| Home | People | Research | News | Contact Us |
|
Part 2: Modeling the rates of urban growth Monitoring and documenting urbanization is only the first step in addressing the issue of global land cover change. The final objective of this study is to determine the mechanisms of land cover change to more fully understand global change processes and their effects. Assessment of land cover conversions caused by human actions requires understanding of the political, social, economic, cultural and environmental factors that motivate behavior and affect the direction and intensity of land and natural resource use. This is generally done by building a model that links human activity and land use (Lonergan 1994). This component of the research will attempt to integrate land cover change information derived from Landsat TM imagery (part one) with county level socioeconomic and demographic data to develop a statistical model of the variables correlated with land conversion. The modeling of land use change mechanisms will rely on a common statistical method for social research, the general linear model. Linear models, primarily regression analysis, are concerned with the study of the dependence of one variable on one or more explanatory variables, with a view to estimating and/or predicting the value of the dependent variable in terms of the known or fixed values of the explanatory variables (Gujarati 1995). The dependent variable, yi, is modeled as:
where xi is value of each independent variable for the ith county, the alpha and beta coefficients are determined empirically using ordinary least squares methods, and ei is the random error term. The advantages of regression models are their readily interpretable format and results, and the substantial array of statistical tools that can be applied to assess the importance of each of the mechanisms (Venables 1994). In this research, the dependent variable yi will be the amounts of land cover change for each metropolitan area determined from remote sensing imagery. During analysis, the task will be to determine which data are appropriate as explanatory variables for land use change, and which model is appropriate and statistically significant given the collected data. There is no doubt that a number of global and local variables are at play in driving land use change in cities across the globe. The question, however, is whether a set of common independent variables can be determined as mechanisms of change for all cities in the population set, at what scale this set of relationships should be modeled, and if this relationship can be generalized across cities in a meaningful way within a predictive model. A general set of independent variables that influence the rate of land cover conversion has been identified in the literature. The table below highlights these variables and potential proxy variables that will be employed.
|
| <- Back | Overview | Part 1: Change detection | Part 2: Modeling |
|
Boston University | Department of Geography | |
|
|
Urban Areas Research |
