cross country high speed corridors

[ralfp]

But the traditional TGV trainsets are not EMU.

Most new HSR designs, including the future TGV design, use EMUs. Even some existing TGV trainsets are moving towards pseudo EMU status: http://en.wikipedia.org/wiki/Image:TGV_duplex.png

And I thought the current practice on the Acela was normally to turn the trainset around via a loop.

Why would Amtrak do this?

 

[Alice]

Does the Tyler, TX airport, at ranking 234 with about 78,482 enplanements a day, have 24 hour car rental? The looks like the largest airport in the 48 states with lower ridership than the New London, CT train station (if we ignore Shore Line East trips).

Good question, I looked (http://www.cityoftyler.org/?TabId=78). Four brands, none are 24 hour, all have longer hours than LAUS.

Airports are usually built away from towns, where downtowns are often built around a rail station. So if there is good local transit (arguably more likely in a downtown area than suburbs), then fewer rail travelers would need a rental car than air travelers.

 

[Joel N. Weber II]

But if you want to compare the US to France, you also have to remember that the US population is roughly 4.7 times the population of France, so if we need 4.7 times as many miles of track, that shouldn't discourage us at all. I do think we may want to end up with nearly twice as much high speed track per capita as the French, but I don't think that's impossible.

Miles of track per capita? That's a silly metric if I've ever seen one. A nationwide HSR network in Canada would require a similar amount of track (probably within a factor of 2 or 3) to one in the US. The track miles per capita metric suggests Canada would need only 10% or so of the US track mileage.

No, it's not a silly metric. If someone wants to argue that building high speed track is too expensive, it makes sense to look at who's going to pay for it: taxpayers. If we have five times as many taxpayers as France, that means if each person pays the same share, we should be able to end up with five times as much track.

Montreal, Toronto, and Vancouver ought to have high speed track to US cities, and maybe there should be high speed track within Canada going from Montreal to Toronto. I don't think there are any other metro areas in Canada with more than a million people.

I don't think the land of the Empire Builder has sufficient population density to justify high speed track, at least once you get west of Minneapolis until you get close to the Pacific. I do think we should be able to get across the country by high speed track via St Louis, Denver, Salt Lake City, and Las Vegas; those cities can be within 3 hours of the adjacent city by high speed rail (though you probably need 300 MPH track on both sides of Denver to make that work), and they each have 1.6 million and greater populations in their metro areas, and I think high speed sleepers from coast to coast make a lot of sense.

It's not clear how you power automobiles or airplanes from a nuclear reactor (or a wind farm or a coal plant).

It's clear to me, at east for cars: batteries. There are other potential energy storage mechanisms, including hydrogen.

But the technology doesn't actually work for replacing fossil fuels at the prices people are typically willing to spend for cars. One cannot go to the local dealership and buy a non-government-subsidized car powered exclusively by batteries with a reasonable range at a cost anywhere near competitive with a gasoline or diesel powered car. (One can buy a government subsidized car that will run on batteries for a distance that I can comfortably walk, but that's not very exciting, especially when my comfortable walking distance generally exceeds that car's battery range.) There's the Tesla Roadster, but that's a $100,000 car. And it's not at all clear that they're going to be able to get the cost for something with a similar range down to $20,000 or so as production ramps up.

Even if Tesla can get their technology into a $20,000 car, does it make sense to widen our highways? Most major cities in the US don't really have enough transportation capacity.

 

[ralfp]

No, it's not a silly metric. If someone wants to argue that building high speed track is too expensive, it makes sense to look at who's going to pay for it: taxpayers. If we have five times as many taxpayers as France, that means if each person pays the same share, we should be able to end up with five times as much track.

No. That means the cost of five times as much track is similar. That just means we might be able to afford constructing five times as much track at a similar cost per person. Just as having twice the income of your neighbors does not support buying a TV that's twice as expensive, having five times the people of France in no way supports the idea that we SHOULD build five times as much HSR track.

