Including the uncertainty in topology and divergence times in historical biogeography reconstruction– two examples from the Coffee family

Jenny Smedmark, Associate Professor

Natural History Collections, University Museum of Bergen

When hypotheses of historical biogeography are evaluated, both the resolution of phylogenetic relationships and age estimates of individual nodes often have a direct impact on what explanation is concluded to be most likely. There is, however, usually some uncertainty associated with the topology, and confidence intervals of estimated divergence times are often large, something that is rarely incorporated in biogeographical analyses. We used two groups in the coffee family, both with disjunct pantropical geographic distributions, to explore how the uncertainty in topology and estimated divergence times affect conclusions in biogeographical analysis. Results from both groups using a single phylogenetic tree, such as the maximum likelihood tree with mean estimates of divergence times, provided clear support for one specific interpretation. Analyses of a large sample of dated phylogenies did, however, show that these results were not consistent, emphasizing the importance of using methods that account for the uncertainty in topology, branch lengths, and estimated divergence times in historical biogeographic inference.