One of the biggest issues facing the development of the built environment, and by extension the natural environment, is the prevalence of low density development extending far from the urban core. This phenomenon is known as urban sprawl and is harmful to both the built and natural environments as it thinly spreads finite municipal resources over large areas and reduces the natural habitats of native plant and animal species. A hotly contested policy that attempts to address the issue of urban sprawl and promote dense development in the downtown core is the implementation of urban growth boundaries, which limits or prohibits city development outside of a proscribed geographical area. There is widespread debate as to whether or not these urban growth boundaries are actually effective at curtailing sprawl, so I conducted a research project that analyzed the effect of such growth boundaries on urban spatial structure.

​

        This project utilized ENVI image processing software to analyze satellite imagery of Portland and Oklahoma City in 1972 and 2018. The purpose of the analysis was to perform an image classification scheme for both cities in each of the time periods to determine the extent of urban land cover and differentiate between high and low density development patterns, with the ultimate aim of quantifying the effect of Portland’s urban growth boundary on the limitation urban sprawl. The study found no statistically significant difference in the growth of the Portland area’s built environment versus that of the Oklahoma City area. Its effects on total expansion of urban space appear negligible. It did, however, determine some benefits of the urban growth boundary such as the protection of critical natural areas at risk of overdevelopment and an increase in higher density development patterns. 

​

          I chose Portland, Oregon and Oklahoma City, Oklahoma as my two case studies. These two cities were selected based on their highly similar populations in terms of both total number and historical growth trends. Portland serves as the experimental factor as in 1979 it became the first US city to adopt an urban growth boundary. Oklahoma City serves as the control factor due to its commonality with the population trends of Portland and its lack of an urban growth boundary. 1972 was chosen as the start of the study timeframe due to its preceding of Portland’s growth boundary and it being a year in which the populations of Portland and Oklahoma City were approximately equal. 2018 was chosen as the end of the study timeframe due to it being one of the most recent periods with available data and again when the cities were roughly equal in population.

​

        I obtained my satellite imagery from the public online repository available through the United States Geological Survey, Earth Explorer website. For the satellite imagery from 1972, I utilized data from NASA’s Landsat 1 satellite and its Multispectral Scanner (MSS) because it is one of the earliest available sources of satellite imagery available and the only for which data from 1972 could be retrieved. For the imagery from 2018, I utilized data from NASA’s Landsat 8 satellite and its Operational Land Imager (OLI) in order to maintain consistency of image characteristics within the long running series of Landsat satellites and to avoid striping errors present in Landsat 9.