Saving Salmon Through Advances in Fluvial Remote Sensing: Applying the Optimal Band Ratio Analysis (OBRA) for Bathymetric Mapping of Over 250 km of River Channel and Habitat Classification

Abstract:

Salmonids are threatened with extinction across the world from the fragmentation of riverine ecosystems from dams and diversions. In California, efforts to expand the range of spawnable habitat for native salmon by transporting fish around reservoirs is a potentially species saving idea. But, strong scientific evidence of the amount of high quality habitat is required to make these difficult management decisions. Remote sensing has long been used in fluvial settings to identify physical parameters that drive the quality of aquatic habitat; however, the true strength of remote sensing to cover large spatial extents has not been applied with the resolution that is relevant to salmonids. This project utilizes hyperspectral data of over 250 km of the Tuolumne and Merced Rivers to extract depth and bed slope from the wetted channel and NIR LiDAR for the surrounding topography. The Optimal Band Ratio Analysis (OBRA) has proven as an effective tool to create bathymetric maps of river channels in ideal settings with clear water, high amounts of bottom reflectance, and less than 3 meters deep over short distances. Results from this study show that OBRA can be applied over larger riverscapes at high resolutions (0.5 m). The depth and bed slope estimations are used to classify habitat units that are crucial to quantifying the quality and amount of habitat in these river that once produced large populations of native salmonids. As more managers look to expand habitat for these threatened species the tools developed here will be cost effective over the large extents that salmon migrate to spawn.