To make that decision you have to look at the cost/benefit ratio.

But the technology doesn't actually work for replacing fossil fuels at the prices people are typically willing to spend for cars.

But HSR doesn't actually work for replacing aircraft at the prices people are typically willing to spend for transcontinental travel.

 

[George Harris]

But the traditional TGV trainsets are not EMU.

Just because something is done a particular way does not mean that is the best way to do it.

The Japanese Shinkansen trains are ALL EMU with varying percentages of axles powered. Remember those guys have been running dedicated high speed trains for a generation before the French even thought about it, and when the Japanese choose to study something, it gets analyzed down to the last eyelash.

The curve of reduction of adhesion with speed combined with the curve for increased power to hold a steady speed with speed say that the faster you go the higher the percentage of axles needing to be powered to reliably operate the train while keeping the axle loads as low as practical. It is highly likely that the recent French speed record could not have been achieved had the rail been wet.

This stuff is from analysis and experimentation by people that do this stuff for real. It is not an opinion.

 

[Joel N. Weber II]

But HSR doesn't actually work for replacing aircraft at the prices people are typically willing to spend for transcontinental travel.

Once the high speed track is built, I'm pretty sure that ticket revenues at prices similar to airplane ticket prices will cover the operating costs just fine.

Passenger rail ticket sales may not pay for the track construction, but airline tickets also do not pay for the costs of the dollar being weakend by petroleum imports, for the costs of any wars we get involved in that we might not fight if we didn't ``need'' the petroleum, or even some of the physical infrastructure on airport property, so that's not really a fair comparison. Nor do gasoline taxes cover the entire cost of maintaining the road network.

 

[ralfp]

But HSR doesn't actually work for replacing aircraft at the prices people are typically willing to spend for transcontinental travel.

Once the high speed track is built, I'm pretty sure that ticket revenues at prices similar to airplane ticket prices will cover the operating costs just fine.

People will still have to pay for the track, be it via fares or tax dollars. There's no track fairy, even in accounting. I repeat my previous statement: HSR doesn't actually work for replacing aircraft at the prices people are typically willing to spend for transcontinental travel.

 

[Guest]

People will still have to pay for the track, be it via fares or tax dollars. There's no track fairy, even in accounting. I repeat my previous statement: HSR doesn't actually work for replacing aircraft at the prices people are typically willing to spend for transcontinental travel.

People in this thread appear to agree that building high-speed track solely for the purpose of transcontinental travel would not be at all worthwhile. What is being discussed is the fact that there are many intercity corridors in the US of 500 miles or less, where HSR would potentially be able to replace aircraft at the prices people are willing to spend. Once enough of these corridors were built, an added benefit might be the ability to run transcontinental sleeper trains over the track, for prices sufficient to cover their operating costs. This is of interest to the sort of people who read this forum, though it does not amount to a justification for building the track in the first place; air-competitive segments of less than 3-4 hours would justify that.

 

[ralfp]

People in this thread appear to agree that building high-speed track solely for the purpose of transcontinental travel would not be at all worthwhile. What is being discussed is the fact that there are many intercity corridors in the US of 500 miles or less, where HSR would potentially be able to replace aircraft at the prices people are willing to spend. Once enough of these corridors were built, an added benefit might be the ability to run transcontinental sleeper trains over the track, for prices sufficient to cover their operating costs. This is of interest to the sort of people who read this forum, though it does not amount to a justification for building the track in the first place; air-competitive segments of less than 3-4 hours would justify that.

Sure, that makes sense. I just doubt that a cross-country network could ever be established with a collection of reasonable intercity HSR networks. A network on the west coast (SAN-SEA) and one in the East, perhaps even to CHI. might make sense.

Crossing the Great Plains, the Rockies, and the Sierra Nevada and Cascades just won't happen. These are stretches of many hundreds of miles, through extraordinarily difficult terrain (well, not the Great Plains), and few population centers.

