HEMA Theme 1: Drivers and Impacts of Land Use Change
Our lab was among the first groups to develop a spatially-explicit land use change model for the purpose of exploring environmental impacts of urbanization and agricultural expansion. The Land Transformation Model dates back to the early 1990s. The first version was developed in Arc/INFO using aml and standard spatial databases.
General Methodological Framework
Over the last 15 years, we have worked to enhance the model and to study drivers and impacts at different scales and at different locations around the world. Drivers are understood using spatially explicit methods of correlation, spatial statistics and a variety of qualitative methods. The drivers of land use change are composed of a host of economic, social and environmental factors that operate in complex ways. We also employ the use of role playing simulations (RPS), key informant interviews, semi-structured interviews, life stories, landscape stories and multi-criteria evaluation methods. We have used stakeholder derived scenarios to test how different policies might impact important outcomes, such as water quality or the quality and composition of biota. A major emphasis in recent years has been the impact of land use change on water quality. We have looked at this in a variety of ways, from employing simple hydrologic models to interfacing our land use models to complex hydrologic and hydraulic models.
Land Change Models
A major focus of our work has also been the development of different kinds of models that can be executed at the same location. For example, we have developed an artificial neural network model that is also compared to agent-based models, multi-criteria models and models generated from focus groups and role playing simulation. We rely on a variety of theoretical frameworks, including resilience ecology, complex social-ecological systems, landscape ecology, human-environment interactions from geography and political ecology. We are also interested in merging qualitative and quantitative sciences.
We have also focused on model calibration and model validation. We have been interested not only what the models produce but how well they create maps of future land use patterns. The science of land use change is still very young. As a community we recognize that a tremendous amount of research is needed to elucidate patterns of land use change that can then be used to formulate effective policy.
Unfortunately, many of the land use change models that exist do not predict all that well, or at least as well as other models (e.g., mechanistic biophysical models) that are used to study landscape dynamics.
We have been fortunate enough to have been funded to study in many areas of the world. We have projects that focus in various areas around the Upper Midwest United States, East Africa, and Eastern Europe. The model that was developed in our lab has been used in six continents by researchers at universities and researcher institutions around the world.
Another focus in the lab has been the development of a backcast land change model that reconstructs historical landscapes that are then used to understand long-term dynamics of land use change and environmental impact. The backcast model can be run for just about any area of the world where census of agriculture and housing data exist to drive the model backwards.
We have also been working on coupling land use and regional climate models through simulations of crop production and vegetation change. Most of this work has occurred in East Africa but more recent work (see the NASA Hydrology Project) focuses on this coupling in the Upper Midwest of the United States.
Output from our models is available for download via the project web sites. A video tutorial on how to use the Land Transformation Model is also posted here along with the executable for the models and sample data to run the model.
Last updated by BCP on February 26, 2011