Interannual variation in nitrous oxide emissions from perennial ryegrass/white clover grassland used for dairy production

Abstract

Nitrous oxide (N2 O) emissions are subject to intra- and interannual variation due to changes in weather and management. This creates significant uncertainties when quantifying estimates of annual N2 O emissions from grazed grasslands. Despite these uncertainties, the majority of studies are short-term in nature (<1 year) and as a consequence, there is a lack of data on interannual variation in N2 O emissions. The objectives of this study were to (i) quantify annual N2 O emissions and (ii) assess the causes of interannual variation in emissions from grazed perennial ryegrass/white clover grassland. Nitrous oxide emissions were measured from fertilized and grazed perennial ryegrass/white clover grassland (WC) and from perennial ryegrass plots that were not grazed and did not receive N input (GB), over 4 years from 2008 to 2012 in Ireland (52°51'N, 08°21'W). The annual N2 O-N emissions (kg ha(-1); mean ± SE) ranged from 4.4 ± 0.2 to 34.4 ± 5.5 from WC and from 1.7 ± 0.8 to 6.3 ± 1.2 from GB. Interannual variation in N2 O emissions was attributed to differences in annual rainfall, monthly (December) soil temperatures and variation in N input. Such substantial interannual variation in N2 O emissions highlights the need for long-term studies of emissions from managed pastoral systems.

Keywords: freeze-thaw cycles; grazing; interannual variation; nitrous oxide; rainfall; regulating factors; white clover grassland.

Fig 1

Monthly rainfall and effective rainfall (rainfall minus evapotranspiration) (mm month⁻¹) at Solohead Research farm during the study period (mm/yy). Dashed vertical lines separate measurement years.

Fig 2

Water filled pore space (%) at 0–5 cm depth (a), daily air temperature (°C) (b) and daily N₂O fluxes (mean ± SE) from unfertilized perennial ryegrass plots (GB) and grazed and fertilized perennial ryegrass/white clover pastures (WC) (c) during the study (mm/yy). Dashed vertical lines separate measurement years. Dashed arrows indicate dates of cattle slurry application; full arrows indicate dates of fertilizer N application. The vertical edges of the shaded boxes represent the beginning and end of each grazing season.

Fig 3

Annual N₂O emissions from unfertilized perennial ryegrass plots (GB) and grazed and fertilized perennial ryegrass/white clover pastures (WC) in 2008/09, 2009/10, 2010/11 and 2011/12. Statistically significant differences (P < 0.05) between year × treatments are indicated by different letters. Emissions on GB in 2010/11 may be underestimated as no measurements were taken during the winter of 2010, which had elevated emissions on WC.

Fig 4

Annual N uptake in herbage dry matter on unfertilized perennial ryegrass plots (GB) and grazed and fertilized perennial ryegrass/white clover pastures (WC) in 2008/09, 2009/10, 2010/11 and 2011/12. Statistically significant differences (P < 0.05) between year × treatments are indicated by different letters.

Fig 5

Relationship between annual N₂O emissions and N input (a) and natural log transformed annual N₂O emissions and soil N balances (Pearson correlation r = 0.29, P > 0.05) (b) using data from unfertilized perennial ryegrass plots (GB) and grazed and fertilized perennial ryegrass/white clover pastures (WC). Annual N input is the sum of fertilizer N, slurry N, N deposited during grazing, biological N fixation and rainfall N deposition. Annual N input was adjusted to account for losses of NH₃ and NO_X assumed to be 10% from fertilizer N and 20% from N in cattle slurry and N deposited during grazing (IPCC, 2006).