Ecological control of nitrite in the upper ocean

Microorganisms oxidize organic nitrogen to nitrate in a series of steps. Nitrite, an intermediate product, accumulates at the base of the sunlit layer in the subtropical ocean, forming a primary nitrite maximum, but can accumulate throughout the sunlit layer at higher latitudes. We model nitrifying chemoautotrophs in a marine ecosystem and demonstrate that microbial community interactions can explain the nitrite distributions. Our theoretical framework proposes that nitrite can accumulate to a higher concentration than ammonium because of differences in underlying redox chemistry and cell size between ammonia- and nitrite-oxidizing chemoautotrophs. Using ocean circulation models, we demonstrate that nitrifying microorganisms are excluded in the sunlit layer when phytoplankton are nitrogen-limited, but thrive at depth when phytoplankton become light-limited, resulting in nitrite accumulation there. However, nitrifying microorganisms may coexist in the sunlit layer when phytoplankton are iron- or light-limited (often in higher latitudes). These results improve understanding of the controls on nitrification, and provide a framework for representing chemoautotrophs and their biogeochemical effects in ocean models.

in main text, but plotted on log scale, and with the R * s of P also plotted without the effects of light limitation (i.e., when R * s are calculated using the constant maximum growth rate µ max instead of µ light , which does not give the increase at the base of the euphotic zone seen in Fig. 4). Since the R * s of P are so low, the log scale reveals their depth variation from grazing (via L). (c): Microzooplankton grazer biomass Z, loss rates L (equation 29), and resulting steady state growth rates µ. The DIN concentrations are lower in the surface than the computed R * s because of the effects of vertical mixing. Phytoplankton also continue to sustain a population at the surface even though their growth rate is lower than the loss rate because of vertical mixing. (In reality, lateral supply of nutrients, or another surface process not represented in this idealized model, may sustain higher surface phytoplankton growth rates in oligotrophic regions.) This vertical mixing supplies additional biomass to the surface from the subsurface maxima, driving the actual resource concentration down from the subsistence resource concentration predicted by equations (3)  oxidation rates (annually averaged). The blue contour indicates where biological growth rates of the AOO and NOO exactly balance loss rates for the AOR and NOR surface means, respectively, encircling the small areas in which the two metabolisms are 'locally sustainable' (stably co-existing with primary production) on average for the course of one year. This was diagnosed with the net biological rate 3 µ N ET .  Chl a and NO − 3 concentrations, photosynthetically active radiation (PAR; scaled so that maximum value at each station appears as 1), amoA gene abundances, NO − 2 concentrations, and NH + 4 oxidation rates. For NO − 2 concentrations and NH + 4 oxidation rates, only two points (125m at Station 70 and 150m at Station 99) were above the detection limit. Error bars denote one s.d.

Supplementary
Supplementary Figure 9: Observed and modeled temperature. Observed temperature at the four stations in the North Pacific sampled for nitrification, and the fit used as the temperature field in the water column model. Table 1: Model parameters for 1D and 3D configurations, including the ranges of uncertainty from which values were randomly sampled in the 1D water column model ensemble.

Parameter
Symbol Value Range Units Nitrifier growth: Gaussian nM Heterotrophic bacterial growth: *These values are those used in the "Both yield and affinity differences" default simulation in the 1D water column model (solid red lines in Fig. 3). The yield and/or affinity differences were removed for the other model experiments (dotted and dashed red lines) by using the listed values of the AOO for the NOO. In the illustrated 3D global simulation, the affinity of the NOO was adjusted by a factor of 2 1/3 instead of 10 1/3 as described in the Methods, giving values of V maxNO 2 ,N OO = 40.3 d −1 and K NO 2 ,N OO = 168 nM.
Supplementary   gives: where J NO 3 represents both advection and diffusion of NO − Rearranging gives: LikeR * , it does not neglect any terms, and so exactly predicts the NO − 2 concentration throughout 34 the model water column. LikeR * , reduced NOO affinity results in a higher NO − 2 concentration.

35
UnlikeR * , the yield is reflected implicitly: the lower yield results in a higher NO − 2 concentration 36 via the lower NOO biomass.

37
In addition to nitrifier activity, this balance emphasizes that phytoplankton activity and circulation and picoplankton. Though a full analysis of the iron limitation to nitrifier growth is beyond the 55 scope of this study, one hypothesis is that small size may allow the nitrifiers a higher affinity for iron 56 than the larger phytoplankton that often populate subpolar waters. Also, evidence that AOO require 57 copper rather than iron for redox machinery makes it plausible that at least NH + 4 oxidation should 58 be favorable in HNLC regions 10 . Thus, we speculate that it may be possible for nitrifiers to thrive in 59 an environment with low iron concentrations despite iron limitation to much of primary production.

60
Future work could examine the hypothesis that the ratio of primary production to nitrification may The framework for the heterotrophic functional type is here described in detail to allow for com-110 parison with other types. Organic matter (OM ) provides the elements and electrons for both the 111 synthesis of biomass (B) and energy production, and oxygen serves as the electron acceptor.
where d is the number of electron equivalents for the generic organic composition correspond to the oxidation states of its inorganic constituents (below as d = 4c + h − 2o − 3n).

114
The growth efficiency (mol B mol −1 OM , or, mol C synthesized mol −1 C consumed) relates to f 115 as: and so y OM = f when assuming the same stoichiometry for both the organic matter substrate and microbial biomass. When assuming the average stoichiometry of marine organic matter The stoichiometries for the three anaerobic metabolic functional types are as follows: ments and electrons for both the synthesis of biomass (B) and energy production, and NO − 3 serves as the electron acceptor, which is reduced to NO − 2 . The full metabolism forming NO − 3 -reducing biomass B HetN O 3 , here written in terms of OM and B HetN O 3 for brevity, is: For marine stoichiometry, the full metabolism is: Denitrifying heterotrophy For the denitrifier, organic matter (OM ) provides the elements and electrons for both the synthesis of biomass (B) and energy production, and NO − 2 serves as the electron acceptor, which we consider here as all being reduced completely to N 2 (here neglecting the formation of N 2 O). The full metabolism forming denitrifying biomass B HetN O 2 is: For marine stoichiometry, the full metabolism is: The organic matter yield for these two anaerobic heterotrophic types is assigned at 0.13 (mol 118 biomass synthesized mol organic matter −1 ). The fact that this yield is lower than that of the aerobic A value of f of 0.05, higher than that of the nitrifiers, results in very similar stoichiometry to that measured and reported by Strous et al.
[13], but perhaps better reflects optimal laboratory conditions rather than the mesopelagic ocean. However, assuming either f = 0.03 (the same value for the nitrifiers) or f = 0.05 results in the exclusion of anammox bacteria from the oxygenated ocean.
With f = 0.03, the anammox metabolism, normalized to one mole of N in order to compare with the nitrifier metabolisms, is: