Hydrobiogeochemical control of temporal variation in atmospheric nitrate export from a temperate forest watershed

Background/Question/Methods

Atmospheric N deposition can have important influences on biogeochemical processes in terrestrial and aquatic ecosystems. However, the mechanisms governing the transport pathways and transformation processes of atmospheric N in forested watersheds remain uncertain, in part because of the lack of an unambiguous tracer of atmospheric N. This lack of understanding limits our ability to predict the response of ecosystems to atmospheric N deposition. The unique Δ¹⁷O values of atmospheric nitrate help to overcome this challenge. We measured nitrate concentrations and Δ¹⁷O and δ¹⁵N of nitrate-N in a stream (n=74 samples) draining a small gaged, forested watershed located in the central Appalachian Mountains over two consecutive water years. Samples were collected across a wide range of hydrologic conditions throughout each year. The relationships among nitrate concentrations, isotopic values and flow conditions were assessed using non-linear regression analysis. A statistical load estimator model was used to estimate daily yields of atmospheric and terrestrial nitrate (NO₃-N_Atm and NO₃-N_Proc) from measured N concentrations and discharge values, and these daily yields were summed to calculate annual yields. Annual flow-weighted mean concentrations were computed from the annual yields and runoff values. We discuss the implications of these results for understanding of N cycling in temperate forests.

Results/Conclusions

Nitrate-Δ¹⁷O values were >0 in 68 of 74 stream samples, confirming the presence of NO₃-N_Atm across a variety of flow regimes.  The statistical load model predicted that the annual flow weighted mean concentration of NO₃-N_Atm ranged from 0.05-0.12 mg N L^-1, and contributed 6-10% of the modeled annual total nitrate yields. NO₃-N_Atm concentrations generally peaked during rising limb periods (up to 0.33 mg N L^-1) associated with summer rainstorm events and winter/spring snowmelt events. The results confirm the direct, rapid transport of atmospheric nitrate to the stream during quickflow periods. Further, we observed non-zero concentrations of NO₃-N_Atm (0.02-0.10 mg N L^-1) consistently in baseflow. Thus, a second pathway allowing export of NO₃-N_Atm may involve its accumulation in the soil, subsequent transport through the soil matrix, and eventual release as groundwater discharge. Significant hyperbolic relationships between NO₃-N_Proc and discharge, δ¹⁵N-NO₃ and discharge, and NO₃-N_Proc and δ¹⁵N-NO₃ suggested that NO₃-N pools become diminished and fractionated at greater depth in the soil. Overall, our results illustrate two pathways by which temperate forests may dynamically respond to declining atmospheric N deposition, which may help explain why many streams draining forested watersheds in eastern North America have exhibited declining nitrate-N yields in recent decades.