What Makes a Megacity?

arch 316 individual independent project | 2019

The dozens of current-day megacities—urban areas with a population above 10 million people—share much in common: they aggregate people in close proximity, facilitating innovation and change; they act as the cultural, economic, and political centers of their respective countries; and they continue to grow. At street level, however, each city is exceptional. The people, institutions, architecture, climate, and underlying geographies of each city influence how people live in and interact with their city in unique, and sometimes strange, ways. This project attempts to aggregate connections between multiple aspects of megacity layouts; including street patterns, land uses, and natural geographies. While some connections are pedestrian and some might be distractions, others might yield insights into the genius of certain cities, and into how other cities can improve the lives of their own citizens.

See the full matrix here.

Megacities and their current populations were determined using the 2018 Demographia list of World Urban Areas. Every current-day megacity was surveyed using Google Maps and Street View to determine its value for 15 attribute categories. These values were then categorized and compared with each of the other 14 attributes, as well as world geographic location, country, population, land area, and population density. The correlations are described below:

Manifest Correlations:

  • Cities with small land areas correlate with significant amounts of informal housing.
  • Street network layout of the CBD generally correlates with the same and similar periphery street network layouts.
  • Industrial land uses are negatively correlated with density.
  • Cities with airports outside their urban area correlate with relatively low populations.
  • Cities with sprawl-like periphery street layouts correlate with non-street-facing mid- to high-rise apartments and with industrial land uses.
  • Cities in Africa correlate with forest as prevailing constricting geography.

Explicable Correlations:

  • High urban land areas are correlated with cities in the US and China, and correlated negatively with India, Pakistan, and Bangladesh.
  • Cities within large megaregions correlate with low densities and high land areas.
  • Prevailing highway layout as ring roads correlates with non-street facing high rises.
  • Cities with airports at the edge of their urban area correlate with mountains as constricting geography.
  • Cities with large separate regions correlate with highway networks which are complex webs.
  • Large megaregions generally correlate with the US, China, Brazil, and Japan.
  • Capital cities and cities with capital cities within their megaregion correlate with national borders which are far from the outside of their urban areas.

Curious Correlations:

  • Rivers as closest significant waterbody correlate with that waterbody being adjacent to the CBD, whereas oceans correlate with near—but not adjacent to—the CBD.
  • Rivers as closest significant waterbody and waterbodies adjacent to the CBD correlate with airports near—but not adjacent to—the CBD.
  • Cities with distorted grid periphery street networks correlate with substantial amounts of informal housing.
  • Cities with organic periphery street networks correlate with agricultural fields as the city’s constraining geography.
  • Cities with small separate neighborhoods correlate with highway networks which are radial toward the CBD, and with agricultural fields as the city’s constricting geography.
  • Cities with water as their prevailing constricting geography correlate with higher populations.
  • Cities with no associated megaregion correlate with significant amounts of informal housing, having agricultural fields as their prevailing constricting geography, and being a national capital.
  • Cities in small megaregions correlate with web-like megaregion geographies and with organic street layouts in their CBDs.
  • Cities in large mega regions correlate with linear megaregion geographies, being in the same megaregion as national capitals but not being capitals themselves, and with mountains as prevailing constricting geometries.
  • Capital cities correlate with small separate neighborhoods and significant amounts of informal housing.
  • Cities within small megaregions correlate with national borders which are outside but nearby the edge of their urban area, whereas cities within large megaregions correlate with national borders which are far outside the edge of their urban areas.

This study finds correlations which might lead to interesting discoveries about how geography and city layout influence each other and the way residents live. Future work should quantify the qualitative values and loose correlations shown here and use samples of more cities to ensure the findings are truly correlated, rather than just coincidence. More detailed research could investigate whether these connections are linked by causation as well by finding explanations for the correlations. More concrete conclusions could allow cities to find solutions to their current problems, and to problems they don’t yet know they have, already implemented in cities all around the world.