Magnitude of oceanic nitrogen fixation influenced by the nutrient uptake ratio of phytoplankton

The elemental stoichiometry of sea water and particulate organic matter is remarkably similar. This observation led Redfield to hypothesize that the oceanic ratio of nitrate to phosphate is controlled by the remineralization of phytoplankton biomass1. The Redfield ratio is used universally to quantitatively link the marine nitrogen and phosphorus cycles in numerous biogeochemical applications2–4. Yet, empirical and theoretical studies show that the ratio of nitrogen to phosphorus in phytoplankton varies greatly with taxa5,6 and growth conditions7–9. Here we present a dynamic five-box ecosystem model showing that non-Redfield utilization of dissolved nitrogen and phosphorus by non-nitrogen-fixing phytoplankton controls the magnitude and distribution of nitrogen fixation. In our simulations, systems dominated by rapidly growing phytoplankton with low nitrogen to phosphorus uptake ratios reduce the phosphorus available for nitrogen fixation. In contrast, in systems dominated by slow-growing phytoplankton with high nitrogen to phosphorus uptake ratios nitrogen deficits are enhanced, and nitrogen fixation is promoted. We show that estimates of nitrogen fixation are up to fourfold too high when non-Redfield uptake stoichiometries are ignored. We suggest that the relative abundance of fastand slow-growing phytoplankton controls the amount of new nitrogen added to the ocean. The notion of a constant nitrogen/phosphorus ratio for both phytoplankton and the deep ocean shapes our current understanding of the balance between nitrogen and phosphorus inventories in the ocean. In this framework, processes that remove oceanic fixed nitrogen (for example, denitrification) and drive the nitrate/phosphate ratio from the canonical Redfield ratio are approximately balanced by nitrogen inputs, primarily from nitrogen fixation3,10. However, field and laboratory data confirm that non-Redfield nutrient utilization is common in phytoplankton, with N/P utilization ratios being below Redfield during blooms6,9 and above Redfield in oligotrophic regions dominated by cyanobacteria8,9 (including non-diazotrophs). One explanation for this N/P plasticity is that fast-growing cells require abundant P-rich growthmachinery and exhibit low cellular N/P ratios7,9. In contrast, resource-limited cells invest in N-rich light and/or nutrient acquisitionmachinery and have higherN/P ratios7,9. Despite the prevalence of non-Redfield nutrient utilization, few ecosystem models11,12, and virtually no geochemical or ocean general circulation models that include biological processes, consider non-Redfield nutrient consumption, assuming instead that phytoplanktonN/P stoichiometry is Redfield (diazotrophs notwithstanding). Considering the predominance of marine environments that support non-Redfield N/P utilization by non-diazotrophic phytoplankton, a better description of how phytoplankton stoichiometry affects oceanic nutrient inventories is critical to understanding the marine N cycle.