Protecting and preserving our natural life-support systems requires the ability to understand the direct and indirect effects of human activities over long periods of time and over large areas. Computer simulations are now becoming important tools for investigating these interactions. Spatio-temporal modeling is essential when simulating heterogeneous environments. These models are necessary for developing the relatively realistic description of past system behavior that is required for making predictions of the impacts of alternative management policies on future system behavior (Costanza, Sklar, & White, 1990); Risser, Karr, & Forman, 1984; Sklar & Costanza, 1991). The data and hardware barriers to building these complex models have begun to erode with the increasing availability of remote sensing data and GIS systems to manipulate it, and the development of parallel computer systems. However, significant advances in modeling tools must take place in order to and in order for non-computer scientists to take advantage of these new resources. This document describes a proposal to develop these much-needed spatial modeling tools.