Natural mortality in exploited fish stocks: annual variation estimated with data from trawl surveys.

Natural mortality (M) is difficult to measure directly in exploited fish stocks. Therefore, in stock assessments, M has usually been assumed to be constant for age classes of fishable size. However, M is likely to vary annually due to temporal changes in food availability, predation, fish size at age, and other biological and environmental factors. We suggest that M of fish at age and year is the sum of length-related mortality and condition-related mortality (M = ML + MC). We propose that MC is directly proportional to PKH, the percentage of fish in the stock with condition factor (K) and liver condition (H) below critical values. This relationship is used to demonstrate likely patterns in annual variation in MC for ages 2–9+ in Icelandic cod (Gadus morhua) based on fish measurements in groundfish surveys in spring and autumn. There were annual fluctuations and temporal trends observed in MC for different age classes attributable to variation in food availability. The described method is suitable for age-structured stock assessments of cod and other fish stocks. Our results indicate that basic stock assessments may be improved by including estimates of variable MC. [ABSTRACT FROM AUTHOR]