|Abstract or Summary
- To best manage Eagle Creek National Fish Hatchery and minimize any negative impacts that the current hatchery program may be having on Endangered Species Act-listed salmonids in the Eagle Creek Basin, I determined if wild fish are being displaced from preferred habitats by hatchery salmonids Oncorhynchus spp. This thesis had two goals. The first goal was to determine the possible effect of hatchery smolts on resident salmonids. I determined the density and distribution of wild juvenile steelhead O. mykiss and coho salmon O. kisutch in Eagle Creek and North Fork Eagle Creek, Oregon. My first objective was to compare summer rearing densities and distributions of wild salmonids in Eagle Creek, which receives a release of hatchery fish, and North Fork Eagle Creek, which does not receive a release of hatchery fish. Next, I determined if residual hatchery winter steelhead were present in Eagle Creek and/or North Fork Eagle Creek and if so whether or not they have an impact on mesohabitat selection, distribution, and density of wild fish in Eagle Creek basin. By conducting a comprehensive snorkel survey I identified significantly higher densities (P < 0.05) of juvenile coho salmon rearing in North Fork Eagle Creek as compared to upper and lower Eagle Creek. Age 0 winter steelhead occurred in significantly higher densities (P < 0.05) in upper Eagle Creek compared to lower Eagle Creek and North Fork Eagle Creek. Residual hatchery steelhead were located only in Eagle Creek and found rearing in the same 15 mesohabitat units that contained the estimated majority of wild fish populations. Residual hatchery steelhead comprised 0.9% of the winter steelhead population (1.1% of age 0 winter steelhead and 9.3% of age 1 winter steelhead), and 2.2% of the coho salmon population estimated to be rearing in Eagle Creek. From these data it is unclear if residual hatchery steelhead are affecting densities, distributions, and mesohabitat selection of wild salmonids in the basin. However, while I was unable to detect any direct impacts of residual hatchery fish on the wild population, these results do suggest a significant potential for ecological interaction between hatchery and wild populations.
I began this study with the intention of constructing a statistical model that would explain microhabitat preference of wild salmonids given the presence or absence of residual hatchery winter steelhead. To produce an unbiased model, ideally fish would behave as if there were no observer present (i.e., undisturbed). This is not always the case. Therefore, I addressed a second goal using underwater video to test the prediction that the presence of an in-water observer can elicit a change in fish movement. I analyzed underwater video recordings to document changes in four metrics that can be used to infer a change in fish behavior, which can ultimately result in collection of erroneous microhabitat use data. My four behavior metrics were upstream movement, downstream movement, total movement, and relative abundance of fish in the field-of-view. I detected significant differences in 9 of 10 replicates (ANOVA, P < 0.05) in at least one of the four behavior metrics. These results suggest that when attempting to document small-scale microhabitat preference by juvenile salmonids, an in-water observer may alter fish behavior thereby producing erroneous results. I suggest that researchers use caution in making inferences to entire populations when using results of models in which data were collected from only "undisturbed" fish by direct observation.