Data from: Joint effects of environmental filtering and dispersal limitation on species assemblage of the Tibetan Plateau

Aim Mountains harbour a rich and non-random cluster of species, yet knowledge on the species' biological attributes that support species coexistence in the montane community is limited. Here, we investigated the association of species occurrence on the Tibetan Plateau with species' morphological, ecological or evolutionary constraints. Location Tibetan Plateau (TP) Taxon Mammals and birds Methods We tested whether species occurrence on the TP correlates with morphological, ecological, or evolutionary constraints using the spatial distribution, phylogeny, dispersal ability, and thermal niche property data for 1,353 terrestrial vertebrates (383 mammals and 970 birds). We used standard (non-phylogenetic) and phylogenetic logistic regressions to disentangle the relative contributions of these attributes of species in explaining the species occurrence on the TP. We assessed the geographical patterns of community structures on the TP and fit linear mixed models to explore the underlying eco-evolutionary forces. Results The TP species exhibited a higher cold tolerance, wider thermal niche breadth, and higher rate of niche evolution than the non-TP species. We supported the assumption that the TP species was not a random subset from the species pool, but was structured jointly by environmental filtering and dispersal limitation. While dispersal and ecological processes underlying species assemblages varied spatially and among taxa, we found that species in stressful environments was limited by environmental filtering, whereas dispersal limitation was more pronounced under favourable climatic conditions. Main conclusions Our study finds that environmental filtering and dispersal limitation jointly shape the species assemblage on the TP. These findings provide significant insights into community assembly processes on the TP and other montane ecosystems on Earth, especially those that are sensitive to global warming.This zip package contains the data supporting the findings of the following paper: He et al. Joint effects of environmental filtering and dispersal limitation on species assemblage of the Tibetan PlateauFunding provided by: National Natural Science Foundation of ChinaCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001809Award Number: 31900324Funding provided by: Guangdong Basic and Applied Basic Research FoundationCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100021171Award Number: 2020A1515011472Distribution data. Geographical ranges of the species were obtained from the expert range maps produced by the IUCN Red List database (http://www.iucnredlist.org) for mammals, and BirdLife International and NatureServe (http://www.birdlife.org) for birds. Dispersal ability. The dispersal ability of birds was described by the hand-wing index (HWI) obtained from Sheard et al. (2020). The dispersal ability of mammals was estimated by the home range size following the methods provided by Tucker et al. (2014). This method fitted linear models between empirical data of the body mass and home range size of mammals separately for carnivores, herbivores, and omnivores. And the home range sizes of the remaining mammals were interpolated using the fitted models. Species thermal niches. Thermal niche properties per species were measured using temperature variables across every 110 km × 110 km grid cell within a species range. Climate data was derived from the WorldClim data set (version 2.0; 2.5 arc min spatial resolution; Fick & Hijmans, 2017). Diversification rate. The species-level diversification rate was calculated based on the equal-splits metric following Jetz et al. (2012). Dated phylogenies were from obtained Upham, Esselstyn, & Jetz (2019) for mammals and Jetz et al. (2012) for birds. Rate of thermal niche evolution. The rates of thermal niche evolution were calculated based on the phylogeny of species within each clade (i.e., genus or family) using the thermal tolerance (i.e., minimum BIO1 values) per species. Ancestral values for each node were reconstructed using the phylogenetic generalized least-squares approach and the phylogenies were transformed based only on the Brownian motion model