The dataset contains 5 files, including: (1) “HLS. new” is a phylogenetic tree constructed with 1,387 species, we used Taxa01, Taxa02 in the phylogenetic tree construction process (refer to Taxa match species file). Please note that I marked outgroups (9 species) in yellow color, you may use “drop tips” function in R to delete them if it’s extra info for you; (2) “Taxa match species” , Taxa name are corresponding to “HLS. new”; (3) “OGU” is a species occurrence file, each gridcell could be regard as “community”, which we can use to analysis species assembling; Gridcell in this file corresponding to the Operational Geographic Units (OGUs). Species occurrence matrix were prepared according to Silva et al.'s (Cardoso da Silva, Cardoso de Sousa, & Castelletti, 2004) method: (a) To leverage the size effect, study area was divided into 50*50 km2 grid cells, covering the land area of China including Taiwan; (b) assign species occurrence into each grid cell; (c) delimit OGUs where contains at least two endemic species and land area covered more than half of grid cells (1,250 km2). (4) “Climate”. bio 1-19 were download from CHELSA: https://chelsa-climate.org/timeseries/; (Karger et al., 2017; Karger, Nobis, Normand, Graham, & Zimmermann, 2021). I also attached the description for chelsa. (5) “Trait”. We tried our best to access to the information regarding to leaf length, height and seed diameter. For few cases, you may still find N.A. data. I believe it’s very common in macroecology research.

The dataset contains 5 files, including: (1) “HLS. new” is a phylogenetic tree constructed with 1,387 species, we used Taxa01, Taxa02 in the phylogenetic tree construction process (refer to Taxa match species file). Please note that I marked outgroups (9 species) in yellow color, you may use “drop tips” function in R to delete them if it’s extra info for you; (2) “Taxa match species” , Taxa name are corresponding to “HLS. new”; (3) “OGU” is a species occurrence file, each gridcell could be regard as “community”, which we can use to analysis species assembling; Gridcell in this file corresponding to the Operational Geographic Units (OGUs). Species occurrence matrix were prepared according to Silva et al.'s (Cardoso da Silva, Cardoso de Sousa, & Castelletti, 2004) method: (a) To leverage the size effect, study area was divided into 50*50 km2 grid cells, covering the land area of China including Taiwan; (b) assign species occurrence into each grid cell; (c) delimit OGUs where contains at least two endemic species and land area covered more than half of grid cells (1,250 km2). (4) “Climate”. bio 1-19 were download from CHELSA: https://chelsa-climate.org/timeseries/; (Karger et al., 2017; Karger, Nobis, Normand, Graham, & Zimmermann, 2021). I also attached the description for chelsa. (5) “Trait”. We tried our best to access to the information regarding to leaf length, height and seed diameter. For few cases, you may still find N.A. data. I believe it’s very common in macroecology research.