Would DEN-SLC, DEN-ABQ, ABQ-PHX, or SAT-ABQ ever make sense? These are 300-600 miles long, with few population centers, and mostly difficult terrain. Are there any reasonable routes crossing the Rockies that have significant population along the way and are practical for building 200mph track?

 

[Joel N. Weber II]

People in this thread appear to agree that building high-speed track solely for the purpose of transcontinental travel would not be at all worthwhile. What is being discussed is the fact that there are many intercity corridors in the US of 500 miles or less, where HSR would potentially be able to replace aircraft at the prices people are willing to spend. Once enough of these corridors were built, an added benefit might be the ability to run transcontinental sleeper trains over the track, for prices sufficient to cover their operating costs. This is of interest to the sort of people who read this forum, though it does not amount to a justification for building the track in the first place; air-competitive segments of less than 3-4 hours would justify that.

Yes.

And part of my point is that if you have a policy of building track from 1.6 million+ metro area to 1.6 million+ metro area everywhere where they are within three hours of each other at a 170 MPH average (easily doable with 25-30 year old French technology with a max speed of 186 MPH), you will almost accidentally end up with a high speed transcontinental route. Add Denver to Kansas City and Denver to Salt Lake City, and you've got a high speed transcontinental route (and those segments can be under 3 hours if your maximum track speed is 300 MPH instead of 186 MPH or 220 MPH). (There are also a handful of cities pairs that meet this criteria that I'm not sure should get direct track connections; Atlanta to Cincinnati may be one, and San Antonio to Houston another; tracks directly connecting those city pairs are not likely to be as useful to the overall network for longer trips as most other city pairs that are within 500 or so miles of each other.)

I suppose we could have a discussion of whether 500 miles of high speed track from a 1.6 million person metro area to another 1.6 million person metro area would be worthwhile if the only people who ever used that track were those 3.2 million people, and if not how big a population would be needed to justify 500 miles of track. The cost of 500 route miles of double track may be somewhere around $100 billion, which is very roughly $30,000 per person. Most segments would be serving larger metro areas on at least one end, though, so the cost goes down from there. Spread out that $30,000 per person over 10 years, and it's $3,000 a person a year. Not a trivial amount, but most of the segments I'm proposing are cheaper than that because they serve more populated areas.

(Las Vegas to Salt Lake City is barely shorter than 500 miles that and barely more populated than 1.6 million per metro area, though. Then again, the total population of the Los Angeles, Chicago, New York City, Boston, and DC metro areas is over 61 million; if you could convince 10% of that population to take the high speed sleeper between those Los Angeles and Chicago/the east coast and Las Vegas to Salt Lake City was the last missing segment, those sleeper riders would tripple the ridership on that segment, and that doesn't even count a large number of smaller metro areas that could also benefit from that track for longer trips.)

 

[Joel N. Weber II]

Crossing the Great Plains, the Rockies, and the Sierra Nevada and Cascades just won't happen. These are stretches of many hundreds of miles, through extraordinarily difficult terrain (well, not the Great Plains), and few population centers.

Why would there be any great need to cross the Sierra Nevada and Cascades? Maybe that would happen with a Salt Lake City to Reno to Sacamento route (which is not among the first routes that I think should be built), but it looks to me like Las Vegas to Los Angeles might go south of the Sierra Nevada mountains.

We've managed to cross the Rockies probably at least a half dozen times between the Interstate Highway system and the existing transcontinental railroads; I don't see why the Rockies can't be crossed again.

 

[p&sr]

Crossing the Great Plains, the Rockies, and the Sierra Nevada and Cascades just won't happen. These are stretches of many hundreds of miles, through extraordinarily difficult terrain (well, not the Great Plains), and few population centers.

