December 1 2016

5:00 pm TLSB 1100

EcoEvoPub Series

Graduate Student Presentations


Tyler McCraney
Department of Ecology and Evolutionary, UCLA

"Phylogeny and diversification of gobies and their relatives"

Understanding causes for the great unevenness in biodiversity across Earth and the Tree of Life is a major goal of evolutionary biology. The invasion of new habitats has been hypothesized to drive diversification because it provides ecological opportunity by reducing competition. If habitat transitions provide ecological opportunity, then rates of diversification should be greatest in clades with an evolutionary history of shifts between marine and freshwater habitats. Here I use a time-calibrated, 915 taxon megaphylogeny of the globally distributed and ecologically diverse Gobiiformes fishes, constructed from a partial matrix of 33 gene sequences to test the hypothesis that habitat shifts drive diversification. I use macroevolutionary modeling of diversification rates and ancestral state reconstruction of habitats to identify lineages in the Gobiiformes radiation that have undergone rapid speciation in marine and freshwater environments. I find strong evidence for diversification rate heterogeneity in Gobiiformes fishes, characterized by a low background rate for most non-goby taxa followed by rate increases at the root of the goby clade and several freshwater and mutualistic goby lineages. In contrast, I found little variation in diversification rates among the marine cardinalfishes or freshwater sleepers, which supports the hypothesis that transitions between habitats drive diversification.

Zack Gold
Department of Ecology and Evolutionary Biology, UCLA

"Message in a Bottle: Designing and Implementing a eDNA Methodology to Compare Biodiversity Across Marine Ecosystems"

Current methods for evaluating marine ecosystem function include fish biodiversity, abundance, and trophic structure assessment. Unfortunately, current visual SCUBA transect methods require taxonomic experts, are labor intensive, and difficult to repeat. However, new environmental DNA (eDNA) techniques offer rapid, affordable, and accurate tools for monitoring fish biodiversity. eDNA includes dissociated cells and DNA in the environment that can be isolated, sequenced, and identified using DNA meta-barcoding. This novel technique has allowed scientists to accurately detect fish species in both aquatic and reef ecosystems at higher rates than traditional methods, allowing for rapid, reliable, and repeatable fish biodiversity assessments. In order to determine the effectiveness and limitations of eDNA techniques in marine environments, I designed and implemented marine eDNA methodology at both the California Science Center and in Mo'orea, French Polynesia. Previous research in Mo'orea through the Smithsonian's BIOCODE project has generated a full reference library of all marine species around the island, providing an ideal location to determine the effectiveness of eDNA in capturing marine fish biodiversity. In addition, samples were collected along a transect with known current profiles from the inner lagoon to 4km offshore to examine the effects of ocean mixing. Preliminary results indicate that ocean mixing plays an important role in the fate and transport of marine eDNA. I found that eDNA signals follow modeled hydrographic patterns in which backreef lagoons with limited water movement were significantly different from fore reef, pelagic, and high velocity backreef habitats. These results strongly suggest that oceanographic variables should be incorporated in eDNA study design in order to accurately reflect resident marine fish biodiversity.











































































































































































































































































































































































































































































































































































































this is idtest: