Samples from the submarine groundwater wells and overlying water column were collected quarterly from November 2022 through March 2024 aboard U.S. Geological Survey (USGS) small boats. This sampling effort was led by collaborator Chris Smith at USGS. Water column samples were collected by lowering the sampling line, acid-cleaned ½” OD Bev-A-Line tubing attached to ¼” synthetic line that was weighed down by a 30 lb kettlebell to target depths. Surface samples were collected at ~1 m depth and bottom samples were collected from ~1 m above the sea floor. A separate sampling line of acid-cleaned 3/8” OD Bev-A-line connected to a custom-made PVC adapter was used for sampling the submarine groundwater wells, which had been installed by USGS. Different Teflon diaphragm pumps were used for the water column and wells samples to avoid cross contamination. Samples for dissolved trace metals were filtered through 0.2 µm Pall Acropak Super membrane filter capsules and collected in acid-cleaned and triple-rinsed 125 mL low density polyethylene (LDPE, Nalgene) bottles. Sample collection and processing for riverine and estuarine end members were accomplished by collaborators from the University of South Florida as described in the related dataset, Conway and Hunt (2026).
All dissolved trace metal samples were acidified within 24 h of collection with ultrapure hydrochloric acid (HCl, Optima) to 0.024 N HCl, double bagged in buckets and stored at room temperature until analyzed at Oregon State University (OSU).
Analysis of dissolved trace metal concentrations
Concentrations of dissolved (<0.2 µm) trace metals, including aluminum (Al), scandium (Sc), manganese (Mn), iron (Fe), cobalt (Co), nickel (Ni), copper (Cu), zinc (Zn), molybdenum (Mo), cadmium (Cd), barium (Ba), lutetium (Lu) and lead (Pb), were analyzed using direct dilution analysis with a Thermo Scientific iCAP-RQ Quadrupole CCT ICP-MS. This instrument is maintained at OSU in the Keck Collaboratory for Plasma Spectrometry.
Samples were diluted 1:20 with 2% nitric acid (HNO3, Optima) and quantified with nine-point calibration curves made in the same acid matrix but spiked with seawater to match matrix salinity of diluted samples; 1 ppb indium was added to all samples and standards as an internal standard. Reference materials (NASS-7, CASS-6) were measured in each run following the same dilution protocol. Blanks were determined from separate measurements of the 2% HNO3 matrix including 1 ppb In. Reference material results, averaged blanks and limits of detection are provided in the supplemental Tables.
Samples were analyzed a second time following the same procedure outlined above but after 90 minutes of UV-oxidation to ensure complete recovery of any organic-bound Cu and Co (Biller and Bruland 2012). The UV-oxidation was carried out on aliquots of samples in 15-mL Teflon vials (Savillex) with custom quartz lids (Hollister et al. 2020). For all elements except Cu and Co, results from analyses after UV-oxidation were combined with no UV results to achieve an additional analytical replicate.
Sample analyses for trace metals were performed by senior researcher Dr. Salvatore Caprara in the College of Earth, Ocean and Atmospheric Sciences at Oregon State University.
