Authors: Dolores Jane Forbes*, U.S. Bureau Of the Census, Zhixiao Xie, Florida Atlantic University
Topics: Urban Geography, Spatial Analysis & Modeling, Land Use
Session Type: Paper
Start / End Time: 8:00 AM / 9:40 AM
Room: Stones Throw 1 - Granite, Marriott, Lobby Level
Presentation File: No File Uploaded
Cities and their environs are examples of complex systems produced by human activities operating at multiple scales. The European Commission has developed a global human settlement model to facilitate comparisons, monitor change, and address UN goals for sustainable development, disaster risk reduction, and humanitarian relief. The U.S. Census Bureau also develops an urban model of the United States (US), serving many purposes including the distribution of federal funds. These goals indicate a need for accurate representation of urban areas.
This study seeks to identify and describe map differences of urban areas using the two models, especially any unrelated to known differences in methods, data sets, or population thresholds. Normalized cross-tabulation (error) matrices calculated at multiple scales describe a continuum of similarities and differences across the U.S., identifying increasingly smaller areas where similarities/dissimilarities are high. Patch count differences identify differences in contiguity, and good-ness-of fit algorithms rank similarities while accounting for differences in class size. All these methods identify the same two areas as most similar/dissimilar. Population density distributions for the cells of disagreement indicates a heavy-tailed distribution in the most dissimilar area. This suggests that the European model follows the geographic property of scaling for urban areas, highlighting the different perspectives produced by the two models and the need for a global definition of urban.