All trace metal-clean seawater samples in this dataset were collected following the prescribed US GEOTRACES protocol (\"GEOTRACES cookbook\": for more information see (Cutter et al., 2010; Cutter and Bruland, 2012)). Briefly, 24 Teflon-coated, acid-cleaned and preconditioned GO-FLO bottles were deployed on an epoxy-coated aluminum rosette and were tripped on ascent at 3 m/min. Once on deck, bottles were immediately capped with plastic shower caps on both ends to reduce contamination and subsequently kept in a clean sampling van under positive-pressure, HEPA-filtered air. All handling and subsampling was performed exclusively by designated, trace metal-clean personnel to preserve homogeneity and further reduce contamination.
\n
During sampling for the GO-Flo samples, salinity and nutrient (unfiltered) samples were taken first, and then all Go-Flo bottles were pressurized to ~0.7 atm with HEPA-filtered air. Each bottle was fitted with an acid-cleaned and seawater-preconditioned 0.2 um AcroPak-200 capsule filter (Pall), and at least 500 mL was passed through each filter to rinse. Samples were collected in acid-cleaned (hydrochloric acid, 1 M (Fitzsimmons and Boyle, 2012)) LDPE bottles (Nalgene) after three 10% rinses of the bottle, cap, and threads and filled to the shoulder or as volume allowed.
\n
In the absence of sea ice, surface samples were collected in trace metal clean fashion by taking a small boat upstream of the ship. The small boat was pointed into the oncoming current to sample \"uncontaminated water\" as it passed the boat, and a trained trace metal personnel wearing trace metal clean gloves and plastic sleeves submerged an acid-cleaned carboy under the water surface from the bow of the boat, always keeping the carboy pointed into the oncoming water. The carboy was immediately capped and bagged for transit back to the ship, where it was immediately filtered (0.2 um, Acropak-200) and sub-sampled into individual bottles, including our acid-clean 250 mL LDPE bottles.
\n
A small subset of samples from designated \"ice stations\" (Stations 31, 33, 39, 43) were collected under the ice (approx. 1, 5, and 20 m) after the ice was drilled with a polypropylene/titanium trace metal coring system. Sampling was done using a polypropylene, battery-powered motor centrifugal pump with \u00bd inch FEP-lined Tygon tubing. Samples were collected and filtered (0.2 um, Acropak-200) into a 25 L acid-cleaned carboy and then sub-sampled back on the ship, including into our acid-clean 250 mL LDPE bottles, within 3 hours of collection.
\n
At these same \"ice stations\", melt pond samples were collected (Marsay et al., 2018) by clearing surface snow with an acid-cleaned polyethylene shovel and then using a polyethylene/titanium trace metal coring system to drill through the upper ice. Melt pond water was pumped using a battery-powered polyethylene pump through pre-cleaned C-flex tubing into a pre-cleaned LDPE carboy. The melt pond sample in this carboy was filtered and sub-sampled into individual bottles, including our acid-clean 250 mL LDPE bottles, within 3 hours of collection.
\n
Subsequently, all samples were acidified under clean air conditions to pH~2 with 0.012 M HCl (Optima, Fisher Scientific) within 1 week of sample collection and stored at room temperature until analyzed.
\n
At least 9 months after acidification, samples were analyzed for trace metal (Fe, Ni, Cu, Zn, Pb, Cd, Mn) concentrations at Texas A&M University after pre-concentration on a SeaFAST-pico system (Elemental Scientific Inc.). This method follows a modified version of Lagerstr\u00f6m et al. (2013) and is described completely for all elements in Jensen et al. (2020). Briefly, 10 mL of acidified seawater sample was weighed and then spiked with an isotope mixture containing 57Fe, 62Ni, 65Cu, 68Zn, 206Pb, and 111Cd. The mixture was then automatically loaded into the SeaFAST system. Monoisotopic metals such as Mn were measured via a 6-point standard curve using standards made up in low-metal seawater (maintained in-house) such that Mn concentrations spanned 0 to 10 nmol/kg; these standard curves were each run through the SeaFAST twice (once to start and once to end each run) as samples. The SeaFAST method buffers the acidified (pH~2) sample with a ~5.90 N ammonium acetate buffer (Optima, Fisher Scientific) to pH ~6.5 and loads the buffered mixture onto a pre-cleaned column containing Nobias PA1 resin (Sohrin et al., 2008). Each sample is then back-eluted with a solution of 10% (v/v) nitric acid and 1 ppb In (Optima, Fisher Scientific) to yield 400 uL, representing a 25-fold pre-concentration of each trace metal. Within days, each eluent is analyzed in low (204Pb, 206Pb, and 111Cd, 114Cd, 115In) and medium resolution (55Mn, 56Fe, 57Fe, 60Ni, 62Ni, 63Cu, 65Cu, 66Zn, 68Zn, 115In) on a Thermo Element XR high resolution ICP-MS in the Ken Williams Radiogenic Facility in the College of Geosciences at Texas A&M University.
