INDEXES OF TOTAL STOCK BIOMASS IN THE NORTHERN AND GULF OF ST-LAWRENCE ATLANTIC COD (GADUS-MORHUA) STOCKS DERIVED FROM TIME-SERIES ANALYSES OF FIXED GEAR (TRAP) CATCHES

The hypothesis that annual catches of fixed gear fisheries are cross-correlated with stock biomass at lags predictable on the basis of the relative ages of fish comprising the catch and biomass was verified for the trapnet fisheries of the northeastern Newfoundland "northern" (NAFO 2J3KL) and northern "Gulf" of St. Lawrence (NAFO 3Pn4RS) Atlantic cod (Gadus morhua) stocks. Time series indices of stock biomass were derived from these cross-correlations. For northern cod, the index was a 3-yr weighted and lagged moving average of catch. For the years 1972-81 (the first half of the available data) the trap index (T(i)) was regressed on the stock biomass (B(i)) determined by sequential population models (SPA) (T(i) = 0.477B(i)0.638, r = 0.99, P < 0.01). Biomass forecasts for 1982-90 derived from this function (inverted) were positively correlated with recent SPA-based estimates (r = 0.94, P < 0.02). For Gulf cod, the index was a 4-yr weighted and lagged moving average of catch. This index was regressed on SPA-determined biomass for the years 1974-81 (T(i) = -3.19 + 0.0217B(i), r = 0.99; P < 0.01). Biomass forecasts for 1982-90 were positively correlated with (but lower than) SPA-based biomass estimates for the Gulf stock (r = 0.91, P < 0.05).