Retrospective Landcover Analysis of Urban Growth and Deforestation in Flagstaff, AZ Using Automated Classification Methods on Landsat Surface Reflectance Imagery
Presentation Time: Mon, 12/12/2022 - 11:30
Keywords: Landcover Analysis, Urban Growth, Deforestation, Automation, Surface Reflectance Imagery
In this project, I describe a method for automating historical land use/land cover change analysis for the Northern Arizona, greater Flagstaff area. The method investigates deforestation as a result of urban growth, the results of which are displayed in a timeseries. The area surrounding Flagstaff, Arizona is interspersed with areas of urban development as well as a diverse population of spruce, fir, and pine trees. In recent years, more forest and grassland areas have been removed to make room for more urban development, both within and outside the city limits. I obtained one surface reflectance raster image per year taken in Spring from 2022 back through 2000. These images were obtained from the United States Geological Survey Earth Explorer data collection and were captured by the Landsat 8 Operational Land Imager/Thermal Infrared Sensor and Landsat 5 Thematic Mapper satellites. After initial processing, scenes were classified using an unsupervised ISO Cluster classification technique. The land cover classifications were identified via manual interpretation techniques aided by high-resolution, Google Earth historical imagery. Analysis of these classifications provided land use and land cover data for understanding the recent extent of urbanization and deforestation in this region. The results of this study demonstrate approximately 3,946.6 acres of forest were lost to urban development, between 2000 and 2022, which equates to a 15.44% loss in forest acreage. In addition, ArcGIS Pro Model Builder models were developed to allow for a reproducible method of performing similar analyses in other study areas.
