ESTIMATION OF VON BERTALANFFY GROWTH-PARAMETERS IN THE PRESENCE OF SIZE-SELECTIVE MORTALITY - A SIMULATED EXAMPLE WITH RED GROUPER

Fish lengths back-calculated to all scale or otolith annuli are often used for estimating parameters of the von Bertalanffy growth equation in an attempt to increase sample size and presumably improve the fitte d model. Lengths at early ages back-calculated from younger age-groups sometimes are greater than lengths at the same age estimated from old er age-groups, referred to as Lee's phenomenon. Lee's phenomenon can b e simulated with use of size-selective mortality and moving averages t o incorporate error dependency in lengths at age. To evaluate the appr opriateness of using back-calculated lengths at formation of all annul i in growth analyses when size-selective mortality is present, we used simulated data based on von Bertalanffy growth parameters and total m ortality for red grouper Epinephelus morio from the southeastern U.S. Atlantic coast. Asymptotic length (L infinity) and the growth coeffici ent (K) showed greater bias when estimated with back-calculated length s at all annuli than when estimated only with back-calculated lengths at the most recent annuli whenever Lee's phenomenon (or its reverse) w as present. One should use only lengths back-calculated to the most re cent annuli when estimating growth parameters representative of a popu lation.