M.Sc. / Please refer to full text to view abstract

Convergence in body form and discrete morphological characters among fishes of similar habits is widespread in both freshwater and marine habitats. The identification of adaptive types of fishes alludes to predictive relationships between morphology and lifestyle. Now, a challenge confronting

Two small streams on Vancouver Island, British Columbia, were examined for patterns of morphological variation in Gasterosteus aculeatus. A progressive analysis beginning with Principle Components Analysis, followed by Nested and Partially Nested Multiple Analysis of Variance and then Duncan's Multiple Range Test was used for pattern determination. This new technique allows the researcher to sequentially isolate the pattern of variation at different levels of generality from species to individual organisms. The pattern of variation for G. aculeatus in Bonsall Creek and Nunns Creek can be summarized as follows: The largest amount of variation accounted for by the analysis is interpreted as individual variation. Populations also account for a large amount of variation and show consistent, fully nested patterns of variation at each of the analysed geographic and microgeographic levels. These populations are probably genealogical units. The so-called "leiurus" and "trachurus" forms on the Pacific coast of North America do not appear to be evolutionary entities, but to be historical artifacts that are best viewed as labels for the extremes of a continuum of variation. In areas where distinct populations meet, different clines are documented in the two stream systems. In Nunns Creek there is a smooth cline between populations, whereas in Bonsall Creek there is a step cline. / Science, Faculty of / Zoology, Department of / Graduate

Freshwater ecosystems and the species that reside therein are disproportionately imperiled compared to terrestrial systems. Over the past 150 years, the Columbia River basin in the western United States has gone from one of the most productive and abundant salmon watersheds in the world, to having just a small fraction of its former salmon abundance. The cause of declines in salmon productivity and abundance are related to overlapping and confounding stressors including changes in large-scale climatic patterns and anthropogenic alterations within and adjacent to the Columbia River. Four main anthropogenic stressors have been identified as the leading causes of salmonid declines: commercial harvest of adult salmon, hydroelectric power generation and associated reservoirs, habitat loss due to impassible dams and restricted access to historical habitat, and hatchery production. My dissertation broadly examined how salmon ecology and population dynamics have been influenced by anthropogenic alterations occurring within the Columbia River basin. This dissertation addresses three main questions: Do the hydroelectric dams on the Lower Columbia River represent a pulse- or press-type ecological disturbance to migrating Chinook salmon (Oncorhynchus tshawytscha) and steelhead (O. mykiss) and what are the biotic and abiotic factors that most influence the survival of fish passing multiple dams? How do wild and hatchery fish differ in regards to the precocious life-history strategy, and which variables influence this strategy across environments (freshwater to marine) and life stages (egg to adult)? What are the morphological differences between wild and hatchery salmonids and how much of that variation is attributable to rearing-environment? I found that high outflow volumes led to involuntary spill in 2011 and created an environment of supersaturated dissolved gas concentrations. In this environment, migrating smolt survival was strongly influenced by barometric pressure, fish velocity and water temperature. The effect of these variables on survival was compounded by multiple dam passages compared to fish passing a single dam. Despite spatial isolation between dams in the Lower Columbia River hydrosystem, migrating smolt appear to experience cumulative effects akin to an ecological press disturbance. In general, Chinook salmon and steelhead respond similarly in terms of survival rates and responses to altered environmental conditions. Management actions that limit dissolved gas concentrations in years of high flow will benefit migrating salmonids during this life stage. Both biotic and environmental factors affect precociousness in hatchery and wild Chinook salmon, across freshwater and marine environments. Wild fish are influenced by density-dependent processes in freshwater, as well as marine conditions that promote growth. Wild Chinook have the highest probability of precocious maturation when large smolt (>150 mm) experience productive marine environments within the first several months of ocean residence. Precocious hatchery fish are broadly influenced by conditions experienced during freshwater residency, outmigration, and in marine habitats. There was no interaction between the size of hatchery fish and environmental variables, suggesting that these fish attained the size required to mature precociously prior to migration. These results indicate that hatchery Chinook salmon do not respond to the same environmental cues that determine life history transitions as wild Chinook salmon, likely as a result of different physiological conditions and environmental exposures during early life stages. There are ecological and economic consequences to the precocious life history strategy including reduced marine-derived nutrients entering freshwater ecosystems and a loss of investment for fish intended for the adult fishery. There are significant differences in body shape between wild and hatchery origin Chinook salmon and steelhead that can be partially explained by rearing environment and variables influenced by smoltification. Hatchery fish of both species are significantly larger (i.e. centroid size, length, weight) than wild fish but have comparable or lower condition factor. In general, hatchery fish have smaller heads and longer, thinner tails (i.e. fusiform) compared to wild fish. Allometric trajectories (i.e. shape change with size) indicate that the shape of wild and hatchery fish are significantly different at small and large sizes. Wild and hatchery Chinook salmon became more morphologically different as size increased, while steelhead became more similar. The overall amount of shape variation was not significantly different between wild and hatchery Chinook salmon. This finding suggests that regardless of significant differences in the way shapes vary, hatchery Chinook have not lost overall shape variation. Total amount of shape variation was significantly greater in wild compared to hatchery steelhead, indicating that hatcheries may have a homogenizing effect on steelhead shape. I recommend a coordinated effort between federal, state and tribal hatcheries to incorporate elements of the natural rearing environment into conventional hatcheries. These elements include in-water structure that promotes the burst swimming mode, increased water velocities to increase dorsal-ventral distance, under-water feeding apparatus that reduce surface feeding behavior, overhead cover and mimicked predators to teach escape behavior. Major advancements have been made in identifying and ameliorating negative effects of anthropogenic alterations within the Columbia Basin, however, many wild populations continue to decline. My research suggests that the physical (i.e. weirs and bypass structures) and operational alterations (i.e. voluntary spill) to the lower Columbia hydrosystem have substantially improved conditions for migrating smolt compared to past years, but involuntary spill during years of high flow continue to create deleterious conditions for migrating smolt. Riverscape-scale management strategies that recognize biotic and abiotic connectivity between dams (i.e. press disturbance) will benefit migrating smolt. My research indicates that hatchery fish are not currently equal or adequate replacements for wild fish. Hatcheries vary greatly in their stated goals and desired outcomes and several coordinated changes to general hatchery practices may help produce fish that are more similar to wild fish in morphology and life histories: 1) Change the goals of production away from producing as many, large individuals as possible, to producing fish that resemble the target wild populations they are intended to support; 2) Alter the hatchery environment to reflect the rearing environments in which salmon evolved and into which they will be released; and 3) Reduce the annual number of hatchery fish released into the Columbia Basin for the sole purpose of harvest. Recovering and rehabilitating critical and diverse habitats utilized by salmonids at each life stage (spawning, rearing, migration, estuary and marine) remains crucial for wild population success into the future.