Understanding the potential for nitrate attenuation from paddock to stream using dual nitrate isotopes : A thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy at Lincoln University

Ruminant urine deposition can result in nitrate (NO₃⁻) leaching from soil to groundwater under agricultural systems. Nitrous oxide (N₂O) emissions from ruminant urine patches are also a major concern as N₂O is a potent greenhouse gas and contributes to ozone depletion. New Zealand agriculture is dominated by grazed-pasture systems, where ruminant urine deposition has become a significant environmental concern. Attenuation (removal) of NO₃⁻ from the system through biological denitrification, reducing it to the inert gas dinitrogen (N₂), is one way to reduce NO₃⁻ leaching losses and N₂O emissions from agriculture. Understanding temporal and spatial variations in NO₃⁻ attenuation capacity that occurs within the landscape, between the site of ruminant urine deposition and stream or groundwater contamination, may provide a key mitigation tool. The general goal of this PhD programme was to use isotope analysis and soil physics to increase our knowledge of when, ‘hot moments’, and where, ‘hot spots’, of NO₃⁻ attenuation occur under grazed agricultural soils. Soil physical properties; soil moisture, defined as soil matric potential (Ψ), and relative gas diffusivity (Dp/D₀) play key roles in the amount and rate of NO₃⁻ attenuation that occurs in the soil. These properties were explored under controlled laboratory conditions for two pasture soils (both A and B horizons), to identify the key Ψ for holding Dp/D0 at a previously identified threshold for peak denitrification to occur (Chapter 4). The identified key Ψ was then used to analyse the effect of soil type on dual NO₃⁻ isotope signatures under laboratory conditions (Chapter 5). Another factor that could affect dual NO₃⁻ isotope signatures is the presence or absence of plants in the soil under bovine urine (BU) patches. This was assessed using lysimeters in a field setting over time, to also better understand temporal isotope dynamics (Chapter 6). All findings were then taken into consideration in a field context, where temporal and spatial variation in N attenuation was measured at two contrasting field sites in Southland, New Zealand using dual NO₃⁻ isotope signatures and soil physical properties (Chapter 7). Significant results were found in laboratory trials indicating that peak N₂O emissions occurred at Dp/D₀ values < 0.006, allowing for the Dp/D₀ values associated with NO₃⁻ attenuation to be extrapolated to field sites, to provide a general overview of what Dp/D₀ values we expect to see in the field when attenuation is occurring. Spatial variation in dual nitrate isotopes was found to be significantly different between soil A and B horizon soils. The A horizon soils played a key role in N processing, showing an isotopic fractionation rate +14‰ greater than in B horizon soils. However, these findings were confounded by diluted expression of NO₃⁻ attenuation, as prolonged water logging shifted the isotopic signature from enrichment to depletion. This was thought to be due to heterotrophic nitrification changing isotopic N signals in the soil under extended periods of high soil water. The lysimeter study showed δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻ to vary temporally, with evidence of denitrification enrichment of isotopes. Plant presence or absence was also found to significantly influence isotope signatures, but only when BU was not applied, indicating that denitrification may be driven by a plant derived C supply at very low NO₃⁻ concentrations. However, the proportion of N leached under such low NO₃⁻ concentrations will be minimal and therefore the influence of plant presence will have negligible impact on interpreting drainage δ¹⁵N values. Field study sites demonstrated highly dynamic NO₃⁻ isotope composition, and that N attenuation hotspots were strongly influenced by spatial variation (soil type) and extended rainfall events, as evidenced by modelled Dp/D₀ values. Dilution of the denitrification isotope signal by mineralisation of soil N and/or nitrification occurred, reinforcing the role of soil processes in realigning the NO₃⁻-N isotope signal back to a soil-N signal as NO₃⁻- moves through the soil profile. This research has clearly shown that δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻ show potential for identifying NO₃⁻ sources and soil processes forming and removing NO₃⁻. Further ground-truthing of in situ NO₃⁻ attenuation, determined by Dp/D₀ values < 0.02 is warranted, as this could potentially provide an economic way for farmers and policy makers to both recognise and even engineer ‘hot moments’ and ‘hot spots’ of N attenuation occurring in the landscape.