Does city living trim greenhouse gas emissions?

A recent study says that it does.

The authors of this study, published in The Journal of Urban Planning and Development, quantified the emissions from building materials and construction, home heating and power demands, and transportation energy, in both urban suburban neighborhoods in the Toronto metro area.  And they found that downtown residents use radically less energy, and consequently emit about two-thirds less climate-warming CO2 than their suburban counterparts.  Take a look:Norman Study Graph

First, they found that the biggest difference between city and suburban living is in transportation emissions. Compared with center-city dwellers, suburbanites used 3.7 times more transportation energy. Low density development requires more and longer car trips compared to higher density areas. Even transit is more carbon-intensive in suburbs.

Second, the greenhouse gas impacts of building materials—all the wood, steel, concrete, glass, and the like—are tiny, compared with the energy used in resident’s daily lives.  In fact, the “embodied energy” in the building materials  isn’t much more than a rounding error. It’s still important to reduce the environmental impacts of building materials and construction, but  when weighing the long-term climate impacts of new construction, the most important consideration is how we live once we’re in the buildings. That means the energy that goes into construction may be less important than the location of the building.

To, me, the most interesting thing about the study is the relative importance of transportation to GHG emissions. Whether the emissions are counted per square foot of living space or per person, low density development generates more emissions and requires more energy in large part because of residents dependence on car trips and longer, diesel-burning transit trips. The study suggests that the kind of transit trips are important with high density dwellers relying on shorter electric powered street cars or subways rather than longer bus trips.

Norman Study Figure 6

From the study:

It is notable that the overall trend between densities has not been fully reversed by changing the functional unit, which suggests a high level of overall energy use and GHG emissions intensiveness for low density development. This is largely due to the significantly higher level of automotive transportation emission and energy use association with low density development compared with high density development.

So while the study has its limits—it compares just two neighborhoods in a single city– it points, as other studies do, to the evidence that sprawl and car dependence are closely linked, and are responsible for a disproportionate share of GHG emissions.

We are a community-supported resource and we can’t do this work without you!

Read more in , , ,


  1. John Niles says:

    Here’s an observation, based on a calculation from the numbers in the bar chart above, about the difference in square meters per person in the two living environments being compared.High density: 43 square meters per person. (=374.1/77.7)Low density: 80 square meters per person. (=8637/107.3)Thus, downtown residents live in radically less space.

  2. Matt the Engineer says:

    [John] I think that’s the point, at least for the Building Operations piece of the chart. But then “radically less space” can still be quite a bit of space. 43 square meters per person works out to nearly a 2000 sf house for a family of 4.

  3. John Niles says:

    [Matt] I agree that’s the point … a push on families to live in a smaller residential footprint, when the trend in housing supply has been the other way. From”Average household size in the United States has dropped steadily from 3.67 members in 1940 to 2.62 in 2002. The average size of new houses increased from about 1,100 ft2 (100 m2) in the 1940s and 1950s to 2,340 ft2 (217 m2) in 2002. Factoring together the family size and house size statistics, we find that in 1950 houses were built with about 290 square feet (27 m2) per family member, whereas in 2003 houses provided 893 square feet (83 m2) per family member (NAHB 2003)—a factor of 3 increase.”—“Small is Beautiful: U.S. House Size, Resource Use, and the Environment” by Alex Wilson and Jessica Boehland from the Journal for Industrial Ecology, Winter/Spring 2005 issue, published July 13, 2005.Maybe somebody reading this thread has data on what’s been happening with the size trend for new housing construction since 2002. Also, the article I quote only covers single family housing: “Because single-family, detached houses account for 63% of total dwelling units in the United States (U.S. Census Bureau, 2001), this study focuses solely on single-family houses. A broader study that examined single-family attached houses, multifamily buildings, and mobile homes would produce somewhat different and probably less dramatic results.”

  4. Matt Herrick says:

    Wish I had remembered this a little earlier, but San Francisco Mayor Gavin Newsom will be chatting online,, at 2 p.m. today, in part on the issue above.

  5. Rod Smelser says:

    How would a widespread coversion to hybrids and electrics alter the transportation picture?Did this study in any way examine prices of accommodation per square meter as between urban and suburban parts of Toronto?

  6. NewmanAntonia34 says:

    When you are in a not good position and have got no cash to get out from that, you would require to take the credit loans. Just because it should aid you emphatically. I get small business loan every time I need and feel fine because of that.

Leave a Comment

Please keep it civil and constructive. Our editors reserve the right to monitor inappropriate comments and personal attacks.


You may add a link with HTML: <a href="URL">text to display</a>