Mapping Past Land Use Changes and Modeling Future Urbanization Patterns in a Mid-Sized American City

On a global scale, humans have been moving from rural to urban areas at an increasingly fast rate in a process called urbanization. It is crucial to understand and predict the impacts of urbanization because city expansion encroaches on farmland and woodland, reducing regional ecosystem services. Therefore, urbanization must be well planned in a forecasting manner. The objective of this study is to map the land use change in Knox County, Tennessee, from 2001-2019 to identify the urbanization trends and understand the dynamic geospatial relationships between the urban and rural areas. The geospatial mapping was done using USGS land use data by classifying land into urban, agricultural, non-agricultural, and waterbody areas. Then, geographic information system (GIS) software was used to analyze the change of each land area in relation to infrastructure change. The results show a decrease in agricultural and non-agricultural land area and an increase in urban use. In addition, normalized difference vegetation index (NDVI) and elevation data, which were collected from the USGS Landsat database, were used to analyze the changes in vegetation, impervious surface area, and surface reflectance during the urban growth. The results were used to create a suitability index model from an environmental view in order to predict the areas in Knox County that are suitable for urban development. Finally, a machine learning model was implemented to compare areas identified as suitable with areas of predicted urbanization. These two models can enable urban planners to develop policies that geospatially optimize urban infrastructure while minimizing environmental footprints.