North Pacific bay gobies inhabit bays, beaches and estuaries of temperate Asia and North America, but are absent from the northernmost latitudes of the central Pacific. Morphological characters have conventionally subdivided the clade into two groups - an elongate infaunal Astrabe group, and a deeper-bodied Chasmichthys group - each with a disjunct East-West (amphi-) Pacific distribution. In chapter 1, I use multi-locus DNA sequence data to examine phylogenetic relationships of bay gobies. Basal divergence of the tree coincides with a dramatic global cooling event at the Eocene/Oligocene transition, and there is no evidence of subsequent trans-Pacific migration. These results suggest that several morphological characters previously used to define the Astrabe and Chasmichthys groups have arisen independently on both sides of the Pacific, revealing convergence of ecologically adaptive characters within a geographically divided clade. Chapter 2 uses inferences of vicariance via biogeographic events to time-calibrate this phylogeny. Divergence time estimates allow me to compare and contrast potential mechanisms of bay goby diversification on either side of the Pacific. Speciation in the West Pacific has been driven largely by interstitial colonization of gravel beaches of varying grain size, and by invasion of freshwater streams around the Sea of Japan. In the East Pacific, diversification appears to be related to an intense upwelling regime combined with isolation in large Miocene-era embayments on the coast of California. These results also provide strong evidence for relictual endemism in the Gulf of California, as the divergence of three out of four Gulf-endemic gobies substantially predates tectonic formation of the Gulf itself. In chapter 3, I use one member of the bay goby clade, Gillichthys mirabilis, to investigate population structure within the Gulf and on the adjacent Pacific outer coast. Phylogeography suggests a complex history of extirpation and colonization driven by Pleistocene sea-level fluctuations. A degree of discordance between mitochondrial and nuclear DNA patterns, however, raises the possibility that differential selection may also contribute to genetic subdivision, precluding north-south migration of mitochondrial haplotypes in the face of more extensive nuclear gene flow.

This dissertation focuses on the transcriptome of the moon jellyfish (Aurelia sp.1). As the chapters progress, larger sets of genes are analyzed, and the work becomes decreasingly comparative in nature.

In the first chapter, I analyze the POU-class genes. I begin by using comparative genomics and "gene fishing" to resolve the topology of the POU gene tree. I then use ancestral state reconstruction to map the most likely changes in amino acid evolution for the conserved protein domains. Four of the six POU families evolved before the last common ancestor of living animals--doubling previous estimates--and was followed by extensive clade-specific gene loss. POU families best understood for their generic roles in cell-type regulation and stem cell pluripotency (POU2, POU5) show the largest number of nonsynonymous mutations, suggestive of functional evolution, while those better known for specifying subsets of neural and hormone producing cell types (POU1, POU3) appear more similar to the ancestral protein.

In the second chapter, I annotate the homeodomain repertoire for Aurelia sp.1, and compare it to data from relatives that lack a medusa life stage (Nematostella, Acropora, and Hydra). Despite having simpler life cycles, the anthozoans Nematostella and Acropora have far more homeodomains than Aurelia, primarily because of clade-specific gene expansions. The one exception to this trend is the non-anterior Hox genes, where Aurelia has seven paralogs compared to Nematostella and Acropora's two. RNA-Seq analyses suggest that these non-anterior Hox genes are expressed dynamically through the Aurelia life cycle, and therefore represent candidate genes for future studies in medusozoan bodyplan evolution.

In the final chapter, I offer a broad analysis of the developmental transcriptome for Aurelia sp.1. Two major shifts in gene expression occur during the life cycle, correlating with formation of the two "adult" morphs (the transition from primary polyp to polyp, and from polyp to strobila). The morphologically complex medusa stage that distinguishes Aurelia from other model cnidarians is not enriched in novel genes, but is enriched in many conserved cell-signaling pathways, transcription factor domains, and neuroactive receptors.

