Determining geochemical and microbial parameters underlying geographic patterns of denitrification and anammox rates in the New River Estuary, North Carolina, USA

Nitrogen removal via denitrification and anammox plays a significant role in mitigating the intensity and duration of eutrophication in estuaries. Environmental conditions and microbial communities mediating N removal vary across space and time, thus the geochemical and microbial drivers of denitrification and anammox were examined to understand the heterogeneity underlying estuarine nitrogen removal. We used 15N isotope pairing incubation experiments combined with quantification of the genes encoding nitrous oxide reductase (nosZ Clades I and II) and hydrazine oxidoreductase (hzo) to determine the underlying microbial and geochemical factors responsible for the spatial and temporal variability of denitrification and anammox rates in the New River Estuary, NC, USA, a shallow, microtidal estuarine system. Mean denitrification and anammox rates were 7.0 ± 11.0 and 0.4 ± 1.2 nmol N g−1 wet sediment hr−1, overall, and slightly elevated in winter. Denitrification accounted for >95% of total N2 production on an estuary scale, during both summer and winter. Percent anammox (ra) was highest where denitrification rates were elevated, reaching >10% during winter at the two uppermost sties. Gene abundance trended higher for nosZ Clade I and hzo genes during winter (with up to 2.47 × 108 and 1.30 × 106 copies g−1 wet sediment, respectively). Elevated nosZ Clade II gene abundance was generally observed mid-estuary, during the summer season and ranged from 2.41 × 104 to 1.72 × 107 copies g−1 wet sediment. Akaike Information Criterion (AIC), used to select the best sets of explanatory geochemical and biological variables for anammox and denitrification, found that nosZ Clades I and II and salinity, which covaries with bottom water nitrate, ammonium and dissolved oxygen, were the best predictors of denitrification rate variability. Anammox rates were best predicted by salinity and ammonium concentrations and disconnected from the hzo gene abundance. We demonstrate that consideration of both the geochemical and microbial features of a system predicts nitrogen removal patterns across estuarine systems.