Paradigm Shift: Does River Metabolism Mask the Isotopic Signal of Nitrate Sources?

Abstract

Nitrate (NO₃⁻) is the most ubiquitous contaminant in surface and groundwaters in Canada. Synthetic fertilizer application and manure production in intensive agricultural areas contribute large quantities of NO₃⁻ to the landscape with a proportion seasonally lost to groundwaters and streams. Elevated concentrations of NO₃⁻ in freshwater systems can result in problems for drinking water supplies and aquatic ecosystem health. The Grand River is the largest Canadian river draining to Lake Erie and the catchment’s land-use is predominantly agricultural (~80%). It receives NO₃⁻ inputs from point (WWTPs) and non-point (agricultural manure and fertilizer) sources. Isotopes of NO₃⁻ are commonly used in ecosystem studies to apportion sources (e.g. manure, septic systems, wastewater treatment plant effluent and synthetic fertilizers) and to determine the important NO₃⁻ transformation processes (nitrification and denitrification). For decades, several assumptions have governed these studies such as: 1) δ¹⁸O-NO₃⁻ from nitrification can be predicted using the 2:1 rule (two O in NO₃⁻ come from H₂O and one O from O₂), 2) NO₃⁻ isotopes indicate denitrification in freshwater environments when elevated in a 2:1 ratio for δ¹⁵N: δ¹⁸O, and 3) The δ¹⁵N- and δ¹⁸O and of NO₃⁻ are conservative in oxic environments and thus if δ¹⁸O-NO₃⁻ is not elevated, the δ¹⁵N-NO₃⁻ can be used for source apportionment. This research indicates that these assumptions may not always be correct. The overall objectives of this thesis are to improve the use of NO₃⁻ isotopes for source apportionment in rivers and streams and if the isotopes cannot be used to separate sources then can a mechanistic model be used to estimate rates of N transformation processes that can ultimately help to determine the fate of NO₃⁻ in rivers. Nitrate isotope data from the Grand River shows no clear denitrification line. A seasonal trend is only observed in δ¹⁵N-NO₃⁻ (high in the summer, low in the spring and fall), not in δ¹⁸O-NO₃⁻. Incubation experiments conducted using two sites on the Grand River with different source inputs demonstrate that the δ¹⁵N- and δ¹⁸O-NO₃⁻ are not conservative and cannot be used to indicate denitrification or to discern source inputs of NO₃⁻. The NO₃⁻ isotopes changed over time even when NO₃⁻ concentrations did not. Results from an in-river experiment were consistent with incubations and confirmed that in a highly productive river, such as the Grand River, source apportionment is difficult as internal N recycling can be rapid, and the effect on the isotopic signal of NO₃⁻ cannot be ignored. Isotopic O-exchange between nitrite (NO₂⁻) and water (H₂O) during nitrification is a mechanism that can alter the δ¹⁸O-NO₃⁻ signal from nitrification. This study found considerable amounts of O-exchange (40-100%) occurring at both sample sites in all incubation experiments under nitrifying conditions indicating that the δ¹⁸O-NO₃⁻ is “reset” toward the δ¹⁸O-H₂O value of the water medium. The δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻ cannot be used as conservative tracers in the river. However, a mechanistic model was developed using these isotopes to explain the results from incubation experiments and include the effects of O-exchange and large kinetic ¹⁸O isotope effects on the δ¹⁸O-NO₃⁻ during nitrification. With this model, gross rates of the N cycling processes nitrification, denitrification, mineralization and uptake were obtained in order to describe the biogeochemical cycling of N in the Grand River. This model helps to determine the variability in N cycling between sites and at different times of the year.