The Mountains don't have to be "difficult terrain" anymore. Just tunnel under the whole range, like they do in Switzerland! All the hi-speed Trains can go underground in the Dark, and leave the scenic routes on the surface for uncrowded enjoyment by those of us who like to look out the windows.

 

[George Harris]

Crossing the Great Plains, the Rockies, and the Sierra Nevada and Cascades just won't happen. These are stretches of many hundreds of miles, through extraordinarily difficult terrain (well, not the Great Plains), and few population centers.

The Mountains don't have to be "difficult terrain" anymore. Just tunnel under the whole range, like they do in Switzerland! All the hi-speed Trains can go underground in the Dark, and leave the scenic routes on the surface for uncrowded enjoyment by those of us who like to look out the windows.

It is still difficult terrain. Some of these Swiss tunnels have taken a LONG TIME to build and been hugely expensive. Not to mention that the elevations of these long tunnels in Europe are significantly lower than foot of the Rockies.

 

[ralfp]

I suppose we could have a discussion of whether 500 miles of high speed track from a 1.6 million person metro area to another 1.6 million person metro area would be worthwhile if the only people who ever used that track were those 3.2 million people, and if not how big a population would be needed to justify 500 miles of track. The cost of 500 route miles of double track may be somewhere around $100 billion, which is very roughly $30,000 per person. Most segments would be serving larger metro areas on at least one end, though, so the cost goes down from there. Spread out that $30,000 per person over 10 years, and it's $3,000 a person a year. Not a trivial amount, but most of the segments I'm proposing are cheaper than that because they serve more populated areas.

$30,000 per person is not $3,000 per person over 10 years. You have to account for things like inflation and interest, among others. It's a lot more than $3,000 per person, but that's not that important, as the numbers are made up, so I'll let it pass.

Anyways the $3,000 is not even per traveler. SLC-DEN, for example, has about 652,000 air passengers a year. If every single one of them took the train, over 10 years, ignoring inflation, etc., the track alone would cost almost $8,000 per trip each way. GIVE ME A BREAK.

The Mountains don't have to be "difficult terrain" anymore. Just tunnel under the whole range, like they do in Switzerland! All the hi-speed Trains can go underground in the Dark, and leave the scenic routes on the surface for uncrowded enjoyment by those of us who like to look out the windows.

Fine, mountains are 'expensive terrain.'

We've managed to cross the Rockies probably at least a half dozen times between the Interstate Highway system and the existing transcontinental railroads; I don't see why the Rockies can't be crossed again.

Sure, they can be crossed, just not at high speed and less than outrageous cost. You can have one or the other, but not both. (We already have the second: so it's basically basically free.)

 

[Guest]

The California High Speed Rail people think they can build 500 miles of track for $30 billion, over some pretty difficult terrain. $50 billion is probably a reasonable estimate, or $100 million per mile. Freeways, by contrast, cost an average of about $30 million per mile (of course varying greatly by area).

Anyways the $3,000 is not even per traveler. SLC-DEN, for example, has about 652,000 air passengers a year. If every single one of them took the train, over 10 years, ignoring inflation, etc., the track alone would cost almost $8,000 per trip each way. GIVE ME A BREAK.

The traffic on I-80 between those two points is a good deal more than that; the whole discussion really only makes sense if most of those people switch to the train.

The rockies would indeed be expensive to cross, and substantial improvements in technology may be required before high speed rail is worthwhile on routes like DEN-SLC. That said, interstate highways across the rockies faced many of the same challenges, and several such have been built, often with less than 30% the potential traffic of DEN-SLC.

 

[Joel N. Weber II]

$30,000 per person is not $3,000 per person over 10 years. You have to account for things like inflation and interest, among others. It's a lot more than $3,000 per person, but that's not that important, as the numbers are made up, so I'll let it pass.

A one time $30,000 is probably really something more like $3,000 a year for thirty years. If you're concerned about the per-year cost, I'm not sure those numbers are all that far off, especially when you consider the amount of potential error in the $30,000.

