Quantification of nitrate sources to an urban stream using dual nitrate isotopes.

Human-engineered landscapes and subsequent altered hydrology affect the fate and transport of reactive nitrogen, particularly in urban watersheds. In this study, we used dual-nitrate isotopes and mixing model analysis (δ(15)N and δ(18)O of NO3(-)) to quantify nitrogen inputs from two sources concentrated in urban systems, sewage and atmospheric deposition. Analysis was conducted on samples collected from Nine Mile Run (Pittsburgh, PA) including over two years of samples collected biweekly and samples collected through the hydrographs of four storm events. Mixing models incorporated uncertainties in the isotopic composition of potential nitrate sources and resolved the relative proportions of nitrate inputs from each source using Bayesian techniques. The results indicate that up to 94% of nitrate in streamwater originated from sewage sources during baseflow conditions. During storms, atmospheric deposition was a substantial nitrate source (∼ 34%) to total event-based nitrate loads, although sewage-derived nitrate remained the dominant source (66%). The potential influence of denitrification was considered by incorporating associated isotopic fractionations into mixing models; up to 19% of sewage-derived samples showed the isotopic effects of denitrification. This study quantitatively delineates proportions of nitrate from different sources to urban streamwater, while incorporating remaining uncertainties in source endmember compositions.

Constraining nitrogen inputs to urban streams from leaking sewers using inverse modeling: implications for dissolved inorganic nitrogen (DIN) retention in urban environments.

Leaking sewer infrastructure contributes nonpoint nitrogen pollution to groundwater and surface water in urban watersheds. However, these inputs are poorly quantified in watershed budgets, potentially underestimating pollutant loadings. In this study, we used inverse methods to constrain dissolved inorganic nitrogen (DIN) inputs from sewage to Nine Mile Run (NMR), an urban watershed (1570 ha) in Pittsburgh, Pennsylvania (USA) characterized by extensive impervious surface cover (38%). Water samples were collected biweekly over two years and intensive sampling was conducted during one summer storm. A nitrogen budget for the NMR watershed was constructed, ultimately inverted, and sewage DIN inputs constrained using Monte Carlo simulation. Results reveal substantial DIN contributions from sewage ranging from 6 to 14 kg ha-1 yr-1. When conservative estimates of DIN from sewage are included in input calculations, DIN retention in NMR is comparable to high rates observed in other suburban/urban nutrient budgets (84%). These results suggest a pervasive influence of leaking sewers during baseflow conditions and indicate that sewage-sourced DIN is not limited to sewer overflow events. Further, they highlight the importance of sewage inputs to DIN budgets in urban streams, particularly as sewer systems age across the U.S.