Aim: We tested the hypothesis that areas that acted as historical refugia during restrictive climate regimes currently harbour higher levels of biodiversity than areas that lacked refugia. Location: The rain forests of Australia's Wet Tropics, the largest remaining fragments of the humid forest habitats that once covered the Australian continent. Methods: We generated a model of climatic suitability for arachnids in the genus Austropurcellia, a group of small, dispersal-limited mite harvestmen that are found throughout the Wet Tropics. We then projected this model onto palaeoclimate data layers from time slices going back to the Last Glacial Maximum and summed suitability over time to arrive at a measure of stability. We compared the power of metrics of present and past climatic suitability and stability to predict diversity (species richness and phylogenetic diversity) across subregions of the Wet Tropics. Results: We found statistically significant correlations between measures of diversity (species richness and phylogenetic diversity) and present climatic suitability, LGM climatic suitability and our stability metric across subregions of the Wet Tropics. Although stability lost predictive power when analyses were corrected for spatial autocorrelation, and present-day mean climatic suitability lost predictive power when corrected for spatial autocorrelation under one of our geographical binning schemes, mean climatic suitability during the Last Glacial Maximum had a positive and significant relationship to both number of species and phylogenetic diversity in all analyses. Main Conclusions: Our results support a model of biodiversity preservation within historical refugia, resulting in higher present-day diversity in refugial areas than in non-refugial areas. Although previous studies of the Wet Tropics biota have demonstrated a relationship between habitat stability and diversity, ours is the first such study to consider phylogenetic diversity in addition to number of species.
Bibliographical noteFunding Information:
Jeremy VanDerWal generously shared all climatic data layers and Catherine Graham provided the Wet Tropics subregions shapefile used in this study. This project would not have been possible without the painstaking taxonomic work of Macalester College students Zach Popkin-Hall, Rachel Quay, Hannah Wiesner and Domi Lauk?, as well as the members of the Fall 2011 iteration of Biology 476: Research in Biodiversity and Evolution (Meghan Davies, Sam DelSerra, Joel Soma, and especially Matt Vance and Whitney Watson). Macalester students Michelle Coblens and Katya Jay generated the new DNA sequences used in this study. Jeff Thole provided technical support for all scanning electron microscopy, and Ashley Nepp provided technical support around ArcGIS analyses. Mike Anderson provided helpful discussion of statistical analyses. The heroic fieldwork of Geoff Monteith and colleagues over multiple decades was essential to the collection of the majority of the specimens included in this study. Mike Rix, Mark Harvey, Gonzalo Giribet, Prashant Sharma and students enrolled in the Spring 2015 iteration of Biology 476: Research in Biodiversity and Evolution (Mito Imagawa, Joanne Johnson, Kaelyn Lemon, Jill Oberski, Kenji Shoemaker and Mo Usavage) all provided helpful conversation about the project. For loans of specimens we thank Robert Raven (Queensland Museum), Gonzalo Giribet (Harvard Museum of Comparative Zoology), Petra Sierwald (Field Museum), Mark Harvey (Western Australia Museum), Charles Griswold (California Academy of Sciences), Peter Schwendinger (Mus?e d'Histoire Naturelle Gen?ve) and Beth Mantle (Australian National Insect Collection). Mike Rix and two anonymous referees provided helpful feedback on an earlier version of this paper. This work was supported by the National Geographic Society, NSF DEB 1020809 to SLB, and NSF DEB 1020426 to KHK.
© 2016 John Wiley & Sons Ltd