You apparently didn’t understand the point I was making. When Cecil’s PE calculated the BTU’s used by a car it already factored in the inefficiencies. When he did the calculation for the mass transit electric train he only used the power necessary to move the train but did not factor in the inefficiency of the power plant and electrical transmission. You need to look at the total energy required, not just the end device.
Incorrect. She either used figures which specifically said they were the gross energy input, or she took the net energy and assumed a conversion factor of BTU to electricity of 0.303 from fuel to wheels using the value from the 2007 EIA Annual Energy Review, page 221 (the 2008 one was not out at the time).
Do the energy consumption estimates factor in the potential for collateral efficiencies? For instance, L.A. has an extensive commuter train (except for the Westside). The catch is that it doesn’t help you much unless your office is near a Metrorail station. If we ever get the “Subway To The Sea”, then West Hollywood, Beverly Hills, Westwood, and Santa Monica will all be just a short subway ride from the downtown Amtrak station. It would then become a practical alternative for many more Inland Empire commuters to leave their cars at home and use mass transit.
If the Big One doesn’t get us first, at least.
Offhand, I’d say that the cost of paying drivers might be the difference. This site lists the starting wage for a city bus driver at $10 an hour, and as high as $25 an hour for experienced drivers (20 years or more). These are government jobs, so I bet there’s healthcare and pensions to pay too. Let’s say a driver earns $17 an hour, and works an 8 hour day (or at least 8 hours on the clock). That’s $136 a day, and if you hired three passenger van drivers to replace him, it’d be close to $400 a day in daily wages alone. I don’t know if three vans are more expensive to fuel and maintain than a big bus; let’s assume it’s a wash. You might be able to save a couple of dollars an hour in wages by not requiring the extra training and licensing for large vehicles, though.
I agree, though, that I (as a car-owning person) would be more likely to ride the bus were it an every-eight-minutes thing, rather than an every-fifteen or every-half-hour thing. And it does seem like the late-night shifts might be better served by vans.
Dear Cecil,
I’m writing here to echo the comment that the use of averages is misleading, in terms of designing solutions to our transportation problems. I ride a bus in Vermont that connects two small cities. It runs about four times a day, and usually has about forty people. At 6 mpg, the bus is getting the equivalent of 240 passenger-miles per gallon.
The context in which the system operates is key to understanding efficiency. Bus rapid transit - with dedicated lanes and terminals, hold promise to improving efficiency.
Finally, on an individual decision level, hopping on a bus that was going to go to a destination anyway obviously uses less fuel than if you were to instead drive your car.
Just FYI, Microsoft has been running its own bus system for a few years now as well: http://www.microsoft.com/presspass/press/2007/sep07/09-06RegionalExpressBusPR.mspx
They’re also building offices in cities with ‘touch down space’, meaning you can come in even if you normally work in a completely different office, and either work your whole day there, or work until the traffic improves. (I have no idea why they call it that.)
Thanks for the information on the Microsoft bus system. I found a more recent article with more details.
http://seattletimes.nwsource.com/html/microsoft/2009025535_msshuttle12m.html
The article says that Microsoft has 3,000 riders a day. It also says that most of the riders use the wi-fi on the bus to do their email and such. I wonder if Microsoft sees the bus system as a way to improve employee productivity? From my experience, email can chew up a huge chunk of every day. Of course, it is 1500 less parking spaces they have to provide also.
Perhaps, but that wouldn’t explain why they created a bicycle shuttle to carry bike riders over a bridge that does not have a bike lane.
Seems to me they are trying to provide impoved services for their employees to reduce traffic into the area and be greener. Making that commute time productive for the employees is win-win.
Dear Cecil,
London Underground used about 1.2 terawatt-hours in 2006. This is the total energy usage of the whole organisation - trains, stations, offices, repairs - absolutely everything.
This gives an efficiency of 155 Wh per passenger kilometer. 1553.41.6 = 843 BTU per passenger mile.
To compare this to cars, we have to include energy for road upkeep and all associated activities.
The Best,
Karl, Norway
In this context when converting from KW to BTU, you usually multiply by 3 to account for the BTUs consumed to produce the electricity. This would mean the actual number would be more like 2,500 BTUs. This is standard for comparing electric vehicles to ones using fossil fuels.
In any case, it is pretty irrelevant to compare mass transit in a city with a population density of over 12,000 people per square mile with the United States. Orange County, Florida where I live is around 1,100 and even New York City is 5,400 for the urban area and 2,800 for the metro.
You are correct; the numbers typically used in studies seem to range from 1/0.25 to 1/0.36, IME. Not converting for the source energy conversion is a frequent problem when comparing energy/transport systems in these sorts of public policy debates.
