In North America 72% of freshwater mussel species are endangered, threatened, or of special concern due to factors such as habitat loss and degradation, biological invasion, and land use change. Propagation of freshwater mussels has been considered a necessary conservation strategy for population restoration where threats have been mitigated but small population sizes limit population viability. Yellow lampmussel is a species of freshwater mussel that is endangered, threatened, or imperiled throughout its range; therefore, I evaluated laboratory techniques (probiotic supplements and secondary rearing designs) to improve culture of yellow lampmussel for population restoration. Several aquaculture facilities commonly use probiotics; thus, I used commercial probiotics to determine if 1) probiotic concentration and 2) type of probiotic mixture improved growth or survival of juvenile mussels during primary culture. I further asked whether probiotics affected mussels by reducing ammonia, thereby improving water quality. Some probiotics increased survival (and, in one experiment, increased growth) of juvenile mussels, regardless of concentration, but results were variable by experiment and probiotic type. Probiotics did not significantly reduce ammonia concentrations, so this was unlikely the mechanism of benefit. I also investigated the effect of different secondary rearing systems at two culture facilities on growth and survival of juvenile mussels in two size classes (5.0 mm). I used five different secondary culture systems that were either indoors (dogpans and baskets) or outdoors (trough, airlift upweller, tank upweller, baskets), where water was either recirculating or flow-through. Survival was exceptional in all larger size class rearing systems, and the baskets in the ponds had the greatest growth rates. Smaller mussels had lower survival than the larger mussels, indicating that when deploying juvenile mussels into outdoor culture systems mussels size should be greater than 5.0 mm. Results of this project will inform future rearing yellow lampmussel in New England and more broadly add to the limited literature on probiotic use and secondary rearing designs in freshwater mussel culture.

The health of freshwater aquatic systems, particularly stream networks, is mainly influenced by water temperature, which controls biological processes and influences species distributions and aquatic biodiversity. Thermal regimes of rivers are likely to change in the future, due to climate change and other anthropogenic impacts, and our ability to predict stream temperatures will be critical in understanding distribution shifts of aquatic biota. Spatial statistical network models take into account spatial relationships but have drawbacks, including high computation times and data pre-processing requirements. Machine learning techniques and generalized additive models (GAM) are promising alternatives to the SSN model. Two machine learning methods, gradient boosting machines (GBM) and Random Forests (RF), are computationally efficient and can automatically model complex data structures. However, a study comparing the predictive accuracy among a variety of widely-used statistical modeling techniques has not yet been conducted. My objectives for this study were to 1) compare the accuracy among linear models (LM), SSN, GAM, RF, and GBM in predicting stream temperature over two stream networks and 2) provide guidelines in choosing a prediction method for practitioners and ecologists. Stream temperature prediction accuracies were compared with the test-set root mean square error (RMSE) for all methods. For the actual data, SSN had the highest predictive accuracy overall, which was followed closely by GBM and GAM. LM had the poorest performance overall. This study shows that although SSN appears to be the most accurate method for stream temperature prediction, machine learning methods and GAM may be suitable alternatives.

In-stream concrete structures were studied through model tests and river tests. The model studies indicated that four designs provided good habitat in the model stream. These structures were the inverted weir, the "V" structure, the slab with legs and the cylinder. Through the river studies it was determined that these structures did not influence enough of the total river area to be effective in providing good fish habitat. Also an appreciable amount of yearly maintenance would be required to free the structures from silting in, debris, and vandalism. The slab with legs was the only promising structure.

