Publication Abstract

Intercalibration of North Sea International Bottom Trawl Surveys by fitting year-class curves

A.J.R. Cotter

Intercalibration of trawl surveys is needed whenever trawling technique, the survey vessel, the season, or the localities of trawling are altered. Statistical modelling offers a generally cheaper method of estimating intercalibration factors than comparative trawling trials. Much variability of whole-survey population abundance indices transformed to natural logarithms can be explained using only the year-class strengths and a coefficient of total mortality, Z, for the species. This modelling approach was therefore used to intercalibrate surveys forming part of the International Bottom Trawl Survey of the North Sea between 1977 and 1997 for four commercially important galoid species, cod, haddock, whiting, and Norway pout. An age-related factor was included to allow for apparently lower catchabilities of young fish. The models fitted satisfactorily, permitting intercalibration factors to be estimated with standard errors. No indications of changes in Z were found over the period or over different year classes for any of the species. Residual errors were positively correlated among-ages-within-years for each survey. Residual degrees of freedom were therefore reduced using an information measure before testing factors in the model for significance or estimating standard errors. A method for comparing the relative precisions of the different surveys given the fitted model is also described.

Reference:

A.J.R. Cotter, 2001. Intercalibration of North Sea International Bottom Trawl Surveys by fitting year-class curves. ICES Journal of Marine Science, 58(3): 622-632.