Water column sampling:
The full water column of 27 stations in the Amundsen Sea was sampled between 8 December 2023 and 13 January 2024 aboard the R/V IB Nathaniel B. Palmer. Depth profile samples were collected from the GEOTRACES trace metal clean rosette sampling system (GTC CTD) outfitted with 24 modified 12-liter (L) GO-FLO (General Oceanics) sampling bottles (Cutter and Bruland, 2012). Surface water (~2 meters) samples for the depth profiles were collected using a trace metal clean surface pump "towfish" system (FISH).

From the trace metal clean rosette and towfish, samples for humic-like substances were filtered (0.2 micrometer (µm), Acropak) inline and collected inside a trace metal clean sampling van. Samples were collected in acid-cleaned, Milli-Q-conditioned, and triple-rinsed narrow-mouth fluorinated high-density polyethylene bottles (FPE; Nalgene) and analyzed shipboard for dissolved iron speciation by Dr. Léo Mahieu in Dr. Kristen Buck's lab before freezing at -20 degrees Celsius (ºC) for shore-based humic-like substance analyses at Oregon State University.

Sample analyses – humic-like substances:
Humic-like substances were measured using cathodic stripping voltammetry following the method first developed by Laglera et al. (2007) and updated by Sukekava et al. (2018). All samples were analyzed using a polarographic 797 VA Computrace (Metrohm) stand equipped with a hanging mercury drop electrode, platinum rod auxiliary electrode, Ag/AgCl reference electrode, and an acid-cleaned Teflon analytical cell. Briefly, frozen samples were set out to thaw overnight at room temperature in the dark. The following day, 10 milliliter (mL) sub-samples were aliquoted into each of 2 paired acid-cleaned, pre-conditioned 50 mL polypropylene vials. The pH of the samples was then set to 8.2 with the addition of 50 microliters (μL) of 1.5 molar (M) boric acid buffer. Iron standard was added to one of the paired vials (vial B, final concentration 60 nanomolar (nM)) to saturate all the humic-like substance binding groups in the solution. Both aliquots were then left to equilibrate for a minimum of 14 hours. After the equilibration period, 500 μL of the 0.4M KBrO₃^-catalyst was dispensed into the cell with each aliquot immediately prior to analysis. Electrochemical analysis was performed using linear sweep voltammetry (-0.1 to -1 volts (V), scan rate 50 millivolts per second (mV s-1)) with a 120-second deposition period with stirring at -0.1V. Measurements were run in triplicate with a 300-second purge step with high-purity N₂ gas prior to analyzing each aliquot. For each sample, the Fe-free aliquot (vial A) was analyzed first, immediately followed by the Fe-saturated aliquot (vial B). Three standard additions of Fe-saturated SRFA standard (0.1, 0.2, and 0.3 milligrams (mg) SRFA per liter) were then added successively to the Fe-saturated aliquot to generate a calibration curve. A 70s N₂ purge was included after each addition to ensure oxygen removal and all additions were measured in triplicate. The Teflon analytical cell was rinsed thoroughly with Milli-Q water between samples to minimize carryover.

Peak heights were extracted from scans using ECDSOFT, then converted into the concentration of electroactive humic substances (micrograms (μg) eq SRFA^-1) using the slope of the curve generated by the three standard additions. The standard addition slope was applied to the peaks obtained for both vials A and B to obtain the ambient Fe-binding humic and total Fe-binding humic concentrations, respectively, in mg SRFA eq per liter. Units were then adjusted to nM Fe eq using the measured binding capacity of SRFA in seawater 14.6 ± 0.4 nM Fe mg SRFA^-1. Variability between triplicate scans averaged 0.02 nM Fe eq for both ambient and total humic results and variability between replicate samples averaged 0.05 (median 0.03) nM Fe eq and 0.06 (median 0.04) nM Fe eq for ambient and total humic-like substances, respectively. The limit of detection (LOD) was calculated as three times the standard deviation of triplicate scans in the lowest measured ambient Fe-binding humic concentration (0.06 ± 0.02 nM Fe eq) and was estimated to be 0.05 nM Fe eq. All measured values were above the LOD.