Comparison of four nonlinear growth models for effective exploration of growth characteristics of turbot Scophthalmus maximus fish strain

Abstract

This study was conducted to compare the effectiveness for non-linear growth models designated as Chapman-Richards, Gompertz, Logistic and von Bertalanffy for selection of fast-growing fish strain of turbot Scophthalmus maximus. These models were compared using the goodness of fit (the coefficient of determination (R²) and the mean square error (MSE)) and the Akaike information criterion (AIC) and the growth characteristics of turbot from 3 to 27 months of age. The results in the present study showed that R² was the highest in Chapman-Richards, but the lowest in von Bertalanffy model. The MSE and AIC values were the highest in Chapman-Richards followed by von Bertalanffy model, whereas Gompertz model is the lowest compared to other models. The Gompertz model had the lowest mean square error (6421.8706) and Akaike information criterion (65.1322) and the second highest coefficient of determination (0.9908) (almost equal to the first highest coefficient of determination), suggesting being the best fit model for description of turbot growth trajectories. Furthermore, the results of turbot growth characteristics explored by the Gompertz model revealed that the fast-growth time interval of turbot were (10.23, 26.78) (unit: months) and the fast-growth time interval distance was 16.55 months. The results of this study suggested that the Gompertz model could be the best fit model for description of turbot growth trajectories, whereas the deduced mathematical formulas of growth intervals could be used in determining the growth characteristics of other fish.

Keywords: Scophthalmus maximus, nonlinear models, comparison, growth characteristics