Balancing watershed nitrogen budgets: accounting for biogenic gases in streams

Abstract

Denitrification is critical for removal of reactive nitrogen (Nr) from ecosystems. However, measuring realistic, scalable rates and understanding the role of denitrification and other dissimilatory processes in watershed nitrogen (N) budgets remains a significant challenge in biogeochemistry. In this study, we focused on the stream reach and network scale in three Mid-Atlantic coastal plain watersheds. We applied open channel methods to measure biogenic N 2 and N 2 O gas fluxes derived from both in-stream and terrestrial nitrogen processing. A large portion of biogenic N 2 flux through streams (33-100 %, mean = 74 %) was a result of groundwater delivery of biogenic N 2 with the remaining portion due to instream N 2 production. In contrast, N 2 O was largely produced in-stream, with groundwater delivery contributing on average 12 % of the total biogenic N 2 O flux. We scaled these measurements across one stream network and compared them to hydrologic Nr export and net anthropogenic N inputs (NANI) to a 4.8 km 2 watershed. The N budget revealed that, during the study period, the biogenic N 2 flux through streams was comparable to the difference between NANI and hydrologic Nr export (i.e. the ''missing'' N). This study provides a methodological and conceptual framework for incorporating terrestrial and in-stream derived biogenic N gas fluxes into watershed N budgets and supports the hypothesis that denitrification is the primary fate of NANI that is not exported in streamflow.

Sediment, water column, and open‐channel denitrification in rivers measured using membrane‐inlet mass spectrometry

“…Using separate microcosm estimates of ER and DN to partition their relative contributions, our results show that the water column may be responsible for a substantial portion of reach‐scale ER in some rivers, and a portion of water column N removal can be attributed to denitrification which permanently removes NO 3 − ‐N from the ecosystem. We note that the microcosm approach excludes the influence of groundwater on sediment DN, which can be substantial in small streams [ Gardner et al ., ] but likely less influential in our study rivers. Incorporating these findings into watershed models focused on N uptake or retention [ Alexander et al ., ; Wollheim et al ., ; Ye et al ., ] will improve the accuracy and mechanistic basis of these models.…”

“…We acknowledge that groundwater can also be out of equilibrium with the atmosphere due to groundwater flowing through soils at different temperatures than the river. Estimates of the groundwater contribution to excess N 2 and Ar can be made using 222 Rn to separate in‐stream and groundwater contributions to N 2 production, and recent research has shown that groundwater can make significant contributions to excess N 2 in headwater streams [ Gardner et al ., ]. We were unable to separate these contributions with our data, but we assumed that the importance of groundwater contributions to N 2 production decreases with increasing stream size, as has been shown for CO 2 [ Hotchkiss et al ., ].…”

“…Additionally, a portion of N 2 found in streams may come from groundwater inflows [Gardner et al, 2016], potentially explaining differences between microcosm and open-channel approaches in the Tippecanoe R, but the contribution of groundwater to excess N 2 in the Tippecanoe R should be minimal compared to headwater streams. In the future, the analysis of a second gas tracer, such as 222 Rn [Gardner et al, 2016], would allow for assessing the groundwater contribution to riverine N 2 production. It is unclear why Journal of Geophysical Research: Biogeosciences 10.1002/2015JG003261 denitrification estimates made using 15 N tracer additions tend to be lower than other reach-scale estimates (Figure 8), but one potential reason could be that the 15 N-NO 3 À additions do not include coupled nitrification-denitrification, which may be particularly important in streams with NO 3 À -N concentrations <300 μg L À1 , thus underestimating denitrification rates as a whole [Mulholland et al, 2008].…”

Gas exchange in streams and rivers

“…It is necessary to know gas exchange using this method, even though the net flux of O 2 between stream and air is of little interest relative to the biological fluxes of O 2 in the stream. This method has been extended to other bioreactive gases, such as N 2 (Baulch, Venkiteswaran, Dillon, & Maranger, 2010;Gardner, Fisher, Jordan, & Knee, 2016). The second area of research interest is the source of streams as greenhouse gases to the atmosphere (Butman & Raymond, 2011;Raymond et al, 2013;Stanley et al, 2016), where gas exchange is needed to estimate gas flux following Equation 1.…”

An Empirical Model for Dinitrogen Gas Emission from Inland Waters

“…denitrified N 2 ) emitted in bubbles is that the fate of anthropogenic N will appear to be uncertain [13] [14] because nitrogen will seem to be "missing", in other words, not enough N output will be found to balance riverine N inputs and outputs. Several studies observed such imbalances [1] [11] [15]- [22].…”