COS 56-10 - Are mountain passes higher in the tropics? A community phylogenetic approach

Tuesday, August 7, 2012: 4:40 PM
F151, Oregon Convention Center
Christopher L. Roy, Brian S. Maitner and Amy E. Dunham, Ecology and Evolutionary Biology, Rice University, Houston, TX
Background/Question/Methods

Understanding how biotic interactions, environment, and geography shape community structure is a central goal in ecology. Daniel Janzen predicted in 1967 that the low seasonal variation in ambient temperature near the equator would cause high-elevation sites to be stronger physiological barriers to tropical organisms than organisms in higher latitudes. This hypothesis has since been central to discussions regarding latitudinal patterns of adaptation and community structure, however many of the predictions of this hypothesis have not been tested. Recent methodological advances in studies of community processes, including phylogenic approaches, have strengthened our ability to examine and predict community patterns and likelihood of species coexistence. If physiological barriers at high elevations become more extreme as you approach the equator, this may result in predictive patterns in the phylogenetic relatedness among community members.  Strong physiological barriers of the tropical mountain peaks may favor organisms which share traits and allow them to overcome these barriers (habitat filtering).  If these traits are conserved phylogenetically, habitat filtering is expected to result in communities that are more closely related. Thus Janzen’s hypothesis predicts a stronger effect of elevation on community phylogenetic structure near the equator.

Results/Conclusions

We use a phylogenetic approach to re-examine Janzen’s hypothesis.  To test the predictions of this hypothesis, we explored the phylogenetic community structure of bird communities along an elevational and latitudinal gradient in the Andes Mountains in South America.  We created phylogenies for 60 different communities which varied in both elevation and latitude.  We calculated measures of relatedness between species in each community and used generalized linear models to test for associations between community relatedness indices and elevation, relatedness and latitude, and any interactive effects of distance from equator (latitude) and elevation. Overall, communities showed a pattern of significant overdispersion (composition of distant relatives).  However, relatedness between community members increased near the equator and at higher elevations.  Finally, we found a significant interaction between distance from the equator and elevation suggesting the effect of elevation on communities was strongest near the equator. These findings are consistent with predictions of Janzen’s hypothesis.