Bulk Aerosol Collection, Digestion, and Leaching Methods (reproduced from Marsay et al., 2022a (GBC vol 36, issue 2), 2022b (GBC vol 36, issue 8):
23 bulk aerosol sample collections were carried out during GP15, with an average of ∼40 hours of total sampling time spread over a period of 1-4 days per deployment. Samples were collected using a high-volume aerosol sampler (∼1.2 cubic meters of air per minute (m^3 air min-1); Tisch Environmental, model 5170V-BL) deployed on the forward rail of the ship's 03 deck, approximately 16 meters (m) above sea level. For each deployment, 12 replicate acid-washed 47-millimeter (mm) Whatman 41 ashless filters mounted on polypropylene filter holders (Advantec-MFS) were installed on the aerosol sampler upon a custom-built PVC adaptor plate, and an air volume averaging ∼240 m^3 was sampled through each filter. To avoid contamination from the stack exhaust, sampling was restricted to periods when relative wind direction was from the front of the ship's bow and at >0.5 meters per second (m s-1) for at least two minutes, as measured by an anemometer and wind vane interfaced with the sampler through a Campbell Scientific CR800 data-logger. All loading and unloading of filter holders took place inside a plastic "bubble" constructed in the ship's main laboratory and supplied with HEPA-filtered air, and filters were handled using acid-washed tweezers. Following sample recovery, filters were stored frozen until further processing back on land.

Triplicate 47 mm filters from each deployment were treated with a three-step strong acid digestion protocol, based on that of Morton et al. (2013), in a Class-100 clean room facility at Skidaway Institute of Oceanography. Briefly, each filter was transferred to a 15 milliliter (mL) perfluoroalkoxy (PFA; Savillex) vial and sequentially digested on a hotplate at 140 degrees Celsius (°C) with: (a) 1,000 microliter (μL) concentrated double-distilled nitric acid (d-HNO3); (b) 500 μL concentrated d-HNO3 with 100 μL concentrated hydrofluoric acid (Fisher Trace Metal Grade) and 100 μL concentrated hydrogen peroxide (Fisher Optima Grade); (c) 500 μL concentrated d-HNO3. Each heating stage was conducted overnight, with samples taken to dryness on the hotplate the following day. Following the third dry-down, residues were dissolved in 0.32 M d-HNO3 for analysis. Unused W41 filters were treated in the same way as samples and average filter blank values for each element were subtracted from every sample to account for the filter matrix and digest reagent contributions. 47 mm filters for each deployment were leached with ultrapure water to provide water-soluble trace elements following the methods of Marsay et al.(2022b), as modified from Buck et al. (2006). Briefly, aerosol filters were loaded onto a Savillex PFA filtration rig on top of a purpose built vacuum chamber, with a 0.2 micrometer (μm) polycarbonate membrane backing filter. A gentle vacuum was applied and 100 mL of ultrapure water was poured over the filter, collected, and then samples were acidfied to 0.024M HCl.

Size-fractionated aerosols (reproduced from Bunnell et al., submitted to GRL):
Size-fractionated aerosol samples were collected on five Whatman 41 cellulose ester stage filters and one Whatman 41 cellulose ester backing filter during 12 deployments using a Tisch Environmental, Inc High Volume Cascade Multi-Stage Particulate Size Fractionator Impactor (Series 230) set to collect 1.2 m^3 air per minute (Tisch Environmental, Inc, 2004; Gurganus et al., 2015). The impactors collect the size-fractionated aerosols onto one of five stage filters or backing filter. Exact size partitioning of each size filter requires additional analytical techniques outside the scope of this work, but the size cut offs for each stage can be estimated as: stage one (>7.2 μm), stage two (3.0-7.2 μm), stage three (1.5-3.0 μm), stage four (0.95-1.5 μm), stage five (0.49-0.95 μm), and backing filter (<0.49 μm) (Tisch Environmental, Inc, 2004; Gurganus et al., 2015). Stage and backing filters were cut into nine (stage 1) or ten (stages 2-5 and backing filter) strips using a ceramic knife and were processed in the same manner as bulk aerosol digests (Marsay et al., 2022a). Briefly, two stage filter strips (one for backing filter) were digested in a series of HNO3, HF-HNO3-H2O2, and HNO3 overnight dry-downs heated to 140oC in a Teflon beaker. After digestion, the samples were dried down before being redissolved in 13 mL of 2% HNO3.

Analysis for Fe, Zn, and Cd trace metals and stable isotope ratios:
Sample aliquots of a subset of digested aerosol laden GP15 filters (bulk or size-fractionated), or DI-water leachates were provided for analysis at the University of South Florida. Digests and DIW leaches were measured for Fe, Zn, and Cd stable isotopes and concentrations. Size fractionated samples were measured for Fe isotopes and concentrations - following MC-ICPMS measurement (see below), size-fractionated samples were binned according to filter stage (size). Data for coarse (filters >0.95 μm) and fine (<0.95 μm) aerosol size-fractions were calculated and reported here to match GEOTRACES parameters.

