Sample collection and preparation:
Each water sample was filtered directly from the trace metal clean GTC rosette/Go-Flo bottle sampler through a 0.45 micrometer (µm) Supor membrane filter (25 millimeters (mm)) or 0.2 µm Pall Acropak-200 Supor cartridge into a trace metal grade acid-cleaned 4-liter (L) polycarbonate bottles. Samples for solid phase extracted trace element ligands (TELs) were pumped at 20 milliliters per minute (mL/min) through Bond-Elut ENV solid phase extraction (SPE) columns (1 gram (g), 6 mL, P/N 12255012, Agilent Technologies) that had been previously activated by passing 6 mL each of distilled methanol (MeOH, Optima LCMS grade, Fisher Scientific), 0.1% HCl (VWR chemical, Aristar Ultra grade), and ultrapure water (qH₂O, 18.2 MΩ) through the column. SPE columns were frozen (-20 degrees Celsius (°C)) immediately after sample collection and returned to the laboratory for processing.

Siderophore Processing:
SPE columns were thawed and concentrated TELs were then eluted with 10 mL distilled MeOH into 15 mL trace metal clean polypropylene tubes. Process blanks were prepared in parallel by eluting activated SPE columns with 10 mL MeOH. The methanol fraction was collected as the process blank. The eluted sample was concentrated down to ~1000 microliters (µL) by vacuum centrifugal concentrator (CentrIVap Evaporating System, LABCONCO).

A 10 µL stock solution of 100 micromolar (µM) cyanocobalamin was added to each sample as an internal standard.

Quantitative analyses of siderophores:
Chromatographic analyses were performed on a bioinert Dionex Ultimate 3000 liquid chromatograph (LC) system fitted with a loading pump, a low flow pump, and a 10-port switching valve. Full loop (50 µL) samples were injected using the 50 µL sample loop, then separated using a BEH C18 column (ACQUITY UPLC BEH C18, 130A^o, 1.7 µm, 2.1 mm X 50 mm, P/N 186002350, Waters) connected with a 5 mm guard column (ACQUITY UPLC BEH C18 VanGuard, 130A^o, 1.7 µm, 2.1 mm X 5 mm, P/N 186003975, Waters) by the low flow pump at 50 μL/min. 5 millimolar (mM) ammonium formate in Milli-Q water and LC-MS-grade methanol were used as solvents A and B, respectively. Samples were separated using a 40 minute gradient flow (95% A/ 5% B to 5% A/ 95% B) followed by 9 minute washout at 95% B. Meanwhile, the loading pump was used to supply an internal standard (4 parts per billion (ppb) Indium in 1% HNO3 as solvent C) at a constant rate and maintain 26% of total organic (LC-MS grade MeoH) flow by mixing loading pump flow (150 µL/min) and low flow (50 µL) post-column before infusing into the ICPMS.

The combined flow from the LC (200 μL/min) was analyzed using a Thermo Scientific iCAP Q ICPMS fitted with a perfluoroalkoxy micronebulizer (PFA-ST, Elemental Scientific), and a cyclonic spray chamber cooled to 0 °C. Measurements were made in kinetic energy discrimination (KED) mode, with a helium collision gas flow of 4-4.5 mL/min to minimize isobaric interferences. Oxygen was introduced into the sample carrier gas at 25 mL/min to prevent the formation of reduced organic deposits onto the ICPMS skimmer and sampling cones.

The metal detector response was calibrated using standards, 10 ppb, 20 ppb, 50 ppb, 100 ppb, and 200 ppb, prepared from a commercial stock solution (InorganicVentures). These standards were analyzed using a method similar to other samples, but no column was connected and eluted with 5% of solvent B and 95% of solvent A for five minutes. A plot of the metal peak areas against concentration yielded a linear relationship (r² ~0.99) for the response of the ICPMS detector as calibration curve. Calibrations and process blanks were made for every 10-20 samples analyzed, with only small changes (RSD ~30%) were observed in the slope of the calibration relationship over the course of the two years of sample analysis. Concentrations of TELs in each sample were calculated by integrating the metal peak area (detector response) of interested time range and using appropriate calibration curve for the peak area.