Measuring Tailbeat Frequencies of Three Fish Species from Adaptive Resolution Imaging Sonar Data

• Published in: Transactions of the American Fisheries Society
• Main Authors: Helminen, Jani, O’Sullivan, Antóin M., Linnansaari, Tommi
• Other Authors: University of New Brunswick, New Brunswick Innovation Foundation, Atlantic Canada Opportunities Agency, Fisheries and Oceans Canada, Emil Aaltosen Säätiö
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2021
• Subjects: Aris; Atlantic salmon; Salmo salar
• Online Access: http://dx.doi.org/10.1002/tafs.10318; https://onlinelibrary.wiley.com/doi/pdf/10.1002/tafs.10318; https://onlinelibrary.wiley.com/doi/full-xml/10.1002/tafs.10318; https://afspubs.onlinelibrary.wiley.com/doi/pdf/10.1002/tafs.10318

Abstract Imaging sonars, such as the Adaptive Resolution Imaging Sonar (ARIS), provide high‐resolution sonar data that are used in fisheries research and management. While sonar methods have enormous potential for making population estimates, species identification via sonar remains an unresolved challenge. One method that may overcome this challenge involves measuring tailbeat frequencies to guide species differentiation. The tailbeat frequencies of three commonly sympatric anadromous fish species of eastern North America, Atlantic Salmon Salmo salar , Striped Bass Morone saxatilis , and American Shad Alosa sapidissima, were measured from imaging sonar data collected in an experimental low‐flow, short‐range setup. The frequencies were significantly different between the species (mean ± 1 SD beats/s: 0.6 ± 0.3 [Atlantic Salmon], 0.9 ± 0.2 [Striped Bass], and 1.4 ± 0.3 [American Shad]) when measured using a previously established manual method. Building on this, an automated method was developed and tested, and the method showed promising results. However, when compared to manually identified number of beats the error was large (on average, 1.1 [Atlantic Salmon], 4.8 [Striped Bass], and −0.4 [American Shad] beats in a fish track), especially in high fish densities. Despite the limitations, the automated method has utility in fisheries management when high‐quality data can be collected for species with differing tailbeat frequencies.