The effect of winter grazing and a nitrification inhibitor on nitrous oxide emissions and denitrification in a stony soil

Intensive grazing of forage crops by dairy cows is common practice during the winter months in the Canterbury region of New Zealand. Such intensive grazing on wet soil increases the risk of denitrification and the production of the powerful greenhouse gas, nitrous oxide (N₂O). However, at the beginning of this research project, no information was available on denitrification and N₂O emission from forage crops grown in free-draining stony soils, which are widespread in Canterbury. The objectives of this thesis were, therefore, to quantify N₂O emissions from winter grazed forage soil, develop methods to mitigate those emissions, and to improve knowledge and understanding of the effects of severe animal trampling on the physical properties of, and denitrifying microbial population in, the soil below forage crops grazed by dairy cows during winter. Three field-based experiments were carried out in 2011, 2012 and 2013, and were complimented by a laboratory-based experiment in 2012. The field experiments used lysimeters and soil blocks collected from a stony Balmoral soil used for forage kale. The nitrification inhibitor, dicyandiamide (DCD), and a carbon-rich biochar were also used as soil amendments to mitigate N₂O emissions. Total N₂O emissions from trampled soil ranged from 1.3% to 1.9% of urine-N applied to the soil. When applied at 20 kg ha⁻¹, DCD reduced total N₂O emissions from trampled soil with urine by 44-68%, but DCD did not significantly affect N₂O emissions from non-trampled soil. Biochar, with or without DCD, had no significant effect on N₂O emissions when incorporated into the soil at 5 t ha⁻¹ (1% of soil dry weight). Trampling increased the bulk density of the fine earth fraction of the stony soil from 0.89 g cm³ to 1.11 g cm³ (P<0.05) in year two, and from 1.03 g cm³ to 1.19 g cm³ (P<0.05) in year three. Trampling decreased the air permeability of the stony soil by more than one order of magnitude (P<0.01). When sampled, the volumetric water content of the trampled soil was, on average, higher than that of the non-trampled soil (P<0.05), which combined with the reduction in air permeability to create anaerobic conditions suitable for the growth of denitrifying microbes. Significant increases in the abundance of NO₂⁻ reducing nirS gene copies were observed in trampled soil that had been treated with urine (under both laboratory and field conditions). Similarly, in the field, significantly more N₂O reducing nosZ clade-I (P<0.05) and nosZ clade-II (P<0.01) gene copies were found in trampled soil than non-trampled soil. The increase in nosZ gene copies in urine amended trampled soil was accompanied by an increase in the proportion of ¹⁵N labelled urine emitted as N₂ gas – the final product of denitrification (P<0.05). The increase in denitrifying gene copies and corresponding increase in urine-derived N₂ gas following trampling is an important finding, as at the time of writing, no authors had described such a relationship. This work has shown that denitrification does occur in free draining stony soil when forage crops are intensively grazed by dairy cows during winter, and is an important N-loss pathway that should not be overlooked.