DIEL HABITAT USE BY JUVENILE PACIFIC SALMON ALONG AN URBAN SHORELINE

Abstract

A mobile dual-frequency identification sonar (DIDSON) was used as an alternate technology to characterize diel juvenile Pacific salmon (Oncorhynchus spp) distributions along an armored and eco-engineered marine shoreline. Eco-engineering modifications to seawall habitat were intended to create an intertidal corridor for shore-oriented salmon and included the addition of marine mattresses (mesh bags filled with rocks) to create intertidal benches, a textured seawall to increase invertebrate colonization, and embedded glass blocks in an overhanging sidewalk to increase ambient light. Acoustic surveys conducted April to August 2019 along the Seattle, Washington, waterfront were used to compare acoustic and snorkel sampling methods and to quantify diel salmon habitat preferences. Acoustic salmon counts were at least twice as high as snorkel counts in most habitats with the difference attributed to the ability of the DIDSON to sample a greater range and volume than visual surveys. Snorkel salmon detections (i.e., salmon counts n ≥ 1) were higher than acoustic detections at new and reference sites, apart from under pier microhabitats, with the difference attributed to the ability of snorkelers to identify lone fish or individuals within mixed groups of fish. Results from generalized linear models and generalized linear mixed models demonstrate that the eco-engineered shoreline creates preferred diel habitat for salmon, while traditional under-pier habitats are less preferred for salmon regardless of time of day. Increased ambient light, fewer shading structures (e.g., pier pilings), and shallow water improve nearshore habitats for juvenile salmon. Pier end salmon counts exceeded under-pier counts and were similar to between pier counts, indicating that some salmon prefer to navigate around pier ends rather than under them. Overall salmon presence was two times higher at night compared to during the day. Salmon presence in the intertidal corridor was more similar to reference sites at night than during the day. Highest salmon counts occurred in transects exposed to artificial light, indicating potential influence of light along the Seattle waterfront. The results of night sampling highlight the need for day/night fish sampling to reduce bias in habitat use and abundance estimates, and to identify diel differences in anthropogenic effects on nearshore communities. This study establishes that the DIDSON is a viable nearshore mobile sampling tool for small fishes (25 mm – 90 mm) that increases sampling efficiency and spatiotemporal knowledge of nocturnal fish distributions and habitat use over snorkeling counts.