Effect of allocation and stratification strategies on precision of survey abundance estimates for Atlantic cod (Gadus morhua) on the eastern Scotian Shelf

One of the primary uses of bottom-trawl survey data, with respect to fisheries management, is the estimation of abundance of the species involved. Stratified-random surveys have been conducted on the eastern Scotian Shelf since 1970. The precision of abundance estimates for Atlantic cod (ages 5-12) from these surveys was found to be inferior to what would have been obtained if simple random surveys had been conducted. These results were primarily due to suboptimal allocation of stations to strata. Historical data were used to devise alternative sampling strategies. This analysis indicated that the number of strata could be reduced considerably without loss of precision. An additional and important benefit of the smaller number of strata would be increased flexibility for allocation. With the present level of sampling, it would be necessary to reduce the number of strata to achieve proportional allocation of stations or to approximate optimal allocation