Methane (CH 4 ) and nitrous oxide (N 2 O) emissions from paddy soils have seldom been estimated when leguminous green manure is applied as a nitrogen source. In this paper, gas fluxes were measured by using a pot sampling device combined with a static chamber method to estimate the effects of Chinese milk vetch (Astragalus sinicus L., CMV) on CH 4 and N 2 O emissions and their integrated global warming potentials (GWP) in a double-rice cropping system. Four treatments (no nitrogen fertilizer, NF; urea as chemical fertilizer, CF; CMV incorporation, MV; 50% CMV incorporation and 50% urea, MVCF) were established. CH 4 flux peaked on the 15th d after treatment application. Total season CH 4 emission was increased by MV and MVCF by 370 and 209%, 152 and 66%, when compared with NF and CF, respectively. Most of the increased CH 4 was emitted in the first two months after incorporation of CMV. N 2 O emission from CF was 17- and 5.6-fold higher than that from MV and MVCF, respectively. Application of CMV restricted N 2 O emission caused by the application of urea. Improved CMV residue management was needed to minify CH 4 emission induced by the input of organic material. Despite the highest GWP being found in MV, we recommend CMV, when applied as an N source in paddy fields, as a potential mitigation tool for greenhouse gas emissions. Methane (CH 4) and nitrous oxide (N 2 O) emissions from paddy soils have seldom been estimated when leguminous green manure is applied as a nitrogen source. In this paper, gas fluxes were measured by using a pot sampling device combined with a static chamber method to estimate the effects of Chinese milk vetch (Astragalus sinicus L., CMV) on CH 4 and N 2 O emissions and their integrated global warming potentials (GWP) in a double-rice cropping system. Four treatments (no nitrogen fertilizer, CH 4 flux peaked on the 15th d after treatment application. Total season CH 4 emission was increased by MV and NF; urea as chemical fertilizer, CF; CMV incorporation, MV; 50% CMV incorporation and 50% urea, MVCF) MVCF by 370 and 209%, 152 and 66%, when compared with NF and CF, respectively. Most of the increased CH 4 was emitted in the first two months after incorporation of CMV. N 2 O emission from CF was 17- and 5.6 -fold higher than that from MV and MVCF, respectively. Applicat ion of CMV restricted N 2 O emission caused by the application of urea. Improved CMV residue management was needed to minify CH 4 emission induced by the input of organic material. Despite the highest GWP being found in MV, we recommend CMV, when applied as an N source in paddy fields, as a potential mitigation tool for greenhouse gas emissions.