Document: JER-3-52-2

Testing simple models of keystone predation.

FOX, J.W.*

Cook College, Rutgers University, New Brunswick, NJ, U.S.A. ¹

Abstract:
Most species occupy intermediate trophic levels, and share both resources and predators with many other species. Simple keystone predation models make testable predictions about how species abundances will shift along a productivity gradient. The identity of the dominant species at any productivity level should be predictable from species' abilities to graze down resources and sustain predators. These predictions are largely untested. I tested these models by assembling replicate protist communities in laboratory microcosms. Without predation, prey species compete for bacterial resources and R* values predict relative success in resource competition, as predicted by the simplest mechanistic models. However, contrary to simple predictions, competing prey species coexist, grazed bacteria increase with productivity, and protist effects on bacteria do not vary with productivity. Bacterial heterogeneity may explain these discrepancies. A phenomenological donor-controlled model predicts the competition data better than simple explicit models of resource heterogeneity. Predation reduces total prey abundance and shifts dominance toward prey with high ratios of per-capita growth rate to attack rate suffered, as predicted by keystone predation models. These predictions are robust to non-equilibrial dynamics. Predator effects on prey cascade only weakly to bacteria, consistent with donor control of prey and suggesting top-down trophic cascades are not robust to trophic level heterogeneity. Future research should focus on the mechanisms leading to donor control. Controlled experiments are essential for distinguishing robust predictions from fragile predictions, and details that matter from details that don't.

Keywords: keystone predation, resource competition, productivity, donor control, protists, microcosms

This abstract is being presented at: 5:00 PM in session:
Oral Session #11: Trophic Cascades.