The GeoGraph 3D Agent-Based Computational Laboratory

Catherine Dibble, Philip G. Feldman

Human interactions of all kinds are increasingly structured by transportation and communication networks. Yet tools to model, understand, and predict dynamic human interactions and behavior on spatial networks and on richly structured geographic landscapes have lagged far behind. Our GeoGraph extensions to the RePast agent-based simulation platform address this gap by supporting models in which mobile agents interact on network and other geographic landscapes. GeoGraphs contribute to geographic science and spatial social science by allowing us to develop models that include not only heterogeneous site-specific characteristics, but also the complementary social and landscape structures that mediate each agent’s dynamic geographic situation. GeoGraph computational laboratory tools have been designed to support controlled experiments for agent-based geographic science through their ability to generate richly structured parameterized families of synthetic geographic landscapes or of complementary Remote Sensing or GIS-derived geographic landscapes.

GeoGraphs support building and testing simulation models grounded in interesting spatial structures such as spatial small-worlds, geographic scale-free networks, hybrids between the two, or GIS representations of real-world landscapes. Then each landscape is populated with homogeneous or heterogeneous distributed mobile agents, including their social networks and context-specific behaviors.

A GeoGraph model can represent any spatial scale, depending for visual clarity only upon suitable cartographic generalization in the representation of the landscape and of its agents. Each GeoGraph node or agent can be scaled visually from the tiniest dots to a fully detailed graphic image of anything from a shrubbery to a house to a Porsche to an elephant to any other graphical avatar. Similarly, nodes in a network landscape may be interpreted, and represented, as anything from tables in a café to world cities or even planets. Agents for each landscape may be scaled either in their correct proportion to the geographic scale, enlarged to facilitate visualization of the model's behavior, or reduced to avoid visual clutter.

Figure Caption: A GeoGraph parameter-driven synthetic

three-dimensional fractal terrain landscape with parameterized

renewable green “tree” agents and small flocks of red and blue

“deforestation” or wildlife agents.

Example publication: Catherine Dibble and Philip G. Feldman (2004) “The GeoGraph 3D Computational Laboratory: Network and Terrain Landscapes for RePast”, Journal of Artificial Societies and Social Simulation 7(1), jasss.soc.surrey.ac.uk/7/1/7.html.