Tracking the removal of buildings in rust belt cities with open-source geospatial data

Urbanization is generally understood as the process of growth in both population and developed areas. However, this perception is not reflective of the type of change that is occurring in the Rust Belt region of the United States where many urban areas reveal shrinkage instead. While much of the research surrounding these shrinking cities is in the realm of socio-economic implications, few studies have investigated how to map these shrinking areas. This research aims to contribute to the growing body of shrinkage research by examining methodology to monitor the fast removal of buildings in the Rust Belt shrinking cities with easily available open source data such as Light Detection and Ranging, aerial orthoimages, and GIS datasets. Our ultimate goal is to develop methodology for improved, generalizable mapping of shrinking cities. We applied our methods to Detroit, Michigan and Youngstown, Ohio which both show significant urban shrinkage and both have a variety of survey datasets available for validation. We map a 5-year change in Detroit as well as a 10-year and 19-year change in Youngstown to provide, in high detail, the process of building removal. For Detroit we found that 12.9% of all land parcels that contained a building in 2009 had lost this building by 2014. New builds were drastically overshadowed by the demolished structures, accounting for <1% of the total number of parcels in the city. We found similar results in Youngstown, were 13.1% of all parcels studied revealed that a structure was removed between 1994 and 2013, with similarly low rebuild percentages (<1%).