The study area is within the Third Pole (Liu et al., 2022), and the occurrence data for all species were acquired from the Global Biodiversity Information Facility (GBIF), National Plant Specimen Resource Center (NPSRC), Plant Photo Bank of China (PPBC), previous studies (Citations) and field surveys (Table S1 and Figure 1). Raw comprehensive datasets may contain duplicated records and spatial clustering, to reduce spatial sampling bias in the occurrence data, we used the spThin R package to perform spatial thinning (Aiello-Lammens et al., 2015) ensuring the occurrences at distances <5 km. Besides, we used the alphahull (V.2.5, https://CRAN.R–project.org/package=alphahull) R package (Pateiro-López & Rodríguez-Casal, 2010) to generate the study area with a buffer range of 200 km (VanDerWal et al., 2009) for each species to avoid overfitting in the following analysis. As a result, 740 occurrence records were used in ENM analysis for 4 species (Table S2).