Hydroacoustic fish stock assessment in the presence of dense aggregations of Chaoborus larvae

Abstract

A new method for eliminating reverberation due to Chaoborus larvae from hydroacoustic recordings is presented based on an assumption of a constant dependence between the area backscattering strength (sa) with a high-volume backscattering threshold (sv) and sa with a low sv threshold for fish. The idea was to analyze data with a threshold high enough to eliminate reverberation and then convert the estimate to coincide with the result that would have been achieved with a low threshold containing all backscattering from fish. The approach was validated with a secondary dataset, and the magnitude of overestimation of fish density by reverberation was evaluated using data from four surveys conducted in a clay-turbid lake, where small planktivorous fish, smelt (Osmerus eperlanus), and larvae of Chaoborus flavicans coexist in the water column. With the presented method, estimation of smelt density was possible even when Chaoborus density was >200 individuals·m–3. The analyses revealed that the overestimation of fish density could be as high as 50% if the reverberation is not taken into account. The presented method might also be applicable for eliminating reverberation due to other unwanted targets, because it is based on the acoustic properties of fish rather than those of unwanted targets.