 

[Joel N. Weber II]

I'm not sure if it was here or on another thread, but some discussion on this board made me realize that there is potential for high speed track in some areas to carry more sleeper passengers than passengers making three hour or shorter trips, even if the three hour and shorter trips capture 90%+ of the year 2008 airplane market, and the sleepers only capture 10% of the year 2008 airplane market, simply because New York City and Los Angeles and other cities are so much larger than some of the cities in the west.

And that might be a good argument for a first high speed transcon going from Fort Worth to Albuquerque to a bit north of Flagstaff to Los Angeles if I'm interpreting the data in Google Maps correctly in terms of what has the lowest terrain.

 

[Joel N. Weber II]

I've been looking a bit at service to US Congressional districts, since having lots of supporters for any plan in the House of Representatives would be good, and it's best if each represetative sees how the plan would benefit their district.

I've been focused on New England so far (simply because 435 districts take a lot of time to sort through, and so I decided to pick somewhere to start), and have been assuming that the only new high speed track in New England would start right near the intersection of Massachusetts, Connecticut, and Rhode Island, head west to pass between Springfield and Hartford, and then curve south towards New York City. I'm not quite sure if the terrain really makes that the best place to build, and geting a straight enough alignment between Hartford and Springfield for Boston trains heading west without stopping to not have to slow down might be difficult. But overall, that's a route that wouldn't require demolishing many buildings.

I'm not sure exactly what counts as enough service to get representatives to support any plan. I'm thinking somewhat tenatively that having existing rail service at some point in a representative's district with one hour travel time (one-way) to a downtown station that provides HSR service might count, but that definition needs a lot of refinement.

Maine has two Congressional districts, and Maine's second district currently has no Amtrak service at all. The first district's travel times to get to that intersection of MA, RI, and CT are well over an hour. So three hours to a major city that Maine doesn't currently have a three hour train to is not likely to be a possible selling point unless a lot more high speed track were added. Maybe funding conventional speed trains much farther north into Maine than the current Downeaster is the answer.

New Hampshire has two districts. There's a proposal to extend the MBTA Lowell Line to Concord and Manchester, which would offer service to both districts, but that may not quite be within an hour of the intersection of MA, RI, and CT, or even within an hour of downtown Boston. There is also existing Downeaster service in New Hampshire.

Vermont has only one district. If we had high speed track from New York City to ALB to Montreal, having a (not necessarily high speed) spur from that to Burlington, VT might be the easiest way to get Vermont improved rail service as part of the deal. The water just to the west of Burlington is several miles wide; Google Maps shows tracks going south to near Cedar Beach, and the outcropping just south of Cedar Beach might be a good place to cross into New York State.

Massachusetts has ten districts. I'm pretty sure disticts 3 through 10 each have at least one subway and/or MBTA Commuter Rail station offering trip times under one hour to BOS and/or BON. The second district includes Springfield, which I think should be directly served by the HSR trains. District 1 may be slightly challenging; it includes Pittsfield (which has a train station which is the only stop along #448/#449's route where the only rail service is #448/#449), but that's more than an hour to both Springfield and ALB. District 1 also includes Fitchburg, but I believe Fitchburg is currently more than an hour to BON; on the other hand, there is some desire to throw money at faster travel times on the MBTA Fitchburg Line to get that time down to about an hour. Filling in some more stops on the Springfield to ALB train might also help with having there be points in district 1 within an hour of high speed rail stations, except it's not clear to me that there are any places that are actually populated to put such stops.

Rhode Island has two districts; PVD appears to be in the first, and the rest of the Northeast Regional's Rhode Island stops appear to be in the second (but I need to check this geography more carefully at some point). Simply building track connections from the MA, RI, and CT intersection to the existing NEC and running trains into Rhode Island along that route ought to provide a good level of service to Rhode Island.

