Diet optimization in a generalist herbivore: the moose.

In an attempt to understand the foraging of a generalist herbivore, a linear programming optimization model was constructed to describe moose feeding in summer. The model attempts to predict the amounts of aquatic vegetation, deciduous leaves, and forbs a moose should consume each day; and to determine whether or not its feeding is constrained by the maximum feeding time available each day, its daily rumen processing capacity, its sodium requirements, and its energy metabolism. The model can be solved for two alternative strategies: time minimization and energy maximization. The energy-maximizing strategy appears to predict the observed diet chosen by an average moose very well. Also, the diets selected by moose of each sex and various reproductive states appear to fit the energy-maximizing strategy. In addition, it is demonstrated that a moose’s body size at weaning, size at first reproduction, and maximum size are related to foraging efficiency. Furthermore, there appears to exist an optimum adult body size for feeding. The general conclusion arrived at is that the foraging of a generalist herbivore can be predicted in a quantitative manner, at least in this case, as has been shown for other types of consumers (carnivores and granivores).