In Defense of Type I Functional Responses: The Frequency and Population Dynamic Effects of Feeding on Multiple Prey at a Time
Márk Novák, Kyle E. Coblentz, John P. DeLong
Abstract
AbstractEcologists differ in the degree to which they consider the linear type I functional response to be an unrealistic versus sufficient representation of predator feeding rates. Empiricists tend to consider it unsuitably nonmechanistic, and theoreticians tend to consider it necessarily simple. Holling's original rectilinear type I response is dismissed by satisfying neither desire, with most compromising on the smoothly saturating type II response for which searching and handling are assumed to be mutually exclusive activities. We derive a "multiple-prey-at-a-time" response and a generalization that includes the type III to reflect predators that can continue to search when handling an arbitrary number of already-captured prey. The multiprey model clarifies the empirical relevance of the linear and rectilinear models and the conditions under which linearity can be a mechanistically reasoned description of predator feeding rates, even when handling times are long. We find evidence for the presence of linearity in 35% of 2,591 compiled empirical datasets and support for the hypothesis that larger predator-prey body mass ratios permit predators to search while handling greater numbers of prey. Incorporating the multiprey response into the Rosenzweig-MacArthur population dynamic model reveals that a nonexclusivity of searching and handling can lead to coexistence states and dynamics that are not anticipated by theory built on the linear type I, type II, and type III models. In particular, it can lead to bistable fixed point and limit cycle dynamics with long-term crawl-by transients between them under conditions where abundance ratios reflect top-heavy food webs and the functional response is linear despite having an inherent upper limit. We conclude that functional response linearity should not be considered empirically unrealistic but also that more cautious inferences should be drawn in theory presuming the linear type I to be appropriate.