BACK-CALCULATION OF WEIGHT AT AGE: IS IT NECESSARY?

Abstract

The back-calculation of the weight at age is analyzed in order to determine whether this procedure should be considerad as a separated aspectof the back-calculation of length in age and growth studiesThe life history of350 fish wassimulated, which were capturad from the age3 until 9. These data were used to evalúate four procedures for the back-calculation of the weight at age. The relative error between the estimates and the true simulated valúes is used as a measure of bias. The methods of back-calculation of the weight were: a) the relationship length-weight (RLW) using the back-calculated lengths; b) the equation proposed by Beyer (1991); c) back-calculation of the weight using the back-calculated lentgth, where the weight is corrected by the ratio between the observed and estimated weight at the moment of the catch; and, d) back-calculation of the weight using a relationship between the weight of the fish and the radius of their otolith without taking into account the length of fish. The back-calculation of the weight using the RLW produces underestimate in the ages that have samples (ages 3 to 9) and overestimation in the ages 1 and 2 (ca. 40% and 10%, respectively). The weights were overestimated in the most younger ages using all the procedures but the positive bias tends to zero in the most oldest ages. The procedure (d) was mora efficient in the most younger ages and the idea of back-calculating the weight at age without taking into account the backcalculation of length is favored.