Simulation of Large Scale Spatial Evolutionary
Processes
Mark P. Miller
Department of Biology and Center for High Performance Computing
Thomas Hauser
Center for High Performance Computing
Utah State University, Logan, UT
March 12, 2007
Computational analyses provide researchers wth the ability to explore diverse
areas of inquiry that may otherwise not be readily amenible to emipirical
investigation. The field of evolutionary biology serves as an excellent example of such
scenarios, as it has a rich history of incorporating mathematics, modeling, and
simulation work. These types of investigations allow researchers to explore processes
that occur over exceptionally long time scales and are therefore difficult to
empirically observe in some instances. Although evolutionary theory is particulary
well developed and can describe processes that occur within well-defined
“populations” of organisms, appropriate computational infrastructures and theory is
not yet in place to completely account for scenarios where species are continuously
distributed over large geographical regions. In this presesentation, we describe
ongoing work by the investigators to develop a simulation infrastucture that
incorporates core elements of the evolutionary process for a continuously distributed
species (organismal movement, reproduction, and mutation). We will demonstrate
how our simulation analyses provide new insights about intrinsic attributes of
spatially structured evolving systems. We further describe the challenges
associated with simulating these processes over landscape-scale geographical
regions and provide illustrations of how resources at the Center for High
Performance Computing at USU are being used to overcome these obstacles.