Adult Sockeye Salmon Assessment in a Tidal, Turbid River: A Comparison of Sonar and Test Fishing Methods

Abstract

AbstractThe feasibility of using sonar in Alaska's lower Kvichak River, a region with a 3.8 m tidal flux, to assess the abundance of adult Sockeye Salmon Oncorhynchus nerka was evaluated as a replacement for an in‐river gill‐net test fishery. Test fish indices are one tool that is used to manage commercial and subsistence fisheries in Bristol Bay. On this river, an observation tower project near the clear headwaters provides accurate abundance estimates for this species, but fishery managers require timelier estimates for in‐season daily management. The lower‐river test fish index is timelier, but it is less precise than tower estimates are. We had to determine whether to replace the test fish project with sonar based on how well each compared with tower estimates and an assessment of their cost and effort. Two dual‐frequency identification sonars (DIDSONs) were deployed along one bank of the lower river (one pointed toward shore and one away from shore) to assess salmon passage across all tidal stages in an attempt to produce an index with higher precision. Compared with the tower estimates, the test fish index showed a slightly stronger correlation in each of the three study years than the sonar index did, despite the stronger sampling power of the sonar. The test fish and sonar indices were strongly correlated. In the lower river, upriver salmon migration (85–98%) occurred primarily during ebb tides and downriver passage (97–100%) during flood tides, indicating that salmon do not use selective tidal stream transport to minimize their energy expenditures in this river. We were able to successfully use fixed‐location sonar in a tidal nearshore region to produce an index of fish abundance. Our recommendation was to continue the test fishing project rather than to replace it with sonar because of the similarity between their indices and the higher cost of the sonar project.