« big news on u.s. federal transit funding | Main | san francisco: "the fuse has been lit" »

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d83454714d69e2012876dd3348970c

Listed below are links to weblogs that reference a visit from the loyal opposition:

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

J

I think that using "hard edge" as a counter is a very bad example.

Here is a hard edge in California. The city cannot grow north of the river (protected land).

http://maps.google.com/maps?f=q&source=s_q&hl=en&geocode=&q=93720&sll=37.0625,-95.677068&sspn=36.452734,79.013672&ie=UTF8&hq=&hnear=Fresno,+California+93720&ll=36.848067,-119.893284&spn=0.017995,0.038581&t=k&z=15

But it's the very definition of sprawl. There are sidewalks, bike lanes and crosswalks.....but where is the supermarket?

Watson

Jarrett,

I agree that the typical suburb in Europe is significantly different from the typical suburb in America. Theirs tend to be more compact and have better transit. But I think that's more a matter of necessity than choice. They have lower incomes than we do, much less land, and more historical constraints on where they can build. But as in America, European suburbs have changed over time. I think that if you look at Europe's newer suburbs, the ones that have developed mainly within the last 10 or 20 years at the outer edges of their metropolitan areas, you'll find that they increasingly resemble American suburbs and exurbs.

And I'm really not trying to argue that suburban, car-oriented lifestyles are objectively "better" than urban, transit-oriented ones. Just that a suburban lifestyle is what most people, all things considered, seem to prefer. And that suburbanization and sprawl are not an American aberration produced by particular zoning laws, parking subsidies, or whatever, but rather a global phenomenon primarily caused by mass affordability of private cars.

Jarrett

Watson. A useful distinction, that. I don't think many of us would argue, or even care, about what caused the current patterns of settlement. We are more interested in the forces that sustain those patterns in the face of the growing desire and need for something else.

Because (a) regardless of what people prefer, it's not sustainable for growing populations to live at such low densities with such high car use and (b) we want people to have choices. Some American cities are still basically nothing but suburbs or slums, so many Americans have never experienced the benefits of urban life to the point that they could even make an informed choice. Europeans are more likely to have choices, so those who live in the suburbs are only those who really want to.

Danny

Its stuff like this that makes me, a libertarian urbanist, not want to bash my head into walls. Sometimes a little support for a cause can be had by just a little more effort to understand perspectives outside of our own.

rhywun
Just that a suburban lifestyle is what most people, all things considered, seem to prefer.

This is a crucial point. But as somewhat of a libertarian myself I appeal to arguments over "choice" too. I suspect that the current arrangements don't reflect the choices Americans would make if they were more free to make their decisions. I like to posit -- maybe as a "thought experiment" -- that all forms of transportation should be unsubsized, both roads and transit, and see what happens.

Winston

J,

It appears that there are markets about .6 of walking distance both north and south of your pin. Given the presence of sidewalks for the whole route and lightly traveled streets for most of it, I'd say that is a pretty walkable suburban neighborhood. Fresno has many flaws but it is not a poster child for unwalkable sprawl in the way that many smaller Eastern and midwestern cities arec.

Watson

Well, we discussed the demand issue before. I see no credible evidence of a growing desire for something else, just anecdotes and wishful thinking. Population growth may cause an increase in the absolute amount of transit use and the absolute amount of high-density housing in America. But as a share of our total transportation system and total housing stock I think they will continue to decline or remain flat.

We haven't really discussed "sustainability." Here's my take: In terms of energy consumption and pollution, the most reputable data and analysis I have seen suggests that transit overall is at best only modestly cleaner and more energy-efficient per passenger-mile of travel than passenger cars. Rail transit is usually cleaner than cars, at least if you only look at operating energy use, but buses are often not. And car technology is not standing still. In fact, we're at the start of a revolution in automobile fuel, propulsion and automation technology. We already have some regular hybrids, and all of the major automakers are planning to introduce plug-in hybrid or electric cars in the near future that are much cleaner and more fuel-efficient than current vehicles. Yes, initially they'll be very expensive and have various limitations. But land-use and transportation planning involve timescales of decades. It does not seem plausible to me that transit will be environmentally competitive against the much cleaner, much more efficient cars we're likely to be driving 20 or 30 years from now.

The other piece of the "sustainability" argument is density. Higher density means less passenger-miles of motorized travel, which means less energy consumption and pollution. I don't dispute that, but again the effect seems to be small. A recent study from the National Academy of Sciences concluded that it would take a huge, unprecedented increase in average density to produce just a small benefit in reduced pollution after a period of decades. I think it's unlikely that most people will consider that small benefit to be worth the costs of a high-density lifestyle that I described in previous comments.

Jarrett


Well, as I said, Watson, it just depends on what's "home" for you, just as it does for me. And when you say you "see no credible evidence" of desire for more urban lifestyles, it just depends on your frame of reference. Readers of this blog who live in places like Portland's Pearl District or Oakland's Fruitvale or similar revitalizing urban places are going to look out the window and not see what you see.

I touched on the sustainability question a bit here:

http://www.humantransit.org/2009/12/yet-another-transit-isnt-green-because-of-empty-buses-story.html?cid=6a00d83454714d69e201287656b6c2970c

The key point is that transit would be much more effective as a sustainability tool if it were operated with that objective, but it's expected to do several competing things.

EngineerScotty

While I'm no Libertarian, some of Watson's comments are I believe correct; in particular, the use of zoning has in many cases exacerbated problems rather then improved things.

Keeping mutually-incompatible land uses apart is a good thing. Here I'm talking about agricultural vs various levels of industry vs ordinary residential and commercial uses. But there really isn't any good reason to keep homes apart from low-impact retail and office uses. The "exclusive residential" zoning has done untold damage to cities, in my opinion. (I should note I'm speaking at neighborhood levels of scale--limiting particular lots within a neighborhood to one use or the other isn't unreasonable).

My second gripe is with designations such as R2 and such--intended to keep very low density neighborhoods from densifying (and often intended to protect wealthy neighborhoods from having poor people move in next door). OTOH, if such restrictions were removed from the zoning plan, they'd probably be added as neighborhood covenants instead.

Alon Levy

Watson, what new exurbs do you speak of? I'm asking because in e.g. Paris, the fastest growth is not in the exurbs, but in the inner-ring suburbs, whose density is high.

I'm also not sure what reputable studies you speak of that say cars and transit are a wash. US subways average the emissions equivalent of about 80 passenger-mpg (link). Buses are less efficient than trains, but overall they're improving relative to cars, because they're adopting hybrid technology faster. Over 20% of new bus orders in the US are hybrid, compared with about 4% of new car sales. Trains are improving as well: first, there are innovations such as regenerative braking; second, they can be powered by renewable resources, which are gaining ground. Since the US will have to convert its grid to zero-carbon anyway, in any sustainable future electrified transit is zero-carbon and zero-pollution.

Finally, what lower incomes are you talking about? The richest cities in Europe - London, Paris, Munich, Amsterdam, Frankfurt - are about as rich as the richer US cities, and are generally the most transit-oriented cities in Europe. The high-income people of London and Paris are much more likely to ride trains than the low-income people of the provinces. This is true in the US as well: within each metro area the poor ride transit more than the rich, but the three richest metro areas - NY, the Bay Area, and DC - also have the highest transit mode shares. People are likeliest to drive in areas too poor for local transit infrastructure.

rhywun
I see no credible evidence of a growing desire for something else, just anecdotes and wishful thinking.

I see it in the fact that city-lovers flee to NYC or SF at the first opportunity, because their hometowns are dying. Yeah, it's anecdotal, but I am old enough to remember at least some signs of life in the towns I grew up in (Rochester and Buffalo), which have largely disappeared. The fact that "urban" cities are booming beyond all expectations tells me that there is a demand for this "lifestyle" that is not being fulfilled in the rest of America.

Michael D

Consider two ways of explaining the proliferation of suburbs in North America.

-People prefer to live in suburbs, and zoning policies and various parking and road subsidies have minimal impact on this.

-People have some preference to live in suburbs, but other factors play a significant role: suburbs are cheaper in part due to subsidies for expensive infrastructure and limits on density, there is a dearth of affordable urban housing (including densely-spaced houses, not just condos), parking is forced to be provided everywhere and so its costs are unseen, and there is little good quality frequent transit that doesn't just go from suburb to CBD. Another one is that agricultural land is undervalued due to the post-war rise in agribusiness that has pushed down wages farmers can earn (often to negative).

