Using stationary acoustic platforms to assess precision and accuracy of acoustical krill surveys

The current CCAMLR protocol specifies daytime acoustic sampling in order to avoid issues of krill migrating too close to the surface to be acoustically detected, as well as potential issues of diel changes in acoustic scattering strengths. However, due to logistic constraints, acoustic data were collected both during day and night for the 2019 Scotia Sea large-scale krill monitoring survey. Data from upward-looking, stationary platforms offer unique opportunities for assessing the effects of krill vertical distribution on survey results. The Institute of Marine Research, Norway has on a regular basis deployed stationary platforms in the main area for commercial krill harvesting at the South Orkney Islands since 2014, and in this report some of these data have been analysed in order to provide an estimate of the magnitude of bias introduced by night-time acoustic sampling. We found great temporal and geographical variation in DVM (Diel Vertical Migration). During February and March, the time period covered by the 2019 survey, the frequency response data obtained using an upward looking NORTEK broadband echosounder suggested that 13 % of the macroplankton backscatter originated above a depth of ~20 m during daytime, while 24 % was found above ~20 m during night. A swarm detection approach suggested a similar pattern, 7 % of aggregated backscatter was found above ~20 m during daytime, and 22 % during night.