An evaluation of the efficacy of radium isotopes as tracers of coastal mixing and submarine groundwater discharge

Eddy diffusion and submarine groundwater discharge (SGD) are critical processes in coastal areas. Over the past several decades, radium isotopes (223Ra, t1/2 = 11 d; 224Ra, t 1/2 = 3.7 d; 226Ra, t1/2 = 1600 yr; 228Ra, t1/2 = 5.7 yr) have been increasingly used as tracers for both eddy diffusion and SGD, in part because these isotopes exhibit near-conservative behavior in seawater. However, there are a number of critical assumptions behind the use of radium as a tracer for these processes; these assumptions must be evaluated in order to ensure the efficacy of a radium-based approach.;223Ra and 224Ra distributions in the Mid-Atlantic Bight measured during January, February, and April of 2007 yield cross-shelf eddy diffusivity (K) values ranging from 0.1+/-0.05 - 1.6+/-0.5 x 102 m2 s-1 (223Ra) and 1.7+/-0.4 - 2.2+/-0.6 x 102 m2 s-1 (224Ra). Non-dimensional scaling reveals that advection and benthic radium input do not have a significant effect on these estimates. The temporal variability in K is low in comparison to the uncertainty of the calculated values. This trend, in concert with the similarity between these values and prior estimates made in the region, suggests that that eddy diffusivity in this area is relatively constant over months, years, and even decades. Observations in the Mid-Atlantic Bight differ from theoretical data corresponding to the tidal dispersion frontogenesis model, suggesting that a different mechanism is responsible for mid-shelf front formation. Multiple interpretations, including reduced cross-shelf mixing and changing mixing regimes, can account for the cross-shelf radium distribution observed near a mid-shelf front located within the study area. Radium isotope measurements in the equatorward jet associated with the front indicate that this jet can effectively transport chemical components over hundreds of kilometers.;The general trends in the K values determined here are comparable to those in previous radium-based studies; 224Ra-based eddy diffusivities invariably exceed those calculated using 223Ra. This runs counter to two views of eddy diffusion. The Fickian model, in which K is constant, dictates that 223Ra- and 224 Ra- based eddy diffusivities should be equal, while the length scale-dependent model indicates that 223Ra-based K values should exceed 224Ra-based estimates, due to the relatively larger length-scale of the 223Ra distribution. Differential diffusion provides an explanation for this discrepancy. A simple finite mixing-length model based on this concept successfully predicts the difference between eddy diffusivities calculated using 223Ra and 224Ra.;226Ra and 228Ra measurements, along with a salt balance, were also used to yield estimates of fresh and saline SGD fluxes to four of Rhode Island's coastal ponds. Fresh SGD estimates made using this methodology fall in line with past values determined using a hydrologic model and a Darcy's law-based approach. The range in saline SGD estimates calculated for each pond is high; maximum and minimum saline SGD values for an individual pond differ by as much as three orders of magnitude. The wide range in saline SGD values coincides with the discrepancy in results from two prior radium-based SGD studies in this location. The high uncertainty in saline SGD estimates is attributed to the variability in groundwater radium activities observed here. Groundwater 226Ra and 228 Ra activities in the ponds are highly variable over both lateral distance and with depth in the sediment. These results indicate that an individual groundwater sample, or average of samples, may not accurately represent the average radium activity of the groundwater actually discharging into each pond, thus affecting the accuracy of radium-based SGD estimates.