One phenomenon that has perplexed biologists for decades is the lack of correlation between genome size and organismal complexity, aka the ‘C-value paradox’. It is now generally accepted that the paradox is caused by repetitive, non-coding sequences such as transposable elements. These ‘selfish’ or ‘parasitic’ sequences have profound impacts on the host genome. For instance, DNA methylation is thought to have originated as a defense mechanism against these genome parasites. As one of the most conserved epigenetic modifications in eukaryotes, DNA methylation plays important regulatory roles in animals. However, the evolution of transposable elements and DNA methylation have rarely been studied in non-bilaterian animals. This dissertation explores genome regulation in Cnidaria, the sister group of Bilateria, and its potential role in the evolution of complex life histories. Chapter 1 studies the distribution of DNA methylation across Cnidaria using published genomes and transcriptomes of over 70 species. By analyzing a proxy of DNA methylation, I show this epigenetic modification is prevalent in Cnidaria with instances of loss in a group of endoparasitic species. I also show cnidarian DNA methylation shares many similarities with bilaterian invertebrates, supporting the hypothesis that gene body methylation is the ancestral state in Eumetazoa. Additionally, cnidarians with complex life cycles tend to have heavier methylation, and both gene body methylation and repeat methylation increase with genome sizes. Chapter 2 examines DNA methylation patterns across the life history of a scyphozoan jellyfish, Aurelia coerulea, via stage-specific whole-genome bisulfite sequencing. This work characterizes the distribution of DNA methylation on the Aurelia genome, the correlation between methylation and expression levels, and highlights that sparsely-methylated genes tend to be expressed dynamically. Moreover, many genes are differentially methylated across life history stages; however, methylation changes do not predict expression changes. This study indicates that the regulatory role of gene body methylation lies more in stabilizing expression, a notion emerging as the primary function of DNA methylation in animals. Chapter 3 studies transposable elements (TEs) in the Aurelia genus. TEs in 12 species are annotated and quantified using whole genome sequencing. One clade shows signatures of ancient TE expansion events, indicating the TEs have undergone different evolutionary histories. Based on the phylogeny within the genus, these expansion events predate the closure of the isthmus of panama which separated the two clades of Aurelia. Additionally, the genome size difference between Aurelia aurita and Aurelia coerulea is unlikely due to activities of known TEs; however, Aurelia genomes harbor many unique TEs that would be worth further exploring. Overall, this dissertation research shows that Cnidaria is one of the most fascinating systems to study the evolution of genome regulation. As the first animals with real tissue layers, neuro-sensory systems, and a wide variety of different life histories, Cnidaria can shed light on how the genome ‘dark matter’ plays a role in the evolution of animal complexity.

Contributions to conservation biology research are traditionally approached from an ecology and evolutionary biology perspective, but education research can be equally valuable. Population genetics research can inform management decision-making while education research can improve how we teach conservation. My dissertation tackles conservation biology from an interdisciplinary framework. This involves studying the population genetics of the endangered tidewater goby and investigating drawing as an active learning tool in a Restoration Ecology course. For chapter 1, I assessed how the genetic structures between tidewater goby (Genus: Eucyclogobius) populations relate to high rainfall events by analyzing rainfall data and 14 microsatellite loci from 526 samples collected in 1990 and 2008 from the central coast of California. There is evidence that supports tidewater goby dispersal is dependent on heavy rainfall that subsequently breaches lagoons. Results also show that two adjacent management subunits are genetically similar in the 2008 samples compared to 1990, which suggests the subunits should be combined. For chapter 2, I investigated how landscape variables inform genetic structures between tidewater goby populations by analyzing satellite imagery data and 14 microsatellite loci from 1346 individuals covering four management units over an 18-year timespan. Results suggest heterogeneous landscapes that include softer substrates may facilitate movement between sites. There is also evidence that harder substrates, headlands, and kelp vegetation may limit dispersal. For chapter 3, I explored how drawing activities in an ecological restoration course affect learning outcomes and whether students continue to use drawing as a study tool after the intervention. Results suggest students use their notes and lecture slides more frequently than drawing activities to study for the midterm. Students who were instructed to draw (drawers) as part of their response for a midterm scored significantly higher than students who did not (non-drawers). Drawers also had a more complex correlation network of paired works in their midterm responses compared to non-drawers. Collectively, this dissertation expands on how we traditionally address conservation biology by offering insights from population genetics biology combined with innovative applications of pedagogy in higher education.

