Heterogeneity of Groundwater Nitrogen Attenuation Elucidated by Combining Isotope Fluxes and Functional Gene Markers

Reactive nitrogen (N) is a ubiquitous freshwater pollutant, but generating accurate catchment-scale measurements of its fate and transport remains elusive. Advances N isotope systematics and the detection of microbial populations provide potentially promising avenues for developing sensitive indicators of in-situ transformations. However, the use of isotopes in terrestrial systems is currently limited by difficulty in distinguishing variations in fractionation from source mixing, while relationships between microbial indicators and fluxes are inconsistent. Hypothesizing that these issues could be overcome by linking the two approaches across hydrologic flux data, we set out to develop an analytical framework for identifying where and how N is attenuated in cryptic groundwater systems.

Variations in the isotopic composition of ammonium, nitrite, and nitrate and the genes associated with their oxidation/ reduction were measured in 50 wells distributed longitudinally across an ammonium plume (from 130 to 0 mg N l^-1) three times over 12 months. Net isoflux calculations revealed the seasonal development of N attenuation hotspots along the plume fringe. The broad correlation of these hotspots with redox transition zones and ammonia oxidizing and nitrite reducing gene abundances corroborated the traditional view of N removal via coupled nitrification-denitrification. It was thus surprising that the relationships between the dual isotope ratios of nitrate and nitrite were not consistent with denitrification in two hotspots. Combining indicators, we found strong empirical proof that attenuation in these zones was driven by seasonal anaerobic ammonium oxidation: both zones had high relative and absolute abundance of nitrite reducing and anaerobic ammonium oxidizing genes alongside high δ¹⁸O_nitriteand nitrite concentrations.

The development and handling of this unique data set provides a practical template for up-scaling process level information to the whole aquifer by integrating variations in nitrate, nitrite, and ammonium isotopic composition. The outcomes emphasize the need for high spatial and temporal resolution data in order to accurately assess groundwater N cycling.