Spawning site selection of coho salmon Oncorhynchus kisutch in Susitna River tributaries, Alaska

Abstract

Coho Salmon Oncorhynchus kisutch are the most widely distributed Pacific salmon species across Alaska. The lack of knowledge surrounding the habitat requirements of this species results in challenges for conservation and management due to natural and anthropogenic pressures. Tributaries of the Susitna River drainage in Alaska support many small and distinct Coho Salmon populations. Heterogeneity of in-stream spawning habitat is an ecological concept known to promote resiliency of salmonid populations. The goal of this study was to investigate the best habitat predictors of spawning site selection and the scale by which spawning habitat should be evaluated for management insights. Scale is particularly important when measuring, assessing, and predicting potential impacts to species from development activities because habitat research at the stream rather than the reach scale can overestimate the amount of available spawning habitat. I investigated a suite of field-measured stream habitat variables paired with empirical Coho Salmon spawning survey data in five tributaries during 2013 and 2014. Physical data was defined as biotic and abiotic surroundings of an organism or population that have an influence on survival, development, and evolution. Mixed-effects modeling results indicated that Coho Salmon spawning-site selection was positively related to gravel substrate and the presence of groundwater flux, and that spawning Coho Salmon avoided cobble substrate. Physical data were analyzed at both the stream and reach scales, and mixed-effects modeling results further concluded that variation in spawning activity at the reach scale (variance = 1.34, SD = 1.16) accounted for more variability and was more predictive than at the stream scale(variance = 0.04, SD = 0.19). This is important because fish habitat-associations identified at the reach scale were not identified at the stream scale. These results highlight the need for multi-scale habitat data collections and analyses to identify the most meaningful fish-habitat associations.