LB 25-213 - Sticks and Stones vs. Satellites in Space: What fine- and broad-scale habitat covariates are most strongly associated with juvenile Oncorhynchus mykiss densities?

Pacific salmon and steelhead are ecologically, economically, and culturally significant in the Pacific Northwest. They are distinctly important in transporting nutrients from marine environments to freshwater ecosystems and are a reliable food source to terrestrial vertebrates, making them a vital keystone species in the Pacific rim; additionally, they are vital Native American harvests and commercial and recreational fisheries. Most salmonids exhibit some degree of anadromy, exposing them to stressors related to climate change and anthropogenic pressure within multiple ecosystems throughout their life cycle, which underlines the importance of understanding the environmental variables which define suitable habitat. Juvenile Oncorhynchus mykiss have flexibility in their life history depending on resource availability, habitat suitability, and density, driving individuals to select either an anadromous (steelhead) or fluvial (rainbow trout) life history. This life history plasticity demonstrates the influence of fine-scale habitat variables at the individual level and how O. mykiss abundance can shift over space and time. Measuring fine-scale salmonid habitat metrics in streams is valuable in assessing habitat quality and availability; while indicative of current habitat conditions and the potential for fish, these metrics do not offer information on overall stream potential. Remotely sensed habitat data, while having their own limitations, can offer more extensive data on a habitat. This project seeks to answer the question of which habitat covariates, both derived from fine- and broad-scale datasets, are most strongly associated with juvenile O. mykiss abundance. This was achieved thus far through an evaluation of a suite of candidate generalized linear models (GLM) and geospatial analyses of the potential for broad-scale environmental variables to predict the presence of fine-scale variables strongly connected with juvenile salmonid abundance in the Oregon Coast. Preliminary analyses have indicated that GLMs with zero-inflated negative binomial and zero-inflated Poisson distributions best fit these data, and scour pool depth and riffle depth are significantly related to juvenile O. mykiss abundance. Further analysis, including potential applications of weighted linear mixed-effects models, is required to understand sampling and temporal effects. The physical and biological variables which compose a habitat are inextricably linked to salmonid life histories and understanding the variables most strongly associated with juvenile abundance will continue to be essential in salmonid conservation.