Individual-Based Spatial Simulations of Ecological Systems
Tania Kostova
Center for Applied Scientific Computing
Lawrence Livermore National Laboratory
Mail code L-561, Livermore, CA 94551, USA
kostova@llnl.gov
Simulation models are gaining increasing popularity in various fields of science and engineering because of the steadily increasing computational power in academic and research institutions. The relative straightforwardness of developing a model of interacting heterogeneous individuals in a heterogeneous spatial environment via the simulation approach allows for studies of emergent behavior as well as for carrying out virtual experiments in a virtual field or laboratory.
While this approach is known in other fields as agent-based modeling, historically the ecological simulations have been referred to as Individual-based models (IBM). IBM include sets of probabilistic and deterministic rules that are nonlinear or of the if-then type.
In certain cases they can be considered as (sometimes infinite-dimensional) discrete dynamical systems. Except in very simple, idealistic cases, these systems do not render themselves to analytical treatment. A large number of simulations and a corresponding statistical evaluation of their behavior is the means to understand their dynamics. Regardless of this, these are the models that can be closest to the real systems they represent.
In this session we invite researchers who explore the world of discrete-time individual-based spatial simulations of ecological systems to present their work and participate in discussions about the usefulness of the method, the output analysis, visualization techniques, the role of space, impact of habitat fragmentation on the survival or time to extinction of a species and other connected issues.