Materials and Methods
Data collection
Plant traits were derived from the large currently available databases (Kattge et al., 2020; Royal Botanic Gardens Kew, 2022) and the published literature (Westoby & Wright, 2003; Díaz et al., 2016; Santini et al., 2017). We took advantage of big data and compiled a global dataset containing 1071 species, 553 genera, 136 families. Plant traits included: 1) seed mass (mg seed-1), plant height (m), genome size (1C, pg), leaf area (mm2), leaf N (mg/g), and growth form. The growth form was split into two functional groups: “woody” and “non-woody”. We obtained 404 woody and 667 non-woody species with known trait values. Following Westoby (1998), all variables were log10-transformed prior to analysis to correct for skewness in trait distributions because trait values can vary by several orders of magnitude, and are often lognormally distributed between species. A full list of species, data and data sources was provided in Appendix S1 in the Supplementary Material.