Abstract: Abstract: Continuous emissions of greenhouse gases (GHGs) will cause further warming and changes of the climate system. Nitrous oxide (N2O) is an important GHG. Irrigation mode is an important factor in regulating N2O emissions from croplands. Controlled irrigation (CI) is one of the major water-saving irrigation practices, and has been widely applied. Compared with traditional irrigation (TI), CI leads to remarkable changes in soil properties and soil biochemical processes, which consequently induces changes in N2O emissions. Whether CI practiced during the rice season has subsequent effects on N2O emission during the following winter wheat season is worthy of further study. Thus, a field experiment was designed to study the effects of CI during the rice season on N2O emissions from the rice-winter wheat rotation systems, with TI as the control. The gas samples were collected by the static chamber technique. The chamber, consisting of 2 separate layers with the same size (0.5 m × 0.5 m × 0.6 m), was made of polyvinyl chloride. Samples were collected by 50 mL syringes, which were connected to chambers and sealed airbags through 3 stopcocks. Irrigation mode during the rice growing season had obvious subsequent effects on N2O emission from the following winter wheat growing season. Compared to TI, CI significantly increased the N2O emission during the rice growing season (P<0.05), but significantly decreased it during the wheat season (P<0.05). During the rice growing season, the mean of N2O emissions from the CI plots was 229.25 μg/(m2·h), which was 1.66 times higher than that from the TI plots. During the winter wheat season, N2O emissions from the CI plots were generally lower than those from the TI plots. High N2O emissions from the CI plots were mainly observed after the regreening fertilizer application. Two or three peaks were observed from the CI plots in the middle and late stage of wheat growth, and the peak values were less than those from the TI plots. In the TI plots, N2O emissions were always high in the early stage of wheat growth and after the regreening fertilizer application, and were low during the overwintering period of wheat growth. Four peaks of N2O emissions were observed from the TI plots. During the winter wheat season, N2O emissions were closely related to rainfall. High N2O emissions were observed from 7 to 10 days after rainfall. Compared to TI, CI significantly increased the cumulative N2O emission in rice growing season by 136.9% (P<0.05), but decreased it in wheat growing season by 47.1% (P<0.05). Over the whole annual cycle, the cumulative N2O emission from the plots under the CI was 761.50 mg/m2, 1.0% lower than that under TI (P>0.05). During the 2009-2010 and 2010-2011 wheat growing season, the cumulative N2O emissions from the CI fields were reduced by 46.7% and 47.6% respectively, compared with those from the TI fields. The cumulative N2O emissions from the rotation systems under CI were 6.88 and 8.35 kg/ha in 2009-2010 and 2010-2011, respectively, reduced by 5.7% and increased by 3.2% respectively, compared with those from the TI fields. The N2O-N loss was used to measure the amount of N2O emissions from the croplands. During the rice growing season in 2009 and 2010, the mean N2O-N losses under CI were 0.97% and 1.13%, respectively, while the mean N2O-N losses under TI were 0.40% and 0.49%, respectively. During the 2009-2010 and 2010-2011 wheat season, the mean N2O-N losses in the CI plots were both 0.86%, while N2O-N losses in TI plots were 1.62% and 1.63%, respectively. Over the whole annual cycle, the mean N2O-N losses were 1.01% and 0.98% in CI and TI plots, respectively. These results suggest that controlled irrigation dose not increase the cumulative N2O emission from the rice-winter wheat rotation systems, compared to traditional irrigation.