This dataset presents the Community-Weighted Mean (CWM) values which were calculated for each site of the TUSDB following the method presented in Garnier et al. (2004). To calculate hte CWM traits, the plant taxa fractional abundances were used for vegetation plots, the pollen fractional abundances were used for plot surface samples. For the plant functional traits inferred at community-scale, plant height (maximum plant height at plant maturity), leaf area (average surface of one side leaf), Leaf N (nitrogen content per leaf mass), seed mass (oven-dry mass of an average set of seeds), specific leaf area (SLA, the ratio between the area and the oven-dry mass of the leaf) and stem specific density (SSD, the ratio between stem dry mass and fresh volume of a stem section) were selected. All trait values have been z-scored.