Comparing the spatial distribution of DCD and urinary nitrogen on well-drained and poorly-drained soils : a thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Agricultural Science at Massey University, Manawatū, New Zealand

Nitrogen (N) losses from urine patches can be significant contributors to greenhouse gas emissions and water quality issues. Nitrification inhibitors may reduce these losses by slowing down the transformation of urine-N to nitrate. Technologies exist that can detect urine patches and target inhibitor applications specifically to the patch area, thereby avoiding the need to apply the inhibitor over the entire paddock. However, the potential time delay between the grazing event and the inhibitor application, and the small volumes of inhibitor used could result in only partial interception of the urine by the inhibitor in the soil. This would limit the potential effectiveness of the inhibitor. Two studies were undertaken to compare the movement of urine to the movement, and therefore potential interception, of the nitrification inhibitor dicyandiamide (DCD). In the first study, patches of urine were created by pouring three different volumes of urine (1, 2 and 3 L) onto two soils of contrasting drainage at two different moisture levels. In the second study, two volumes of DCD (the equivalent of 10 and 20 kg DCD/ha) were sprayed using a Spikey® spray unit onto urine (2 L volume) patches created within 80 cm diameter chambers in two soils of contrasting drainage at two different moisture levels. The variation in urine-N concentration both within and between individual urine patches was substantial. Total urine N recovery averaged 38%. On average, 67% of the recovered N was recovered from the top 5 cm, 14% from 5-10 cm and 19% from 10-20 cm. On average, 78% and 69% of the DCD applied at 30 mL and 60 mL, respectively was recovered from the soil. Of this, on average 67% was present in the 0-2 cm, 8% in 2-5 cm and 24% in 5-10 cm soil depths. DCD concentrations in the top 2 cm varied greatly and average concentrations of 15.5 and 11.4 mg DCD/kg soil were measured for 30 and 60 mL applications. There was little difference in DCD (1.45 mg DCD/kg soil) measured below 2 cm between application rates. Concentrations were significantly higher with a higher application rate at 0-2 cm on the Tokomaru soil but not on the Manawatū. After five days, following 24 mm rainfall, DCD recovery remained the same but its distribution and concentrations among the soil depths changed indicating its downward movement. About half of the recovered DCD remained in the 0-2 cm soil, one-third accumulated in 2-5 cm depth and the remainder was in 5-10 cm depth. The difference between urine and DCD distributions suggests that the DCD applications used in this experiment only intercepted 35-50% of the urine patch, without rainfall. With at least 24 mm of rainfall and 60 mL of DCD (13.8 kg DCD/ha) the DCD could be intercepting 80% of the urine-N. This will limit the effectiveness of DCD to reduce N leaching. It’s impact on N₂O emissions is less certain