Comparison of age-length keys of Baltic cod derived from Polish commercial and research data

Abstract

This paper compares the age-length keys (ALKs) of cod sampled from Polish commercial and research catches during 1992–1993 with respect to time (in quarters 1–4), place (subdivisions 25 and 26) of catch and fishing gear (trawl, gillnet, research vessel). Two methods have been applied to determine similar ALKs: cluster analysis and statistical tests. Statistical testing was based on the χ 2 test and Fisher's exact test. The general conclusion from the analysis is that ALKs sampled by different fishing gear are not significantly different. Similarly, those sampled in both subdivisions 25 and 26 do not show significant differences. However, the ALKs differ in respect of sampling time. Although there are no significant differences between ALKs sampled in the 1st and 2nd quarter, those sampled in the 2nd, 3rd and 4th quarters are generally significantly different. The results of the analysis show which ALKs can be interchanged and which may be helpful in both filling the gaps in sampling for age and for planning the sampling method. The results of the χ 2 test were very similar to the outcome of Fisher's exact test when sample size was >20. The conclusions based on cluster analysis were confirmed by statistical tests – the distance between the compared ALKs in the cluster analysis and the probability from statistical tests showed high negative correlation.