Connecticut has five districts. District 1 includes Hartford, which could easily be served by connecting the new high speed tracks to Amtrak's line from New Haven to Springfield. Those trains could continue from Hartford to New Haven, serving district 3. Perhaps the trains could continue beyond New Haven towards New London to serve district 2; this would not put district 2 within an hour of the high speed tracks, but on the other hand, my recollection from living in district 2 is that there's probably no significant commercial air service in district 2, and thus the rail connection wouldn't be any worse than the air connection. There are also tracks along the Thames River that someone mentioned had been used for passenger service in the past; I'm not sure if that would provide an attractive connection to the high speed tracks. Districts 4 and 5 are at least as challenging as district 2, plus there's probably not much spare capacity in Metro-North territory. Having trains from districts 4 and 5 that head into New York City and continue on high speed tracks to Washington DC and Pittsburgh might be a way to provide useful high speed service to those districts if the Metro-North capacity problems didn't exist.

I started looking a bit at New York State, and was reminded that some of the residents of eastern Long Island actually don't want good transporation to the rest of the country.

 

[Neil_M]

Most new HSR designs, including the future TGV design, use EMUs. Even some existing TGV trainsets are moving towards pseudo EMU status: http://en.wikipedia.org/wiki/Image:TGV_duplex.png

That was/is a proposal to create extra long Duplex TGV sets by removing the power cars and having the powered bogie under a coach, so gaining 2 extra coaches per 2 unit formation. The original PSE 'Orange' TGVs had the same kind of layout, with 3 powered bogies each end. 2 under the power car and 1 under the outer end of the end trailer cars. Not heard any more about it, so maybe its not happening.

 

[Neil_M]

It is highly likely that the recent French speed record could not have been achieved had the rail been wet.

That is a bit of a pointless statement. The speed record was just a bit of grandstanding, showing off what they could do after a long period of testing. No point in doing it if the weather was going to upset the plan , is there?

The whole point of LGV Est was to provide faster trips from Paris to Eastern France and beyond.

A reduction in journey time for Paris to Strasbourg (305 miles) from 4 hours to 2 hours 20 minutes (with further reductions to come) is the daily and hourly reality.

It even does that if its raining.

 

[Joel N. Weber II]

Sometime in the last 18 hours, I started thinking that maybe the key to funding a high speed transcon is to convince California's federal legislators that a high speed transcon would be a great way for them to get between home and DC. Google Maps tells me that Los Angeles to DC is 2,671 miles by highway; if most of the track is 300 MPH, that might be achievable in 10 hours, barely enough for two meals and a good night's sleep. The legislators who have to travel to northern California might actually get a chance to also get a little bit of work done on the train in addition to sleeping and eating.

 

[George Harris]

It is highly likely that the recent French speed record could not have been achieved had the rail been wet.

That is a bit of a pointless statement. The speed record was just a bit of grandstanding, showing off what they could do after a long period of testing. No point in doing it if the weather was going to upset the plan , is there?

The whole point of LGV Est was to provide faster trips from Paris to Eastern France and beyond.

A reduction in journey time for Paris to Strasbourg (305 miles) from 4 hours to 2 hours 20 minutes (with further reductions to come) is the daily and hourly reality.

It even does that if its raining.

Having seen pictures of the insides of the cars and talked to one of the French engineers that was part of the process, these high speed runs were anything but grandstanding. These vehicles were instrumented to the eyeballs. Much of what they learned will probably be kept highly secret unless it is to their advantage to publish. There is a lot to be learned from pushing the envelope. Why do you think automobile manufactures, and automotive componenet manufacturers have always been so involved with automobile racing? There is a lot to be learned when you put your stuff into extreme conditions. I would love to get access to the data and analysis of these test runs.

