Density and in situ target strength of herring and sprat: a comparison between two methods of analyzing single-beam sonar data

Abstract

We analyzed sonar data obtained with a 70-kHz single-beam echosounder with two methods for removing the beam pattern effect: one based on the Craig and Forbes logarithm; the other based on a z-transformation and deconvolution operation. The sonar data were collected from 4 July to 23 October 1985, in a coastal area of the northern Baltic proper where herring ( Clupea harengus L.) and sprat ( Sprattus sprattus L.) dominated the pelagic fish populations. The two methods gave similar total fish densities, but the proportion of different size classes sometimes differed by up to 30 percentage units. The acoustic data are compared with size distributions and catch per unit effort in simultaneously collected vertical gill-net samples. Size distributions in nets and acoustics were fairly similar, but the relationship between catch per unit effort in gill nets and acoustic density estimates was poor. Target strength to fish-length regressions from the two acoustic methods were similar even though individual estimates differed by up to 4 dB on some sampling occasions. When the slope of the regression is set to 20 we obtained the equation TS=20 log L-69.9 with both methods ( TS is target strength in dB and L is fish length in cm).