CTD data:
The temperature and conductivity probes are calibrated annually by the manufacturer while the oxygen sensor was calibrated against discrete seawater samples analyzed for dissolved oxygen concentrations by Winkler titration (Carpenter 1965; Grasshoff et al. 1983).

Nutrients:
Duplicate samples were measured for nitrate+nitrite on different days, and the standard deviation for duplicates was <0.5 µM, with a lower standard deviation for lower concentration samples
Duplicate samples for phosphate and nitrite were measured in duplicate on different days of analysis, yielding a standard deviation for duplicates of ≤0.1 µM. 

Nitrate isotopes:
The N and O isotope ratios of nitrate were measured in triplicate in separate batch analyses and standard deviations for both δ15N and δ18O were <0.3‰. Certified nitrate isotope ratio reference materials in nutrient-free seawater were measured in all batch analyses. These included IAEA-NO3-, with δ15N of 4.7 ± 0.2‰ vs. N2 air (Gonfiantini et al. 1995) and and δ18O of 25.6 ± 0.4‰ vs. VSMOW (Böhlke et al. 2003), and USGS-34, with δ15N of -1.8 ± 0.1‰ vs. N2 air and δ18O of -27.9 ± 0.3‰ vs. VSMOW (Böhlke et al. 2003). Nitrate isotstandards in individual runs were diluted in nutrient-free seawater to concentrations similar to those of the samples to account for potential matrix effects on the δ18O measurements (Weigand et al. 2016). Reproducibility was monitored by analysis of an internal seawater nitrate standard from the deep North Atlantic.