The first French high speed run to 515 km/h several years back was pretty well limited by the length of track available. That is fairly obvious if you look at a speed/distance plot, which was published in the Railway Gazette not long after the run occurred. This one had a limit set by certain concerns that were not being evaluated in their tests. I think we can look for more high speed tests out of the French later after they get through chewing their way through the results of this one.

My statement on wet rail was merely a comment on the difference between wet rail and dry rail adhesion, and by extension a comment about the advantages of distributed application of traction power, that is multiple unit trains with, as speeds go up, a higher and higher proportion of the axles being powered.

 

[TVRM610]

Sometime in the last 18 hours, I started thinking that maybe the key to funding a high speed transcon is to convince California's federal legislators that a high speed transcon would be a great way for them to get between home and DC. Google Maps tells me that Los Angeles to DC is 2,671 miles by highway; if most of the track is 300 MPH, that might be achievable in 10 hours, barely enough for two meals and a good night's sleep. The legislators who have to travel to northern California might actually get a chance to also get a little bit of work done on the train in addition to sleeping and eating.

I'm as pro-rail as they come but why would you spend the money involved in this project when airplanes can make this treck already? Rail Travel is great... and I'm all for corridors of higher speed (90-110 is all we can realistically hope for right now) but a transontinental 300MPH railway just does not seem like a good idea to me.

 

[George Harris]

Sometime in the last 18 hours, I started thinking that maybe the key to funding a high speed transcon is to convince California's federal legislators that a high speed transcon would be a great way for them to get between home and DC. Google Maps tells me that Los Angeles to DC is 2,671 miles by highway; if most of the track is 300 MPH, that might be achievable in 10 hours, barely enough for two meals and a good night's sleep. The legislators who have to travel to northern California might actually get a chance to also get a little bit of work done on the train in addition to sleeping and eating.

I'm as pro-rail as they come but why would you spend the money involved in this project when airplanes can make this treck already? Rail Travel is great... and I'm all for corridors of higher speed (90-110 is all we can realistically hope for right now) but a transontinental 300MPH railway just does not seem like a good idea to me.

You build it as a railway connecting beads on a string from coast to coast. Think Washington - Pittsburg - Columbus - Indianapolis - St. Louis - Kansas City - Topeka - Amarillo - Albuquerque - Los Angeles or Pittsburg - Cleveland - Chicago - Des Moines - Omaha - Salt Lake City - San Francisco, or any of several of several other variants for intermediate points. Say, maybe Washington - Charlotte NC - Atlanta - Montgomery AL - Jackson MS - Dallas - El Paso - Phoenix - Los Angeles

 

[Joel N. Weber II]

I'm as pro-rail as they come but why would you spend the money involved in this project when airplanes can make this treck already?

High speed trains run on electricity. Airplanes run on petroleum.

The majority of petroleum consumed in the US needs to be imported, which is not ideal from a national security perspective. While there is probably something that can be done in the way of better intracity mass transit (more efficiently than building high speed intercity rail, even), we will eventually exhaust all those possibilities, and probably still want to be using less petroleum; far too many buildings in the US have been built in places and dimensions not compatible with mass transit.

Furthermore, electricity has the potential to use energy sources that are more environmentally friendly than burning fossil fuels.

Maybe biofuels will someday power airplanes, or maybe someday batteries will power enough automobiles to leave a plentiful supply of petroleum for the airplanes, but those technologies are unproven. Meanwhile, 220 MPH trains are a working technology, and I wouldn't be surprised if a lot of the reason trains in other countries aren't going faster is a lack of imagination for how fast things might get when the curves were laid out on the existing high speed track.

Airport and highway capacity are also issues that can be addressed with high speed trains. I think Chicago, New York City, and Boston are all cities where the main airports are near capacity, and there isn't necessarily space readily available for expansion in ways that the neighbors are happy with. Moving some of those passengers to rail could help.

If high speed track is built along lesser-used routes, there's the potential for high speed Auto Train like services on those routes which could simultaneously save people time while reducing petroleum imports and reducing air polution.