Mountain ranges harbour exceptionally high biodiversity, which is now under threat from rapid environmental change. However, despite decades of effort, the limited availability of data and analytical tools has prevented a robust and truly global characterization of elevational biodiversity gradients and their evolutionary origins1,2. This has hampered a general understanding of the processes involved in the assembly and maintenance of montane communities2,3,4. Here we show that a worldwide mid-elevation peak in bird richness is driven by wide-ranging species and disappears when we use a subsampling procedure that ensures even species representation in space and facilitates evolutionary interpretation. Instead, richness corrected for range size declines linearly with increasing elevation. We find that the more depauperate assemblages at higher elevations are characterized by higher rates of diversification across all mountain regions, rejecting the idea that lower recent diversification rates are the general cause of less diverse biota. Across all elevations, assemblages on mountains with high rates of past temperature change exhibit more rapid diversification, highlighting the importance of climatic fluctuations in driving the evolutionary dynamics of mountain biodiversity. While different geomorphological and climatic attributes of mountain regions have been pivotal in determining the remarkable richness gradients observed today, our results underscore the role of ongoing and often very recent diversification processes in maintaining the unique and highly adapted biodiversity of higher elevations.
We thank R. E. Ricklefs, P. V. A. Fine, T. Price, C. D. Cadena, J. Beck, E. Spriggs, N. Upham and R. Freckleton for manuscript comments; members of the Future Earth Global Mountain Biodiversity Assessment, including C. Körner, E. Spehn, M. Fischer, and D. Payne for feedback; B. Klempay, A. Houston and A. Ranipeta for help collecting the elevational data; and the ‘Monitoring Häufige Brutvögel (MBH)’ project of the Vogelwarte Sempach (M. Kery) for sharing data on the Swiss breeding bird survey from 2007. We acknowledge the World Climate Research Programme’s Working Group on Coupled Modelling, which is responsible for CMIP, and we thank the climate modelling groups for producing and making available their model output. This material is based upon work supported by NSF grants DGE-1122492, DEB-1441737, and DBI-1262600.
Extended data figures
Elevational ranges for each species for each mountain range they inhabit. The fields are: Species, following the taxonomy of (Jetz et al. 2012); Country, specifies, for Andean species only (i.e., Mountain ID 404), which country the elevational range corresponds to; Mountain ID, specifies the mountain system the elevational range corresponds to (Mountain ID can be matched with the name in Table S3); Minimum Elevation and Maximum Elevation, corresponds to the elevational range used in this study; Source and Notes, species-specific references and, sometimes, specific decisions for certain elevational ranges.
Field surveys along elevation we used to compare and validate the assemblage richness data (Extended Data Fig. 3). The fields are: ID, the ID of each independent elevational transect; N_sites, number of localities where species richness was assessed; min_ele and max_ele, the elevational extent of the survey; Effort_months, an approximate measure of time spent in the filed during the survey (in months); Bird Group, the focal subset of bird species surveyed, ALL = all species, F = Forest species, B = breeding species only; Locality; Country; Reference; Notes, transect specific notes for some of the surveys.
Compilation of mountain ranges uplift times from the literature. The fields are: Continent; Mountain ID, specifies the mountain system; Mountain Range, specifies the mountain range the Mountain ID corresponds to; Uplift Ages, the first pulse of uplift (Start) and the most recent one (Last) and other significant ones in between (Other significant); Notes, specific notes related to the information compiles; Reference, numeric correspondence with the articles below where the information was obtained.
Specific Global Climate Models and Modeling groups used in this study.
Cross tables summarizing parameter changes from the sensitivity analyses.