This paper uses actual krill length frequency data in an attempt to address the problem of determining adequate sample size to obtain representative krill length frequency distributions from commercial catches. In addition, the possible effect of two other factors (within-trawl variability and sample decomposition) which may influence length frequency data quality are also considered. The attention of WG-Krill is drawn to the following:
* All measurements of length from commercial krill catches should as far as possible be made by a single observer/vessel;
* WG-Krill should consider the minimum length differences between-catches which should be detectable. As far as possible, meaningful account must be taken of the desired biological characteristics which the Group wishes to discern (e.g. length-with-age). At present, and in the interests of both statistical rigour and the efficiency of measurement, it is proposed that length measurements from commercial catches should be grouped into 2 mm size classes;
* The biological implications associated with the detection of specific differences in length should be considered along with the need to collect maturity stage information.
* For most purposes a minimum sample size of at least 100 animals/trawl is necessary to obtain statistically meaningful differences between-samples.
* The problem of how frequently to sample commercial krill catches still requires consideration.
Abstract:
We have no standard as to how many krill we should measure to obtain reliable estimates on its body length distribution. Krill samples collected from layers in the northern waters of Livingston Island were used in this study to get fundamental information on this problem. Based on the measurements of 2000 animals per sample, comparisons were made in mean body length and cumulative frequency distribution with less number of measurements. Measurements on 50 animals seem enough to obtain reliable estimates on mean body length. At least I000 measurements are desirable to get reliable estimates on cumulative frequency distributution.
