Submarine groundwater discharge data at meter scale (223Ra, 224Ra, 226Ra, 228Ra and 222Rn) in Indian River Bay (Delaware, US)

Submarine groundwater discharge (SGD) was sampled at high-spatial resolution in Indian River Bay, DE, USA, in July 2016 to characterize the spatial variability of the activity of the radium and radon isotopes commonly used to estimate SGD. These data were part of an investigation into the methods and challenges of characterizing SGD rates and variability, especially in the coastal aquifer transition from freshwater to saltwater (Hydrogeological processes and near shore spatial variability of radium and radon isotopes for the characterization of submarine groundwater discharge (Duque et al., 2019)). Samples were collected with seepage meters and minipiezometers to obtain sufficient volumes for analytical characterization. Seepage meter samples (for 223Ra, 224Ra, 226Ra, and 228Ra) were collected at two-hour intervals over a semi-diurnal tidal cycle from 30 seepage meters. Samples for 222Rn characterization were collected with a minipiezometer from 25 cm below the bay bed at each seepage meter location. All samples were analyzed with standard and state of the art procedures.

below the bay bed at each seepage meter location. All samples were analyzed with standard and state of the art procedures.
© 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons. org/licenses/by/4.0/). Table 1 contains the locations, sample volumes, and radioactive tracer activities at each of 30 sampling locations (Fig. 1) where a seepage meter was installed in Indian River Bay (DE) on July 2016. Groundwater was collected from seepage meter bags for later laboratory analysis. A porewater sample from 25 cm depth was also collected during this sampling period near each seepage meter. Sample Specifications Table   Subject Water Science and Technology Specific subject area 223 Ra, 224 Ra, 226 Ra, 228 Ra, 222 Rn and specific conductance data collected from submarine groundwater discharge Type of data The field site was monitored at high spatial resolution for the collection of groundwater directly discharging to a shallow bay Description of data collection A grid of 6 Â 5 seepage meters was installed with 3 m distance between seepage meters and used to collect samples for measurement of SGD radium activity. Minipiezometers were used to collect porewater samples adjacent to seepage meters to analyse radon activities. All samples were collected over a semi-diurnal tidal cycle.  volumes depended on SGD rates, so collected volumes varied between the 30 locations ( Table 1). Activities of 223 Ra, 224 Ra, 226 Ra, 228 Ra, and 222 Rn were later determined in the laboratory and are presented, along with the total propagated uncertainty of the analytical methods ( Table 1). The specific conductance (SC) is reported for each sample, because it affects Ra activity and also act as a salinity proxy and indicator of the origin of discharging groundwater. A bay water sample, collected in proximity of the study area, is provided for comparison.

Experimental design, materials, and methods
This dataset presents the activities of natural radioactive tracers measured in groundwater discharging directly to Indian River Bay, DE, USAdnot water sampled from nearby wells or from surface water. This distinction is important, because when sampling wells at the coast, chemical processes that occur during flow through aquifers or mixing with surface water can generate differences between wells and discharging water-a full discussion can be found in Duque et al. [1]. The dataset shows the scale of variability in activity of 223 Ra, 224 Ra, 226 Ra, 228 Ra, and 222 Rn in this natural system, and the range of spatial variability that can be detected over short distancesdinformation that is essential for defining end members needed to estimate SGD with radioactive tracers.
The study area was selected for being relatively geologically homogeneous [3]. Aquifer salinity has important implications for Ra mobility [1], so seepage meter locations were selected to capture the fresh/saline transition of the submarine aquifer. Seepage meters were arranged in a 5 Â 10 grid (3 m spacing) covering a 180 m 2 (12 Â 27 m) area that was shallow (0.3e1.5 m water depth), bathymetrically approximately flat, and nearshore (within 20 m of the shore line). Inshore of the first seepage meter, a low hydraulic conductivity layer prevents SGD between the shoreline and study area. Each seepage meter was positioned with high precision real time kinematic (RTK) GPS.  Table 1) modified from Duque et al. [1].
Seepage meters were installed one week in advance of sample collection in the field site to allow the flushing of the seepage meter chamber that, based on the fluxes measured, was completed several times in all seepage meters. Each seepage meter was sampled five times collecting water for two hours over a semi-diurnal tidal cycle (9:00e11:00, 11:00e13:00, 13:00e15:00, 15:00e17:00, 17:00e19:00). Empty, labeled bags were used for sample collection. The use of empty bags may have an effect on SGD flux measurements, but this practice avoids contaminating collected samples with prefill water.
Samples for 222 Rn are sensitive to degassing, which likely occurs in seepage meters, so shallow porewater samples were collected from 25 cm depth (below the seabed) for 222 Rn analysis. We assumed this shallow porewater was representative of water being discharged from the aquifer. We collected the minimum volume required for the analysis to avoid drawing surface water into the sample. Samples were collected slowly with a minipiezometer (MHE products) and syringe using minimal suction to avoid degassing. Sampled water was immediately stored in 250-mL gas-tight bottles that were filled from the bottom and overflowed prior to capping to minimize degassing and atmospheric exchange of gases. A campaign laboratory was installed near the field area to immediately measure 222 Rn activities using a RAD7 radon detector with RAD H2O accessory. The analysis protocol was adapted to decrease the uncertainty of the 222 Rn content of the samples.
For analysis of Ra isotopes, Ra was pre-concentrated by adsorption from each water sample onto 20 g (dry weight) of Mn-oxide coated acrylic fiber which adsorbed Ra from the water in the field site directly after collection by passing the water via gravity feed at < 1L/minute through a cartridge containing the fiber [4e6]. 223 Ra and 224 Ra were measured on a Radium Delayed Coincidence Counter (RaDeCC) system using the protocols described by   [7] and   [8]. Initial 223 Ra and 224 Ra measurements were made within 10 days of collection, and all samples were run again within 3e6 weeks of collection to correct for 228 Th-supported 224 Ra activity. The error associated with each short-lived Ra isotope measurement was calculated using methods described by Garcia-Solsona et al. (2008) [9]. For 226 Ra and 228 Ra, the Mn-fiber samples were ashed at 700 C and sealed in polypropylene vials. Two high-purity Ge gamma spectrometers measured sample gamma emissions. Data were normalized to known quantities of certified NIST Ra solutions of 226 Ra and 228 Ra adsorbed to 20 g of Mn-oxide coated acrylic fiber. Specific activities and one sigma errors were calculated using standard counting techniques [10].