Predicting spawning habitat for coho salmon (Oncorhynchus kisutch), Chinook salmon (Oncorhynchus tshawytscha), and steelhead (Oncorhynchus mykiss) using geospatially constructed stream morphology from high-resolution lidar-derived digital elevation model and field survey data in the Indian Creek watershed, Mendocino County, California

Restoration of anadromous salmonid habitat is of primary importance to the economic, historical, and cultural geography of the Pacific Northwest. Derivation and use of geospatial habitat models as guides to pinpoint key areas where limited restoration funding can be cost-effectively employed is of great importance. To this purpose, 1 meter resolution lidar-derived Digital Elevation Model data was acquired for the Indian Creek and neighboring watersheds in Mendocino County, California, and used together with field-acquired geomorphic stream data to geospatially model stream widths, depths, and streambank morphology. These geospatial covariates were field-verified in selected locations and then used in conjunction with field surveyed habitat presence data and substrate data to model potential anadromous salmonid species spawning habitat. Probability surfaces, each comprising the areal extent of the Indian Creek stream system and representing the probability for spawning habitat occurrence, were developed for each of the species of interest. The mean area under the curve (AUC) for 100 model replications for Chinook, Coho, and Steelhead were 0.954, 0.951, and 0.958, with standard deviations of 0.036, 0.034, and 0.036, respectively. In contrast to other models that solely use linear lengths of stream, the models developed in this work incorporate modeled stream bankfull widths and modeled stream corridor morphology, thus allowing additional interpretation and prediction involving the amount of species’ use of specific streams and watersheds. Models were field-verified by California Department of Fish and Wildlife fisheries biologist staff and Pacific Watershed Associates engineering geologists and field scientist staff as being representative of actual field conditions, thus assuring the value of modeling results and methodology in future projects and research