All of those latter factors are ubiquitous in North America. In cities where those factors are decreased, I bet there are substantially more people living in the urban area.

anonymouse

One non-anectodal example: the recent Manhattan baby boom is a fairly well documented example. Some time around the early to mid 2000's, the number of children born to families living in Manhattan increased sharply, catching some people by surprise and leading to crowded daycares, schools, and so on, because it was a reversal of the trend of urban families having relatively few children, or moving to the suburbs to have children, a trend that is still the case in, for example, San Francisco. And while this might seem like a fairly marginal trend in the big picture, remember that in economics, what happens at the margin often matters a whole lot.

anonymouse

Also: it's pretty much impossible to say very much empirically about what the US would look like if not for restrictive zoning and the various suburban subsidies, because there just aren't very many examples of places without zoning. Pseudo-libertarians love to use the example of Houston, but while Houston doesn't have land-use-separation zoning, it has minimum lot sizes, minimum setbacks, minimum parking requirements, minimum street widths, and massive investments in highway infrastructure. And most land is subject to covenants which restrict land use, providing de-facto zoning, especially because the city actually enforces those covenants on behalf of private parties.

In the realm of conjecture, I'd say that suburbanization made some sense in the 1950s, when a full 50% of households had children thanks to the baby boom. Now that this number has reduced considerably, there is a relative overabundance of suburban sprawl, and a relative shortage of urban and inner-suburban housing. This is going to get worse as those same baby boomers get older and less able to drive everywhere.

Alon Levy
And most land is subject to covenants which restrict land use...

...which is encouraged by the city. Houston's standards for streets have a minimum width of 60', reduced to 50' if all properties on the street are deed-restricted to single-family.

Watson

Watson, what new exurbs do you speak of? I'm asking because in e.g. Paris, the fastest growth is not in the exurbs, but in the inner-ring suburbs, whose density is high.

Here's a summary for Europe as a whole, from Slate's review of Robert Bruegmann's "Sprawl: A Compact History":

Most American tourists spend their time visiting historic city centers, so they may be unaware that suburbs now constitute the bulk of European metropolitan areas, just as they do in America. We marvel at the efficiency of European mass transit, but since 1950, transit ridership has remained flat, while the use of private automobiles has skyrocketed. Just as in America. "As cities across Europe have become more affluent in the last decades of the twentieth century," Bruegmann writes, "they have witnessed a continuing decline in population densities in the historic core, a quickening of the pace of suburban and exurban development, a sharp rise in automobile ownership and use, and the proliferation of subdivisions of single-family houses and suburban shopping centers." Despite some of the most stringent anti-sprawl regulations in the world and high gas prices, the population of the City of Paris has declined by almost a third since 1921, while its suburbs have grown. Over the last 15 years, the city of Milan has lost about 600,000 people to its metropolitan fringes, while Barcelona, considered by many a model compact city, has developed extensive suburbs and has experienced the largest population loss of any European city in the last 25 years. Greater London, too, continues to sprawl, resulting in a population density of 12,000 persons per square mile, about half that of New York City.

Since you mention Paris specifically, here's Bruegmann's own description of the postwar population shift in the Paris area:

during much of the mid-20th century, as Paris was losing population, the ring of suburbs immediately around it was gaining. However, since the early 1970s even these inner suburbs have started to decline in population and density as the outer suburbs and exurbs have boomed. Between 1962 and 1990, as the city of Paris slipped steadily in population from 2.79 million to 2.15 million, the inner suburbs first gained in population, overtaking the population of the city and reaching just over 3 million by 1975, but then declined again, slipping back to 2.94 million by 1990. During the same period, the outer suburbs witnessed an accelerating growth rising from 1.66 million to 2.62 million. Beyond that an "exterior zone," including the rest of the large Parisian region, the Ile-de-France, with its comprehensively planned new towns, exurban developments and still-rural areas, grew from 1.2 million to 2.9 million. By 1999 the Ile-de-France had nearly 10 million people, meaning the city of Paris accounted for fewer than a quarter of all Parisians. ... the outer Parisian suburbs and exurbs, with their low-density subdivisions of single-family houses, shopping centers, industrial parks, and freeways, function and look increasingly like those of the United States.
anonymouse

And one more point. There's a difference between car use and car dependence. Europeans, and many Americans, use cars because they are convenient. But in many parts of America, people use cars because they have literally no alternative, or no alternative that is even remotely as usable. I've been to Paris, including the outer suburbs (well, one outer suburb at near the west end of the RER B line), and while auto ownership seemed reasonably high there, it was also perfectly possible to get around by foot, on bike, or by bus and train. That just isn't the case in most of the US.

Alon Levy

How does Bruegmann define inner and outer suburbs? The conventional definition in Paris is that the inner suburbs are the three departments contiguous to Paris, whose population densities range from 5,000 to 8,000 per km^2, and whose total population today is 4.4 million; the outer suburbs are the departments not contiguous to Paris, whose densities range from 200 to 900 km^2, and whose total population is 5 million. Those inner suburbs have averaged 1% annual population growth in recent years, faster than both Paris's 0.4% and the outer suburbs' 0.8%. Even then, in the outer suburbs most territory is not urbanized, so urban density is far higher than 200-900/km^2.

In addition, the definition of "look like exurbs in the US" is suspect. Paris has an extensive commuter rail system, whose ridership makes any US commuter rail look puny. Overall, Greater Paris's total annual rail ridership is a little more than 200 times total population; the highest comparable number in the US is for New York, where the same ratio is about 90. Unlike US cities, Paris has invested in suburb-to-suburb transit, and is planning to invest even more; this is why even the edge city that's formed around La Defense has a high transit mode share, unlike the edge cities in the US.

Watson

I'm also not sure what reputable studies you speak of that say cars and transit are a wash.

Here is the latest edition of the Transportation Energy Data Book. Table 2.12 on page 2-14 lists the operating energy intensities of different modes of passenger transportation. These numbers indicate that rail transit is indeed significantly more efficient than passenger cars, but that transit buses are significantly less efficient. And most transit is buses, not rail.

It's true that transit agencies have been moving to hybrid buses, but that doesn't seem to have been happening widely enough to close the efficiency gap with cars. Given the state of transit agency finances, a switch to an all-hybrid national fleet in the near future doesn't seem very likely. In the long term, we can expect the energy efficiency of cars to increase dramatically. Even first-generation plug-in hybrid and electric vehicles are expected to be at least three or four times as energy efficient as conventional vehicles of equivalent size and power. Perhaps transit buses will be able to match this increase in efficiency with cars, but I see no serious prospect for rail transit to keep up. Rail transit already benefits from the efficiency advantages of electric propulsion.

I should add that a better analysis would compare not just operating energy use, but life-cycle energy use (including energy used for infrastructure and vehicle construction, maintenance, etc.). Under that analysis, rail's advantage on operating energy may disappear. This recent study from UC Berkeley found that the ratio of life-cycle energy use and emissions to operating energy use and emissions is about 1.4 for automobiles, but about 2.6 for light rail and 2.1 for heavy rail.

Perhaps transit really is significantly cleaner and more efficient than cars, but reputable sources seem to indicate otherwise. Even if cars are less efficient, consumers seem to be willing to accept that cost in return for the benefits of car travel over transit, such as speed, comfort, convenience and coverage.

Jarrett

Watson

The Transport Energy Data Book to which you refer has a Table 2.12, showing that urban buses have a higher btu/passenger mile than cars (4315 vs 3514). But they're using an average load factor for buses of 9.1, and that, in turn, reflects the fallacy that I discussed here:

http://www.humantransit.org/2009/12/yet-another-transit-isnt-green-because-of-empty-buses-story.html

Re sprawl, Bruegmann and I are always going to pass in the night because he's talking about very aggregated density figures (as is Wendell Cox when he discusses county-level population changes in an earlier link). Bruegmann even describes Versailles as sprawl. I, on the other hand, am forming my impression by poking around the edges of European cities on Google Earth looking at the actual urban texture, including apparent density and relationship to transport infrastructure, and comparing these things to typical American forms.

What you call "sprawl" in Europe often looks like what I'd call good New Urbanist design. At its lowest density extreme, e.g. Stadtwald pictured above, it's car dependent, but there are surprisingly few places like that, and surprisingly many places, like the Valencia image I used above, where very dense urban fabric abuts farmland directly. Ultimately, European cities haven't done anything comparable to Texas-style sprawl because they never had the illusion of limitless land that defines so much of 20c American culture.

Anyway, readers should explore this for themselves and form their own impressions.

Alon Levy
Here is the latest edition of the Transportation Energy Data Book. Table 2.12 on page 2-14 lists the operating energy intensities of different modes of passenger transportation.

The BTU analysis gives a different picture from a straight emissions analysis. The Bureau of Transportation Statistics gives the average emissions of a bus as 3.8 mpg (link). At 9 passengers per bus, it works out to an emissions equivalent economy of 29 passenger-mpg, about 50% more than for the average single-passenger car. This accords with FTA statistics saying buses are 50% more emissions-efficient than a single-passenger car.

