Urban green spaces provide important ecosystem services, such as amenity, biodiversity, productivity, climate amelioration, hydrological and biogeochemical cycling. Intensively managed urban gardens can sequester carbon through vegetation growth and soil C increase, but may experience nitrous oxide (N 2 O) emissions and reduced soil methane (CH 4 ) uptake from irrigation and fertiliser use. Soil atmosphere exchange of N 2 O, CH 4 and carbon dioxide (CO 2 ) was measured in lawn and wood chip mulched garden areas in Melbourne, Australia in winter, spring and summer under various water and fertiliser regimes. Gas exchange before and after lawn fertiliser application was measured continuously for three weeks using an automated chamber system. Applying fertiliser led to a peak N 2 O emission of >60 μg N m −2 h −1 , but overall only weekly irrigation (10 mm) significantly increased mean soil N 2 O emissions above that in other treatments. Under mulch, mean soil N 2 O emissions (14.0 μg N m −2 h −1 ) were significantly smaller than from irrigated lawn (27.9 μg N m −2 h −1 ), whereas mean soil CH 4 uptake under mulch (−30.7 μg C m −2 h −1 ) was significantly greater (p<0.01) than in any lawn treatment. Lawns were either a weak CH 4 sink or source. Soil C density (0-25 cm) under mulch (12.5 kg C m −2 ) was greater that under lawn (8.0 kg C m −2 ). On a carbon dioxide equivalent (CO 2 -e) basis, soil N 2 O emissions offset the benefits of soil CH 4 uptake. Mulched garden areas provide greatest C sequestration potential in soil and vegetation and the smallest non-CO 2 emissions, as soil CH 4 uptake offsets a large fraction of soil N 2 O emissions. Results of this study suggest that reducing the irrigation and fertiliser application to lawns can help mitigate GHG emissions from urban garden systems, and increasing the area of mulched perennial garden beds can also provide net GHG benefits;