|dc.description.abstract||The loss of nitrogen (N) through nitrate (NO₃⁻) leaching and nitrous oxide (N₂O) emissions from pastoral dairy systems is one of the largest challenges facing the New Zealand agricultural industry. Nitrate leaching contributes to nutrient enrichment and accelerated eutrophication of streams, lakes and estuaries, while N₂O is both a greenhouse gas and the dominant anthropogenic emission contributing to stratospheric ozone depletion. Urine patches are the primary source of N loss from pastoral systems due to the high N loading that occurs over a relatively small area. Excessive or inappropriately timed fertiliser applications can also add to N loss.
Few studies have sought to determine the effect of concurrently deposited urine and fertiliser on the fate of N in pastoral systems, even though the application of fertiliser soon after grazing is commonly practised, while no studies have examined seasonal effects of any interaction. It is generally assumed that fertiliser applied over a urine patch will simply exacerbate the total N losses, as the urine-N saturates plant-N utilisation rates in pasture. This study, therefore, aimed to quantify the additional losses (if any) that occur as a result of fertiliser being applied concurrently onto a urine patch, and furthermore, determine the fate of the fertiliser N within the urine patch.
To determine this, a two year lysimeter study was undertaken where urine patches were applied in either autumn or spring. Urea fertiliser enriched with ¹⁵N was applied to these lysimeters at rates of either 200 or 400 kg N ha⁻¹ according to the standard regional practice. The amount of fertiliser derived N was measured in the leachate, N₂O emissions, pasture and soil. Fertiliser 15N recovered in leachate and N2O emissions was <2.2% and <0.1%, respectively. Urine and fertiliser at the 400 kg N ha⁻¹ rate did increase total NO₃- leaching by up to 55 kg ha⁻¹ (p < 0.001), but this was as unlabelled N. Pasture uptake accounted for up to 52% of the fertiliser 15N recovery and increased plant uptake was observed under increasing fertiliser rates (p < 0.001). Recovery of fertiliser ¹⁵N in the soil at the end of the experiment was 22% with the majority of this in the top 10 cm soil. Total fertiliser ¹⁵N recovery ranged from 68-81% and it is suggested most of the unaccounted for ¹⁵N was lost as N₂ emissions and/or leached as dissolved organic N.
To further this work, data from the lysimeter experiment was validated against simulated results from a dynamic, process based model, APSIM (Agricultural Production Systems Simulator). The APSIM simulation was designed to mimic the experimental conditions and management of the lysimeter trial for the purpose of later using it to extrapolate the lysimeter data beyond its current confines to a larger range of treatment scenarios, climatic regions and soil types. Most of the modelled outputs were within the 95% confidence interval of the experimental data. However, the model significantly overestimated N₂O emissions from under urine patches, and as a result, the existing simulation was not deemed suitable to immediately carry out extrapolative predictions from the lysimeter study. The key parameters potentially responsible for the overestimation of the modelled N₂O emissions were identified, and a sensitivity analysis was performed on these parameters.
The lysimeter trial raised further questions as to how urine patches affect pasture N uptake and soil N dynamics in the field. Thus a field trial was established with the primary objective of determining the ‘effective area’ of a ¹⁵N labelled urine patch. Circular plots consisting of a urine patch ‘wetted area’ (0.28 m²) and the potential ‘effective area’ were established and the pasture and soil N pools were monitored inside three distinct zones. A total of 22% of the urinary 15N was recovered in pasture outside the wetted area, mainly due to surrounding plant root proliferation. Recovery of urinary ¹⁵N in the soil was much smaller than in the pasture and was short-lived. Urinary N was recovered in the pasture up to 0.5 m from the edge of the wetted area, resulting in a total potential effective area of up to 2 m² (up to 6 times the wetted area).
In summary, this work has shown that when N fertiliser is applied over a urine patch at a rate of 400 kg N ha⁻¹ yr⁻¹ there is an increase in the amount of N leached, compared to urine alone, but that fertiliser N applied over a urine patch at 200 kg N ha⁻¹ yr⁻¹ does not increase the N loss. This study also suggests that improved validation and parameterisation of N₂O emissions is required before the APSIM model can be used to produce accurate representations of the fate of N under urine and fertiliser deposition. Finally, using ¹⁵N recovery, the results suggest that the area affected by a urine patch is larger than the wetted area, primarily due to root extension and proliferation of the surrounding pasture.||en