Aller au contenu principal

    A tool for creating simulated survey outputs from longline data

    Demander un document de réunion
    Numéro du document:
    WG-FSA-2022/55
    Auteur(s):
    M. Kerr and T. Earl
    Soumis par:
    Georgia Robson (Royaume-Uni)
    Approuvé par:
    Chris Darby
    Résumé

    This paper describes the continued development of a tool for generating simulated multiyear surveys from longline C2 data containing catch, effort and observer sampling observations, first described in WG-SAM 22/16. Following feedback from CCAMLR-SAM-22 additional features are described in this paper.

    The tool resamples historic catch and observer data according to randomly selected or prespecified survey haul locations, applying user specified conditions (e.g. gear type). It allows statistical analysis in R to be applied to each bootstrapped data set providing summary statistics, model fits and bootstrap confidence intervals from the resampling procedure. An example application of the tool using anonymised Patagonian toothfish (Dissostichus eleginoides) data from Subarea 48.3 is presented.

    The tool has applications in testing the design of future survey proposals in areas where longline fishing has previously occurred within the CCAMLR Convention Area. It can test whether a design is likely to achieve the stated survey objectives. The tool will be further developed based on Working Group feedback and can then be made available to CCAMLR scientists for exploration of their own data sets.