Proper rice straw management in paddy fields is necessary in order to sustain soil productivity and reduce greenhouse gas emissions. A field experiment was carried out from 2008 to 2011 in subtropical China: (1) to monitor rice yield, soil available nutrients, CH4, and N2O emissions and (2) to evaluate the effects of timing of rice straw incorporation and joint N application rate in a double rice cropping system. The total amount of rice straw from one cropping season was incorporated in winter (WS) or in spring (SS) and mineral N was jointly applied with rice straw incorporation at rates of 0, 30, and 60 % of the basal fertilization rate (N0B, N30B, and N60B) for the first rice crop. Soil water was naturally drained during the period of winter fallow (PWF) and controlled under intermittent irrigation during the period of first rice growth (PFR). Compared with SS, WS significantly (P < 0.05) increased the first rice yield only in the flooding year (2010), and increased the soil available K concentration after PWF and PFR in 2008–2009 and the hydrolysable N concentration after PWF in 2010–2011. Meanwhile, WS significantly decreased the total CH4 emission by about 12 % in 2009–2010 and 2010–2011, but increased the total N2O emission by 15–43 % particularly during PWF in all 3 years, resulting in a lower GWP in WS in the flooding year and no differences in the nonflooding years. Compared with N0B, joint N application (N60B and N30B) increased the soil hydrolysable N after PWF in all 3 years. Meanwhile, it decreased the total CH4 emissions by 21 % and increased the N2O emissions during PWF by 75–150 % in the nonflooding years, but the net GWP was lower in N60B than in N0B. The results suggested that the rice straw incorporation with joint N application in winter is more sustainable compared with the local practices such as rice straw incorporation in spring or open-field burning.