I think a lot of the HST skeptics would do well to spend some time in the northeast. In just 10 years, Amtrak has captured 75% of the traffic between NYC and DC, and over 50% of the traffic between NYC and Boston. It's phenomenal. At my office, nobody takes flights between NYC/PHL/DC anymore. Why go out to LaGuardia when you can walk over to Penn Station, enjoy WiFi and a ton of leg room, not to mention a snack car, and get there in about the same amount of time as flying? Oh, and you're dropped off right in downtown DC, instead of Reagan or in the god forsaken suburbs if you're unlucky enough to land in Dulles.
And Amtrak isn't even very good! Years of neglect and bad regulation have left the service with old cars (except Acela), crumbling infrastructure, etc. But it beats the hell out of air travel, and the differential will only get more dramatic as fuel prices continue to increase.
Sure would be nice if the prices were more competitive. You can still get in & out of DC round trip for about half the price of a one-way Acela trip (from Boston). Not worth the premium in most cases... but I want to like it!
So booking a random trip: Wed Mar. 6 to Fri. Mar. 8. For business hours flights (leaving NY at 2 pm, returning at 9AM) I can get UAL for $175 round trip. But that's JFK <-> IAD. But Midtown to JFK is a $50 35 minute cab ride, and Dulles to downtown DC is an $80 40 minute cab ride. Add in an hour at the airport in JFK and half an hour getting out of IAD, you're looking at $305 round trip, and 4 hours for each leg. The Acela Express will get you downtown to downtown for $300 even ($150 each way) in three hours including boarding time/disembark time at each station. The Northeast Regional will get you there for $100 round trip in three and a half hours including boarding/disembark time.
Yep, I was talking about DC, which (from memory) was $279 each way on Acela, but only $140 round trip on Jet Blue into Reagan.
Sounds good to NYC though—I will consider it for sure.
Also, you highlight an interesting deficiency of the Acela: it's only 30 minutes difference in time. On the BOS->DC route it's only a 70 minute advantage on a ~6.5 hour trip. Sort of makes HSR seem unnecessary when compared to "slow" standard regional train service. It's just not that much better.
They need true grade-separated modern HSR on this route. It would be excellent.
I was talking DC to NYC. It can be $250, but there are $150 Acela trains spaced throughout the day. There is a 10 am Acela for $150 that gets you there into DC in time for a late lunch or afternoon meeting. The best part is: you can actually schedule that lunch and count on probably being able to make it.
As a skeptic, I look at the northeast, and say that may be the one place in the country it makes sense to have high speed rail because of the population density, and close proximity of the major cities.
You don't care so much about the density in the abstract, but the size, density, and proximity of the major cities. Several other regions of the U.S. compare favorably to Europe in this regard. E.g. the Great Lakes mega-region has about the same population as France. It's longer (St. Louis to Montreal is about 1,000 miles versus Toulouse to Metz which is about 6,00 miles), but it's also more linear versus France which is squarish. Most importantly, it has 11 metro areas with a population greater than 2 million (leaving out Minneapolis), while France has only 2.
If France can have high speed train from Paris (about the size of Chicago) to Marsaille (about the size of Milwaukee) why can't we have high speed rail between Chicago and Milwaukee (which are much closer together)? Or for that matter Chicago and Detroit, the latter of which is bigger than the 2-4'th largest French metro areas put together?
A more apropos comparison is Paris-Lyon-Marseilles (480 miles, 14 million people) to Chicago-Milwaukee-Madison-Minneapolis (430 miles, 14.9 million people).
A train from Detroit-Chicago-Milwaukee-Minneapolis would in a single 700 mile line cover 1/3 the population of France.
NYC is the largest city in the US and the economic capital. DC is the political capital. Both are huge tourist destinations. There are myriad reasons people travel between them. Minneapolis to Detroit, much less so. The 16 non stop flights (8 each way) between Detroit and Minneapolis probably cover most the needs without an infrastructure project that would cost tens or hundred of billions of dollars.
Since when do we only connect the largest city and the capital? If that's what we did, there would be no trains in France because Paris is basically the only economically important city in the whole country.
Detroit, Chicago, and Minneapolis and their metro areas together have $900 billion in GDP (about 1/3 the whole GDP of France) and several dozen Fortune 500 companies. There aren't three cities together anywhere in France that have comparable economic importance.[1] Yet, there are trains between a whole bunch of these French cities.