Samples were processed for dissolved Fe, Zn and Cd isotope and concentration analysis at the University of South Florida following Sieber et al. (2019), modified from Conway et al. (2013) and Conway et al. (2019). For all samples Fe, Zn, and Cd double spikes were added prior to chemical processing, in an approximate 1:2 sample:spike ratio based on expected concentrations from other analysis. Total digested aliquots were then evaporated to dryness, while leached samples were put through a batch extraction using Nobias PA-1 chelating resin and evaporated to dryness. All samples were then refluxed with a HNO3-H2O2 solution to digest organics, followed by purification by anion-exchange chromatography using AG-MP1 resin. Isotope analyses were then performed on a Thermo Neptune Plus MC-ICPMS in the Tampa Bay Plasma Facility at the University of South Florida using the double spike technique via a ~100 uL min-1 PFA nebulizer and Apex Ω (Fe, Zn) or Apex Q (Cd) introduction introduction system, Pt Jet (Fe, Cd) or Ni H (Zn) Sampler cone and an Al X (Fe) or Ni H (Cd, Zn) skimmer cone (Sieber et al., 2021; Sieber et al., 2023a; 2023b).

We express Cd stable isotope ratios in delta notation (δ114Cd) relative to the NIST SRM-3108 Cd standard. A secondary standard, BAM-I012, was analyzed over 8 sessions on the same timescale as the samples to provide an estimate of long-term instrumental precision. We obtain a value of −1.32 ± 0.06‰ (2SD, n = 172), in agreement with consensus values (Abouchami et al., 2013). We consider a 2SD uncertainty of 0.06‰ as a conservative estimate of analytical precision, and have applied it to all samples, except for low concentration samples where the larger internal error is considered a more conservative estimate of uncertainty. Concentrations were calculated using the isotope dilution technique based on on-peak blank, interference and mass-bias corrected 114Cd/111Cd ratios measured simultaneously with isotope analyses (Sieber et al., 2019). We express uncertainty (1SD) on Cd concentrations as 2% (Conway et al., 2013).

We express Fe stable isotope ratios in delta notation (δ56Fe) relative to the IRMM-014 standard. A secondary Fe standard, NIST-3126, was analyzed over 44 sessions to provide an estimate of long-term instrumental precision. We obtain a value of +0.36 ± 0.05‰ (2SD, n = 604), in agreement with consensus values (Conway et al., 2013). We consider a 2SD uncertainty of 0.05‰ as an estimate of analytical precision, and have applied it to all samples, except for low concentration samples where the larger internal error is considered a more conservative estimate of uncertainty. Concentrations were calculated using the isotope dilution technique based on on-peak blank, interference and mass-bias corrected 57Fe/56Fe ratios measured simultaneously with isotope analysis. We express uncertainty (1SD) on Cd concentrations as 2% (Conway et al., 2013).

We express Zn stable isotope ratios in delta notation (δ66Zn) relative to the JMC-Lyon standard. A secondary Zn standard, AA-ETH, was analyzed on the same timescale as the samples (over 10 sessions) to provide an estimate of long-term instrumental precision. We obtain a value of +0.28 ± 0.03‰ (2SD, n = 147), in agreement with consensus values (Archer et al., 2017). As a second estimate of external precision, we use the 2SD of offsets from the mean of full replicate measurements based on 26 pairs of replicate analysis using separate seawater samples collected at the same depth (0.03‰), which is similar to the analytical precision. Therefore, we consider a 2SD uncertainty of 0.03‰ as an estimate of analytical precision, and have applied it to all samples, except for low concentration samples where the larger internal error is considered a more conservative estimate of uncertainty. Concentrations were calculated using the isotope dilution technique based on on-peak blank, interference and mass-bias corrected 67Zn/66Zn ratios measured simultaneously with isotope analysis. We express uncertainty (1SD) on Zn concentrations as 5%, based on replicate analysis on separate seawater samples collected at the same depth (n = 26).

Fe stable isotope ratios are expressed in delta notation (δ56Fe) relative to the IRMM-014 standard. A secondary Fe standard, NIST-3126, was analyzed over 44 sessions to provide an estimate of long-term instrumental precision. We obtain a value of +0.36 ± 0.05‰ (2SD, n = 524; runs = 37), in agreement with consensus values (Hunt et al. 2022; Conway et al., 2013). We consider a 2SD uncertainty of 0.05‰ as an estimate of analytical precision, and have applied it to all samples, except for low concentration samples where the larger internal error is considered a more conservative estimate of uncertainty.