The Geography of Speciation Across Andean-Centred Clades

Aim: Mountains harbour a disproportionate share of global biodiversity, yet the geographic pathways through which this diversity arises remain debated. In the Andes, diversification has been linked both to geographic isolation generated by complex topography and climatic dynamics, and to ecological divergence across steep environmental gradients. Here, we assess the predominant geographic modes of speciation across Andean-centred clades using a large-scale comparative framework, while evaluating how alternative representations of species distributions influence macroevolutionary inference. Location: Andes, South America. Taxon: A total of 102 Andean-centred genera of animals, plants, and fungi. Methods: We combined time-calibrated phylogenies with species distribution data derived from two alternative spatial representations: concave polygons and potential distribution models. For each genus, we estimated 13 summary metrics capturing range overlap, range asymmetry, and phylogenetic structure. We then used simulation-trained, likelihood-free classification approaches, including linear discriminant analysis and two approximate Bayesian computation methods, to assign each clade to its most probable geographic mode of speciation: vicariant, peripatric, parapatric, sympatric, or mixed geographic scenarios. Results: Across clades, overlap-associated scenarios, corresponding to parapatric or sympatric classifications, were inferred as predominant in 57–67% of genera, whereas strictly allopatric scenarios, that is, vicariant or peripatric speciation, accounted for 23–30%, depending on the spatial representation and classification method used. Sympatric scenarios alone were inferred in 38–52% of genera, particularly when potential distribution models were used. At least one-third of clades showed congruent classifications across both analytical methods and spatial representations. However, several genera were sensitive to how species distributions were represented, with some shifting between allopatric and sympatric classifications depending on the spatial input used. Main Conclusions: Our results suggest that Andean diversification has not followed a single dominant geographic pathway, but instead reflects a mosaic of spatial contexts involving both isolation-driven and overlap-associated scenarios. However, classifications as parapatric or sympatric should be interpreted as hypotheses of geographic overlap during lineage divergence, rather than as direct evidence of speciation with gene flow. By shifting from single-clade case studies to a broad multi-taxon comparative perspective, this study suggests that the geography of Andean speciation reflects a heterogeneous process shaped by multiple spatial pathways to lineage divergence, while identifying candidate clades for future phylogeographic, ecological, cytogenetic, and population genomic studies of mountain biodiversity.