(Yes, cars on average have more than one passenger. For commute trips, the average is 1.2-1.3 passengers. For other trips they're not competing with transit buses.)

It's true that transit agencies have been moving to hybrid buses, but that doesn't seem to have been happening widely enough to close the efficiency gap with cars. Given the state of transit agency finances, a switch to an all-hybrid national fleet in the near future doesn't seem very likely.

Well, the switch is already happening. In New York the fleet is already 19% hybrid.

Perhaps transit buses will be able to match this increase in efficiency with cars, but I see no serious prospect for rail transit to keep up. Rail transit already benefits from the efficiency advantages of electric propulsion.

Regenerative braking is a feature of new trains, not old train. And as renewable energy gains ground, the advantage of electric traction will increase. For example, the C-Train in Calgary is 100% wind-powered, which means its emissions are zero.

This recent study from UC Berkeley found...

...that for hydro-powered transit such as that in the Bay Area, operating emissions are so low that most emissions are generated by construction. See page 14 of the FTA link above. So BART turns out to be not 10 times more efficient than cars, but only about 3 times. This doesn't scale to less efficient transit because powering trains by coal rather than hydro power isn't going to make construction emissions increase.

Even if cars are less efficient, consumers seem to be willing to accept that cost in return for the benefits of car travel over transit, such as speed, comfort, convenience and coverage.

The consumers aren't paying that cost - they're externalizing it. The people who are paying live in asthma-ridden US ghettos and in flood-prone lowlands in South Asia.

anonymouse

Jarrett makes a good point about aggregating. If you have a city of a couple dozen square miles taken up entirely with single family housing, that's very different from having the same area, with the same population, and the same number of single family houses, but half the land in the municipal boundary is left as forests and fields and parks. In much the same way that I'm sure many cities have enough green space in roadway medians to match NYC's Central Park, but the two are obviously in no way comparable in their usefulness as public space.

Watson

Jarrett,

On the energy use/emissions issue, we seem to be talking past each other. I have been responding to the statement you made in your first comment that "regardless of what people prefer, it's not sustainable for growing populations to live at such low densities with such high car use." For the reasons I have explained (my discussion of energy efficiency and pollution), I don't think this claim is supported by the evidence.

The argument you make in the post you link to in your most recent comment, if I understand it correctly, is different. In that post, you seem to be saying that since transit serves other important goals besides efficiency, such as improving mobility for poor and disabled people, it is inappropriate to judge transit against cars on efficiency or "sustainability" grounds alone. I agree with that proposition. But it's different from the other proposition, which I do not agree with, that low-density, car-based lifestyles are not "sustainable."

On the issue of sprawl and transportation in the U.S. vs. Europe, it seems to me that you're focusing on minor differences and ignoring the fundamental similarities. On both continents, cars were once rare luxuries but have now become the overwhelmingly dominant form of transportation. On both continents, transit has declined dramatically to become just a small component of the transportation system. In the U.S. transit seems to have bottomed out at about 1-2% of passenger-miles of travel. In Europe, transit is a slightly larger part of the transportation system, but it's still shrinking. On both continents, the population distribution has undergone massive sprawl. Yes, there are differences, but the fundamental patterns and trends are the same.

Watson

Alon,

I can't make sense of your discussion of bus emissions. The chart on page 10 of your link labeled "FTA statistics" suggests that, on average, autos and transit buses produce about the same amount of CO2 emissions per passenger-mile. Buses are better than autos with a single occupant, but the average occupancy of cars is about 1.6.

Regarding rail transit, again, it doesn't make sense to compare energy consumption and emissions for the average car with the corresponding numbers for BART. BART is one of the most efficient rail transit systems in the country. The average rail transit system is much less efficient and much dirtier than BART (see the chart on page 11 of your "FTA statistics" link). Any new rail transit systems we build in the future are more likely to reflect the averages for existing rail transit systems than the most efficient existing rail transit system.

In any case, what matters to "sustainability" and long-term transportation policy is not the efficiency of today's average car, but the efficiency of the average car two or three decades from now. And there is every indication that, if we want it or need it to be, the average car of the future will be much more efficient than today's models.

Jarrett

Watson. To the extent that current sprawl development patterns in North America appear sustainable, it's because so FEW people in the world, relatively speaking, want those patterns and can afford them. Even if you believe that we'll get electric cars in time to adapt to the decline of petroleum sources, the real issue is the impact on the rest of the world.

America has been transmitting its idea of the ideal lifestyle to the whole world for decades now. Everyone who has seen a television has been fed that idea of affluence. (Notice, for example, what a high percentage of music videos feature cars.) The real issue for the planet is what happens when most people in China and India can afford those patterns. See, for example, this Jared Diamond essay from the NYT a couple of years ago.

http://www.nytimes.com/2008/01/02/opinion/02diamond.html?_r=4&pagewanted=2

These are large and well-worn debates. I appreciate your perspective, but for the reasons I outlined in this post we are unlikely to convince each other.

Watson

Jarrett,

I don't really have the expectation of convincing anyone in the pro-transit camp. I'm just laying out the reasons why I think the "sustainability" argument for a shift to transit doesn't work and isn't likely to be persuasive to people in general, including people in other countries. You mentioned China and India. It's worth noting that last year China became the world's largest market for new cars, pushing the U.S. off the #1 spot that it has occupied for decades. India, Brazil and other large developing nations are also rapidly embracing the car. Over the next few decades, hundreds of millions of people in these countries are likely to become rich enough to afford to buy their first car. The Indian carmaker Tata last year introduced the Tata Nano, a small, inexpensive car that may become to India what the Model T Ford was to America. Whether or not you think Chinese and Indian people should embrace the automobile as Americans and Europeans have, it seems pretty clear that very large numbers of them are likely to do so. It also seems pretty likely that cars will continue to dominate the transportation systems of America and Europe for the foreseeable future. So it really is in our own interests to promote new automobile technologies that will make future cars cleaner and more efficient than today's vehicles.

I think Jared Diamond makes this same point in his NY Times piece you link to. The car-loving people of the developed world will not accept what they consider to be a big drop in their standard of living. Poor people in the developing world will not stop pursuing what they consider to be a better life for themselves and their families. So we have to find ways of allowing people to live a car-oriented lifestyle that is more friendly to the environment.

EngineerScotty

One other relevant difference between the developed world, and developing nation, is that in the US, the largest expense for transit agencies is labor. Elsewhere, rolling stock and fuel probably dominate the price, but in the US, drivers are the limiting factor on how much service can be provided.

Placemaking Institute

Awhile ago, I came across this article http://americancity.org/magazine/article/the-french-revolution/ which right now may be a worthy addition to the commentary:

“(Strasbourg, France’s) downtown is filled with department stores, teenagers of any ethnicity sporting a European style that takes a lot of inspiration from their American counterparts of five years ago, and shwarma shops competing with McDonald’s for their attention. But walk around Strasbourg’s charming medieval city center and you will see that one thing is virtually unchanged from its medieval origins: the absence of automobiles. This is not, however, an uninterrupted history. In fact, it is the direct result of actions recently taken by Strasbourg’s government — ones that should inspire comparably sized older American cities, from Buffalo to St. Louis. Just like most American cities, the car’s midcentury domination had largely forced public transportation out of Strasbourg. The once-extensive tram lines fell into disrepair, and the last one was taken out of service in 1960. But by 1989 traffic and parking had become major headaches for residents and for businesses in the dense warren of downtown streets. Rather than see retail flee to suburban malls, as it did in America, the city decided to take action…Some might attribute this phenomenon entirely to a cultural difference, arguing that the French will take advantage of bike paths and trains but Americans will not. But the Strasbourgers I interviewed, whether politicians, pedestrians or businesspeople, all told me that the French, like Americans, have an emotional attachment to their automobiles, and that it is ultimately a political choice to encourage or discourage driving. Absent the incentive structure set up by Strasbourg, the French will take the path of least resistance — a car, whenever possible — just like Americans.”

John

Frankly, I'm less concerned about the environmental effects of sprawl and cars than I am with public health. Building active transportation like walking and biking into daily routines would reduce average obesity rates, heart disease, diabetes, etc... Even an all-electric national car fleet powered only by happy thoughts can't help you get exercise like a walkable place with mixed land uses.

J

Winston, the pin in the map I linked to is in a slightly older (1980s) area and is very walkable.

