Aims Plants have been suggested to have significant effects on methane (CH 4 ) and nitrous oxide (N 2 O) fluxes from littoral wetlands, but it remains unclear in subtropical lakes. Methods We conducted in situ measurement of CH 4 and N 2 O fluxes for two years. To distinguish between the effects of shoots and roots, three treatments (i.e., intact plants as control, shoot clipping, and root exclusion) were used. Effects of plant biomass, temperature, and soil moisture on CH 4 and N 2 O fluxes were analyzed. Results The mean ecosystem CH 4 emission rate was 36 μg CH 4 m −2 h −1 for drying periods, but 8219 μg CH 4 m −2 h −1 for drying-wetting transition periods. CH 4 fluxes were positively correlated with below-ground and total biomass, but not with above-ground biomass. Clipping did not significantly alter CH 4 flux rate, but root exclusion decreased the CH 4 flux by 116 % as compared to the control. N 2 O emissions were similar for both the drying and drying-wetting transition periods, with a mean rate of 20 μg N 2 O m −2 h −1 . Both clipping and root exclusion significantly increased N 2 O fluxes as compared to the control. Conclusions There was no significant correlation between CH 4 and N 2 O fluxes. Roots dominated plantmediated enhancement in CH 4 fluxes, but played almost an equal role as shoots in plant-regulated suppression on N 2 O fluxes in this Carex meadow during drawdown periods.