\n
Each sample run included ~80 aliquots from the SeaFAST. Every 10 SeaFAST samples on the XR, pure 10% (v/v) nitric with 1 ppb of 115In was run to extract a potential instrument blank coming from the instrument itself. Prior to running on the ICP-MS, the machine was \u201ccleaned\u201d with this mixture as a precaution against contamination from other samples, in addition to using a separate set of cones, nebulizer, spray chamber, and sample probe for SeaFAST samples only.
Concentrations of dissolved micronutrient trace metals (Fe, Zn, Ni, Cu, Cd, Pb, Mn) in seawater, sea ice, and melt ponds\u00a0collected on\u00a0the US GEOTRACES Arctic cruise (HLY1502, GN01) from August to October 2015.
\n
These data have been published in the following:
\nMarsay et al., 2018 \u2013 Melt pond metal concentrations
\nKadko et al., 2019 \u2013 Surface seawater metal concentrations
\nJensen et al., 2019 \u2013 Dissolved Zn seawater concentrations
\nZhang et al., 2019 \u2013 Dissolved Cd seawater concentrations
\nCharette et al., 2020 \u2013 Upper ocean dissolved metal concentrations
\nJensen, 2020 (PhD Dissertation)
Data Processing:
\nThe raw counts from the HR-ICP-MS analyses were first corrected for any instrument blank derived from the 10% nitric acid and In solution repeatedly analyzed during the instrumental analyses. Following this, each sample was corrected for machine drift and variation in instrumental sensitivity during analysis using the added 115In as well as instrument mass bias using elemental standards for Fe, Zn, Ni, Cu, Pb, and Cd. Subsequently, all counts for isotope spiked elements were converted to concentration in nmol/kg based on the weight of the sample and known concentrations of the added isotope spike mixture. The counts for the monoisotopic elements were converted to concentration units using the matrix-matched standard curves, which were run through the SeaFAST themselves as samples. It should be noted that all Cd concentrations are reported based on the 114/111Cd ratio.
\n
For every 40 seawater samples at least 24 \"Procedure blanks\" were included throughout the run (every 10 seawater samples). These process blanks consisted of freshly acidified milli-Q (MQ) water spiked with a diluted version of the isotope spike mixture. Each blank was processed identically to a sample in an attempt to account for contamination coming from solutions (buffer, eluent) used or the SeaFAST itself. The MQ water is typically collected from various supplies at Texas A&M to ensure that the \"procedure blank\" is not derived from the MQ water itself. After conversion to concentration units, the most consistent procedure blank values for each metal are used to subtract an average procedure blank from all sample concentrations determined during that run. Each run, an internal standard was analyzed six times and an external standard (such as SAFe water) was analyzed two times as a \"check\" of our long-term precision and accuracy.
\n
Quality Control:
\nFor detection limits, accuracy, and precision information, refer to the Supplemental File Detection_Limits_Accuracy_Precision_817259 (PDF)
\n
Quality Flags:
\nData were flagged using the SeaDataNet quality flag scheme. For more information on SeaDataNet flags, see: https://www.geotraces.org/geotraces-quality-flag-policy/ and https://www.seadatanet.org/Standards/Data-Quality-Control
\n
SeaDataNet quality flag definitions:
\n0 = No quality control;
\n1 = Good value;
\n2 = Probably good value;
\n3 = Probably bad value;
\n4 = Bad value;
\n5 = Changed value;
\n6 = Value below detection;
\n7 = Value in excess;
\n8 = Interpolated value;
\n9 = Missing value;
\nA = Value phenomenon uncertain.
\n
BCO-DMO Processing:
\n- renamed fields;
\n- added date/time columns in ISO8601 format;
\n- 2021-04-15 (v2): made the following corrections:
\nchanged Event_ID to 6449 for Sample_ID numbers 12147, 12151, 12155, and 12161;
\nadded Start_Time_UTC, Start_Latitude, and Start_Longitude to Event_ID 6405 (these values were obtained from dataset https://www.bco-dmo.org/dataset/762056)