On predicting the growth of cultured bivalves

A simple, limited ecosystem model focused on bivalve growth in a coastal aquaculture site is presented. The model is based on a system of coupled, nonlinear ordinary differential equations which predict the temporal evolution of the following state variables: individual bivalve weight, bivalve numbers, zooplankton biomass, phytoplankton biomass and non-plankton seston. A limiting nutrient is also included to constrain the overall system. The equations are based on population mass budgets and allow for particle exchange with adjacent regions. The model structure is general and designed to be applicable to a variety of bivalve species or sites. In order to test the feasibility of the model for predicting growth, it is applied to a blue mussel Mytilus edulis culture site in a coastal inlet near Lunenburg, Nova Scotia, Canada. Idealized forcing functions for the annual cycles of light, temperature and the far-field concentrations of the state variables are used. It is shown that the model is able to reproduce the general features of observed mussel growth in the different regions of the inlet at the relatively low stocking densities found there. Numerical experiments with high stocking densities are carried out in order to estimate carrying capacity for the inlet. Sensitivity analysis shows that predicted mussel growth is highly influenced by small chan in the physiological parameters which describe the mussel energy budget. It is suggested that this feature may prove to be an important limitation in using such models as predictive tools for managing the development of shellfish aquaculture. (C) 1997 Elsevier Science B.V.