Document Type

Thesis

Degree

Master of Science

Major

Biology, Ecology

Date of Defense

7-18-2014

Graduate Advisor

Robert E. Ricklefs, Ph.D.

Committee

Ivan Jiménez

Peter Stevens

Peter Jørgensen

Abstract

Species rarity is often defined in terms of: local abundance, geographic range size and habitat breadth. It is thought that each of these rarity axes has distinct effects on extinction risk. Thus, understanding phylogenetic patterns of these three axes is of considerable interest because they provide insights into the extent to which rarity is phylogenetically conserved and, in turn, how extinction risk is distributed across phylogenies. Here I examine the extent to which the three axes of rarity show phylogenetic signal (the tendency of related species to resemble each other more than species drawn at random from the same phylogenetic tree), and phylogenetic conservatism (stronger phylogenetic clustering than expected from a Brownian motion model of evolution) across a regional pool of woody plants in the Madidi region of the tropical Andes of Bolivia. I measured local abundance, geographic range size, and habitat breadth for 806 species in 101 families occurring in a set of 48 1-ha tree plots, and for 1,739 species in 141 families occurring in a set of 442 0.1-ha tree plots. I used three approaches to describe phylogenetic patterns of rarity: 1) hierarchical variance partitioning across taxonomic levels, 2) Blomberg’s K statistic, and 3) disparity through time. I compared observed patterns described by these three approaches to patterns expected from a tip randomization null model that randomly assigns values of the axes of rarity to species. In addition, I compared observed patterns described by Blomberg’s K statistic and disparity through time to patterns expected from a Brownian motion model of evolution. The hierarchical variance partitioning analysis showed that the three axes of rarity display phylogenetic signal: species belonging to the same genus tended to be more similar than expected from the tip randomization null model. At deeper phylogenetic levels the axes of rarity exhibited little or no phylogenetic signal and did not display phylogenetic conservatism. These findings suggest that, given the currently changing environment of the Tropical Andes, extinction risk could be phylogenetically clustered because certain genera may contain an unusually high number of threatened species.

Share

COinS