Doctor of Philosophy (PhD)


Biological Sciences

Document Type



For nearly two centuries, it has been known that biological diversity is not homogeneously distributed across the planet, and yet a general explanation for this variation remains elusive. Hundreds of studies have found that taxonomic diversity (i.e., richness) is strongly associated with characteristics of the environment. Measures of energy or climate seem to be particularly strong correlates of richness, while measures of environmental heterogeneity are typically of secondary importance. The pervasiveness of richness-environment correlations has been seen as evidence that diversity gradients result from environmental gradients. However, effects of environment on diversity may have been overestimated. In this dissertation, I used comparative analyses and computer simulations to test whether environmental energy and climate are truly general determinants of richness gradients at broad scales. I found that energy and climate are strong correlates only of diversity patterns of species with broad distributions. Richness of narrowly distributed species is only weakly associated with environment, and it is heterogeneity (not energy or climate) that better accounts for richness of these species. This questions the idea that environment (particularly energy/climate) is an important and general determinant of diversity gradients, as richness of a large proportion of species (those with narrow distributions) are not strongly associated with environmental characteristics. Moreover, I found that the stochastic diversification of clades can produce frequent, but spurious, richness-environment relationships. Thus, pervasiveness of empirical richness-environment correlations might not be reliable evidence for environmental effects on the production of richness gradients. This also suggests that re-evaluation of richness-environment relationships using appropriate null models, which incorporate the diversification process, could lead to important new insights about the determinants of richness patterns. For example, I found that when comparing empirical richness-environment relationships to those expected by random clade diversification, the perceived effects of energy/climate are reduced while the effect of environmental heterogeneity is increased. This suggests that heterogeneity might play a more significant role in the formation of richness gradients than previously assumed. My results have important implications for theories developed to explain diversity gradients, and for efforts to predict the future of biodiversity in the face of large scale changes in climate.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Stevens, Richard