Mathematical Biology Seminar

Kevin Lafferty
USGS, UC Santa Barbara
3:05PM, FRIDAY, October 1, 2010
LCB 225
Stochastic Ecological Network Occupancy Models

Abstract: I am interested in how to approximate complex ecological systems with simple mathematical tools. Stochastic ecological network occupancy (SENO) models predict the probability that species will occur in a sample of an ecological network. SENO models are a means to fill a gap in the theoretical toolkit of ecologists. As input, SENO models use a topological interaction network and rates of colonization and extinction (including consumer effects) for each species. A SENO model then simulates the ecological network over time, resulting in a series of sub-networks that can be used to identify commonly encountered community modules. The proportion of time a species is present in a patch gives its expected probability of occurrence, whose sum across species gives expected species richness. SENO models can be used to investigate how topological complexity, species interactions, species traits, and spatial scale affect communities in space and time. They can categorize species as biodiversity facilitators, contributors, or inhibitors, making this approach promising for ecosystem-based management of invasive, threatened, or exploited species.