The unique habitats of the Northern Gulf of California are home to many endemic vertebrate and invertebrate taxa. Owing to their limited geographic range, these taxa are often vulnerable to extinction. Located at the tip of the Northern Gulf is the Colorado River Delta. Upstream damming and water diversions in the past century have turned the Delta into an inverse estuary. The resulting rise in salinity and environmental impacts in the region have raised concerns for the local endangered fauna. Among these is the endemic silverside fish, Colpichthys hubbsi, which is narrowly restricted to the tidal channels within the Delta. Its sister species, Colpichthys regis, inhabits lagoons and estuarine channels on both sides of the Gulf. The two species share a region of syntopy at the southwestern edge of the Delta where hybridization is suspected to occur. Combining phenotypic and molecular data, this study shows evidence for introgressive hybridization between C. regis and C. hubbsi. While we find no evidence for recent population decline in C. hubbsi, the reduction of river outflow may lead to diminished ecological separation between the sister species, putting C. hubbsi at risk of extinction by introgression. Future work on this system will need to focus on identifying genes under divergent selection and assessing the full extent of genomic introgression using high-throughput sequencing approaches. Molecular analyses reveal a surprising lack of genetic diversity in the less geographically restricted species (C. regis). Results from neutrality tests based on mitochondrial DNA suggest that both species of Colpichthys may have undergone recent demographic expansion. Recent studies have demonstrated profound effects of Pleistocene climate fluctuations on population structure of estuarine fish, prompting the need for additional investigations to elucidate the complex demographic histories of these taxa.

Over 150 years of investigation has yielded knowledge of the patterns and mechanisms of biological evolution. Yet rarely do such studies integrate the physical mechanisms that drive this evolution on a timescale that is biologically meaningful. Without integrating physical and biological processes, we risk overlooking the co-evolutionary nature of Earth and life. This thesis presents first a broad synthesis of how geologic, climatic, and environmental mechanisms drive patterns of evolution on long (> 5 Myr), medium (1–2 Myr), and short (10s–100s kyr) timescales. It secondly presents a detailed assessment of how estuaries and their inhabitants co-evolve through time and space in response to changing sea levels and the physical landscape. Chapter 1 is a meta-analysis and review of the biological and geological histories of the Sonoran Desert and Gulf of California (Gulf) from 15 Ma to present. We suggest a middle-Miocene marine embayment could explain the deposition of reworked marine microfossils and speciation ages and distributions of Gulf endemics. Assessment of the Pleistocene-age mid-peninsular seaway hypothesis reveals that the uplift age of the Baja peninsula and strong genetic discordance of highly dispersive taxa render this hypothesis unlikely. Finally, we document the distribution patterns of 527 plants disjunct between the mainland and Baja California peninsula and suggest postglacial responses that underlie these patterns. In Chapters 2 and 3 I develop a paleohabitat modeling technique that estimates the size and distribution of estuarine habitat from 20 kya to present on a near-millennial timescale using physical parameters. I apply this model to ~4,600 km of coastal distance from San Francisco, USA to Sinaloa, MX. Independent of this is an assessment of the genetic history of 524 individuals of three co-distributed fishes using mtDNA and large microsatellite datasets. Together, there is statistically significant agreement from these findings; lowstand greatly reduced estuarine habitat and individuals evolved independently in these isolated refugia. Tectonic and oceanographic processes have shaped the regional geomorphology of coastlines and thus the degree to which these refugia and refugial populations are isolated. This isolation-recolonization pattern is likely global, and is not restricted to glacier-adjacent coastlines at high latitudes.

