Performance of a glider‐mounted multifrequency echosounder for measuring the vertical distribution and abundance of large pelagic copepods

Abstract The difficulty of sampling zooplankton communities in situ has driven advancements in autonomous, remote sensing technology. The goal of this paper was to perform a gear comparison study testing the performance of one such piece of technology—a glider‐mounted four‐frequency echosounder—against traditional shipboard methods of measuring zooplankton, with a particular focus on stage C5 Calanus finmarchicus distribution. Abundance data from concurrent vessel (nets, optical plankton counter) and glider‐mounted acoustic backscatter were collected over a 9‐d period in two deep basins in the outer Bay of Fundy, Canada. Zooplankton were generally found in depth‐stratified layers, with C. finmarchicus in particular being concentrated near the bottom of both basins, and smaller copepods near the surface. The echosounder was typically able to discriminate between the presence and absence of C.finmarchicus layers vertically in the water column and resolve patchiness within the deep layer. The echosounder was unable to detect layers of small copepods. We found moderate correlation between expected and observed backscatter that depended primarily on echosounder frequency. C. finmarchicus concentration was best predicted ( R 2 = 0.26) by observed backscatter using the 455 kHz frequency filtered using the dB difference between the 200 and 455 kHz frequencies. Absolute concentration estimates from the echosounder after correction with the derived linear model remained consistently lower than net‐derived estimates by approximately one order of magnitude. Overall, the glider‐mounted instrument performed well at its given objectives, and we recommend its use in other areas where patch location is an important attribute to determine.