Characterization of nitrogen isotope fractionation during nitrification based on a coastal time series

Isotopic enrichment factors are key to using stable isotope signatures in biogeochemical studies. However, these are typically determined in laboratory experiments and their applicability to environmental conditions is difficult to test. Here, we analyzed nitrogen stable isotope changes associated with nitrification in a coastal basin using weekly time-series measurements of delta N-15 in particulate nitrogen, ammonium, nitrite, and nitrate. Two year-long time series were selected as contrasting natural experiments in the ammonium-rich, aphotic bottom water of Bedford Basin, Nova Scotia, Canada. In 2014, ammonia oxidation (AO) was associated with Thaumarchaeota and nitrite concentrations remained low (< 0.5 mu mol kg(-1)). In contrast, transient nitrite accumulation (similar to 8 mu mol kg(-1)) and a more rapid delta N-15(NH4) increase in the fall of 2017 were likely caused by ammonia-oxidizing bacteria, associated with higher AO rates and, possibly, stronger nitrogen-isotope enrichment ((15)epsilon(AO)). Estimates of (15)epsilon(AO) (21.8 +/- 2.2 parts per thousand, 24.1 +/- 1.1 parts per thousand) were derived empirically using Rayleigh models applied to field data from restricted periods during which the bottom waters approximated a closed system and influence on (15)epsilon(AO) from other processes was demonstrably insignificant. Using a numerical reactive-transport model, we found that the best fit for the delta N-15 data was obtained with (15)epsilon(AO) values (18.9 parts per thousand, 25.1 parts per thousand) close to those determined by the Rayleigh models. The time series also revealed substantial (similar to 7 parts per thousand) N-15-enrichment of the particulate nitrogen due to light-independent assimilation of partially nitrified ammonium. Consistent with previous studies, these field-based nitrogen isotope fractionation experiments suggest that the range of (15)epsilon(AO) values relevant for marine systems may be narrower than determined in laboratory studies.