Robles will use this SGER award to assemble a workgroup of colleagues to use innovative modeling techniques and progressive empirical approaches to model spatially structured dynamics in marine communities. The workgroup will consist of Hal Batchelder (UC Berkeley), Doug Donalson (UC Santa Barbara), Roger Nisbet (UC Santa Barbara), Robert Desharnais (California State at Los Angeles), Mark Denny (Stanford University), and Carlos Robles along with his students and Post Docs. The workgroup will utilize the NCEAS format.
The paradigm that governed the thoughts of many researchers for a generation - and which therefore underlay explanations of such diverse phenomena as intertidal zonation, maintenance of diversity, and stable coexistence of predators and prey - has begun to be questioned as new observations and experiments have been made. The original paradigm consists of the keystone predator hypothesis and the refugia hypothesis, which emphasize the role of predation in determining prey population structure. In this study, the researchers will examine the alternative paradigm of Spatially Structured Dynamics, in which new spatially explicit population models (SEPMs) must consider shifts in the relative levels of predation and prey recruitment over environmental gradients, rather than hiatuses in predation caused by physical stresses alone. Prior attempts at such spatial modeling of intertidal communities have been limited in focus or else have been based on hypothetical conditions with no empirically based parameterizations. Robles and the workgroup will construct SEPMs based on twenty years of experimental data that Robles has collected on mussel population structure at rocky shores along Pacific coast. Two alternative modeling approaches will be used, a "stochastic arena model" using cellular-automata, and an individual-based model. The dual thrust will allow the researchers to take advantage of the strengths of each approach. The workgroup will validate the models in several ways including the comparison of predicted maps with observed GIS maps, thus pushing modeling practice to greater levels of realism. This project will be a unique opportunity to bring together a group of PIs with strong ecological modeling expertise. The incorporation of strong field data into new spatially explicit population models may have substantial implications
