COS 81-4
Drivers of trait patterning in a classical niche assembly model

Wednesday, August 12, 2015: 2:30 PM
320, Baltimore Convention Center
Rafael D'Andrea, Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI
Annette Ostling, Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI

Following the realization in the past decade that communities under niche assembly may look neutral, we need refined tests that can distinguish between the two. In this trait-based exploratory analysis, I simulate a stochastic niche assembly model where competition between consumer species derives from overlap in resource use as determined by consumer traits, aiming to examine how factors such as mass effects, the shape of the resource utilization curve, and the degree of specialization of different consumer species contribute to create pattern in species abundances on the niche axis. By comparing the observed abundances in simulations to expected abundances from a neutral model, I look for telltale signs of niche assembly, such as the partitioning of the niche axis into clusters of similar species, each cluster representing a niche strategy. In addition to furthering our understanding of how specific factors drive pattern, the goal is to get an indication of when trait-based approaches are likely to be useful overall. 


Pattern is low to indistinguishable from neutral when consumers adopt similar resource utilization strategies, i.e. when different consumer species differ only with respect to preferred resource but not in scope of resource use (in other words, when utilization curves have the same shape). Pattern is more visible when consumers differ with respect to their degree of specialization to available resources. The shape of the utilization curve itself also matters for pattern formation. In particular, pattern is more visible if the transition in attack rates from preferred to non-preferred resources is sharp rather than smooth. Overall, it seems that a large region of parameter space leads to weak or distinguishable pattern. In contrast, pattern is almost always clear in purely deterministic closed-community formulations of the model. I conclude that A) even when competition is decisively impacted by resource overlap, drift and mass effects may obliterate the trait structure expected of niche assembly; B) only under certain circumstances can we expect niche assembly to overcome stochastic forces in driving species abundances on the niche axis.