Comparison of Acoustic and Trawl Methods for Estimating Density and Age Composition of Kokanee

Abstract

Three acoustic and three trawl methods of estimating density and age composition of kokanee Oncorhynchus nerka were compared. Density estimates from the three acoustic methods were significantly correlated with each other and with the trawl estimates. Age-classes 1–3 were not distinguishable from each other in acoustic records, but their combined peak could be separated from that of age-0 kokanees on the target strength frequency distributions produced by both the dual-beam and the deconvolution analyses. Dual-beam age composition estimates were significantly correlated with deconvolution (r = 0.97) and trawl (r = 0.70) estimates. Deconvolution density estimates for age-1–3 fish were 34% higher than the dual-beam estimates, but neither of the acoustic density estimates differed significantly from the trawl values due to higher variance associated with trawl estimates. Acoustic estimates of total densities were less consistent than those for age-1–3 fish because of the problem of establishing a consistent lower threshold for age-0 fish. At the extreme, duration-in-beam, dual-beam, and deconvolution estimates of total density were 3.3, 1.9, and 1.8 times the trawl estimates. Differences in towing speed (1.0 versus 1.5 m/s) for the slow and fast otter trawls had little effect on the estimates of either density or age composition, but the beam trawl was less efficient than the otter trawls. Estimates of densities from the beam trawl averaged 54%, 35%, and 21% of otter trawl estimates for age-0, age-1, and age-2–3 kokanees (which averaged 42, 130, and 198 mm in length, respectively), suggesting serious size selectivity for one or both trawl designs.