[1] We present a new nitrogen isotope model incorporated into the three-dimensional ocean component of a global Earth system climate model designed for millennial timescale simulations. The model includes prognostic tracers for the two stable nitrogen isotopes, 14 N and 15 N, in the nitrate (NO 3 − ), phytoplankton, zooplankton, and detritus variables of the marine ecosystem model. The isotope effects of algal NO 3 − uptake, nitrogen fixation, water column denitrification, and zooplankton excretion are considered as well as the removal of NO 3 − by sedimentary denitrification. A global database of− observations is compiled from previous studies and compared to the model results on a regional basis where sufficient observations exist. The model is able to qualitatively and quantitatively reproduce many of the observed patterns such as high subsurface values in water column denitrification zones and the meridional and vertical gradients in the Southern Ocean. The observed pronounced subsurface minimum in the Atlantic is underestimated by the model presumably owing to too little simulated nitrogen fixation there. Sensitivity experiments reveal that algal NO 3 − uptake, nitrogen fixation, and water column denitrification have the strongest effects on the simulated distribution of nitrogen isotopes, whereas the effect from zooplankton excretion is weaker. Both water column and sedimentary denitrification also have important indirect effects on the nitrogen isotope distribution by reducing the fixed nitrogen inventory, which creates an ecological niche for nitrogen fixers and, thus, stimulates additional N 2 fixation in the model. Important model deficiencies are identified, and strategies for future improvement and possibilities for model application are outlined.

The ocean is an important source of nitrous oxide (N 2 O) to the atmosphere, yet the factors controlling N 2 O production and consumption in oceanic environments are still not understood nor constrained. We measured N 2 O concentrations and isotopomer ratios, as well as O 2 , nutrient and biogenic N 2 concentrations, and the isotopic compositions of nitrate and nitrite at several coastal stations during two cruises off the Peru coast (~5-16°S, 75-81°W) in December 2012 and January 2013. N 2 O concentrations varied from below equilibrium values in the oxygen deficient zone (ODZ) to up to 190 nmol L À1 in surface waters. We used a 3-D-reaction-advection-diffusion model to evaluate the rates and modes of N 2 O production in oxic waters and rates of N 2 O consumption versus production by denitrification in the ODZ. Intramolecular site preference in N 2 O isotopomer was relatively low in surface waters (generally À3 to 14‰) and together with modeling results, confirmed the dominance of nitrifier-denitrification or incomplete denitrifier-denitrification, corresponding to an efflux of up to 0.6 Tg N yr À1 off the Peru coast. Other evidence, e.g., the absence of a relationship between ΔN 2 O and apparent O 2 utilization and significant relationships between nitrate, a substrate during denitrification, and N 2 O isotopes, suggest that N 2 O production by incomplete denitrification or nitrifier-denitrification decoupled from aerobic organic matter remineralization are likely pathways for extreme N 2 O accumulation in newly upwelled surface waters. We observed imbalances between N 2 O production and consumption in the ODZ, with the modeled proportion of N 2 O consumption relative to production generally increasing with biogenic N 2 . However, N 2 O production appeared to occur even where there was high N loss at the shallowest stations.

Bacteria of the NC10 phylum link anaerobic methane oxidation to nitrite denitrification through a unique O 2 -producing intra-aerobic methanotrophy pathway. A niche for NC10 in the pelagic ocean has not been confirmed. We show that NC10 bacteria are present and transcriptionally active in oceanic oxygen minimum zones (OMZs) off northern Mexico and Costa Rica. NC10 16S rRNA genes were detected at all sites, peaking in abundance in the anoxic zone with elevated nitrite and methane concentrations. Phylogenetic analysis of particulate methane monooxygenase genes further confirmed the presence of NC10. rRNA and mRNA transcripts assignable to NC10 peaked within the OMZ and included genes of the putative nitrite-dependent intra-aerobic pathway, with high representation of transcripts containing the unique motif structure of the nitric oxide (NO) reductase of NC10 bacteria, hypothesized to participate in O 2 -producing NO dismutation. These findings confirm pelagic OMZs as a niche for NC10, suggesting a role for this group in OMZ nitrogen, methane and oxygen cycling.