I however was trying to show this area, perhaps the link sent you to the wrong place? This area was built in 2003-2006. There is nothing in walking distance, but there are sidewalks, bike lanes and crosswalks.

http://maps.google.com/maps?f=q&source=s_q&hl=en&q=N+Carnegie+Ave+%26+W+Bluff+Ave,+Fresno,+California+93722&sll=36.848968,-119.891707&sspn=0.004499,0.009645&ie=UTF8&cd=1&geocode=FZxIMgIdeJXa-A&split=0&hq=&hnear=N+Carnegie+Ave+%26+W+Bluff+Ave,+Fresno,+California+93722&ll=36.848204,-119.889915&spn=0.008998,0.01929&t=k&z=16

Alon Levy

Watson, the use of BART is significant in two ways. First, BART's efficiency is really the Northern California grid's efficiency; since the US will need a zero-carbon grid soon anyway, BART represents the reality of what future transit will look like, i.e. what future cars will have to compete with.

And second, your point about construction emissions becomes moot. BART doesn't have lower construction emissions per passenger-mile than other transit systems. Its high construction-to-operations emissions ratio comes from low operations emissions. You can't extrapolate this ratio to higher-emissions systems. Again, running trains on coal rather than hydro power doesn't increase construction emissions.

The most relevant statistic from the FTA chart is that per-passenger-mile construction emissions are lower for BART than for a car. They're just lower by a smaller amount than the operating emissions.

On another note, India isn't like the US of 100 years ago in one way: the Indian federal government is promising a 50% match to any city building a subway, whereas beginning in the 1910s the US federal government promised the same match to highways.

Watson

Alon,

BART and other electric rail transit systems are more efficient than today's average car because they benefit from the greater efficiencies of electrical power and propulsion. New automobile technologies are starting the bring the efficiency benefits of electrification to cars. Even first-generation plug-in hybrids and electric vehicles are expected to be at least three times more efficient than conventional gasoline-powered vehicles of equivalent size and power. The Chevrolet Volt is expected to get the equivalent of more than 100 mpg in average use. The Nissan Leaf all-electric car is expected to get the equivalent of more than 300 mpg. Rail transit already benefits from the efficiencies of electrification. I see no new rail transit technologies on the horizon that could improve the efficiency of rail transit even remotely as much as new automobile technologies are likely to improve the efficiency of cars.

As for non-operational energy consumption and emissions (e.g., energy and emissions attributable to the construction and maintenance of vehicles, roads, tunnels, railtrack, stations, etc), the Berkeley study I cited earlier suggests that taking these non-operational costs into account wipes out much, or even all, of rail transit's operational efficiency advantage over cars today. The researchers concluded that taking into account non-operational costs increases total energy consumption and greenhouse gas emissions for heavy rail by as much as 2.1 times and for light rail by as much as 2.6 times, but for cars only by as much as 1.6 times.

The bottom line is that when total energy consumption and emissions are taken into account (not just operational energy and emissions) the average rail transit system seems to be barely competitive with the average car even today. It seems unlikely that rail transit will be competitive at all with the much cleaner, much more efficient cars we'll be driving thirty or forty years from now.

Alon Levy
I see no new rail transit technologies on the horizon...

...except regenerative braking and more aerodynamic and transmission-efficient equipment. (Cars have been improving transmission efficiency, too, but in the US the efficiency gains have mostly gone to higher vehicle weight, not higher fuel economy.)

But if what you're saying is that the US grid will always run on coal, then you shouldn't be arguing either sustainability or economics. You should be arguing how to accept hundreds of millions of immigrant refugees from low-lying or rainfall-dependent developing regions.

The Chevrolet Volt is expected to get the equivalent of more than 100 mpg in average use.

No, it's not. The figure comes from taking an average of zero emissions while running on electricity and 40-50 mpg in charge-sustaining mode, weighted for projected miles driven. See explanation on The Oil Drum.

The researchers concluded that taking into account non-operational costs increases total energy consumption and greenhouse gas emissions for heavy rail by as much as 2.1 times and for light rail by as much as 2.6 times, but for cars only by as much as 1.6 times.

No, they only concluded this for the lowest-emissions systems. You're repeating the same assertion without actually addressing the fact that Chester studies systems in California, which have unusually low operating emissions because they're powered predominantly by hydroelectric energy. The construction emissions are not so low, so on a relative basis they appear high.

You can't use ratios here. NorCal hydroelectricity means that electrified rail more efficient. It doesn't mean construction requires pouring a proportionately smaller amount of concrete.

Ben

I think Germany is a great case study on efficient human behaviours towards transportation.

Germany is one of the few places on earth where there are no speed limits. This means that (in theory) luxury/sport cars can become more than status symbols, as you can drive them to their limits legally. You can even purchase exotic supercars, with insane performance characteristics.

However, Germany has some of the most efficient transit systems in the world. Most cities have extensive urban and suburban rapid transit lines, frequent and reliable local bus and tram routes, and commuter and intercity road and rail services as well. Stations are sleek and modern, and vehicles are clean and comfortable.

People realize that while exotic sport cars are great for driving in the country, transit is the better way of getting around the cities. Transit is so fast, efficient, and affordable that it is the preferred choice for urban mobility. In contrast, many North American systems are dirty, slow, and unreliable. This has led to the mindset that transit is for "poor people", a social service for those who cannot afford a car. Even in some cities where it is faster to take transit over driving, people may still drive since the entire transit experience is almost degrading.

If we want people to park their cars at stations and take transit into core areas, we need to make it the clear choice. It needs to be affordable, it needs to offer a customer oriented experience, it needs to be comfortable and reliable, and it needs to be faster than driving. If you can provide a transit experience which is better than driving, then people will take it over driving.

Michael D

"BART and other electric rail transit systems are more efficient than today's average car because they benefit from the greater efficiencies of electrical power and propulsion."

Other reasons: Steel wheel on steel rail has 10-70 times less rolling resistance than do car tires on concrete (according to Wikipedia). The per-passenger weight of transit vehicles serving worthwhile transit purposes is much lower on average than the per-passenger weight of a car, however electric the latter is. And of course, those who don't drive walk more and make shorter trips on transit than they would by car, so there is efficiency in the trip not taken.

Ed O

Watson, one problem with your vision of the future is that people in their electric cars will still be sitting in traffic and whining about congestion; and cities will still be pondering the necessity of providing transit to improve travel times - exactly where we are right now. (Perhaps "sustainability" includes sustained dissatisfaction?) The cities seriously thinking and investing in non-car travel at the moment are those that are looking to get off this endless treadmill.

Alon Levy
The per-passenger weight of transit vehicles serving worthwhile transit purposes is much lower on average than the per-passenger weight of a car, however electric the latter is.

Um, no. You're underestimating how heavy trains are and overestimating how many people ride them. In New York City, the average subway car weighs about 36 metric tons and carries 28.5 people (the national average is 25 people per car and, I think, about equally heavy cars). That's 1,250 kg per person, which would be about a wash with a US-sized sedan at average occupancy and slightly better than an SUV. The advantage of subways over cars is the traction and the central power generation, not the weight.

Bear this depends on crowding, which is typically low in the US off-peak. The rush hour Lexington trains in New York average 130 passengers per car and run 31-ton cars, for an average of 240 kg of unladen weight per passenger, comparable to four people in a very small car. In East Asia, such high ridership levels last all day, increasing the efficiency of transit.

Watson

Alon,

except regenerative braking and more aerodynamic and transmission-efficient equipment.

No, not except that. The complete sentence I wrote, including the part you chopped off in your partial quote, is "I see no new rail transit technologies on the horizon that could improve the efficiency of rail transit even remotely as much as new automobile technologies are likely to improve the efficiency of cars." As I said, even first generation plug-in hybrid and fully-electric vehicles are expected to be at least three or four times as energy efficient as comparable conventional vehicles. Improvements in train braking, aerodynamics and transmission efficiency could not possibly deliver efficiency improvements of that magnitude.

The figure comes from taking an average of zero emissions while running on electricity and 40-50 mpg in charge-sustaining mode, weighted for projected miles driven. See explanation on The Oil Drum.

No, the figure comes from the expected average mix of driving in charge-sustaining and charge-depleting mode, using the EPA's methodology for converting electrical power into an mpg-equivalent. As your own citation indicates, the Volt is expected to get 50 mpg even if it is operated exclusively in charge-sustaining mode (i.e., if it is never plugged in and all its energy comes from gasoline). That is twice the efficiency of a conventional sedan of equivalent size and power. In reality, no one is going to buy a Volt and never plug it in. There would be no point in paying the huge price premium for the plug-in technology if you're not going to use it. In average use, the Volt is expected to get the equivalent of over 100 mpg, which is four times the efficiency of a comparable conventional vehicle.

