News for North Texas
Play Live Radio
Next Up:
0:00
0:00
0:00 0:00
Available On Air Stations

How Do You Define A 'Megaregion' Like Dallas? By Analyzing 4 Million Daily Commutes

GARRETT DASH NELSON AND ALASDAIR RAE
/
PLOS ONE
This map shows the megaregions of the U.S. (represented by colors) based on an algorithmic analysis of four million commutes (represented as lines).

When people say they live in Dallas, they could actually mean Plano, Irving or Rockwall. It’s a common practice to group urban and rural areas and label them as the nearest big city. It's easier to recognize by a broader scope of people. Analysts do this, too. For several reasons, the economic geography of the U.S. is made up of and described as “megaregions.”

The term has been around for decades, but the definition has been based on loose interpretation. Megaregions are large, regional areas that often cut across state lines.

To determine these regions, analysts rely on “interlocking economies, transportation links, shared topography, or a common culture,” according to National Geographic. But that information doesn’t translate into concrete boundaries and renders a largely subjective map.

An attempt to mathematically define the ambiguous term was recently published in the open-access journal Plos One by Garrett Dash Nelson, a postdoctoral student in geography at Dartmouth College, and Alasdair Rae, an urban data analyst at the University of Sheffield in England.

They looked at commutes. Their study is called "An Economic Geography of the United States: From Commutes to Megaregions."

How it looks vs. how the math works

Nelson and Rae gathered data from the U.S. Census of more than 4 million commuter paths to define megaregions across the country. Then, they applied two different methods to analyze the data: a visual interpretation and another based on an algorithm developed at MIT.

You can see the difference between the two methods in these images of the Minneapolis-St. Paul megaregion. 

Map from visual interpretation

The visual approach looks at commutes of 50 miles or less — represented by straight lines between the start and end points — to capture the vast majority of commuters within major metro areas. Short, high-volume commuter flows are in yellow and longer, low-volume commutes are in red. 

As you can see, the map shows a major concentration on the Twin Cities and largely ignores satellite cities like St. Cloud and Rochester. 

Credit GARRETT DASH NELSON AND ALASDAIR RAE / Plos One
/
Plos One
The visual approach used to draw the Minneapolis-St. Paul megaregion shows concentration in the Twin Cities that extends outward.

Nelson and Rae say this approach could suffice for economic planning because the big cities are likely the driving economic force for the area. But it's hard to know how functionally separate these outlying cities are from Minneapolis and St. Paul based on this approach alone. 

Map from algorithm

The algorithm, designed by MIT’s Senseable City Lab, mathematically recognizes communities and ignores physical location. The algorithm considers the strength of the connection (the volume of commuters) between nodes (70,000 census tracts). This map shows how self-contained the Minneapolis-St. Paul region really is, in terms of commuter flows. Few commutes stretch to or from neighboring communities. Unlike the visual method, the algorithm assigned more outlying cities, like Fargo, North Dakota, to the megaregion.

Credit GARRETT DASH NELSON AND ALASDAIR RAE / Plos One
/
Plos One
This map based on the algorithm shows concentrations in smaller cities in the megaregion based on commuting paths.

The study focuses visually on California and Minnesota. But here are maps showing Texas commutes:

Credit Rae, Alasdair; Garrett.G.D.Nelson@dartmouth.edu (2016): United States Commutes and Megaregions data for GIS. figshare.

Credit Rae, Alasdair; Garrett.G.D.Nelson@dartmouth.edu (2016): United States Commutes and Megaregions data for GIS. figshare.

The researchers limited the algorithm to 50 megaregions in the U.S., where every node color-coordinates to the region it belongs to — see the map below.

According to National Geographic, 50 seems like an arbitrary number but “it makes sense mathematically because a very high percentage of commutes lie entirely within a megaregion relative to paths that cross boundaries between regions.” 

This map shows census tracts color-coded to their assigned community. 

Credit GARRETT DASH NELSON AND ALASDAIR RAE / Plos One
/
Plos One
The algorithm assigned individual census tracts (dots) to 50 megaregions (colors).

The algorithm — operating without geographical information and an understanding of the country's cultural boundaries — didn't get everything right. So Nelson and Rae combined the two methods to draw a more conclusive map.

National Geographic explains how they accomplished the final rendering of U.S. megaregions:

They started by drawing lines around the dots on the map above. They then overlaid those shapes on the flow map at the top of the post and reinterpreted the boundaries to eliminate outliers and emphasize geographic continuity. The result is the map below, which eliminates some visual oddities. For example, the splotchy green area has been absorbed by the New Orleans-Delta megaregion, and a big swath of the west with relatively low population isn’t included in any megaregion.

Credit GARRETT DASH NELSON AND ALASDAIR RAE / Plos One
/
Plos One
A combination of visual and algorithmic approaches to define megaregions in the U.S.

Both approaches have their strengths and limitations, according to Nelson and Rae, and interpreting these regions concretely is a work in progress.

Learn more

Despite some shortcomings, the report shows a new way to define America’s largest regions. How we define those regions have implications for the economy, infrastructure, transportation and culture. Explore the full study