An Evaluation of the Feasibility of the Time-Lapse Electrical Resistivity Tomography Method in Quantifying Submarine Groundwater Discharge in Fine Sediment and Highly Saline, Shallow Embayments

There is uncertainty surrounding the application of the time-lapse electrical resistivity tomography (ERT) method in quantifying submarine groundwater discharge (SGD). The technique has been proven effective in areas with significant differences in salinity between surface water and discharging groundwater. However, there are inherent limitations associated with the method when studying embayments with fine bottom sediments, highly saline porewaters, and shallow surface waters. To evaluate this approach, a 68-hour time- lapse ERT study was conducted at University Beach (UB) in Corpus Christi, Texas, constrained by concomitant measurements of naturally occurring isotope tracers and groundwater characterizations along the ERT profile. Surface water continuous 222Rn measurements were conducted to calculate SGD rates. Subsurface fluid conductivity measurements along the profile were constrained using inversion models via Archie’s Law (AL) and the Waxman-Smits equation (WSE). The average ERT-derived SGD rate for the study period was 26±1cm∙d-1. Average tracer-derived rates using the deep well endmember were 69±26, 49±1, 80±2, and 21±1 cm∙d-1 corresponding to 222Rn, 223Ra, 224Ra, and 226Ra, respectively. Over the course of the study 222Rn SGD estimates increased, while ERT and radium isotope estimates tended to decrease. There were no significant differences between the model accuracy of AL (R2=0.5, p<0.01) and the WSE (R2=0.5, p<0.01), thus there is no evidence that negatively charged clay particles in the subsurface matrix had any latent influence on the ERT measurements. Temperature, however, was found to be the predominant source of error in the ERT. The influence of temperature is often evident in the uppermost 1-2 m of images as temperature inflections drive the conductive recirculation of seawater. When temperature corrections were well constrained, the inversion models were significantly more successful at deriving modeled subsurface fluid conductivities closer to those observed, reducing error by up to 20%. Consequently, precise constraints on temperature are necessary to perform an effective experiment. ERT ultimately produced conservative SGD estimates relative to the radiogenic tracers. This is related to a lack of contrast between ambient and groundwater subsurface fluid conductivities. Thus, in the investigated environment, ERT may be used for qualitative and exploratory purposes, in the absence of other validating in-situ measurements.