Because of the long temporal scales and the large spatial scales over which landscape processes operate, it is often difficult to implement observational studies or controlled experiments at the landscape level. Simulation modeling has therefore arisen as an important methodological approach in landscape ecology, complementing traditional field-based studies and large-scale empirical work based on remotely sensed data. Landscape simulation models provide a framework for synthesizing knowledge about ecological relationships derived from many different studies, and assessing the outcome of multiple, spatially interacting processes as they are played out over time and space. The Landscape Ecology Lab is currently working on the development and application of two landscape simulators.

The LADS model was developed by Mike Wimberly (Warnell School of Forestry and Natural Resources) as a generalized simulator of landscape dynamics and ecological responses in a disturbance-driven mosaic of age-structure habitats. The model was initially developed to simulate the spatial and temporal dynamics of forest age classes under pre-European fire regimes in the Pacific Northwest, but can be applied to a variety of different forest types. Recent enhancements include a module for tracking the landscape-level dynamics of live and dead tree biomass, and a species model that simulates metapopulation responses to habitat dynamics. Future work will include a timber harvesting submodel, and a refined forest growth submodel that represents a broader range of disturbances, environments, and successional pathways.

The LANDIS model was originally developed by David Mladnenoff (University of Wisconsin - Madison) and Hong He (University of Missouri - Columbia). We have parameterized and applied LANDIS to simulate disturbance regimes and forest community dynamics in the southeastern United States. Research to date has analyzed shifts in landscape-level species composition under changing disturbance regimes, and examined the sensitivity of these results to various modeling assumptions. Future work may involve revising LANDIS or developing a new model that better captures key elements of Southeastern fire regimes, such as short (< 10 years) fire return intervals and strong relationships between species composition, fuel quality, and fire behavior.