Data for the manuscript "Unveiling the ghosts of landscapes past: changes in landscape connectivity over the last decades are still shaping current woodland plant assemblages"

<strong>SA_Matrix.txt</strong>: matrix of species abundance (i.e. percent cover <em>per </em>species) for the six or eight quadrats performed in the 50 woodlands. <strong>STAw_Matrix.txt</strong>: matrix of the total abundance of each species (i.e. sum of the percent cover recorded in the six or eight quadrats <em>per </em>species) weighted by the number of quadrats for the 50 woodlands. <strong>AFH_matrix.txt</strong> = matrix of the species affinity to forest habitats (France, Atlantic region) with the following typology: "1" = specialist forest species (species that occur only in forests), "2" = generalist forest species (species that occur both in and outside forests) and "O" = species of open habitats. <strong>TD_Matrix.txt</strong>: taxonomic diversity [total species richness (richness) and Piélou's evenness (evenness)] matrix for the 50 woodlands. <strong>TD_GSFS_Matrix.txt</strong>: taxonomic diversity [total species richness (richness) and Piélou's evenness (evenness)] of specialist and generalist forest species for the 50 woodlands. <strong>CT_matrix.txt</strong>: species trait values matrix for each continuous trait considered [number of seeds per plant (NSPP), seed dry mass (SDM), and seed germination rate (SGR)]. <strong>T_matrix.txt</strong>: species trait values matrix for each trait considered [number of seeds per plant (NSPP), seed dry mass (SDM), seed germination rate (SGR) and type of reproduction (ToR)] following a MICE procedure to estimate missing continuous trait values. <strong>FS_Matrix.txt</strong>: functional structure [community-weighted mean (CWM) and Rao’s quadratic entropy (RaoQ)] matrix for each trait considered [number of seeds per plant (NSPP) seed dry mass (SDM), plant vegetative height (PHV), seed germination rate (SGR) and type of reproduction (ToR)] in the 50 woodlands. <strong>Connectivity_matrix.txt</strong>: matrix of the connectivity [Potential Dispersal Flux (PCflux)] for each time step (1952, 1974, 1985, 2000, 2006 and 2016)] for the 50 woodlands. <strong>Temporal_dynamics_connectivity_matrix.txt</strong>: matrix of the temporal dynamics of connectivity [the degree of connectivity (“potential Dispersal Flux” index obtained for 1952, PCflux_1952), the magnitude (the absolute value of the slope of OLS regression model, Abs_slope) and the variability (the residual standard error of OLS regression model RSE)] for the 50 woodlands and their associated trends (estimated from the value of the slope of OLS regression model, Slope). <strong>Data_REG.txt</strong>: data* to perform regression models (that systematically include the degree of connectivity or not) using model-averaging methodology *Coordinates (X and Y) of the centroids of the 50 woodlands, and merging of Temporal_dynamics_connectivity_matrix.txt, TD_Matrix.txt and FS_Matrix.txt. <strong>Script_TD.R</strong>: script to obtain the matrix of the total species richness (richness) and Piélou's evenness (evenness) for the 50 woodlands (TD_Matrix.txt) <strong>Script_TD_GSFS.R</strong>: script to obtain the matrix of the total species richness (richness) and Piélou’s evenness (evenness) of specialist and generalist forest species for the 50 woodlands. <strong>Script_CT.R</strong>: script to obtain species trait values matrix for each continuous trait considered [number of seeds per plant (NSPP), seed dry mass (SDM), plant vegetative height (PHV) and seed germination rates (SGR)] (CT_matrix.txt). Please, note that the data obtained following our request to the TRY global plant trait database cannot be made publicly available. <strong>Script_FS.R</strong>: script to obtain functional structure [community-weighted mean (CWM) and Rao’s quadratic entropy (RaoQ)] matrix for the 50 woodlands (FS_Matrix.txt) <strong>Script_REG.R</strong>: script to perform regression models using model-averaging methodology. <strong>Script_REG_DoC_included.R:</strong> script to perform regression models that systematically include the degree of connectivity using model-averaging methodology.