Greenhouse gas fluxes from an Australian subtropical cropland under long-term contrasting management regimes

The long-term effects of conservation management practices on greenhouse gas fluxes from tropical/subtropical croplands remain to be uncertain. Using both manual and automatic sampling chambers, we measured N2O and CH4 fluxes at a long-term experimental site (1968-present) in Queensland, Australia from 2006 to 2009. Annual net greenhouse gas fluxes (NGGF) were calculated from the 3-year mean N2O and CH4 fluxes and the long-term soil organic carbon changes. N2O emissions exhibited clear daily, seasonal and interannual variations, highlighting the importance of whole-year measurement over multiple years for obtaining temporally representative annual emissions. Averaged over 3 years, annual N2O emissions from the unfertilized and fertilized soils (90 kg N ha(-1) yr(-1) as urea) amounted to 138 and 902 g N ha(-1), respectively. The average annual N2O emissions from the fertilized soil were 388 g N ha(-1) lower under no-till (NT) than under conventional tillage (CT) and 259 g N ha(-1) higher under stubble retention (SR) than under stubble burning (SB). Annual N2O emissions from the unfertilized soil were similar between the contrasting tillage and stubble management practices. The average emission factors of fertilizer N were 0.91%, 1.20%, 0.52% and 0.77% for the CT-SB, CT-SR, NT-SB and NT-SR treatments, respectively. Annual CH4 fluxes from the soil were very small (-200-300 g CH4 ha(-1) yr(-1)) with no significant difference between treatments. The NGGF were 277-350 kg CO2-e ha(-1) yr(-1) for the unfertilized treatments and 401-710 kg CO2-e ha(-1) yr(-1) for the fertilized treatments. Among the fertilized treatments, N2O emissions accounted for 52-97% of NGGF and NT-SR resulted in the lowest NGGF (401 kg CO2-e ha(-1) yr(-1) or 140 kg CO2-e t(-1) grain). Therefore, NT-SR with improved N fertilizer management practices was considered the most promising management regime for simultaneously achieving maximal yield and minimal NGGF.