The role of plant community structure and plant functional traits for above- and belowground carbon (C) fluxes was studied for 2 years in a mesocosm experiment with grassland monoliths, using continuous gas exchange measurements and soil analyses. Here we test the response-and-effect trait hypothesis, by applying a mathematical framework used to predict changes in C fluxes after a change in disturbance through the community response (R) and effect (E) traits. Monoliths were extracted from two contrasted long-term field treatments (high vs. low grazing disturbance) and exposed to both low and high (simulated grazing) disturbance during a 2 years experiment. Carbon dioxide exchanges were measured continuously in an open flow system. Net ecosystem productivity and ecosystem C balance were positively correlated at low disturbance with plant species richness. Aboveground net primary productivity (ANPP) and soil C sequestration were, however, unrelated to these variables. Community aggregated leaf (specific leaf area, leaf dry-matter content) and root and rhizome (specific length, tissue density, diameter) traits responded (R) significantly to changes in disturbance, indicating an increased dominance of conservative plant growth strategies at low compared with high disturbance. Applying the mathematical framework, ANPP was predicted by distribution of leaf traits within the community (functional divergence), while mean root and rhizome traits had significant effects (E) on soil C sequestration, irrespective of the experimental disturbance and of the year. According to highly significant linear regression models, between 6% and 61% of the transient changes in soil C sequestration resulted from community root and rhizome (response-and-effect) traits after a change in disturbance.