Fish production can lead to discharge of wastes and have negative impacts on the environment. It is therefore important to carefully monitor and plan the development of aquaculture. A model that can simulate fish growth on the basis of available fish species and local conditions (like water quality and quantity and quality of fish feed) could therefore be extremely useful. Since growth strongly depends on feed intake, a growth model should be able to predict maximum feed intake of fish under ad libitum feeding. This thesis aimed at investigating some potential mechanisms for feed intake regulation and predicting maximum feed intake in the light of these hypothesized mechanisms using a theoretical modeling approach. Among a wide range of factors involved in feed intake regulation, glucose and oxygen are two factors of particular interest. The overall hypotheses were: (1) high blood glucose level has a negative effect on feed intake in fish; and (2) dissolved oxygen concentration of the water is a major determinant of feed intake in fish.
To investigate the effects of blood glucose on feed intake in Nile tilapia (Oreochromis niloticus), an experiment was done with feeds containing different combinations of carbohydrate and energy levels. The results showed that fish fed a diet high in carbohydrate but low in energy failed to achieve their digestible energy requirement, suggesting that a high level of blood glucose suppresses feed intake in Nile tilapia. A dynamic explanatory model for fish growth which incorporated metabolite pools as potential regulators of feed intake was then developed. After parameterization and calibration for rainbow trout (Oncorhynchus mykiss), the model gave good prediction of fish growth. The results showed that feed intake regulation could be simulated based on the glucose static theory. The parameterization and simulation results showed the gaps in our quantitative knowledge about the intermediary metabolism in fish.
To investigate the effects of dissolved oxygen (DO) concentration and body weight on maximum feed intake and growth of Nile tilapia, three experiments were conducted with different DO levels and body weight classes. The results provide strong support for the theory that limitation of oxygen supply through the gill surface area results in lower feed intake and growth of fish at low DO concentration. The allometric relationship between gill surface area and body weight results in lower relative feed intake, and therefore lower relative growth in bigger fish than in smaller fish. The study also demonstrated that the incipient DO (DO above which feed intake of fish remains unchanged) depends to a large extent on body weight and probably on feed composition.
The short-term effect of DO on ad libitum feed intake was then simulated based on the balance between oxygen demand and oxygen supply. The model was calibrated and validated using the experimental data from the present study and some additional data from earlier published experiments. The calibration and validation results showed that the model could predict satisfactorily maximum feed intake and growth of Nile tilapia when DO was limiting. However, feed intake when DO was not limiting was not simulated satisfactorily. This was attributed to the empirical equation for maximum feed intake determined by a factor other than DO.
In the last chapter, the modeling approaches, the role of glucose and oxygen in feed intake regulation, implications for aquaculture management, reliability of data on maximum feed intake and implications for further research are discussed.