between the three parameters being estimated, a watershed simulation of a pollutant breakthrough to a local stream illustrates that the estimated transport parameters can still substantially help to constrain and predict the nature and timing of solute transport. The combined use of multiple environmental tracers with this dual-domain approach could be applicable in a wide variety of fractured-rock settings. Plain Language Summary The movement of dissolved chemicals, such as pollutants, underground has been notoriously difficult to predict in terrains composed of fractured rock because of the complex way in which the chemicals mix between fractures and the host rock. One evidence for this has been that when different atmospheric-derived tracers are measured in springs in karst terrains they give different apparent times of travel between the land surface and the spring. The authors of this paper have presented a new approach that can both partially characterize the nature of how such chemicals are mixing underground and reconcile the difference in apparent times of travel. The authors took a widely used model that can simulate the flow of groundwater and transport of chemicals without mixing, and applied it the upper Potomac River Basin where they had tracer data from 28 springs. They wrote a new routine that calculates the mixing of the chemicals in transit and added it to the flow lines predicted by the simulation model. They were able to reproduce the tracer concentrations in the springs, partially constrain the amount of void space in both the fractures and solid rock, and estimate the rate at which chemicals exchange between these two.