Frontiers of Biogeography

Recent climatic models suggest the late Pleistocene was colder and had different precipitation regimes from the present. If this climatic shift occurred more rapidly than species could adapt, species likely shifted their ranges as populations moved in concert with suitable environmental conditions. We examined changes in altitudinal and horizontal distribution in response to past climate change of amphibian species from different elevational zones and habitat requirements in the Eastern Cordillera of the Colombian Andes. We used environmental information and species occurrence data to model the distribution of 14 amphibian species (seven highland and seven lowland) which we projected to the Last Glacial Maximum (LGM) using two past climatic reconstructions. For these 14 species, we studied the predicted elevational and areal shifts. In agreement with palynological-derived models for Andean flora, we predicted that the elevation of montane amphibians shifted downwards increasing their total altitudinal range. We did not detect any evidence of compression related to drier lowlands. In some cases, the wider distribution areas of high-elevation amphibians during the LGM overlapped with contemporary distributions implying that these areas are present-day refugia for some species. Lowland species showed little or no elevational changes across time, but their areal distributions changed depending on habitat requirements. Four lowland frog species occurring in present-day xeric environments showed substantial range expansion across the lowlands during the LGM, while two species occurring in humid habitats likely expanded their ranges since the LGM. Since the LGM ended, ranges of mid‐ to high-elevation species shrank and shifted to higher elevation. Lowland species in xeric or open habitats have also experienced shrinking ranges, with some evidence that they have been moving upwards. Thus, low- and high-elevation species may be at risk under predicted anthropogenic climate change. Our results generate spatial hypotheses about amphibian responses to climate change that can be tested with phylogeographic data.