Biological and physical regulation of the oceanic fixed nitrogen reservoir

Fixed nitrogen (N) limits phytoplankton growth throughout the low latitude ocean. The oceanic reservoir of this essential nutrient is therefore a critical determinant of the fertility of the ocean as a whole, and the strength of the biological pump that sequesters carbon in deep waters out of contact with the atmosphere. Two biological processes dominate the ocean's N budget: a source of N through N2 fixation by diazotrophic plankton, and a sink through bacterial denitrification when oxygen is depleted. These processes are subject to external climate-related forces that push the N budget out of balance, but also coupled through internal feedbacks that strive to restore a balance and stabilize the N reservoir. The feedback mechanism can be thought to operate like a "nutrient thermostat", in which the nitrogen-to-phosphorous (N:P) requirement of phytoplankton represents a setpoint towards which the ratio of N and P reservoirs is restored. This dissertation comprises four projects that address different aspects of the general question: how flexible is the oceanic N reservoir? Chapters 2 and 3 examine the flexibility of the setpoint towards which the N reservoir is restored. In Chapter 2, a diagnostic model is used to show that the N:P of nutrient drawdown by marine phytoplankton, long considered to have a universal "Redfield ratio" of 16:1, varies significantly at the scale of marine biomes due to stoichiometric diversity between taxa. Thus the mean N:P of marine phytoplankton, previously thought of as a fundamental biological property, only reflects the balance of high and low N:P biomes under modern-ocean conditions. Chapter 3 explores the role of this variability in the nutrient thermostat feedbacks. An ecosystem model is used to show that the ratio of nutrient reservoirs it is biased upwards towards the high N:P requirements of subtropical phytoplankton that cohabit and compete directly with N2-fixers. This resolves a discrepancy between the observed N reservoir and the predictions of "Redfieldian" models, which lose too much N before reaching equilibrium. We also demonstrate an important role in the nutrient thermostat for ocean circulation, which communicates stoichiometric signatures between biomes. Changes in the N reservoir may thus be driven, over millennial timescales, by a restructuring of plankton biomes and the circulation pathways that connect them.Chapters 4 and 5 examine the potential for variations in the N reservoir via direct forcing of the oceanic N budget. Chapter 4 focuses on the process of N2-fixation, which may be limited either by an external supply of Fe through the atmosphere, or the internal generation of N deficits that allow diazotrophs to compete with faster-growing plankton, which are susceptible to N-limitation. Multiple geochemical constraints are used to show that N2-fixation in the modern ocean is limited by Fe inputs at the community scale, but strongly coupled to N loss through denitrification at the basin scale. These results reconcile biochemical evidence for Fe-stress in diazotroph communities with the existence of regulatory feedbacks that maintain a balanced the N budget. Within the regime of intermediate Fe-control, the oceanic N reservoir would respond weakly to the enhanced dust fluxes hypothesized for a glacial climate, but strongly to the reduced fluxes expected under climate warming. Chapter 5 uses a biogeochemical model to derive a simple framework for understanding the strength of N cycle feedbacks. A feedback response to any N-budget forcing comprises two components: a fast initial adjustment period involving only the upper ocean, in which changes in the N reservoir are minor, and a slow millennial response involving deep ocean circulation. The magnitude of changes in the N reservoir depend on the fraction of forced anomalies that "leak" into the deep ocean, and the timescale over which those deep-ocean anomalies are removed. Feedbacks through N2-fixation, rather than denitrification, are most efficient at preventing leakage and removing deep-ocean anomalies, and are strong enough to prevent major perturbation to the N reservoir, unless Fe shortages confine diazotrophs to small regions of the ocean