Predation hazard or intimidation plays an important role in the ecology of freshwater communities by structuring habitat use by member species. Locally, intimidation may suppress growth and reproduction, thereby directly affecting the population dynamics of member species. While intimidation from predators is known to produce local effects, no experimental work has addressed the important question of how intimidation is expressed on larger scales; i.e. the scale of whole streams or watersheds on Iotic systems. The goal of this proposal is to explore that question in a tropical stream fish community. We will test the hypothesis that predators, through intimidation and direct killing, can generate observed patterns across a large (geographic) scale, and also a hypothesis that competition is responsible for an observed pattern. We will test specific predictions derived from the hypotheses in controlled experiments in an tropical first- and second-order experimental stream system, and by conducting a large scale survey of the Paria drainage in the Northern Range Mountains of Trinidad.
