The recent article “Life-cycle assessment of high-speed rail: the case of California” (Chester, Horvath) analyzes the entire life-cycle GHG emissions of California’s proposed high-speed rail project. The study has gotten a lot of attention in the media lately, because it claims high-speed trains may have negative benefit in reducing GHG emissions compared to automobiles and airplanes.
The rather obvious point of the article is that train load-factors determine emission savings. Here the authors show this in graph form for the various occupancy scenarios: 100% load-factor and 10% load-factor (single and double trainsets).
Nobody would deny that running nearly empty trains would be wasteful – but is that a plausible scenario? TGV load-factor is 71%. And even the German ICE (which favors clock-face scheduling over load-factors) has around 50%.
The article also states that infrastructure construction must be included in GHG accounting:
The energy and GHG performance of CAHSR is dominated by active operation but shows significant contributions from infrastructure construction and fuel (electricity) production. The primary contribution to the infrastructure construction component is from concrete and steel material production. Construction of retaining walls and aerial track segments are the two largest concrete requirements in the inventory (PB 1999). The production of concrete is energy intensive and releases CO2 in cement production from both fossil fuel use for kilns in clinker production and the calcination of limestone.
The paper makes a major blunder here because it assumes the infrastructure is only intended to provide the HSR express service. In fact, it will be a shared resource and ideally should generate a huge number of commuter trips (over 15 million annual trips in the case of Caltrain). Yes, let’s count the GHG cost of concrete production, but make sure to divide by the total number of users.
One erroneous assumption causes the entire Chester/Horvath thesis go up in smoke.
GIGO.
The other issue with it (something I just asked the authors about) is the assumption that a high-speed train consumes 170 kWh per VKT / 271 kWh per VMT. This parameter is attributed to the CHSRA, and underlies just about every conclusion they make, since the energy to run the trains dominates the results.
Take a Paris – Lyon TGV. We know:
Distance = 425 km
Travel time = 2 hours
Rated power = 8800 kW (at the rail– say 10 MW at the pantograph)
425 km at 170 kWh / VKT = 72 MWh for the trip
2 hours at full throttle (10 MW) = 20 MWh for the trip
We know they don’t run at full throttle, all the time.
What’s going on?
On the one hand, cars rarely run with 5 people – the average is 1.57.
On the other hand, heavy rail transit frequently does run with more than 350 people. In New York the average is about 250, and rush hour peak loads are over 1,000.
Wikipedia says:
“A 1997 EC study[25] on page 74 claims 18.00 kWh/train-km for the TGV Duplex assuming 3 intermediate stops between Paris and Lyon. ”
Here is the original source:
http://www.inrets.fr/ur/lte/cost319/MEETDeliverable17.PDF
It should be noted that speeds (and aerodynamic costs) will be higher for the California project.
Alon:
If memory serves, the studies are saying average 2.2 riders per car in the LA-SF market. Of course, that number is closer to 1 for commute market.
Totally damning evidence, which reveals an order of magnitude error in their calculations. I hope they have the scientific integrity to amend their conclusions.
The other trouble with this sort of study is where do the users come from? Acela on the east coast has taken a huge bite out of DC-NY-Boston air shuttle traffic. Less measurable are cars taken off the road. If there were a reasonably fast train to LA from SF, I’d take it over an airplane.
I don’t know the Acela load factors, but it can be hard to find a seat and sometimes they are all booked up and you can’t get a ticket.
I’m afraid I don’t find your rebuttal to be well founded. There are good reasons for load factors on rail being lower than load factors on air. You can look at some real world occupancy figures, and some of the reasons behind them, in this paper:
http://www.rssb.co.uk/pdf/reports/research/T618_traction-energy-metrics_final.pdf
If you want to do a better job of a rebuttal, try making a detailed comparison between the two papers. I’ll point out one issue and leave the other one for you to find:
Firstly, Chester doesn’t include radiative forcing, which the DfT report includes at a level of 1.3 for turboprop and 2.0 for jet air. (Removing those, Chester and DfT’s figures are very similar). Now there are arguments for and against including radiative forcing, but if you do it things look rather different.
A bigger issue is in the energy requirements of rail. Have a look into the supplemental data in Chester, and compare it with the numbers in DfT. There is an oddity there, although I think the mistake is not by Chester, but in the CAHSR 2008 figures which he uses. Now maybe CAHSR have a reason for specifying startlingly inefficient trains, but it certainly doesn’t make the proposal look good.
A more general issue is that going fast at ground level is really inefficent. If efficiency is the number one concern, electric rail at conventional speeds is the way to go (but of course it is less likely to get people off planes).
Oops, how did I not I spot that Clem already spotted the same issue! I was hoping for someone else to confirm my figures, given the number of unit conversions.
CAHSR 2008 includes the 924K btu/mile figure for a 16 car, 1200 seat vehicle. That’s 168kWh/VKT.
The most energy hungry modern European HST is the ICE-3, with 0.065kWh/seat km for 400 seats, or 26 kWh/VKT, which agrees with figures from DfT, wikipedia and a number of other sources. Even if we assume 3 ICE-3′s to get 1200 seats, with no aerodynamic savings from coupling them together, that’s still only 78kWh/VKT, or about half the CAHSR figure used by Chester.
A double decker (e.g. TGV duplex) uses less much energy for more passengers – so the actual error is probably a factor of 3-4. The bizzare thing is that the error is not in Chester, but in the CHSR proposal. Why are they trying to wreck their own proposal?