Watson

Alon,

No, they only concluded this for the lowest-emissions systems. You're repeating the same assertion without actually addressing the fact that Chester studies systems in California, which have unusually low operating emissions because they're powered predominantly by hydroelectric energy. The construction emissions are not so low, so on a relative basis they appear high.
First, your argument here simply does not apply to energy efficiency. A kWh of electricity is the same amount of energy whether it's generated by a hydroelectric plant or a coal-fired plant. With respect to CO2 emissions, the effect you describe may exist, but it seems to be small. As reported in your own "FTA statistics" citation, the ratio of operating emissions to total emissions for BART is very similar to the ratio for Caltrain, even though BART runs on relatively clean electricity whereas Caltrain runs on diesel. Other factors seem to have a far larger effect on that ratio than the technology used to produce electricity. So the national average for rail transit is probably close to the California average for rail transit. In any case, this is a quibble. Even if rail transit today is on average significantly more "carbon emissions efficient" than cars, it is unlikely to maintain that advantage for very long given the burgeoning advances in auto technology.
Alon Levy

As reported in your own "FTA statistics" citation, the ratio of operating emissions to total emissions for BART is very similar to the ratio for Caltrain, even though BART runs on relatively clean electricity whereas Caltrain runs on diesel.

Caltrain is a separate type of service - a predominantly rush hour-only train, for one. This increases load factors, and reduces operating emissions.

Pay attention to the fact that like the other rail transit services described in the study, Caltrain produces less construction emissions per passenger-mile than cars. So no, there's no ratio - just a lot of fuzzy math.

On the other hand, I could argue it the other way. Electric cars reduce operating emissions, but not construction emissions. So the advantage of transit over cars is still there.

Improvements in train braking, aerodynamics and transmission efficiency could not possibly deliver efficiency improvements of that magnitude.

Noncompliant equipment for commuter rail can halve fuel consumption. Caltrain equipment is already half as efficient as it can be because of FRA regulations, which Caltrain is about to get an exemption from. Caltrain also plans to electrify, further reducing emissions.

So on the one hand, you have relatively efficient Caltrain, and on the other hand, Caltrain's efficiency is about to more than double.

Watson

Alon,

Caltrain is a separate type of service - a predominantly rush hour-only train, for one. This increases load factors, and reduces operating emissions.

The point is that other differences between rail transit systems (including, yes, load factor differences) swamp the effect of the technology used to generate electricity on differences in the ratio of total emissions to operating emissions. And as I said, the technology used to generate electricity has no effect at all on the ratio for energy consumption.

Pay attention to the fact that like the other rail transit services described in the study, Caltrain produces less construction emissions per passenger-mile than cars. So no, there's no ratio - just a lot of fuzzy math.

Yes, there is a ratio. As reported in your "FTA statistics" citation, if you look only at operating emissions, the average sedan produces 3.1 times as much CO2 per passenger mile as Caltrain. But if you look at total emissions, the average sedan produces only 2.4 times as much as Caltrain. Thus, looking only at operating emissions significantly overstates the emissions benefits of Caltrain compared to the average sedan.

According to the charts on pages 11 and 14 of your citation, Caltrain is much cleaner in operating emissions than the average heavy rail transit system. Indeed, the numbers on page 11 suggest that many heavy rail transit systems, including the Washington Metro, Philadelphia SEPTA, Chicago "L", and Boston "T" emit about as much carbon per passenger-mile from operations as the average sedan. In fact, the Chicago "L" seems to be much dirtier than the average sedan.

Finally, if you really believe that there are technologies on the horizon that are likely to increase the efficiency of rail transit as much as PHEV and other electrification technologies are likely to increase the efficiency of cars, please show me your evidence for them.

Alon Levy

Watson, you've produced no evidence for widespread adoption of any emissions-saving technology in cars. The most widespread, hybrid-electric operation, is now being adopted more quickly in buses than in cars. More technologies the US is about to adopt making transit much more efficient include,

1. Bus to rail conversion.
2. Low-emissions grid.
3. A rewrite of FRA compliance regulations, allowing US commuter trains to be half as heavy as they currently are.
4. Commuter rail electrification.

All are currently in existence - in fact, the most sweeping, the FRA rewrite, is based on a positive train control mandate that's going to kick in by 2015. Low-emissions energy is gaining ground right now, and some electrification projects, e.g. for Caltrain, are going to be complete by 2020.

This is not the same as with plug-in hybrids, which are slated to cost $40,000 and are not getting cheaper. Battery technology isn't progressing very quickly. If $22,000 Priuses are too expensive to have a market share higher than 1-2% nearly ten years after they were launched, $40,000 Volts will be even more inaccessible.

Nor have you produced any evidence that we should think in terms of "ratios" and not "Caltrain and BART have lower construction emissions per passenger-mile than cars, but the difference is smaller than in operations."

Do you have any evidence that the Chicago L has higher construction emissions than BART? No. The studies you quote don't say anything about it. So the L isn't worse than a sedan; it's about 20% better than the average sedan at average commute occupancy.

CroMagnon

I'm curious to know whether the cost, energy consumed and emissions generated by the oodles of parking spaces to support sprawl were counted in the automobile statistics sited!

Watson

Alon,

Watson, you've produced no evidence for widespread adoption of any emissions-saving technology in cars.

Considering that I haven't claimed that emissions-saving technology has been widely adopted in cars, this isn't terribly surprising. Obviously, PHEV and pure-electric vehicles are still rare. Even regular hybrids and alternate-fuel vehicles are still comparatively rare. But that's because there is at present little economic incentive to adopt these new technologies. The rate at which they penetrate the market will strongly depend on future gasoline prices and government incentives. The point is that these technologies are capable of providing a huge increase in automobile efficiency - a 300-400% increase just for first-generation models. Again, if you really believe there are technologies on the horizon that could improve the efficiency of rail transit to a comparable degree, produce your evidence. And I do mean evidence, not merely a list of unsubstantiated, unquantified possibilities that may produce only small improvements in efficiency.

Do you have any evidence that the Chicago L has higher construction emissions than BART? No.

I don't know why you think I need any such evidence. According to page 11 of your own "FTA statistics" citation, the Chicago "L" produces more than six times as much CO2 per passenger-mile of operation as BART. According to page 14 of your citation, BART produces 64 grams of CO2 equivalent per passenger-mile of operation. This means that the Chicago "L" produces something like 380 grams per passenger-mile of operation. The corresponding figure for a sedan is only 230 grams. In fact, the numbers in your cited document suggest that the Chicago "L" produces as much CO2 per passenger-mile from operation alone as a sedan produces from both operational and non-operational sources combined.

The Washington Metro, which is the second largest heavy rail transit system in the U.S., is not as dirty as the Chicago "L", but according to your source is still much dirtier than BART, producing 3.8 times as much CO2 per passenger-mile of operation as BART. Again, that means the Washington Metro produces as much CO2 per passenger-mile of operation as the average sedan.

Watson

CroMagnon,

I'm curious to know whether the cost, energy consumed and emissions generated by the oodles of parking spaces to support sprawl were counted in the automobile statistics sited!

The Berkeley study I cited on the life-cycle energy consumption and emissions costs of different transportation modes includes parking in its estimates for automobiles. See pages 30 and 31 of the study for details.

Alon Levy
The point is that these technologies are capable of providing a huge increase in automobile efficiency - a 300-400% increase just for first-generation models.

It works for both cars and buses' operating emissions, but not for construction emissions. Unless you believe that the magical ratio is such that going electric will cut the emissions of highway construction by a factor of 4.

And together with other improvements in rail utilization, noncompliant equipment responsible to much lower emissions from mainline rail in rail-intensive countries such as Japan and Switzerland, on the order of 300-500 passenger-mpg.

According to page 14 of your citation, BART produces 64 grams of CO2 equivalent per passenger-mile of operation. This means that the Chicago "L" produces something like 380 grams per passenger-mile of operation.

No! You're assuming that the L emits 6 times as much CO2 as BART in construction just because it creates 6 times as much in operation. You must provide evidence it does. Harping about ratios without showing why they're a meaningful concept doesn't cut it.

calwatch

I think the example of Germany pointed to above is instructive, because when you start sizing highways for rush hour load, you're basically leaving plenty of capacity for 75-85% of the day. Transit can ramp up to peak loads easier because people can stand on buses and trains. In an ideal world, a "human-operated PRT" where strangers would get into random vehicles so that they would always be at capacity and "transfer" to other vehicles might be more efficient, but people are much less tolerant of the occasional stranger entering their own personal space than on a bus or train operated with professional supervision.

And the cost of highway construction in urban areas is only growing. For one lane on the 405 freeway, the cost is roughly $90 million a mile. Considering the delays caused during construction, and the increase in traffic that will saturate the lane within a few years of construction, that is not sustainable. Most rail systems operate at far below their "real" capacity. We need to stop building parking lots for Black Friday crowds and freeways for 5 p.m. traffic.

samussas

Sorry, I didn't read the whole conversation but for what I've seen I think Watson (and the others) just forgot something. And I will start by pointing that out with a quote from Watson.

