IMPORTANCE OF SPATIAL SCALE AND PREY MOVEMENTS IN PREDATOR CAGING EXPERIMENTS

Abstract

Field experiments involving local manipulations of predator densities are an important tool for studying the role of predation in natural communities. I use mathematical models to investigate how treatment effects should vary with the size of the area manipulated in open predation experiments, i.e., in studies that allow prey to migrate in and out from experimental units. The most general result is that the influence of prey movements on prey densities decreases with increasing spatial scale of an experiment, while the effect of predation rate is independent of scale. Thus, the results seen in small-scale experiments tend to reflect prey movements, while the effects seen in large-scale experiments are due mainly to predator-related mortality. If predation rates are spatially variable (e.g., the habitat consists of refuge patches and predator-rich patches) the problem becomes more complex, because movement rates influence predation rates. Movements that are independent of predator density can swamp the effects of predation mortality in small-scale experiments. However, in larger scale experiments, the effect of movements is to increase the predation rate. This is because a larger proportion of the prey is exposed to predators at high movement rates. Predators can also affect local densities through effects on prey behavior. If prey increase movement rates out from predator patches, this will lead to increased treatment effects in small-scale experiments. In contrast, the result of such behaviors is to decrease treatment effects in large-scale experiments, because most prey will be in refuge patches and predation rates will be low. Alternatively, sometimes predators cause prey to decrease activity when close to a predator. This kind of behavioral response leads to decreased treatment effects in small-scale experiments. If the response is strong, the result can even be a reversed effect, i.e., reduced movements out from predator cages lead to increased prey densities in such cages. I argue that the effect of this type of behavior in large-scale experiments should also be to decrease treatment effects. This is because predation rates are usually low on inactive prey. However, freezing behaviors cannot lead to reversed predator effects on a large scale.