Methodology:

Surface seawater for particulate and dissolved metals, radium, dissolved micronutrients, salinity, and stable oxygen isotopes were collected by hand using zodiac boats during the JR19002 Icebergs 3 cruise (Supplementary Table S5). Acid-cleaned bottles were rinsed and filled with surface seawater over the bow to avoid contamination from the Zodiac. Bottles were immediately capped and returned to the ship and processed.

For total particulate digest, 4 L of seawater was pressure filtered onto 25mm Supor (PES) filters (0.45 µm pore size) and frozen immediately. filters were frozen immediately at -20 degrees C. Dissolved phase samples were filtered using a 0.2 micrometers Acropak cartridge filter and peristaltic pump and collected into acid-washed 125mL HDPE bottles (dissolved metals). All sampling and handling was performed according to GEOTRACES protocols.

Three PES filters per site were subject to complete acid digestion using 10 percent HF/50 percent HNO3 (UpA, distilled) at 135 degrees C for 4 h following GEOTRACES protocol. Elemental composition was quantified by sector-field ICP-MS (Finnigan MAT Element 1) analysis. Nine multi-element external standard dilutions were created gravimetrically from a stock calibration standard to calibrate the range of sample concentrations. Samples were corrected for blank signals using blanks for acids, PTFE vials, and filters. Indium (1 ppb) was used as internal standard to correct for drift and matrix effects.

Concentrations for dissolved metals were determined using SeaFAST preconcentration followed by ICP-MS analysis. Iron was quantified using isotope dilution, while Mn used external matrix-matched standards. Analytical replicates were run around every sixth sample and were typically 1-3 percent deviation about the mean. Precision, from repeat measurements of an in-house seawater standard was typically 3 percent or better, and accuracy was assessed by repeat measurements of consensus standards SAFe S and SAFe D2.

Salinity measurements from Zodiac boat surface samples were analysed on a Guildline 8400B salinometer. Salinity is here presented on the Practical Salinity scale.

Samples for stable oxygen isotopes were stored in 50 ml glass vials sealed with stoppers and crimps. Oxygen isotope samples were analysed around 9 months later at the UK's National Environmental Isotope Facility at the British Geological Survey, using the CO2 equilibration method with an Isoprime 100 mass spectrometer plus Aquaprep device. Isotope measurements were calibrated against internal and international standards including VSMOW2 and VSLAP2. Based on duplicate analysis, analytical reproducibility was less than 0.05 percent for all samples.