"So we have to find ways of allowing people to live a car-oriented lifestyle that is more friendly to the environment."

But that's not possible, even with an electric-car-oriented lifestyle this life style will not be environment friendly. It might be slightly better than the fuel based one but not so much because you will still have to produces the cars, those endless 'burbs, the energy, the infrastructure... which is very environmentaly and economocally consumptive and destructive. Just look at the general state (of decay) of US infrastructure and tell me that a urban-sprawling, car-oriented society is sustainable. It's well documented that infrastructure (in general) is more efficient in places where density is high than in low density suburbs/cities/countries. It's also true with public commodities and ammenities.

But anyway, we are speaking about the environment. And a car oriented society in the form adopted in most of the western world (wether electric or fuel based) is and won't never be sustainable environmentaly speaking. It have too big an effect on our ressources and our environment.

And in the long term the only possible solution is to have a cultural change toward more dense and public transit/low impact means of transportation friendly cities...

I know this can sound a bit extremist to some but you can't speak about sustainability and only focus on the carbon emission produced during transportation. That's just a small very visible part of the problem.

I will finish on a fun fact found on US EPA website:
- What is the total amount of water used to manufacture a new car, including new tires?
39,090 gallons per car.


Placemaking Institute

Great point, samussas (on top of that each Prius manufactured consumes 90 barrels of oil); it takes a lot of water to make electricity, and a lot of electricity to make water. Each day our nation’s thermoelectric power plants (90 percent of all power plants in the United States) draw 136 billion gallons of water to cool the steam used to drive turbines, according to the Department of Energy. In recent years, the energy department says, plans for new power plants had to be scrapped because water-use permits could not be obtained. For their part, water- and wastewater utilities consume at least 13 percent of the electricity drawn nationwide each day and, according to a 2007 report from the U.S. Geological Survey, the United States uses 47% of all its water to produce energy.

In a sense, the hope of electric vehicle-ing our way out of this congestion mess is similiar to a heroin junky deciding to "clean up" by getting hooked on morphine instead - Because even if alternative fuels do become feasible for individual commuters to use, those who use those vehicles will still be stuck in traffic and complaining about congestion because we as a society can no longer afford to build enough roads. Simple as that. Culled from the Texas Transportation Institute: If a region’s vehicle-miles of travel were to increase by five percent per year, roadway lane-miles would need to increase by five percent each year to maintain the initial congestion level. It would be almost impossible to attempt to maintain a constant congestion level with road construction only. Over the past 2 decades, less than 50 percent of the needed mileage was actually added. This means that it would require at least twice the level of current-day road expansion funding to attempt this road construction strategy. An even larger problem would be to find suitable roads that can be widened, or areas where roads can be added, year after year.

CroMagnon

Re: Berkely Report,

The study cites 105 million parking spaces. I'm skeptical that this total includes all private parking spaces in the US. I can't believe there's only 1 parking space per every 3 people, when most people have a car and most people have either a garage, carport, carpad, or parking space on street in front of their house. Many people have multiple cars, many rusting away on some on an impervious constructed surface.

Also, though I haven't had time to read the all 126 pages of the Report, I do not believe emission per passenger mile is the appropriate metric. I believe the emissions per capita per unit time is more appropriate. PPM isn't a true comparison because suburban sprawl inherently requires more miles to be driven over a range of geography that is not transposable with the range of transit in most cases. It would be better to measure transit emissions against auto emissions with comparable operating areas, although may be impractical to assess. PPM should only be used when comparing identical service geography.

The report also goes out of its way to assign every indirect emission for rail, the buildings for the admin for rail's insurance and liability, but doesn't appear, as far as I can tell, to compute all the emissions generated by the insurance companies for private autos or the zillions of emissions generate by the innumerate number of car dealers constantly pedaling their products.

Watson

CroMagnon,

The parking spaces estimate comes from the International Parking Institute and the EPA. There are about 250 million registered motor vehicles in the United States. At any given time, many of these vehicles will be in motion rather than parked, and many others will be parked in places other than designated parking spaces. So I don't understand why you think the IPI/EPA estimate is implausible.

I also don't understand your argument for measuring emissions as "emissions per capita per unit time." Emissions per passenger-mile allows us to compare different modes for a given trip. Emissions differences attributable to differences in land use patterns is a different issue. Transit does not necessarily promote higher density. Commuter-oriented transit, for example (commuter rail, BRT, and subway/bus/light rail routes connecting suburbs to city centers) may promote sprawl by making it easier for workers to live in the suburbs and work downtown.

CroMagnon

Corrections, you're right that many cars are in motion and not parked. But most cars are parked most of the the time. Many cars are parked on the street in front of the house, so that would count as roadway infrastructure. But I postulate that the parking space total doesn't include private garages, carports and driveways.

I don't claim that transit promotes higher density development, but most transit exists where the development is denser because that development existed prior to mass auto use. But the typical trips aren't the same between transit and auto, because the auto has many trips where transit doesn't operate. The cost of automobiling would be more accurately compared to transit if only the autos that operated within the service zone of transit were factored. Certainly gas mileage for the auto average would decline significantly.

I understand the study is used to understand how things actually are. But doesn't mean the results can be used to determine the value of one mode against another. If airlines were far and away the most emission efficient (or whatever), that would mean, "let's make airlines our primary mode of travel for any trip." Because you're never going to be able to practically take an airplane uptown to downtown.

Watson

CroMagnon,

The parking estimate certainly includes private commercial garages. It is not clear whether it includes private household parking. But it is arguable whether such space should be counted as dedicated motor vehicle infrastructure for the purpose to estimating environmental effects anyway. If households that use space on their private property for parking did not want or need to do so, they might use the same amount of space for some other purpose, such as a larger yard or an extra bedroom.

I agree that the relative "value" of different modes of transportation is more than just a matter of their relative environmental costs. The point I am making here is that the claim that transit has lower environmental costs than automobiles per unit of transportation benefit is highly dubious.

Watson

Alon,

No! You're assuming that the L emits 6 times as much CO2 as BART in construction just because it creates 6 times as much in operation.

No, I'm not assuming that. You seem to have misunderstood the sentence of mine you quoted. The numbers in that sentence refer only to emissions attributable to operation. They have nothing to do with "construction." The emissions figures on pages 11 and 14 of your citation suggest that the Chicago "L" emits as much carbon from operation alone as the average sedan emits from operational and non-operational sources combined. Even if the Chicago "L" had zero emissions from non-operational sources (what you keep erroneously referring to as "construction"), it would still be as dirty or dirtier than the average sedan.

Alon Levy
The emissions figures on pages 11 and 14 of your citation suggest that the Chicago "L" emits as much carbon from operation alone as the average sedan emits from operational and non-operational sources combined.

No, they do not. They suggest that the Chicago L emits two thirds as much carbon from operation alone as the average single-occupant car.

Watson

Alon,

No, they do not.


Yes, they do. Either you're misreading the numbers or you're getting the math wrong.

(1) The chart on page 11 reports that the Chicago "L" emits 6.7 times as much CO2 per passenger-mile of operation as BART (0.60 for the "L" vs. 0.09 for BART).

(2) The chart on page 14 reports that BART emits 64 grams of CO2 equivalent per passenger-mile of operation. Applying the ratio from (1) above, this means the Chicago "L" emits 429 (64*6.7) grams of CO2 equivalent per passenger-mile of operation.

(3) The chart on page 14 also reports that the total emissions of the average sedan, from both operation and non-operational sources combined, is 380 grams of CO2 equivalent per passenger-mile.

So the Chicago "L" emits more CO2 per passenger-mile from operation alone (428 grams) than the average sedan emits from both operation and non-operational sources combined (380 grams). Even if the Chicago "L" had zero emissions from non-operational sources, it would still be dirtier than the average sedan.

JP

The Chevy Volt will supposedly use 0.25 kWh/mile. According to the pdf Alon linked, the Chicago L uses 0.132 kWh/seat mile. So if you drive a Volt with at least one other passenger (in Chicago), it will be cleaner than the L. If you drive it in San Francisco it's a lot cleaner than the L, but you need two other passengers to be as clean as BART.

Alon Levy

Watson, you're still assuming constant ratios in everything. Please, stop. The page-11 chart says the L emits 0.6 pounds of CO2 per passenger-mile, compared with 0.97 for an average single-occupancy sedan.

