Growth of cod (Gadus morhua) estimated from mark-recapture programs on the Scotian Shelf and adjacent areas

Abstract

Growth of cod (Gadus morhua) was estimated from length increment data compiled from tagging programs conducted between 1978-1985 in the Southern Gulf of St Lawrence, on the Scotian Shelf, in the Western Bay of Fundy and on Georges Bank. A model developed by Francis was used to estimate growth parameters gα and gβ which represented mean annual growth rate at chosen reference lengths 45 and 65 cm respectively. Individual growth variability was modelled as proportional to the expected change in length, which in turn was a function of time at liberty and initial size. The model was fit using a maximum likelihood technique. The tails of the distribution of residuals were longer than normal, reflecting the scatter of the data. The residuals did not show any pattern with either of the independent variables (time-at-liberty, length-at-release). Mean growth was greatest in the Western Bay of Fundy and Georges Bank, intermediate on the Scotian Shelf (Brown's Bank, Halifax and Banquereau Bank) and least in Sidney Bight and the Gulf of St Lawrence. The regional trend in growth estimated from tagging data was consistent with trends in growth estimated from size-at-age distributions. Regional growth differences reflected the underlying hydrographic regime. The correlation between gα and bottom temperature was r=0.84 (df=7, p=0.004) while that between gβ and bottom temperature was r=0.88 (df=7, p=0.002). The estimate of individual variability of growth is large, and suggests that size-at-age distributions among regions would overlap. Based on the geographic distribution of recaptures, fish from the Scotian Shelf migrated shorter distances than did fish from the Gulf of St Lawrence, Brown's Bank, Western Bay of Fundy and George's Bank; fish from adjacent regions were more likely to mix than fish from more distant regions.