The objective of this study was to evaluate CO2, N2O and CH4 fluxes and C fixation in the soil-plant system, in an area with different managements of eucalyptus harvest residues. Before the experiment was set up, a plot was selected at the end of the third rotation (5 years) and harvested using the Harvester + Forwarder system. After harvest, the coppicing system was adopted and three types of residue management were established: 1. Maintenance of all harvest residues and litter from the previous rotation (HR+L), 2. Maintenance of litter from the previous rotation (L), 3. Removal of harvest residues and litter from the previous rotation (WR). Soil gas flux was sampled 5 days before harvest and at 4 times (distributed along the dry and rainy seasons) over 22.3 months. To collect the gases, PVC chambers were introduced into the soil to a depth of 5 cm in the planting rows and inter-rows. Subsequently, the gas samples were analyzed by gas chromatography. To evaluate C inputs in the system, the quantity of C in the plants and organic matter content in the first ten centimeters of soil were quantified at the end of the experiment (22.3 months). There was higher CO2 equivalent flux in the soil in the area with HR+L at 0.3 and 9.3 months after harvest, with no differences between the treatments WR and L. CO2 equivalent fluxes were similar among all treatments in the samplings performed at 15.8 and 22.3 months after harvest. No differences were observed in the treatments regarding organic matter content in the soil and C fixation in plant biomass.