Seawater samples for nitrate isotope analyses were filtered through a 0.2-um pore-size polyethersulfone membrane into pre-rinsed 60-ml high density polyethylene bottles and were stored frozen until analysis.

The naturally occurring isotope ratios of nitrogen (15N/14N) and oxygen (18O/16O) in nitrate (NO3-) were analyzed by the denitrifier method (Casciotti et al., 2002; Sigman et al., 2001). Briefly, 20 nmol of NO3- were quantitatively reduced to nitrous oxide (N2O) gas by denitrifying bacteria that lack an active terminal N2O reductase (P. chlororaphis f. sp. aureofaciens; ATCC #13985). The product N2O was analyzed by continuous flow isotope ratio mass spectrometry on a Thermo Delta V Advantage isotope ratio mass spectrometer interfaced with a purpose-built, gas chromatography-based device for N2O extraction, concentration, and purification (Casciotti et al., 2002; McIlvin & Casciotti, 2011). Nitrite (NO2-), which interferes with the NO3- isotope analyses, was removed from samples with sulfamic acid (Granger & Sigman, 2009) prior to analysis in the few samples where it was detected. Individual analyses were referenced to injections from a laboratory standard N2O tank and calibrated using the NO3- reference materials IAEA-N3 (4.7‰vs. N2 and 25.6‰vs. VSMOW; Böhlke et al., 2003; Gonfiantiniet al., 1995) and U.S. Geological Survey-34 (+1.8‰vs. N2; -27.9‰vs. VSMOW; Böhlke et al., 2003), with monitoring of reproducibility by analysis of an internal seawater NO3- standard from the deep North Atlantic. NO3- standards in individual runs were diluted in nutrient-free seawater to concentrations equivalent to those of samples to account for potential matrix effects on δ18ONO3 measurements (Weigand et al., 2016). In order to ensure measurement accuracy, samples were analyzed in duplicate within runs, for a minimum of three discrete runs, yielding average standard deviations of 0.2‰ for N and 0.3‰ for O, although with a lower precision averaging 0.4‰ for δ18ONO3 at lower NO3- concentrations (<10 μM).