UC San Diego

Understanding the processes that have given rise to observed patterns of species distributions is a central question in biogeography. These patterns include spatial distributions of species richness, phylogenetic relatedness, phenotypic similarities and differences, and the geographic ranges of species. Much work in inferring process from pattern is based on qualitative expectations, but by modeling processes that are of interest, one can test such expectations and generate a firmer intuitive foundation. Dynamic models can also lead to more powerful methods of data analysis, allowing stronger inferences from available data. This dissertation brings quantitative methods to bear on topics in biogeography at two scales. First, I develop neutral macroevolutionary models to address questions of regional diversity, endemism, and lineage ages with the goal of estimating region-specific rates of lineage origination, extinction, and dispersal. In Chapter 1, I introduce a model and show that a region with high diversity or high endemism need not have a rapid rate of taxon origination, as is often assumed, but that estimation of regional rates is possible when the ages of extant lineages are known. In Chapter 2, I emphasize that dispersal must be explicitly included in attempts to estimate origination and extinction rates of different regions. Comparing model results with empirical data suggests that the nature of macroevolutionary and biogeographic processes may differ substantially between marine and terrestrial groups. Second, I employ more detailed coevolutionary models to investigate the formation of geographic borders between species, thus connecting ecological and evolutionary processes with observable patterns of species distributions and phenotypic variation. In Chapter 3, I find that character displacement may be common on an environmental gradient, but that it would often not be recognized by the customary methods of looking for greater difference in sympatry than allopatry. In Chapter 4, I find that species borders maintained by hybrid inviability or interspecific competition may be attracted to regions of reduced dispersal, while those maintained by local adaptation and gene flow are repelled from dispersal barriers. These results show that species interactions may be quite important in limiting geographic ranges and potentially in forming biotic provinces