Friday, July 22, 2011

How we (under)estimated the GHG reductions due to "Carmageddon"

Last Thursday afternoon, CREC got an e-mail from a reporter from the Huffington Post, asking us if we could help calculate the GHG reductions due to "Carmageddon," the closure of a 9.9 mile stretch of the 405 freeway in Los Angeles.  As geeky as this is, it made my day: not only is doing a back-of-the-envelope carbon calculation always a fun exercise in whole-systems thinking, but this is also the type of question that, in a better (and probably cooler) world, we'd be asking ourselves frequently as we planned our cities.  Unfortunately, in my enthusiasm to help meet the Post's deadlines, I made a rookie mistake that led me to underestimate the GHG reductions by roughly tenfold.  For those curious about how I arrived at an estimate of a reduction on 1.6 million pounds CO2 due to the 405 closure, and why it should have been closer to 16 million pounds, you can learn more below.

In an ideal world, we'd know how long the average weekend trip that traveled the closed stretch of the 405 was, and we'd have some guesses about how people would react to the closure: what portion of them would stay home, what portion would make a trip by transit, what portion would drive an alternate route, etc.  But either Caltrans and the other transportation agencies hadn't done their homework, or they weren't willing to share.  So instead I looked for an example of a similar situation--a weekend closure of an important stretch of freeway in a large metro area--and found one close to home: the closure of the Bay Bridge over Labor Day weekend of 2009.  It's not a perfect analogy, because the Bay Area has different land use patterns, higher transit ridership and better transit service along the closed corridor, and because a lot of people leave town on a holiday weekend anyway.  Nonetheless, as my high school physics teacher used to say, it's "good enough for government work," especially since the governments involved weren't sharing their work.

I remembered reading in a report by Lauren Michele of Policy in Motion that when the Bay Bridge, which represents 1.2% of the freeway lane miles in the nine-county Bay Area, closed, vehicle miles of travel (VMT) dropped by 3.7%.  So this means that for every one percent decrease in lane miles, short-term VMT decreases by roughly three percent (this 3:1 ratio between the percent of change in two variables is called an elasticity).  Based on data from the Texas Transportation Institute, I estimated that the closed stretch of 405 represents about 0.1% of the total "freeway equivalent" lane miles in Greater LA--and that's where I made my rookie mistake.  I arrived at this figure by dividing the 9.9 miles of closed freeway by the total lane miles in the LA metro area (roughly 9400), but most of the closed stretch 405 has 10 lanes, so it's actually around 99 lane miles that were closed, which is around 1.05% of the total lane.  This error then propagated its way through the rest of my work.  Using the elasticity from the Bay Bridge closure, I originally estimated a 0.32% reduction in VMT, but it actually should have been 10 times that, 3.2%... and so on.

Based on a Brookings Institute report on annual per capita VMT in major metros, I estimated the average weekend VMT in LA at around 520 million, and 3.2 percent of that is almost 17 million.  If you divide this by the average mileage of the U.S. passenger vehicle fleet (20.4) and multiply by the amount of CO2 in a gallon of gasoline, you get a 16 million lb. reduction in CO2.  This doesn't account for the extra CO2 generated by the additional bus service that LA Metro ran over the weekend or JetBlue's bargain-basement flights from Burbank to Long Beach, and as my colleague Bill Eisenstein discussed in the article, all sorts of other caveats apply to this analysis.  But the freeway closure does seem to have had some effect on driving, at least according to LA County Supervisor Zev Yaroslavsky, who doesn't say where he got his numbers from. 

The biggest challenge with an analysis like this is that climate change is a systematic, global, and long-term issue, but a one-time event like Carmageddon only produces short-term changes in behavior, so it's difficult to come to any meaningful conclusions about whether it combats or contributes to climate change.  That's why CREC focuses on sustainable planning and community design: because urban environments last a long time.  Changing the way that we plan our cities has the potential to yield serious long-term GHG reductions.  By all accounts, Carmageddon seems to have been almost a non-event; people stayed home with their families, enjoyed their neighborhoods, watched the new Harry Potter movie.  But if, during future decisions about how to spend transportation funds, LA residents think back on the past weekend and remember that not driving was actually kind of fun, we may see some long-term benefits to Carmageddon.