Not to be a nitpicker, but assuming by NYC’s urban area you mean the portion with the corporate limits, the density is 8,363,710 divided by 305 sq.mi. = ~27,400. The residential density of Manhattan is ~71,000 sq.mi., which is high even by European standards; Paris is ~56,000/sq.mi., and London as you say is ~12K for the city as a whole.
Good catch. I didn’t notice Wikipedia actually divided New York into city and urban and metro.
http://en.wikipedia.org/wiki/New_york_city
BTW, does anybody know much about the Tri-Rail system in Miami? I’m puzzled that it only get 16,000 passenger a day. The population is concentrated into a long narrow strip that should be ideal for commuter rail.
A factor of 3 seems extremely excessive. A vast organization like the LU has economies of scale that will make that factor a lot lower. The LU gets most of its electricity from natural gas-fired powerplants, where the gas is piped directly from the North Sea. There is little loss from transportation, and modern power plants have become very efficient. Comparing it to the inefficiencies of a teeny electric car is just wrong.
In any case, there must also be a multiplier applied to cars, to account for inefficiencies in transport of fuel. To compare it with the LU number, you must also include all other energy usage - highway police, car wash, and lots of other stuff.
It’s not irrelevant if you are in a city, is it? In rural areas and most towns, a car will be the most efficient transport - in a city, it’s buses or the subway.
Cecil should include a number for urban subway, it will be a lot better than the number for heavy rail. I’ve also heard that US federal regulations make the trains twice as heavy as in europe - for no safety benefit whatsoever.
By the way, London is not a particularly dense city. There are vast stretches of suburbia, and the LU goes there. The center is a lot more efficient than the outlying areas.
Another statistic that we can compare, is the emission of CO2. According to Cecil, an average SUV emits 1,050 pounds of CO2 per thousand miles, while the average scooter emits 259 pounds. In comparision, the London Underground emitted 198 pounds per thousand passenger miles.
The Federal Transit Administration classifies subways and elevated trains as heavy rail, as distinct from light rail, meaning streetcars. REALLY heavy rail (suburban trains) is classified as commuter rail. I agree the terminology is confusing.
Transit usage of course tends to vary with population density, but transit energy efficiency doesn’t, at least not in a direct way. To a large extent it’s a function of load factor (riders per car), and that in turn reflects scheduling. Suburban train efficiency in Chicago for example is much higher than “L” efficiency, in part because commuter trains operate relatively infrequently and are usually full. Although Cecil and I are still exploring this, we think the CTA’s relatively poor showing is a result of light usage on some west and south side lines.
Unless you are working in the energy industry (as I do), what it seems is of limited value, so allow me to share some facts to help the discussion.
The facts are that in 2008 energy inputs for the UK electrical grid were a total of 965 TWh, of which natural gas only made up 377.2 TWh. When we discount conversion, T&D, and auxiliary power losses, we end up with total losses of 622.2 TWh. This results in a net fuel heat to electricity efficiency of 35.5% - roughly 3:1. (I’m not counting as a deficit 1.3 TWh of electricity exports, and regardless that is an insignificant value compared to the whole).
The London Underground cannot pick and choose where it gets its electricity from - it comes from the grid, regardless of how they might try to link themselves to some spark-spread scheme. Thus the 3:1 number is a very good estimate.
Your claims as to the efficiency of modern power plants and such are not relevant, as the installed base generation is not modern, neither in the UK nor the US, and this is where the error lies.
Source: Digest of United Kingdom energy statistics (DUKES), 2009 edition, Chapter 5, pg. 116.
OK. The terminology doesn’t really matter. It’s just that there may be a difference in regulations that require US trains to have more steel per passenger than in Europe, and that this may affect energy efficiency.
Yes. Customers may expect a minimum frequency below which they will not use the service. A subway line usually has at least 4 trains per hour all day all week, so many lines are a lot less full in the suburban outskirts outside of the peak. I’ve been alone in a subway train several times, and that’s of course ridiculously inefficient - it would have been a lot cheaper to give me a free taxi ride.
I always like facts, thanks! Is it fair to use this average for an electricity customer that is large enough to build their own power plant if they wanted to? They are certainly so large that their suppliers have to set distribution networks up especially for them.
I don’t know their methodology, but LU says they emit 198 pounds of CO2 per thousand passenger miles. Unless they are trying to fool us, this number should be a good indicator of how much fuel they burnt to generate their electricity. I still think that compares favorably with a scooter at 250 pounds or so.
What customers expect is a matter of education. If a commuter rail service operates trains every two hours on Saturdays but posts a schedule and sticks to it, customers are OK with that in my experience. Many systems also run less frequent service to outlying stations while maintaining high frequency in the core through turnbacks and the like; they started doing this in Chicago a few years ago.