Kyle G. Dexter
Ph.D. Candidate
Biology Department, Duke University
Advisors: Cliff Cunningham, John Terborgh
Curriculum Vitae html pdf
Contact information
Biology Department
Duke University
Box 90338
Durham, North Carolina 27708
(919) 660-7431
kyle.dexter at duke.edu

General Research Interests

I am broadly interested in community ecology and biogeography. The basic question I strive to answer for any group of organisms is this: why are the species that are found in a given place those species as opposed to any other species? There are many factors that can determine this including the regional species pool, the dispersal abilities of species, the adaptations of species for particular habitats, and interactions between species. I use molecular methods (phylogenetics and population genetics) to approach these basic topics in a new and informative manner.

Specific Research Projects

1) Community Ecology of Inga species in Southwestern Amazonia

As part of my dissertation research, I surveyed communities of the tree genus Inga (Fabaceae: Mimosoideae) across southern Peru. This genus is the most numerically dominant (~240 individuals reaching breast height per hectare) and diverse (>20 sympatric species) tree genus in Madre de Dios. I collected soil samples across all survey sites to assess the influence of edaphic factors upon species distributions. I also conducted population genetic sampling to determine the role of dispersal in limiting where species were found. Finally, I have constructed a phylogeny for the species I am studying to find out if competition between closely related species can explain why some species do not coexist.

2) Relating phylogeny to spatial and temporal distribution patterns of Inga species on BCI
Collaborators: Toby Pennington (Royal Botanical Garden, Edinburgh), Lissy Coley (University of Utah), and Tom Kursar (University of Utah)

Species distributions (including those of Inga spp.) have been mapped across 50 hectares at Barro Colorado Island (BCI), and this 50 ha plot has been recensused every 5 years since 1982. A species level phylogeny of Inga species on BCI is currently being completed. This offers the rare opportunity to relate local distribution patterns and temporal trends in abundance to phylogeny at a low taxonomic level. For example, most Inga species have been showing consistent upwards or downward trends in abundance since 1982, probably a response to climate change on BCI. Are the species showing an upward trend clustered in the phylogeny?

3) Species delimitation and identification of tropical trees based on vegetative morphology versus DNA sequences

During my dissertation research in southern Peru, I originally delimited all Inga individuals I collected based on vegetative morphology. This is the conventional identification method used by most tropical plant ecologists. However, taxonomists generally delimit and identify species based on characters of flowers and fruits. This begs the question of how often ecologists are wrong and if their errors seriously affect their results. To address that question, I used DNA sequence from two genetic markers from over 750 Inga individuals to assess error and how it might affect conventional ecological results. This study has also provided information about the utility of DNA barcoding in diverse plant genera.

4) Distribution patterns in the Lesser Antillean avifauna

Bird communities on islands have long provided a model system for research in evolutionary biology and ecology. I am using population genetic data from land birds in the Lesser Antilles, a particularly well-studied archipelago, to determine if differential dispersal ability of species explains the varying patterns of distribution found in different groups of birds.

5) Long-term vegetation monitoring in Great Smoky Mountains National Park
Collaborator: Erin Tripp (Duke University)
Climate change and other anthropogenic impacts upon plant communities are inevitable in the future. One important issue is how these communities will respond to changing environments. Through annual surveys of a plot in Great Smoky Mountains National Park (North Carolina, USA), we document how plant communities are changing through time. We survey several levels of plant diversity (trees, saplings, herbs, and lichens) to determine if trends are correlated across species and how different plant groups may be impacting each other through time (e.g. if trees are leafing out earlier, will herbs respond in their phenological timing?). Note: for the purposes of the discussion here, lichens are referred to as plants.