Data from: High-resolution and large-extent mapping of plant species richness using vegetation-plot databases

The recent increase in the availability of large vegetation-plot databases has created unprecedented opportunities for analysing and explaining patterns of fine-scale plant species richness across large areas and for individual habitat types. Here we demonstrate how these data can be used to (1) prepare country-wide high-resolution maps of species richness and identify national diversity hotspots for grassland and forest vegetation; (2) compare diversity patterns of all, native, alien and Red List species; and (3) identify potential environmental drivers of these patterns. At the same time we examine and quantify the stability of predicted species-richness patterns with respect to the most common biases that are inherent to large vegetation-plot databases. Vegetation-plot records were obtained from the Czech National Phytosociological Database and the Random Forest method was used to map fine-scale spatial diversity patterns of all, native, alien and Red List vascular plant species, separately for grasslands and forests across the Czech Republic. The stability of the predicted species-richness patterns was tested using differently resampled datasets in which we either reduced or increased local oversampling and preferential sampling of more species-rich communities. Models for grassland and forest vegetation explained 40–65% of variation in fine-scale species richness. Spatial patterns of all and native species richness differed considerably between grasslands and forests, whereas alien and Red List species showed a higher congruence between these two vegetation types. Patterns of modelled species richness were highly stable with respect to all resampling strategies applied to the initial datasets. We conclude that vegetation-plot databases are a valuable source of data for high-resolution mapping of the plant species richness of different vegetation types and species groups, because each of them can exhibit a different diversity pattern. The resulting maps provide robust representation of the spatial patterns of fine-scale species richness and can be used both for testing scientific hypotheses about the controls of diversity patterns and for conservation planning.Predicted species richness of grassland and forest vegetation in the Czech Republic (shapefiles)ESRI shapefiles containing maps of species richness of grassland and forest vegetation in the Czech Republic modelled by Random Forest. Numbers of all, native, alien and Red List species were modelled for 16 sq. m plots for grasslands and 400 sq. m plots for forests. Model results were projected to a grid of 37,763 cells, each spanning 1.25' of longitude and 0.75' of latitude, which corresponds to approximately 1.5 × 1.39 km (2.09 sq. km).Divisek_Chytry_Diversity_maps_data2.zi