How fuel-efficient are Amtrak long-distance trains, really?

[Vermontanan2]

Does anyone know of (or knows of a link to) an actual study which actually tackles how fuel-efficient Amtrak long-distance trains are?

 

Here’s an interesting article penned in Canada:

 

https://www.cbc.ca/news/science/what-on-earth-green-travel-train-1.6396327

 

It concludes that while corridor trains tend to be fuel efficient, long-distance trains are not, using the obvious examples of the Ocean and Canadian.   Basically, they take too long to get where they’re going, have low-occupancy and equipment with low capacity relative to corridor-route passenger trains, and tend to have a lot of non-revenue cars, such as baggage cars, dining cars, and lounge cars.  In other words, the best part about long-distance train travel is the amount of room available to get up and move around, but it is at a cost.

 

And then there’s this one from the Oak Ridge National Laboratory:

 

https://afdc.energy.gov/data/10311

 

This shows that “Intercity Rail” gets about 80 passenger miles per Gasoline Gallon Equivalent, which is the second-best behind “Transit Rail.”  But then it says, “All forms of rail achieve relatively high values because of high ridership, proportionally low drag, and high electrification rates (electricity is inherently more efficient than combustion-engine propulsion).”  Well, we know this isn’t true because the only place in the U.S. where intercity trains are electrified is in the Northeast Corridor, so not all “Intercity Rail” is the same.  It also references the Transportation Energy Data Book from the Oak Ridge National Laboratory, specifically table 2.13:

 

https://tedb.ornl.gov/wp-content/uploads/2021/02/TEDB_Ed_39.pdf

 

This again lumps all Intercity rail as “Amtrak” which has an energy intensity of 1,535 BTUs per passenger mile.  This is the figure Amtrak uses on its website:

 

https://www.amtrak.com/travel-green

 

(There’s an edition 40 available, but Amtrak uses edition 39.)

Regardless, Amtrak’s claim is untrue, simply because the source of the figure is assuming electrified, fast, high-occupancy, high-capacity trains with few non-revenue cars.  Long-distance trains are likely much less fuel-efficient for the reasons given in the Canadian study.

 

Of course, there are many variables.  Long-distance passenger trains serve many markets with varying degrees of occupancy.  Planes, for the most part, have their passengers traveling 100% of the distance.  But not all planes are the same, either.  Many Essential Air Service flights are sparsely occupied and costs millions annually per destination to operate.  And just think how incredibly fuel-inefficient the Capitol Limited just have been in the summer of 2023 with its three cars (only two revenue cars).  Should we consider the amount of fuel burned by freight trains (when locomotives are idling) when waiting to meet an Amtrak train?  Also, since all long-distance trains are slower than their pre-Amtrak counterparts in 1971, they must certainly use more fuel than they would if they were on a schedule without so much excessive dwell, as is often the case now.  And it gets even more muddled when a combination of modes exists.  An automobile trip is usually one mode of transport, but a trip by another means of motorized transportation requires consideration of how the person gets to and from the station at the origin and destination respectively.

 

If the data in the Oak Ridge report about passenger trains is any indication, maybe all their conclusions are suspect.  At the very least, the figure for Intercity passenger trains (the generic “Amtrak”) is obviously too rosy considering it mentions electrification and consistently high-occupancy and ignores trains with much non-revenue space.

 

Does anyone know of another possibly more-comprehensive study?

[CMStPnP]

Never heard about this so I have no other study but agree with one of the root causes, which is Amtrak space utilization and interior car design is among one of the most non-commercial approaches I have seen among any other transportation carrier that carries the public.    It seems like with the design they are using is what they think has worked in the past (which is classic railroad methodology) or they are not giving much thought to interior design and how it impacts costs or revenue.     So much waste or bad design of interior space on the cars Amtrak tows is driving their expenses and maintenance budget.

On another tangent, Long Distance trains because they are the sole passenger train on the track in many cases have high fixed costs and no economies of scale so the fuel efficiency per passenger measure does not surprise me.    The lack of frequency in addition to all other items, keeps ridership low as compared to a corridor train running at peak usage time.   Most LD trains only operate with less than 300-400 on board at any given moment.   Amtrak would even consider those numbers as high demand.     Once upon a time when the Pope visited Milwaukee or Chicago (forget which), chartered amtrak train was above or nearing 1000 riders.   Unusual but I bet that was efficient on a fuel spent per passenger basis.

[Overmod]

There were European studies, both around the time the LGV were being built and more recently, that discuss the issue in terms of 'tare weight' per passenger.  I do not have copies of them to consult, but one I remember indicated something like six tons per passenger for HSR at that time.  This had more impact on trains with multiple starts and stops than on long-distance sustained high speed (where much of the transient acceleration can be done with momentum) and of course the shorter actual trip time results in lower 'average' energy consumption compared with slower modes.

