Geographic Responses of Groundfish to Variation in Abundance: Methods of Detection and Their Interpretation

Abstract

Recent published studies have used data from bottom trawl surveys of groundfish populations to test whether distributional area and abundance are correlated. Two studies that used different indices to represent the distributional area of Georges Bank haddock (Melanogrammus aeglefinus) yielded conflicting results. To determine whether this is an example of different distributional indices measuring different things, both indices were regressed against estimates of abundance of haddock from a different but neighbouring location on the southwestern Scotian Shelf. Positive correlations were observed for immature age-classes using both indices whereas only one of the two indices resulted in positive correlations for mature age-classes. The following factors contributed to the lack of agreement among distributional indices: (1) age-aggregated indices potentially obscure correlations between distributional area and abundance for individual age-classes; (2) distributional indices that depend on the magnitude of catch rates confound variation in the large-scale horizontal distribution of stocks with diurnal variation in the three-dimensional distribution of schools; (3) distributional indices that scale positively with abundance generate spurious correlations. The results suggest that the outcome of any test of whether distributional area and abundance are correlated depends on the index chosen to represent distributional area.