In situ N2 production rates with and without the addition of sinking particles were obtained from the Eastern Tropical North Pacific on the R/V Sikuliaq cruise SKQ201617S in January 2017, which sampled two stations: St P2 (16.5ºN 107ºW) and St P1 (20.1 ºN 106.2ºW), and the R/V Revelle cruise RR1805 sampled approximately the same two stations in 2018: St P2 (16.9ºN 107ºW) and St P1 (20.3 ºN 106.1ºW).

The unpoisoned trap-incubator design collected particles using either a hard plastic cone or mesh net design similar to (Peterson et al., 2005). Despite these differences, the plastic cone and mesh net trap styles were demonstrated to have similar collection efficiencies in the ETNP in 2017 (Cram et al., 2022). The gate valves at the tops and bottoms to incubation chambers, as well as the plungers of 15N tracer-injecting syringes, were held open with tensioned rubber tubing restrained with nylon-jacketed electronic dissolving links (“burn wires”) prior to deployment. The chamber gate valves and syringes were controlled using a preprogrammed onboard Arduino microcontroller that directed current from two 9V alkaline batteries to individual electronic dissolving links. The nylon jacket on the electronic dissolving link was stripped at 1-cm wide sections which held open a gate valve or syringe plunger. When current was applied by the Arduino microcontroller, the electronic dissolving links corroded in seawater and released either the attached gate valves or syringe plungers. Sediment trap incubators were deployed with all gates open to facilitate the dissipation of any trapped air bubbles during an 8-hour acclimation period in anoxic waters at the target depth.

Particles collected in the trap fell into a vertically-oriented 1 L incubation chamber (“+particles chamber”) that was initially open on both ends. Another vertically-oriented 1 L incubation chamber (“water column chamber”) sat to the side of the sediment trap and was deployed open on both ends, but closed immediately before incubation. The bottom of the +particles chamber was programmed to close after the 8-hour acclimation period at the target depth, beginning a collection phase of 12-36 hours. After the collection phase, the top of the +particles closed, along with both top and bottom of the water column chamber. At the beginning of the incubation phase, 15N-labeled nitrite was injected into both +particles and water column chambers. A second 500 mL collection chamber (“top collector”) immediately above the +particles chamber collected additional sinking particles during the incubation phase for use in flux calculations.

Immediately after returning the sediment trap-in situ incubator systems shipboard, the experimental chambers were sampled using single-use needles into 12 mL septum-capped glass Exetainer vials (LabCo, UK) that had been purged with helium gas for 5 minutes. Duplicate vials were sampled for each incubation. Vials were poisoned with 50% (w/v) zinc chloride and stored in the dark at room temperature. Water from CTD casts was collected into Exetainers to measure the background isotopic composition of N2 gas by overfilling the exetainer four times and closing the exetainer with a needle inserted to just below the septa, to cause any bubbles to be removed during closing. Each Exetainer was checked for bubbles. Exetainers were measured for 29N2 and 30N2 accumulation on a Thermo Delta V isotope ratio mass spectrometer in continuous flow mode using helium gas at the University of Washington. Each datapoint was background corrected. A tank of N2 gas was the internal standard and was measured three times during each sample measurement. Each sample was in turn measured three times. Air was used as an external standard and was added to helium purged Exetainers using a range of volumes. These air standards were measured regularly throughout the run to quantify any drift. Rate measurements were calculated following de Brabandere et al. (2014) using incubation durations and the volumes of the experimental chambers.