In the Southeastern U.S. rapid urbanization is a major challenge to developing long-term conservation strategies. The SAMBI DSL project used predicted urban growth models described herein to inform future landscape conditions that were also based climate change impacts and vegetative community succession. These future landscape conditions were then applied as a context for land use and management decisions in conservation planning.
SLEUTH, named for the model input datasets (Slope, Land use, Excluded, Urban, Transportation and Hillshade) is the evolutionary product of the Clarke Urban Growth Model that uses cellular automata, terrain mapping and land cover change modeling to address urban growth (Jantz et al, 2009; NCGIA 2011). SLEUTH provides urban growth projections which are useful across a range of applications; including wildlife habitat analysis, conservation planning, and land cover dynamics analysis. SLEUTH incorporates four growth rules (Spontaneous Growth, New Spreading Centers, Edge Growth and Road-Influenced Growth) to model the rate and pattern of urbanization. The model simulates not only outward growth of existing urban areas, but also growth along transportation corridors and new centers of urbanization. SLEUTH incorporates five parameters (Dispersion, Breed, Spread, Slope and Road Gravity) into the growth rules which project future urbanization. Possible parameter coefficient values range between 1 and 100. During calibration every possible combination of these five parameter coefficients (between defined start and stop values and by a defined step size) is applied to the growth rules, in order to find the combination that best matches past urbanization patterns observed in the training data. Once found, the model is run in prediction mode using these parameter values in the growth rules. The model produces one urban growth cycle per year. For each growth cycle, a GIF image is produced showing the probability of urbanization for each pixel.
This project utilized the SLEUTH-3r version of the model taking advantage of added new functionality and substantially increased performance (Jantz et al. 2009) over previous versions.