Browsing by Subject "Gene flow"

This dissertation focuses on the implications of habitat change in freshwater ecosystems. Habitat change has three primary components that are inextricably connected; habitat loss, alteration, and fragmentation. Habitat loss is the physical removal and/or replacement of “core” habitat such that a new “matrix” habitat exists in its place. Habitat alteration is the modification of core habitat that causes a quality change (either positive or negative depending on the target species). Habitat fragmentation affects the connectivity between core habitat patches in the landscape. Rivers are highly fragmented both naturally and anthropogenically, so they represent a system readily available to study the impacts of habitat change on ecosystems.

Four approaches were used to evaluate the impacts of habitat change on freshwater ecosystems. First (Chapter 1 as published in the Annals of the New York Academy of Sciences with co-authors Dr. Martin Doyle and Dr. David Strayer; see Fuller et al. (2015)), a review of the major causes and consequences of habitat change in river networks was conducted with the goal of also bridging a theoretical gap between terrestrial and freshwater systems related to habitat change ecology. Second (Chapter 2), an empirical evaluation of a fragmented (dammed) river reach was used to evaluate the local impacts of habitat loss and alteration on physical (sediment), biogeochemical (dissolved oxygen), and biological (freshwater mussels) response variables. Third (Chapter 3), gene flow model simulations were used to identify the genetic impacts of habitat fragmentation at the river-network scale. This simulation effort contrasted the impact of habitat fragmentation with species longevity to see how organisms using different life history strategies related to lifespan respond genetically to habitat fragmentation. Fourth (Chapter 4), an empirical landscape genomics evaluation of a species of freshwater mussel (Elliptio complanata) was conducted to identify its genetic response to a river network with a long history of habitat change.

Conclusions from this research make several contributions to the ecological theory of habitat change. First, by applying the habitat change lexicon in terrestrial systems to freshwater systems, sharing results and theory across the terrestrial-aquatic literature becomes simple and may advance the theory behind habitat change ecology more rapidly with more empirical results to draw upon. Second, temporally dynamic matrix habitat and species capitalizing on altered edge habitat were identified surrounding a local habitat fragmentation agent (a dam), suggesting some species may strongly benefit from the presence of edge habitat in river networks. Third, from the gene flow model simulations, the life history of a species played an important role in how organisms respond genetically to habitat fragmentation where long-lived species appear buffered from the genetic diversity loss caused by habitat fragmentation. Finally, the empirical evaluation of a freshwater mussel species that has experienced a long history of anthropogenic-driven habitat change via water quality alterations, inundation losses, and dam fragmentation appears to have maintained a population genetic structure unrelated to the expected habitat change in the system.

Meiotic recombination creates genetic diversity by shuffling combinations of alleles across loci, yet alleles at neighboring loci often remain non-randomly associated. This non-random association is known as linkage-disequilibrium (LD), and it has evolutionarily important effects both within and between species. Nucleotide diversity at a given locus may be reduced by directional selection on the locus, or by selection on neighboring linked loci. Recombination rates and nucleotide diversity are positively correlated across loci within many species, which can be explained by linked selection reducing nucleotide variation disproportionately in regions of low recombination. However, the independent contributions of different types of linked selection are difficult to disentangle. Between species, chromosomal inversions have been proposed to suppress recombination in hybrid inversion heterozygotes and thereby maintain LD and species distinction, but many models of how this happens are overly simplistic—they often ignore non-crossover gene conversion, which reduces LD. Little direct empirical data exist on gene conversion with respect to inversions in hybrids, so despite existing models, inversions may be quite ineffective at keeping hybridizing species distinct. Here, I examine the evolutionary consequences of LD at two levels: nucleotide diversity within species, and recombination-suppression in hybrids between species. I present three investigations driven by this overarching goal of understanding how LD plays into fundamental evolutionary mechanisms. First, I examine nucleotide variation in Drosophila pseudoobscura, and I present a novel test for evidence that particular kinds of selection at linked sites (background selection and/or soft sweeps) may reduce nucleotide variation even in the absence of hard selective sweeps. Second, I show that inversions are permeable to non-crossover gene conversion, which occurs throughout inverted regions in intra- and inter-specific hybrids. I provide a genome-wide empirical analysis of gene conversion rates both within species and in species hybrids, and I estimate that gene conversion occurs at a rate of 1 x 10-5 to 2.5 x 10-5 converted sites per bp per generation in experimental crosses within D. pseudoobscura and between D. pseudoobscura and its naturally-hybridizing sister species D. persimilis. Finally, I use extensive whole-genome sequence data to re-examine patterns of introgression and divergence in the D. pseudoobscura / D. persimilis system. I show how failing to consider variation in evolutionary rate can lead and has led to misinterpretations regarding effects of introgression. Through these genomic examinations, I refine our understanding of how recombination and linkage disequilibrium have shaped the divergence and speciation of Drosophila pseudoobscura and D. persimilis.