Wild-trapped and laboratory-reared Peromyscus maniculatus (Wagner) were raised as confined colonies indoors with various stocking densities ranging from 0.026 to 0.100 mice per sq. ft. in pens of 40, 77, and 154 sq. ft. The animals were individually tagged and all pens were censused at weekly or semi-monthly intervals to record animals present, body weights, food consumption, and overt reproductive condition. At the termination of each experiment, all mice were necropsied and organ weights of gonads and adrenals recorded. Histological sections were made of the testes and ovaries. Half the colonies were subjected to density-independent exploitation by removal of 50% of the mice in each litter before weaning. Colonies stocked with the wild-trapped P. m. rufinus failed to establish a regular pattern of successful breeding and were terminated after one year. The succeeding colonies were stocked with the lab-reared P. m. sonoriensis and these did breed regularly, and the young survived to maturity. Reduction of the population growth rate to zero at the equilibrium density was accomplished by a cessation of breeding by the original females after an average of 2.9 litters was produced by each female. This, combined with a failure of the progeny born into the colony ever to produce young, caused the population growth rate to remain at zero for up to the maximum of 52 weeks permitted in this experiment. The exact mechanism appears to be psycho-physiological in nature, in that regression of the germinal tissues of the adults occurs, and these tissues failed to mature in the progeny in all experiments. Reduction of the population growth rate to zero at equilibrium density is not accomplished by mortality of either young or adults, except when wild-trapped mice are used. Density-independent exploitation of the colonies increased the number of litters born, the litter size, and the total number of young born. There was a tendency for higher pre-weaning mortality. The net effect of these changes was a tendency for reduction in the equilibrium density of the exploited colonies, which may indicate that this species under these conditions is not totally self-regulatory; i.e., equilibrium density may be determined by both intrinsic and extrinsic factors. There was no discernible effect of exploitation on the measures of body weight, food consumption, adrenal weight, or adrenal weight per gram of body weight. Many parameters were found to be negatively correlated with stocking density: the length of the reproductive period, numbers of litters born in a colony, total number of young born in a colony, litter size, number of nest boxes occupied by the mice at equilibrium, and the number of mice present in a colony at equilibrium. Some physiological measures proved to be positively correlated with density: progeny weight at three months, adrenal weight, and adrenal weight per gram of body weight.

Juvenile rainbow trout Oncorhynchus mykiss stocked into mid-elevation reservoirs in Utah are vulnerable to predation from piscivorous fish and birds. I determined how effectively juvenile trout used cover to avoid these predators by (1) direct observations (snorkel transects) of habitat selection in two reservoirs and (2) measurements of survival and growth rates in a pond experiment where adult brown trout Salmo trutta were predators. Observations of juvenile trout were conducted within five weeks of stocking in 1988 and 1989. During the day, juvenile trout were abundant in complex inshore habitats and avoided simple habitats such as sand and gravel. Measurements of gut fullness indicated that juvenile trout fed during the day but not during the night. Large Daphnia comprised more than 95% of the diet of juvenile trout. Because large Daphnia were often higher offshore than inshore in both reservoirs, selection of inshore cover is believed to be primarily a response to reduce predation risk. At night, trout in both reservoirs selected more exposed areas and rested on the bottom. In the pond experiment, the presence of brown trout significantly increased mortality of juvenile trout, decreased their growth rates, and caused them to avoid offshore areas. The presence of cover significantly decreased predation rates but did not affect growth of the juvenile trout.

The Bonneville cisco (Prosopium gemmiferum), a small planktivorous whitefish, is an important part of the distinctive fish community of Bear Lake, Utah-Idaho. The Bonneville cisco plays a key role in the trophic structure by converting zooplankton to fish biomass and providing a major forage sour ce for cutthroat and lake trout. Aspects of cisco feeding ecology studied include characterization of the zooplankton community composition and dynamics and cisco feeding habits and prey select ion.

Early in 1952, walleye fry were stocked for the first time in Utah Lake. Subsequent plants were made in 1954, 1955 and 1956. Altogether, over two million fry have been stocked. By January, 1958, it was felt sufficient time had elapsed for the species to establish patterns of growth, reproduction, food habits and consequent relationships to the new environment . Thus, on January 1, 1958, a Federal Aid to Fisheries project, number DJ F-4-R-5, job T, was initiated. This project was entitled "investigations of Yellow Pikeperch and Channel Catfish of Utah Lake, Utah County." Although the closing date was December 31, 1958, dat a were collected until October, 1959. Funds were allocated by the Utah Fish and Ga.me Department after December 31, 1958, for the continued study. Collection of data on the walleye was taken entirely from March, 19 58, to April, 1960, and all the work included herein is from that period.