Yes, if you apply the same ratio as in page 11 to the page 14 data, which are drawn from a separate source, you get a different answer. This is sloppy - you're assuming that the Chester study's higher emissions figures for BART (0.14 pounds/passenger mile, versus 0.09 for the NTD data) would raise the figure for the L by an equal amount.

(P.S. Note also that the Chester study gives average-occupancy emissions numbers for cars, which are lower than the single-occupancy numbers given on the page 11 chart; it's important to distinguish them to avoid double-counting)

Watson

Alon,

Watson, you're still assuming constant ratios in everything.

I'm not "assuming" any ratio in the emissions comparison I just showed you. The only ratio involved is reported right there on page 11 of your citation: the Chicago "L" emits 6.7 times as much CO2 per passenger-mile of operation as BART.

Yes, if you apply the same ratio as in page 11 to the page 14 data, which are drawn from a separate source, you get a different answer.

There is no "different" answer. The ratio of operating emissions per passenger-mile for BART vs. the "L" is provided on page 11, and the quantities of operating and total emissions for BART vs. the average sedan are provided on page 14.

This is sloppy - you're assuming that the Chester study's higher emissions figures for BART (0.14 pounds/passenger mile, versus 0.09 for the NTD data) would raise the figure for the L by an equal amount.

No, I'm not assuming. Both those emissions numbers refer to operating emissions. The ratio of operating emissions for the "L" vs BART is 6.7. The quantities and ratio involved all come from your own citation. If those numbers are even remotely accurate, the Chicago "L" emits far, far more CO2 per passenger-mile than the average sedan. And the Washington Metro, Los Angeles Metro and Boston "T" all emit about the same amount of operating emissions (actually, somewhat more) per passenger-mile as the average sedan.

Alon Levy

Watson. The page-11 chart shows precise emissions numbers for each mode. The page-14 chart shows slightly different but equally precise numbers for each mode. They're not comparable.

Placemaking Institute

Now now, Alon, let's keep it civil for Jarrett by keeping in mind the definition of statistics: The science of producing unreliable facts from reliable figures...furthermore, the greenhouse gas argument is currently being construed by me as the so-called power structure maneuvering us into time-wasting battles with others when we could be involved in something that’s constructive and getting a real job done. Because much sooner rather than later we the people are going to have no choice but to break our auto addiction and develop more compactly.

Watson
The page-11 chart shows precise emissions numbers for each mode. The page-14 chart shows slightly different but equally precise numbers for each mode. They're not comparable.

Of course they're comparable. Both pages contain estimates of operating emissions. Page 11 also shows the ratio of operating emissions for the Chicago "L" vs BART. They don't have to match exactly. Even if the numbers are only roughly accurate, they still indicate that the Chicago "L" is much dirtier than the average sedan. There's no other reasonable interpretation.

Watson

Alon,

And together with other improvements in rail utilization, noncompliant equipment responsible to much lower emissions from mainline rail in rail-intensive countries such as Japan and Switzerland, on the order of 300-500 passenger-mpg.

I have asked you several times for evidence supporting your claim that the efficiency of rail transit could be improved to a degree comparable to the improvements in automobile efficiency provided by PHEV and electric technology. I'm still waiting for that evidence.

Alon Levy

Okay, I was out of line in my previous comment, now deleted. Let's try again:

The page 11 chart states emissions figures for various transit systems, coming from the NTD. The page 14 charts states slightly different figures for a subset of those transit systems, coming from an independent peer-reviewed study. You can't say "study #2 gives a 50% higher emissions figure for BART than study #1, so let's assume all emissions figures for transit are 50% higher than study #1 states." This is what you're doing with the L emissions.

As for evidence for the improvement in transit efficiency, I'm not sure what you're looking for. Do you want a link for the claim that non-FRA-compliant trains consume half as much energy as FRA-compliant ones? Or a link to claims about three-figure efficiency figures in Switzerland and Japan? Do you want a link to data about transit efficiency in areas where most electricity does not come from coal (well, other than BART)? I'm not exactly sure what you're looking for.

Watson

Alon,

I'm looking for links supporting the proposition that the efficiency of rail transit could be improved to a degree comparable to the increase in automobile efficiency provided by plug-in hybrid and all-electric vehicles. That is, an efficiency improvement of at least 300-400%. You keep mentioning possible ways of improving rail transit efficiency (regenerative braking, reduced vehicle weight requirements, etc.), but you haven't produced any documentation indicating how much improvement in efficiency these methods would produce if they were implemented, or even that implementing them is economically or politically feasible at all.

Watson
You can't say "study #2 gives a 50% higher emissions figure for BART than study #1, so let's assume all emissions figures for transit are 50% higher than study #1 states." This is what you're doing with the L emissions.

We don't need to assume they're exactly 50% higher. In fact, we don't need to assume they're higher at all, although they almost certainly are. The reason for the discrepancy in the BART figures is probably the different definitions of "operating emissions" used in the two studies. But even if we just use the numbers on page 11, and adjust the "Private auto (SOV)" figure for occupancy (1.6 passengers instead of 1) and mpg (25 mpg instead of 20), your citation still indicates that the Chicago "L" is dirtier than the average sedan in operating emissions.

CroMagnon

"The parking estimate certainly includes private commercial garages. It is not clear whether it includes private household parking. But it is arguable whether such space should be counted as dedicated motor vehicle infrastructure for the purpose to estimating environmental effects anyway. If households that use space on their private property for parking did not want or need to do so, they might use the same amount of space for some other purpose, such as a larger yard or an extra bedroom."

A study like the Berkeley one should definitely count these spaces. The opportunity cost reasoning you're using is weak b/c most homeowners use the garages at least some of the time for parking their car. Even when they don't, they're still using the driveway. One can't authoritatively argue from the point that if the garage were not constructed, the footprint or materials for the garage would have be used for another room. Some houses would simply be smaller and cost less. All told, this is a huge cost, similar to the cost of building a new auto. I don't know what the emissions of this would be, but I'd venture to say they're not insignificant.

I've seen the reference from the parking organization for 105 million spaces. I agree that that total is representative of the total number of off-street parking spaces that are at issue. That figure is simply unreasonable and implausibly low to me. I'm conducting online research to gather additional figures. FWIW, there was the Purdue University study, I think, that counted the number of non-residential parking spaces in Tippecanoe County, Indiana, and found 3 times as many spaces as registered vehicles.

CroMagnon

Strike the 2nd line in the last paragraph. It should be more like: I agree that that total is likely a conclusive count of some selected parking spaces, but not the total number of off-street parking spaces that are at issue.

Watson

CroMagnon, I'm not authoritatively arguing anything about the parking estimate. I'm just saying (1) it's not clear, to me at least, that household private parking is not included in the Berkeley estimate, (2) it's not clear that it should be included anyway, and (3) even if it isn't included but should be, it probably wouldn't have more than small impact on the total energy and emissions numbers for autos anyway. In other words, I think it's a quibble. One could probably also quibble over the numbers for rail and buses.

Alon Levy

Yes, it's a quibble. And in every quibble, you make sure to pick the highest emissions figures for transit and lowest figures for cars. The average economy for cars in the US is 20 mpg, but you choose 25 mpg, which excludes SUVs (why? Are they being banned?). The average load factor for commute trips, which is where transit competes with cars, is 1.2, but you pick 1.6, which includes intercity trips. There are two emissions figures for BART, but for some reason you extrapolate from them to get an emissions figure from the L that's pulled out of thin air.

Alon Levy

As for a 300-400% increase, do you have any documentation that electric cars are about to have widespread adoption? No. Even hybrids don't. So the increase isn't 300-400%. In fact, average fuel economy in the US has been stagnant for nearly 20 years.

In contrast, high energy efficiency for transit already exists and is going to be adopted soon in the US, in the form of noncompliant trains. Higher load factors can increase efficiency dramatically - this is why Japan and Switzerland have higher figures.

Is there any region of the world where cars routinely get three-figure passenger-miles per gallon?

Watson

Alon,

Yes, it's a quibble. And in every quibble, you make sure to pick the highest emissions figures for transit and lowest figures for cars.

No, I don't. In fact, I think I generally try to give the benefit of the doubt to transit, precisely to try and avoid getting bogged down in debates over quibbles and to focus on my central point that under any reasonable examination of the evidence the argument for transit on environmental grounds is at best very weak.

The average economy for cars in the US is 20 mpg, but you choose 25 mpg, which excludes SUVs (why? Are they being banned?).

I use 25 mpg for "the average sedan" as defined by the authors of the Berkeley study because that is the correct figure, not 20 mpg.

The average load factor for commute trips, which is where transit competes with cars, is 1.2, but you pick 1.6, which includes intercity trips.