Genetic diversity, population structure, and connectivity influence interactions among communities and populations. At hydrothermal vents in the western pacific, population structure in vent-associated species could occur at spatial scales ranging from vent sites separated by a few hundred meters to oceanic basins separated by more than 3000 kilometers. The spatial scale of population structure has important conservation implications; species that are well-connected across large geographic regions are more resilient to natural and anthropogenic disturbance. This dissertation examines the genetic diversity, population structure, and connectivity of 3 vent-associated species in the western Pacific. It first presents results from the development of microsatellite primers for Ifremeria nautilei, a deep-sea vent associated snail, then uses mitochondrial COI sequences and a suite of microsatellite markers to examine the broader connectivity of three vent-associated species, Ifremeria nautilei, Chorocaris sp. 2, and Olgasolaris tollmanni, across three back-arc basins in the western Pacific.

Within Manus Basin, no significant genetic differentiation was detected in populations of Ifremeria nautilei (based on COI and microsatellite), Chorocaris sp. 2 (based on COI and microsatellite), or Olgasolaris tollmanni (based on COI). A previously documented low-abundance cryptic species, Chorocaris sp. 1, was detected from a single site, South Su (based on COI). The population of I. nautilei in Manus Basin was found to be significantly differentiated from a second population that appeared to be panmictic across North Fiji and Lau Basin (based on COI and microsatellites). Chorocaris sp. 2 was also found to be significantly differentiated between Manus and North Fiji Basin (based on COI). Both I. nautilei and Chorocaris sp. 2 showed signs of potential low-level migration between Manus and other southwestern Pacific basins. O. tollmanni was undifferentiated between Manus and Lau Basin (based on COI). It is likely that a variable impedance filter exists that limits the realized dispersal of some, but not all species between Manus Basin and other western Pacific back-arc basins. The presence of a variable filter has implications for the conservation and management of hydrothermal vents in Manus Basin, as it is unclear what effects sustained anthropogenic disturbance will have on isolated populations of I. nautilei and Chorocaris sp. 2.

Understanding the ecological and evolutionary responses of plant species to shifts in climate (and other rapid environmental perturbations) will require an improved knowledge of interactions between ecological and evolutionary processes as mediated by reproduction and gene flow. This dissertation research examines the processes of seed dispersal, intra- and inter-specific gene flow, and reproductive success in two red oak populations in North Carolina; the variation in these processes from site to site; and their influence on genetic structure, population dynamics, and migration potential.

Using genetic and ecological data collected from two large long-term study sites, I develop a hierarchical Bayesian model to identify the parents of sampled seedlings and characterize the scale of effective seed and pollen dispersal. I examine differences in scale of dispersal between the Appalachian and Piedmont sites in light of the spatial genetic structure and ecological differences of the two sites. I then use the pedigree and dispersal estimates derived from these analyses to examine variation in reproductive success and to test hypotheses about the causes and consequences of such variation. Using parentage estimates and measures of genetic differentiation between species, I study the likely extent of hybridization in these mixed-species secondary forests. Finally, using the SLIP stand simulator, I explore the implications of new genetic dispersal estimates for migration potential in oaks.

I find that effective seed dispersal distances are longer than estimated using seed trap data. While at the Piedmont site the large number of seedling found >100 m from their mother trees suggests that animal dispersers play a vital role, at the Appalachian site seedling distributions conform more closely to the original gravity-created pattern of seed density. Individual trees vary widely in their reproductive success. Seedling production was found to be positively associated with annual seed production, but exhibited hump-shaped or reversing relationships with age (suggesting the effect of senescence) and growth rate (suggesting tradeoffs in allocation). Germination fraction was negatively associated with fecundity, suggesting that density-dependent mortality may be acting on the high concentrations of seeds near highly fecund adults. Due to overlapping generations and variation in individual reproductive success, effective population size is estimated to be less than half the size that numbers of "adult" individuals would suggest, with consequences for the relative strength of drift and selection. Hybridization may boost effective population size somewhat; my analyses suggest that inter-specific gene flow is common at both study sites. Finally, simulations show that dispersal has a relatively stronger effect on migration rate and population growth than fecundity or size at maturity, and that genetic estimates of seed dispersal can yield significantly higher rates of migration and/or population persistence than seed-trap based estimates under both competitive and non-competitive conditions.