Authors: Michelle Stuhlmacher*, DePaul University, Matei Georgescu, Arizona State University, Yina Hu, Peking University, B. L. Turner II, Arizona State University, Ran Goldblatt, New Light Technologies, Sarthak Gupta, Arizona State University, Amy Frazier, Arizona State University, Nicholas Clinton, Google
Topics: Urban Geography, Remote Sensing, Land Use and Land Cover Change
Keywords: urbanization, urban, pattern, remote sensing, Google Earth Engine, United States, India, China, urban sustainability, LULC
Session Type: Virtual Paper
Start / End Time: 8:00 AM / 9:15 AM
Room: Virtual 33
Presentation File: No File Uploaded
Globalization—the worldwide interaction of people and institutions—has long shaped global cities. Its effects have been hypothesized to generate a homogenization of urban form and associated impacts. This hypothesis, however, has rarely been tested across multiple countries. Using machine learning to classify urban areas from satellite imagery, we test the urban form homogenization hypothesis on 150 of the most populous cities in China, India, and the United States, examining how the area and configuration of built-up land within cities have changed over time (1995-2015). We find evidence of urban configuration homogenization in the highly populous cities of each country in terms of increasing connectivity (i.e., greater infill) and increasing shape complexity (i.e., increased sprawl), which corresponds to higher daytime surface urban heat island (SUHI) intensity. Our results highlight the potential utility of employing urban configuration to inform planning and management, permitting shared best practices among cities to facilitate sustainable development.