Abstract: Nitrous oxide (N₂O) was undoubtedly a key element with respect to global warming and climate change. As about 60% of global anthropogenic N₂O emissions have originated from agricultural soils, it has been a huge challenge to maintain high crop yields while reducing N₂O emissions under the continued application of nitrogen fertilizers. Limited available field data have explored the effects of fertilizer application regimes on soil N₂O emissions in the Sichuan Basin, one of the most intensive agricultural regions in China. Thus the objective of this research was to compare the effects of different N fertilizer application regimes on soil N₂O emissions so as to find the best N fertilizer management practice which not only decreased N₂O emission, but also maintained high crop yields. Static chamber-gas chromatographic techniques were used to measure soil N₂O emissions from purple soil croplands with long-term N fertilization in treatments of pure synthetic N fertilizer (N), pig manure (OM), synthetic NPK fertilizer (NPK), pig manure combined with synthetic NPK fertilizer (OMNPK) and returned crop residues combined with synthetic NPK fertilizer (CRNPK). No fertilizer was applied under the control experiment (NF). The in situ field measurements were conducted during wheat season from November 2012 to May 2013. The results showed that cumulative N₂O fluxes in different fertilization regimes were significantly different (P < 0.05) even when the same rates of total N were applied 130 kg(N)·hm^-2. Cumulative N₂O fluxes in treatments of N, OM, NPK, OMNPK and CRNPK were 0.38 kg(N)·hm^-2, 0.36 kg(N)·hm^-2, 0.29 kg(N)·hm^-2, 0.33 kg(N)·hm^-2 and 0.19 kg(N)·hm^-2, respectively. N₂O emission factors in the respective treatments were 0.25%, 0.23%, 0.18%, 0.21% and 0.10%. Cumulative N₂O flux in treatment of NF was 0.06 kg(N)·hm?2. Nitrogen fertilizer application significantly enhanced N₂O emission. Inorganic N (ammonium and nitrate) content was the main controlling factor of soil N₂O emission. Also rainfall significantly enhanced soil N₂O emission. Wheat grain yields in treatments of N, OM, NPK, OMNPK and CRNPK were 1.34 t·hm^-2, 3.71 t·hm^-2, 2.80 t·hm^-2, 3.13 t·hm^-2 and 3.40 t·hm^-2, respectively. Wheat grain yield in CRNPK treatment was not significantly different from that of OM treatment, which had the maximal yield that was much higher than yields of all the other treatments (P < 0.05). Furthermore, we assessed the global warm potential (GWP) of soil N2O emission on the basis of crop grain yield under the different treatments. The results showed that the yield-based GWPs of treatments of N, OM, NPK, OMNPK and CRNPK were 132.57 kg(CO₂ eq)·t^-1, 45.70 kg(CO₂ eq)·t^-1, 49.07 kg(CO₂ eq)·t^-1, 48.92 kg(CO₂ eq)·t^-1 and 26.41 kg(CO₂ eq)·t^-1, respectively. The yield-based GWP of soil N₂O emissions in CRNPK treatment was the lowest of all the treatments. In contrast with conventional fertilization regime (NPK), the yield-based GWP decreased by 46%. This implied that returned crop residues combined with synthetic NPK fertilizer not only increased grain yield, but also decreased soil N₂O emission due to the application of N fertilizer. For wheat season, the fertilizer application regime of returned crop residues with reduced synthetic fertilizer was therefore recommended as the best fertilization measure in purple soil croplands.