Islands, archipelagos, and beyond: population genetics and phylogeography of Hawaiian coral reef echinoderms

Abstract

Genetic connectivity determines the evolutionary independence of populations and plays a primary role governing intraspecies evolution. Patterns of genetic connectivity can also be used to make inferences about macroevolutionary processes. For my dissertation I used mtDNA sequences in phylogeographic and population level analyses to examine connectivity within and between Pacific Archipelagos, utilizing four species of tropical echinoderms. First, I review and prescribe non-lethal sampling techniques for use in genetic studies of many species of marine invertebrates. Second, I showed that two species of sea cucumber, Holothuria atra and Holothuria whitmaei, with a similar life history, species range, and habitat usuage cannot be used as proxies for each other in order to predict phylogeographic patterns, degree of connectivity, and population genetic structure within the Hawaiian Archipelago. Third, I examined the genetic population structure of H. atra across the central tropical Pacific to show that despite its large range, H. atra has hierarchical, fine-scale population structure driven primarily by between-archipelago barriers, but with significant differences between sites within an archipelago. Finally, I compared population genetic patterns in two congeneric brittle stars, Ophiocoma pica and Ophiocoma erinaceous, across Hawaiʻi and Central and Eastern Polynesia. I found contrasting phylogeographic patterns in these two similar species, as was the case with H. atra and H. whitmaei. Given the real-world constraints of limited time and money in marine ecosystem management it would be ideal if model species could stand in as proxies for a host of similar species. This dissertation shows that this ideal scenario is unlikely to be the case; similar life histories and close phylogenetic relationships do not appear to predict population connectivity. Generalizations based on a few representative taxa are unlikely to offer much in terms of delineating boundaries for spatial management areas. Though a more inclusive multi-species approach is bound to cost more in terms of time and resources, it should ultimately payout as more informed, if complex, management.