Skip to main content

    Describing krill: swarms or integration intervals?

    Request Meeting Document
    Document Number:
    SG-ASAM-17/02
    Author(s):
    M.J. Cox
    Submitted By:
    Dr Martin Cox (Australia)
    Approved By:
    Dr Dirk Welsford (Australia)
    Abstract

    CCAMLR anticipates that acoustic data collected from krill fishing vessels will be incorporated into krill feed back management.  Acoustic data collected from fishing vessels may have different characteristics to data collected from research ships.  For example, acoustic data collected from fishing vessels may contain seabed aliased echoes and instrument cross-talk.  Analysing acoustic data on a krill-swarm, rather than the current integration grid approach, may facilitate the semi-automated processing of acoustic data and reduce post-processing time.  Using a validated aggregation detection algorithm, I estimated mean areal krill density for a small (65 km by 60 km) survey conducted in the waters off Mawson research station in the East Antarctic (66o 25' S 63o 14' E).  I found 61 % overlap between the variance estimates for conventional grid-based swarm-based krill density.  The processing time of the swarm-based appraoch was half that of the standard grid-based technique. Whilst additional test data sets are required, the techniques presented here can be used to further explore the efficacy of swarm-based analysis.