Mesoscale eddies in Oxygen Minimum Zones (OMZs) have been identified as important fixed nitrogen (N) loss hotspots that may significantly impact both the global rate of N-loss as well as the ocean's N isotope budget. They also represent "natural tracer experiments" with intensified biogeochemical signals that can be exploited to understand the large-scale processes that control N-loss and associated isotope effects (ε; the ‰ deviation from 1 in the ratio of reaction rate constants for the light versus heavy isotopologues). We observed large ranges in the concentrations and N and O isotopic compositions of nitrate (NO 3 À ), nitrite (NO 2 À ), and biogenic N 2 associated with an anticyclonic mode-water eddy in the Peru OMZ during two cruises in November and December 2012. In the eddy's center where NO 3 À was nearly exhausted, we measured the highest δ 15 N values for both NO 3 À and NO 2 À (up to~70‰ and 50‰) ever reported for an OMZ.Correspondingly, N deficit and biogenic N 2 -N concentrations were also the highest near the eddy's center (up tõ 40 μmol L À1 ). δ 15 N-N 2 also varied with biogenic N 2 production, following kinetic isotopic fractionation during NO 2 À reduction to N 2 and, for the first time, provided an independent assessment of N isotope fractionation during OMZ N-loss. We found apparent variable ε for NO 3 À reduction (up to~30‰ in the presence of NO 2 À ).However, the overall ε for N-loss was calculated to be only~13-14‰ (as compared to canonical values of 20-30‰) assuming a closed system and only slightly higher assuming an open system (16-19‰). Our results were similar whether calculated from the disappearance of DIN (NO 3 À + NO 2 À ) or from the appearance of N 2 and changes in isotopic composition. Further, we calculated the separate ε values for NO 3 À reduction to NO 2 À and NO 2 À reduction to N 2 of~16-21‰ and~12‰, respectively, when the effect of NO 2 À oxidation could be removed. These results, together with the relationship between N and O of NO 3 À isotopes and the difference in δ 15 N between NO 3 À and NO 2 À , confirm a role for NO 2 À oxidation in increasing the apparent ε associated with NO 3 À reduction. The lower ε for N-loss calculated in this study could help reconcile the current imbalance in the global N budget if representative of global OMZ N-loss.

Assessment of the global budget of the greenhouse gas nitrous oxide (N2O) is limited by poor knowledge of the oceanicN2O flux to the atmosphere, of which the magnitude, spatial distribution, and temporal variability remain highly uncertain. Here, we reconstruct climatologicalN2O emissions from the ocean by training a supervised learning algorithm with over 158,000N2O measurements from the surface ocean—the largest synthesis to date. The reconstruction captures observed latitudinal gradients and coastal hot spots ofN2O flux and reveals a vigorous global seasonal cycle. We estimate an annual meanN2O flux of 4.2 ± 1.0 Tg N⋅y−1, 64% of which occurs in the tropics, and 20% in coastal upwelling systems that occupy less than 3% of the ocean area. ThisN2O flux ranges from a low of 3.3 ± 1.3 Tg N⋅y−1in the boreal spring to a high of 5.5 ± 2.0 Tg N⋅y−1in the boreal summer. Much of the seasonal variations in globalN2O emissions can be traced to seasonal upwelling in the tropical ocean and winter mixing in the Southern Ocean. The dominant contribution to seasonality by productive, low-oxygen tropical upwelling systems (>75%) suggests a sensitivity of the globalN2O flux to El Niño–Southern Oscillation and anthropogenic stratification of the low latitude ocean. This ocean flux estimate is consistent with the range adopted by the Intergovernmental Panel on Climate Change, but reduces its uncertainty by more than fivefold, enabling more precise determination of other terms in the atmosphericN2O budget.