A field experiment was conducted at the Key Field Station for Monitoring Eco-environment of Purple Soil of the Ministry of Agriculture of China in the farm of Southwest University,Chongqing. The static chamber and gas chromatography method was used to study the effect of the cropping systems on greenhouse gases from rice-fallow ( RF), rice-rapeseed rotation ( RR), and rice-brussel mustard rotation ( RV) cropland in situ for a year. An opaque chamber was used for CH_4 and N_2 O observations and a transparent chamber was utilized for CO_2 observations. The results show that the annual cumulative CH_4 emissions from different crop rotations were ( CH_4,kg·hm~(- 2) ) RF ( 422.87 27.1) > RR ( 132.05 23.11) > RV ( 50.91 3.83) . The RV and RR were significantly lower than RF ( P < 0.05) . The annual cumulative emissions of N_2 O[N_2 O,kg·hm~(- 2)]were RV ( 21.38 6.51) > RR ( 20.02 5.23) > RF ( 0.48 0.02) . The RV and RR were significantly higher than RF ( P < 0.05) . The annual net cumulative emissions of CO_2 were ( CO_2,t·hm~(- 2) ) RR ( - 55.43 5.04) > RV ( - 29.1 3.00) > RF ( - 14.08 1.81) . The RV and RR were significantly higher than RF ( P < 0.05) . At the time scale of 100 a,the integrated global warming potentials ( GWP) of CH_4,N_2 O,and CO_2 were ( CO_2,t·hm~(- 2) ) RR( - 46.43) > RV( - 22.01) > RF( - 2.11), indicating that converting flooded paddy fields to paddy-upland crop rotation systems notably increases the potential increment of carbon sinks. Compared with RV,RR has a better effect,which suggests that rice-rapeseed rotation is the most effective measure for the escalation of carbon sinks of ecosystems in the southwestern area.