But I'm not talking about commute trips only. Any fully-loaded subway car probably is more efficient per passenger-mile than any conventional motor vehicle with a single occupant. But on average, a subway car is not fully loaded. And on average, a motor vehicle has more than one occupant. The evidence indicates that overall, transit is no more efficient or only slightly more efficient than private autos. And that is why the "sustainability" argument for shifting from cars to transit probably isn't likely to persuade many people.

Watson

CroMagnon,

By the way, to see why your argument about parking is a quibble, look at Tables 31 and 32 on pages 40 and 41 of the Berkeley study. The contribution from parking to the total life-cycle energy use and emissions of autos is represented by the tiny dark red slivers. It's just a tiny component of the totals.

CroMagnon

I know, which is why I'm questioning the inputs of the study so much.

FWIW, our good friends at CostHelper.com have estimates of $8400 to $10,800 for a 240 sq. ft. single car garage. I doubt that includes any driveway that most homes would have. Now I'm not sure how that translates into either energy or emissions, but cost is often a proxy for energy--and least on a rough order of magnitude.

Maybe the Parking Institute meant to type 1.05 billion spaces? :D

The more I look at the Berkeley study, the more it appears superficially comprehensive and concise, but in actuality it's rather rough with crude inputs. I'm still scrutinizing where it claims it dismisses the maintenance costs of cars on roadways and only considers buses, and not cars or freight trucks. The study also seems to claim, by my tentative inpretation, that only weight matters, not usage. It neglects time, which is certainly a factor. Weathering will absolutely deteriorate a road over time, even if no trucks or buses pass through.

Alon Levy
I generally try to give the benefit of the doubt to transit...

...except when you make up a 50%-higher-than-reported emissions figure for the Chicago L.

Any fully-loaded subway car probably is more efficient per passenger-mile than any conventional motor vehicle with a single occupant. But on average, a subway car is not fully loaded. And on average, a motor vehicle has more than one occupant.

Yes, and on average, the subway train emits one quarter as much CO2 per passenger-mile as a single-occupant car. That's about one third as much as the average car at average commute occupancy.

(Buses are barely better than cars; however, buses are adopting hybrid technology faster than cars, so the advantage is growing over time. Similarly, green energy generation will increase the gap between transit and cars, not decrease it; Calgary already has a zero-emissions light rail system, powered 100% by wind power.)

And this is beside the point, which is that transit scales better. Double the number of cars on the road and congestion increases, so energy consumption more than doubles. Double transit usage and load factors increase, so the total vehicle-miles and thus energy consumption numbers less than double (in fact, the reduced congestion would also reduce bus emissions). Right now US subways are only three times as efficient as cars, but as drivers shift to transit, subways get more efficient, whereas as transit riders shift to driving, transit gets less efficient.

Placemaking Institute

Good point, Alon, transit does scale better...And if one further thinks about it, due to hybrids requiring massive batteries which, in turn, require massive amounts of heavy metals which, in turn, require massive amounts of energy to extract and transport, the "environmental footprint" of a hybrid may very well be higher than that of a gas-driven comparable model, no?

Watson
...except when you make up a 50%-higher-than-reported emissions figure for the Chicago L.

For the umpteenth time, I didn't "make up" anything. I applied the operating emissions ratio from page 11 of your citation to the operating emissions quantities from page 14 of your citation. As I just pointed out to you, even if we completely ignore the figures on page 14 and use only the emissions figures on page 11, the Chicago "L" is still dirtier than the average sedan. There is simply reasonable way of interpreting the document you yourself cited to support any other conclusion on this point.

Yes, and on average, the subway train emits one quarter as much CO2 per passenger-mile as a single-occupant car.

No it doesn't. There is no data for a single-occupant "car." And comparing the subway average to a single-occupant car would be a meaningless exercise, anyway. The average occupancy of cars is about 1.6, not 1. I've been over all this already.

(Buses are barely better than cars;

Buses are worse than cars. Again, your own source indicates that the average transit bus emits more CO2 per passenger-mile than the the average automobile. The Transportation Energy Data Book indicates that the average transit bus uses about 23% more energy per passenger-mile than the average car, and about 10% more than the average personal truck. And most transit is buses, not rail.

Watson
And this is beside the point, which is that transit scales better. Double the number of cars on the road and congestion increases, so energy consumption more than doubles. Double transit usage and load factors increase, so the total vehicle-miles and thus energy consumption numbers less than double (in fact, the reduced congestion would also reduce bus emissions).

There is no basis for the claim that an increase in transit usage would cause an increase in load factors. Load factors would increase only if the growth in usage was higher than the growth in capacity. Expanding transit capacity with new bus routes and rail lines would almost certainly reduce transit efficiency, because of diminishing marginal returns. All of the high-density, high-demand travel routes already have transit. There will almost certainly never be another heavy rail transit system in the United States. All of the cities that have enough density and population to support heavy rail transit already have it. New transit would consist mainly of bus services and light rail lines on low-density, low-demand routes with correspondingly low efficiency. Meanwhile, the efficiency of automobiles keeps increasing. The current CAFE standards require automakers to increase average fleetwide fuel efficiency to 35 mpg by 2020. The "sustainability" case for transit is weak now and will be even weaker in the future.

Watson
As for a 300-400% increase, do you have any documentation that electric cars are about to have widespread adoption?

I haven't claimed that electric cars are "about" to have widespread adoption. The adoption of PHEV and electric vehicles will be gradual. It will likely take decades before a majority of the nation's car fleet consists of such vehicles. But as the share gradually increases, so will the average efficiency of the nation's auto fleet.

In contrast, high energy efficiency for transit already exists and is going to be adopted soon in the US, in the form of noncompliant trains.

I ask for the umpteenth time: show me your evidence. Show me evidence supporting the proposition that the efficiency of rail transit could be improved to a degree comparable to the increase in automobile efficiency provided by plug-in hybrid and all-electric vehicles.

Alon Levy

Watson, when I show you the evidence, you say it means something it doesn't. I show you a chart saying that the average bus emits two thirds as much CO2 per passenger-mile as the average single-passenger car; you say it means buses are worse than cars. I show you a chart saying the same about the Chicago L; you say it means the L is worse.

Environment changes

The current CAFE standards require automakers to increase average fleetwide fuel efficiency to 35 mpg by 2020....

Nathanael

Watson:
'But even if we just use the numbers on page 11, and adjust the "Private auto (SOV)" figure for occupancy (1.6 passengers instead of 1) and mpg (25 mpg instead of 20), your citation still indicates that the Chicago "L" is dirtier than the average sedan in operating emissions.'

While this is totally unreasonable (occupancy is 1.2 on commute trips, and nobody can use the Chicago "L" for long-distance trips; using a 25 mpg average assumes the banning of SUVs), it's more reasonable than the "dividing apples by bananas and claiming you got kumquats" that you used in the rest of your incoherent "argument" -- at least it has *some* meaning.

So I will point out what it means:
(1) Electricity in the Chicago area is almost entirely coal-generated, which makes it have more carbon emissions than gasoline.
(2) And the coal plants are old and inefficient, generating relatively little electricity per pound of coal burnt.

In other words, switch to electric cars in *Chicago* and you'll find that your emissions don't get much better -- the energy efficiency is largely cancelled out by the coal plants. The way to improve emissions in this area is simply to upgrade the electric grid.

In hydro-powered areas like California, battery electric cars will make a huge emissions improvement. At that point construction emissions start to dominate (as previously noted). And at that point you discover that *for high-usage paths* it's more environmentally sound to build a train line rather than a bunch more road lanes.

Fundamentally, the determiner of whether you want roads with private cars or rail lines with huge trains is *number of people travelling*. If your "plain old" roads have congestion anywhere apart from the occasional intersection at rush hour, you want trains, and that's all there is to it. If they don't, you probably don't (at least from an efficiency perspective), unless you're trying to get a network synergy, or trying to induce growth, or airplanes are an alternative (crowded short-hop planes -> replace with HSR).

Watson:
"All of the high-density, high-demand travel routes already have transit. There will almost certainly never be another heavy rail transit system in the United States."
Absolute, utter garbage. You have never looked at either a density map or a travel demand map of anywhere in the US -- you haven't even looked at current plans! Heck, the Second Avenue Subway in NYC alone proves that you're wrong about "All of the high-density, high-demand travel routes", but in fact the crush-loaded zillion-lane highways in cities from Cleveland to Los Angeles prove you wrong. LA will build its Subway to The Sea, construct multiple additional light rail lines, and then start converting the light rail to heavy rail (which is already badly needed for the Blue Line). San Francisco has needed a Geary Subway for a very long time and may get one if they ever get some sense and money at the same time.

The comments to this entry are closed.

the firm

Jarrett is now in ...

Related Posts Plugin for WordPress, Blogger...
Related Posts Plugin for WordPress, Blogger...