Fifty-seven Trumpeter swans (Cygnus buccinator) were translocated to the Bear River Migratory Bird Refuge and the Bear River Club Company in northern Utah. The purpose of this effort was to encourage dispersal of the Rocky Mountain population of Trumpeter swans during the winter, and to reestablish a migratory route to southern wintering grounds. I assessed the success of the translocation by evaluating 13 translocation criteria proposed in the literature. In this study I addressed two of these criteria in detail by evaluating habitat quality at the translocation sites and by analyzing potential competitive interactions with Tundra swans (Cygnus columbianus). Habitat quality was determined by analyzing the spatial distribution of sago pondweed (Potamogeton pectinatus) tubers in wetland sediments before the fall and after the spring migration of Tundra swans. Sixty-four transects were established within the study sites with 10 sediment cores per transect. Geostatistical procedures were employed to account for autocorrelation between samples. Tuber biomass was not randomly distributed within the studied wetlands. In fact, discrete areas of high values appeared to exist before and after swan foraging . It is not likely that sago pondweed tubers are limiting swan abundance in this system. Thus, the habitat quality of the study sites is sufficient for Trumpeter swan translocation. Potential competitive interactions with Tundra swans were evaluated by examining differences in resource utilization patterns of the two species. I measured body size differences, dietary overlap, resource availability, and the efficiency of extracting available resources. Trumpeter swans appear to benefit from a larger body size and a longer neck because they are more efficient in extracting tubers from the sediment, and are able to exploit tubers to a greater sediment depth than Tundra swans. However, Trumpeter swans incur higher traveling costs due to the larger body size. The trade-off between higher foraging efficiency of Trumpeter swans and higher traveling efficiency of Tundra swans may be a potential mechanism for coexistence. Lack of support by governmental and non-governmental agencies did not allow for more than one year of translocation. Even though the Trumpeter swan translocation in 1996 was successful, I concluded that the Utah translocation program failed because the translocation did not meet translocation goals.

Five groups of 400 hatchery rainbow trout, (Salmo gairdneri), were stocked in a small, mountain stream at )-week intervals from June to September, 1972. A fish trap captured any fish moving out of a 500 m study section. Fish began moving at high levels during the first day of each stocking and continued at high rates for 5-8 days (Early Phase), after which movement decreased to low levels for 6-9 days (Late Phase). Early Phase fish moved primarily at night, possibly due to their disoriented state and high subjectivity to stream conditions. Fish moving during Late Phase did so mainly during daylight, probably in response to diurnal periodicity of a day-active food organism in the drift. Forced movement due to social behavior did not seem to be an influencing factor, but the duration of visible light seemed important to moving fish.

Effects of exercise and the relative condition of exercised fish were viii evaluated for two strains of rainbow trout. Circular tanks fitted with a paddle were used to continuously exercise the fish and a stamina tunnel was used to evaluate the condition of exercised trout. Trout were exercised continuously at one fish-length per second and one-half fish length per second. Serum LDH activity in Shephard-of-the-Hills (Missouri) rainbow trout acclimated to 10 C increased two fold during the first four days of exercise. Serum LDH activity returned to control levels by the tenth day of exercise. Serum LDH activity in Shephard-of-the-Hills rainbow trout exercised at one-half length per second did not increase significantly. Serum LDH activity in Fish Lake stock acclimated to 17 C did not increase significantly. Shephard-of-the-Hills and Fish Lake rainbow trout exercised at one length per second were better conditioned to water velocities of two lengths per second than were controls.