A combination of complex geological and climatological history in California have created a biodiversity “hotspot” with numerous endemic species. Most of the wetlands in the region, such as rivers and coastal lagoons, were shaped by uplift, sea-level change and climatic fluctuations creating a physical template on which genetic diversification of the aquatic biota could unfold. Coastal lagoons, for example, provide important ecological services as well as habitat and nursery for endemic and endangered animals, such as the steelhead (Oncorhynchus mykiss), red-legged frog (Rana aurora draytonii), and the tidewater goby (Eucyclogobius newberryi). However, anthropogenic impacts on the landscape, such as urbanization and water management for human use and/or flood control, have had major impacts on the landscape threatening much of the fauna associated with these wetlands. Such is the case of the rare and endangered unarmored threespine stickleback, Gasterosteus aculeatus williamsoni, found in only a few drainages in southern California. Most of the populations are now extirpated and those that persist have suffered isolation and bottlenecks, as well as inadvertent mixture of distinct lineages and introduction of non-native predators. Many conservation and restoration efforts led by numerous governmental units and NGOs are underway in California. These require decisions informed by more comprehensive understanding of current and extirpated populations, as such information will lead to more successful and effective planning and management. Advancements in molecular techniques, particularly the development of high-throughput sequencing, has made it possible to not only deepen our knowledge on evolution but also to use it as a tool in conservation biology. This revolution allows us to unravel genomic signals of demography and selection, investigate historic data from museum material, and even look at environmental samples collected from soil and water. All the work presented here explores the use of high-throughput sequencing and bioinformatic tools to target relevant conservation issues related to endangered species and habitats in southern California. We expect that results will have an impact on conservation efforts, and techniques can be applied to other systems and fields of interest.

This project is directed towards implementing aspects of the tidewater goby recovery plan in coordination with, and funded by, the US Fish & Wildlife Service (USFWS) through a Section 6 Cooperative Agreement awarded to the University of California, Los Angeles on May 15, 2015. The primary focus of this dissertation was to developed a quantitative framework to complete a metapopulation viability analysis (MVA) for the endangered tidewater gobies in the genus Eucyclogobius. Modeling tidewater goby metapopulation dynamics is an essential component in constructing long-term management plans rangewide throughout the California Coast. This dissertation examines more closely how these dynamics affect viability, connectivity, and long-term persistence of tidewater goby metapopulations throughout the California coast. In the first chapter of this dissertation, I conducted annual population surveys (2014, 2015, and 2017-2018) in 117 estuaries and lagoons to assess the current health and status of the tidewater gobies in five of the six Recovery Units, spanning from Bodega Bay to San Diego, CA. This massive effort has provided continuous coastal surveys over four years, and over 300 observations, which helped create the framework for a robust and comprehensive presence/absence dataset to help inform metapopulation management and recovery actions. In the second chapter of this dissertation collated all existing rangewide occupancy data, metapopulation descriptors, wetland site characteristics, and repository specimen collections into an open access database. This database will provide critical information relative to the federally endangered tidewater gobies and help inform the metapopulation viability analysis model developed in this study, as well as support continued research on the conservation and management of these incredible fish species and the coastal wetland ecosystems they inhabit. In the third chapter of this dissertation I review the general biology, conservation status, habitat impacts, and metapopulation dynamics of the northern tidewater goby (Eucyclogobius newberryi) and southern tidewater goby (Eucyclogobius kristinae). In addition, I demonstrate the effectiveness of a Bayesian approach to provide a flexible method to generate metapopulation viability analyses and provide a detailed summary of the MVA model framework, including limitations, required corrections, and future amendments that need to be addressed in order to meet the recovery criterion envisioned in the recovery plan.