MEAN SIZE AT AGE - AN EVALUATION OF SAMPLING STRATEGIES WITH SIMULATED RED GROUPER DATA

The influence of sampling protocol on estimation of mean lengths at ag e was evaluated by computer simulation of a population of red grouper Epinephelus morio. Variation in length at age was simulated with plato ons of different lengths within each year-class. Mean length of each p latoon was assigned with the normal distribution and mean size at age from a previous growth study. Natural mortality was assumed to be 0.2. Simulated samples were obtained at random or with length stratificati on from either the population or its fishery. Fishing mortality was as sumed to be a function of either fish age or fish length. Estimates of mean length at age contrasted with known true mean lengths indicated that reliable estimates of mean size at age requires random sampling o f lengths within ages. Stratification of samples by length biases the estimates of mean length at age. Similarly, samples drawn from size-se lective gears or fisheries yield biased estimates of mean length at ag e. Growth models fitted to such data will not generally reflect the me an growth of individuals in the population, and even slight changes in sampling protocol can result in misleading temporal shifts of estimat es of size at age.