Aims For tropical and subtropical soils, information is scarce regarding the global warming potential (GWP) of no-till (NT) agriculture systems. Soil organic carbon (OC) sequestration is promoted by NT agriculture, but this may be offset by increased nitrous oxide (N2O) emissions. We assessed the GWP of a NT as compared to conventional tillage (CT) in a subtropical Brazilian Ferralsol. Methods From September 2008 to September 2009 we used static chambers and chromatographic analyses to assess N2O and methane (CH4) soil fluxes in an area previously used for 3-4 years as a field-experiment. The winter cover crop was ryegrass (Lolium multiflorum Lam.) while in summer it was silage maize (Zea mays L.). Results The accumulated N2O emission for NT was about half that of CT (1.26 vs 2.42 kg N ha(-1) year(-1), P = 0.06). Emission peaks for N2O occurred for a month after CT, presumably induced by mineralization of residual nitrogen. In both systems, the highest N2O flux occurred after sidedressing maize with inorganic nitrogen, although the flux was lower in NT than CT (132 vs 367 mu g N m(-2) h(-1), P = 0.05), possibly because some of the sidedressed nitrogen was immobilized by ryegrass residues on the surface of the NT soil. Neither water-filled pore space (WFPS) nor inorganic nitrogen (NH (4) (+) and NO (3) (-) ) correlated with N2O fluxes, although at some specific periods relationships were observed with inorganic nitrogen. Soils subjected to CT or NT both acted as CH4 sinks during most of the experiment, although a CH4 peak in May (autumn) led to overall CH4 emissions of 1.15 kg CH4-C ha(-1) year(-1) for CT and 1.08 kg CH4-C ha(-1) year(-1) for NT (P = 0.90). The OC stock in the 0-20 cm soil layer was slightly higher for NT than for CT (67.20 vs 66.49 Mg ha(-1), P = 0.36). In the 0-100 cm layer, the OC stock was significantly higher for NT as compared to CT (234.61 vs 231.95 Mg ha(-1), P = 0.01), indicating that NT resulted in the sequestration of OC at a rate of 0.76 Mg ha(-1) year(-1). The CO2 equivalent cost of agronomic practices was similar for CT (1.72 Mg CO(2)eq ha(-1) year(-1)) and NT (1.62 Mg CO(2)eq ha(-1) year(-1)). However, NT reduced the GWP relative to CT (-0.55 vs 2.90 Mg CO(2)eq ha(-1) year(-1)), with the difference of -3.45 Mg CO(2)eq ha(-1) year(-1) (negative value implies mitigation) being driven mainly by OC sequestration. The greenhouse gas intensity (GHGI, equivalent to GWP/silage yield) was lower for NT than CT (-31.7 vs 171.1 kg CO(2)eq Mg-1 for silage maize). Conclusion As compared to CT, greenhouse gas emissions from a subtropical soil can be mitigated by NT by lowering N2O emissions and, principally, sequestration of CO2-C.
