Nitrous oxide emissions from grazed hill land in New Zealand

Abstract

Sheep and beef cattle grazed hill land represents a potentially large source of nitrous oxide (N2O) emissions globally. However, N2O emissions and associated emission factors for the dominant nitrogen (N) source of excreta N (EF3) are thought to be highly variable due to spatial differences in soil conditions across hill land units (HLUs; defined according to slope, aspect and soil type). Variability is also determined by animal grazing and resting behaviour affecting excretal-N deposition. Knowledge of spatially different EF3 values could be used to improve estimates of N2O emissions from grazed hill land. This paper presents N2O emission factors for sheep urine (SU) and dung (SD) and for beef cattle dung (BD) determined in four regions in New Zealand (NZ) (Waikato, Southern Hawkes Bay, Manawatu and Otago). Urine (spring 2009) or urine and dung (autumn 2011) was applied to low (<12°) and medium (12–25°) slopes in each region. N2O emissions were measured for 3–4 months for urine and for a whole year for dung using a static chamber technique. There were large variations in EF3 between seasons, between regions and between slope classes within a region and season. Over all regions, there was a marginally significant (P=0.08) difference in EF3 for spring 2009-applied SU on low and medium slopes, with EF3 values averaging 0.46% and 0.08%, respectively. In the autumn 2011 trial, there was no significant slope effect, with EF3 averaging 0.12% and 0.11% on low and medium slopes, respectively. By combining the datasets, EF3 for low slopes (0.24 with 95% confidence intervals of between 0.14 and 0.40) was significantly greater (P<0.05) than for medium slopes (0.07% with 95% confidence intervals of between 0.02 and 0.14). EF3 values for BD and SD were not significantly different. The contribution of sheep excreta to NZ national N2O emissions, based on a spatial framework model that disaggregates excreta deposition according to slope class and using the current inventory EF3 values of 1% for urine and 0.25% for dung and assuming that all NZ sheep grazed on hill land, was 6.08Gg N2O in 2012 in NZ. This is considerably higher than the 1.02Gg N2O estimated using the measured EF3 values, 0.24% for urine and 0.06% for dung, from this study. These results suggest that the current IPCC GHG inventory methodology is likely to overestimate N2O emissions from animal grazed hill land.