Project Gigalopolis: Urban and Land Cover Modeling
Dr. Keith C. Clarke, University of California at Santa Barbara, Department of Geography
The UC Santa Barbara Gigalopolis Project (http://www.ncgia.ucsb.edu/projects/gig/v2/About/bkOverview.htm) aims to "develop tools to best predict urban growth on a regional, continental and eventually global scale." It was developed by Dr. Keith C. Clarke of the Department of Geography at UC Santa Barbara and has been sponsored by USGS and NSF. The program can be used on various different data sets and can cover nearly any given amount of time. Thus, the project is not on a specific research subject, but rather devoted to a tool that can be used for such.
There are two models involved in the project: the Urban Growth Model (UGM), which can run independently of the urban code, and the land cover deltatron model (LCD), which is driven by the UGM and which is included in the code. Together they are referred to as SLEUTH, an acronym for the image input requirements: slope, land cover, exclusion, urbanization, transportation, and hillshade.
The Gigalopolis Project has created a hypothetical data set, Demo_city, to illustrate the use of the model, and once downloaded and understood, the model can be used on any properly formatted data set. The time frame is an input, and is represented by a growth cycle, where one growth cycle is one year of growth. A simulation is a series of growth cycles that can model urban growth over time. It must be started with a set of conditions, including a seed that initializes the random number generator, a value for each of the five growth coefficients, and SLEUTH input images.
An example of this program used a South Coast study area. It mapped out the urban growth boundary, parks, and agriculture, but the images I found for it were part of a presentation, and thus unattached to any sort of explanation. It does, however, show that this model can be used to map the growth rate of past urban sprawl and to predict the future patterns of urban growth in a given area. Since this project is a visual interpretation of growth patterns, it cannot be done without mapping or spatial analysis. Stating a growth rate as a figure and describing its direction would not give the reader an accurate interpretation. Questions to be explored using a model like this include how much the transportation system is affected by growth and vice versa, which does not need to be analyzed spatially, but perhaps temporally. Other questions include the change in size and location of agricultural land, which is certainly spatial, and where in a specific state or country is urban growth occurring the most.