Relationships Between Landscape Characteristics, Coho Salmon, and Their Channel Habitat: Spatial Extent, Spatial Variability, and Modeling Species Occupancy

Landscape patterns influence the distribution and abundance of aquatic species and habitats.  To explore this principle, relationships were evaluated using long-term monitoring datasets, collected by the Oregon Department of Fish and Wildlife, and geospatial landscape data (e.g., climate, geology, and land use).  Three long-term in-channel datasets were available; spawning survey data for coho salmon sampled at index reaches (long-term, non-random, consistently productive locations) and at probability reaches (based on a random, spatially balanced sample), and stream habitat data sampled at probability reaches.  Our goals were to summarize landscape characteristics upslope of the fish and habitat sampling reaches and identify relationships among landscape predictors and local fish abundance, occupancy, and habitat.  We apply a series of spatial and statistical analyses to (1) incorporate the co-varying nature of landscape characteristics and partition the relative influence of both natural and anthropogenic landscape predictors on in-channel habitats, (2) predict the relative abundance and occupancy of adult coho salmon based on landscape characteristics, and (3) compare model performance among index and probability datasets.  Overall, our results support the premise that the conservation of freshwater resources can often be best guided by investigating and managing stream systems from a landscape perspective.  However, we are still challenged with modeling a species that is often irregularly distributed, temporally and spatially.  We suggest improvements to modeling and adjustments to sampling designs that may aid current or future monitoring programs.  Predicting the spatial distribution of species and in-channel habitat based on landscape data can guide recovery efforts over large geographic areas, and restoration efforts that may consider implications at broader spatial scales.  For future work, we plan to explore similar relationships between juvenile salmon at probability reaches and landscape characteristics, as well as to ultimately synthesize results from all analyses to develop a holistic model integrating the landscape, in-channel habitat, and multiple coho salmon life stages.