Linking Remotely Sensed Hydromorphological Conditions With Biological Observations to Estimate Salmon Spawning Capacity

ABSTRACT Performing ecohydraulic assessments of salmon habitat is essential for planning and evaluating management actions. Current in situ and modelling approaches, although accurate, can be time‐consuming and expensive, which limits the areal extent of the study. To address this spatial limitation, we developed and evaluated a remote sensing methodology to map salmon spawning microhabitat and predict spawning capacity using remotely piloted vehicle (RPV, i.e., drones) collected imagery. We used RPV‐based aerial imagery to model local hydraulics (depth and velocity) and substrate size ( D 50 and D 84 ). We assessed model accuracy using in situ measurements for each characteristic. Using modelled hydraulics, we predicted habitat quality and evaluated the model via comparison to known redd locations. We found redds were concentrated in areas predicted to have high habitat quality, indicating the habitat model performed well. We then coupled the hydraulic model with a substrate biophysical model to estimate reach‐level spawning capacity and found the site could accommodate 4730–8344 redds depending on habitat condition thresholds, compared with 414 redds that were counted at the site in the same season. Our findings indicate that RPV‐collected imagery can model fluvial characteristics and can be used to efficiently model spawning habitat and capacity. This approach provides a methodology for researchers to conduct further salmon habitat modelling studies using remote sensing to map habitat quality and quantify capacity at larger spatial extents while maintaining high resolution.