Amtrak is lucky to get most of the LD trains up to 79mph, is or was constantly stopping to take siding for other traffic, and has a stated raison d'etre of stopping at a plethora of underpopulated and underserved stations to 'provide transportation services' to them.  It should not be difficult to take a typical Amtrak consist, determine the appropriate Davis factor for it over a given route, and calculate the cumulative fuel consumption from locomotives of its type.  A methodological guide would be the Sikorsky proposal to NYC for use of the TurboTrain, which went into considerable detail with curve and grade information to make its case.

Once you have that, you can game the result with whatever metric of passenger miles vs. perceived convenience that satisfies political criteria.  But we'll know consumption per ton averaged for a trip.

[BaltACD]

With the BULK of CSX mileage being single track - I often wondered how many trains per day actually operated under clear signal indications for TWO hours during the day.  My observations, garnered at the Dufford Dispatching Center, indicated TO ME that very few were doing it, either freight or passenger.  The busier the railroad the less likely it was to happen.  Holidays, when CSX would shut down freight traffic - all the Amtraks were arriving early and had to hold at stations until the scheduled departure times (except AutoTrain - which routinely leaves origin ahead of schedule).

[Erik_Mag]

Having spent a couple of years commuting on the Surfliner with a sprinkling of LD superliners with the bi-level California cars - I can't see riding a LD train with the short haul seating density needed to get better fuel efficiency. OTOH, the AT&SF bi-levels were built specifically to reduce tare weight er passenger while maintaing chair car seating pitch.

Battery technology has now reached a point where a hybrid passenger locomotive is technically feasible, which should give at least a 20% improvement in ppassenger miles per gallon.

[CMStPnP]

Erik_Mag

Battery technology has now reached a point where a hybrid passenger locomotive is technically feasible, which should give at least a 20% improvement in ppassenger miles per gallon.

Because it is feasible does not mean it is a good idea or it represents the future.   In my view, still too early to tell.   Lets have the discussion again in 2030.

[CSSHEGEWISCH]

Hybrids are feasible.  The earliest ones that come to mind are the NYC and RI diesel-battery locomotives that switched La Salle Street Station in the 1930's and were patterned after the NYC and DL&W tri-powers. It will be some time before they are practical beyond yard limits. 

[Erik_Mag]

I've been wondering aboout hybrid locomotives for commuter trains since the F69 experimentals were mentioned in Trains in the early 1990's. Holdback until recently was a suitable energy storage medium - something that would have a specific power on the order of 1MW/ton, specfic energy at least 50kWhr/ton and several 10's of thousands cycle life with about 50kWhr for each cycle. LFP batteries have the necessary energy and power, though not sure about cycle life (which would be 10 to 20% depth of charge). I have seen reports about silicon anode batteries that can take very high charging rates (think regenerative braking) with hints of high cycle life. AFAIK, Li-ion batteries are a non-starter for this application (think battery fires).

For commuter locomotives, the main advantage of a hybrid would be getting close to straight electric acceleration. Fuel savings would be greater than LD applications, and the locomotives would likely have lower emissions of NOx and particulates due to less cycling of the prime mover. The increased acceleration of a hybrid would be useful for LD as well.

Hybrid technology may make gas turbine locomotives practical, the battery would allow the turbine to run at a constant power and thus could be sized for a lower power than a non-hybrid locomotive. Justification for a gas turbine would lowering emissions by running on CNG, LNG or ethanol.

Other than cycle life of the batteries, the main obstacle for hybrid locomotives is economic feasibility.

[dpeltier]

Vermontanan2

Does anyone know of (or knows of a link to) an actual study which actually tackles how fuel-efficient Amtrak long-distance trains are?

I don't, which means I really shouldn't be replying here, but your message reminded me of so many flame wars on Usenet's misc.transport.rail.americas from my college and post-college days that I was overwhelmed with nostalgia.

Those used to center around the Bureau of Transportation Statistics' "Energy Intensity of Passenger Modes" dataset:

https://www.bts.gov/content/energy-intensity-passenger-modes

It is possibly the most useless aggregation data ever devised. The fact that they calculate a number for BTUs per passenger mile of semi trucks is pretty indicative of how much thought went into it.

Dan

[Overmod]

The whole of the relevant Amtrak data is apparently sourced from "personal communications".

They had so much fun aggregating different groups' data into their Excel table that they did not filter for purely passenger fuel consumption.  Someone who gives a crap could probably download 4-20 and make up some sort of report that shows just the significant passenger modes.  I was too lazy to go back to the APTA data to see how much of it was electrified, but BTS does at least give us their basis for energy... in BTU (which they don't capitalize correctly).

[JL Chicago]

Considering that the average new jetliner gets approximately 100 mpg per passenger it seems that Amtrak and Via have a lot of work to do.  

[CMStPnP]

JL Chicago

Amtrak and Via have a lot of work to do.  

Which is why they should both start marketing to the mortuary market.   A dead person won't complain as much and still counts as a rider. 

Railfans would love it because then they could slip in "Waffle House" instead of the more expensive formal dining service Amtrak provides.

[Gramp]

They could offer free roof seating, too, a la Pakistan and India. That'd help the numbers.