SHOULD TOTAL LANDINGS BE USED TO CORRECT ESTIMATED CATCH IN NUMBERS OR MEAN WEIGHT-AT-AGE

Many ICES fish stock assessment working groups have practised Sum Of P roducts, SOP, correction. This correction stems from a comparison of t otal weights of the known landings and the SOP over age of catch in nu mber and mean weight-at-age, which ideally should be identical. In cas e of SOP discrepancies some countries correct catch in numbers while o thers correct mean weight-at-age by a common factor, the ratio between landing and SOP. The paper shows that for three sampling schemes the SOP corrections are statistically incorrect and should not be made sin ce the SOP is an unbiased estimate of the total landings. Calculation of the bias of estimated catch in numbers and mean weight-at-age shows that SOP corrections of either of these estimates may increase the bi as. Furthermore, for five demersal and one pelagic North Sea species i t is shown that SOP discrepancies greater than 2% from the landings ar e very unlikely. Larger discrepancies probably are indications of prob lems with the sampling design. The proper action is to reexamine the s ampling programme and to revise it where needed. (C) 1997 Internationa l Council for the Exploration of the Sea.