Big Spatial Data & Land Use Progression in the United States 1945-2016

The complexity of large spatial data and the lack of cyberinfrastructure able to process it has presented a significant obstacle to advancing knowledge across multiple domains. This research presents a new methodological approach for processing large spatial data on a desktop computer. Additionally, novel statistical summaries of the spatio-temporal progression of land use in the United States are presented using the land use data product generated by this methodology. Using the land use data, novel analytical methodologies for handling large complex data derivations are explored through a preliminary, in-depth case study that assesses the material composition of the built environment with using a spatial lens and multiple clustering paradigms. We find that there is remarkable consistency in land use composition both regionally and across the rural-urban continuum. Furthermore, there is clear attributional data structure and strong spatial signals in the material aspects of the built environment presenting new avenues of inquiry for the built environment as a complex system