[1] France's second largest city, Lyon, has about the same GDP as Milwaukee or Kansas City.
Because of their unique positions of power, NYC and DC are useful to connect. Minneapolis and Detroit are already connected by rail, train, bus, and car. If you want to spend a hundred billion dollars, the you need to justify it by showing that there some pent up demand for travel along the route, or some gross inefficiency that HSR will correct. GDP isn't a statistic you can substitute for number of daily trips. It is ultimately about how many people ride the train vs how much it costs, right?
One of the major observations about transport is that it's a "build it and they will come" sort of thing. Creating efficient transport links can create demand for travel (just as how if you build more roads, people drive more).
GDP is a better thing to look at then current demand because it gives you an idea of the economic capacity of the region to create demand for travel, on the presumption that demand for travel is intrinsically roughly proportional to economic activity. It's a counter-point to the common refrain that American cities are insufficiently dense and near to support high speed rail. If I can get a high speed train from Frankfurt (GDP: $220 billion) to Paris (GDP: $560 billion), distance 460 miles, why can't I get a high speed train from Detroit (GDP: $250 billion) to Chicago (GDP: $570 billion), distance 280 miles? Why would the demand for travel between the latter pair of cities be intrinsically lower than the demand for travel between the former pair?
The larger point is that while the U.S. may be too big to support HSR criss-crossing from NYC to LA or Atlanta to Seattle, it is divided into several mega-regions, each of which have economic density not that different from that of European countries.
Boston-NYC-PHL-DC, Vancouver-Seattle-Portland, Detroit-Chicago-Minneapolis, Birmingham-Atlanta-Raleigh, even Houston-Dallas-Oklahoma City all link population centers that are roughly the same size and same distance apart as major European cities.
"I think a lot of the HST skeptics would do well to spend some time in the northeast"
Taking "the northeast" to be New England + New York + Pennsylvania + New Jersey, it covers ~135,000 square miles and has a population of ~56 million people.
New Mexico alone has an area of ~121,000 square miles and has only ~2 million people.
Yeah. It makes it a lot easier to build a truly high-speed train.
It's harder for a train to go 200mph through densely populated areas where NIMBYs don't want anything built. Cows, on the other hand, don't care. The train also doesn't have to stop very often -- a train can stop in Albequerque and Santa Fe, and zip north to Colorado Springs and Denver.
The biggest challenge to high-speed trains in the US is not population density or technology; the problem is political.
You're forgetting the immense networks of rural roads. With high speed trains you can't just throw a level crossing up, and wait for a pickup full of chickens to get t-boned at 220mph. And then there is the wildlife - you'd need high fences the entire distance to stop wildlife and stock from wandering onto the tracks, but that also causes other issues with wildlife corridors.
Those rural areas are also largely divided into private property holdings, which have to be resumed meaning compensation for all the property owners.
You're right, the problem is political, in that the politics of the country is set up in such a way that a centralised body cannot just go and throw railways everywhere. But that is a design feature, not a bug. The same governmental processes that stop people ending up with a nuclear waste dump next to their town also apply to railway lines. You can't just plan this stuff from a central office in Washington and make it so.
"You're forgetting the immense networks of rural roads. With high speed trains you can't just throw a level crossing up, and wait for a pickup full of chickens to get t-boned at 220mph. And then there is the wildlife - you'd need high fences the entire distance to stop wildlife and stock from wandering onto the tracks, but that also causes other issues with wildlife corridors."
Right. Because they don't have vehicles, chickens or wildlife in any of the other nations with high-speed rail networks. These are clearly unsolvable problems.
Sometimes it seems like people are just reaching to find any possible excuse to dismiss the idea based on ideology alone, doesn't it?
I really don't think people don't want HSR. It's pretty clear they do. I think it just makes sense to center the HSR around the most valuable areas, not just build it out in rural areas because it's easier to go faster, which is basically equivalent to "because we can" logic. I don't think we should build huge transportation projects just because we can; that's my problem. It has nothing to do with ideology, it's just logic and economics.
As far as the problem with wildlife and trucks and rural roads—you just need a grade-separated track with passages above and below. Expensive but necessary and plausible.
There was talk of a new rail line to the south of Rochester, MN a number of years ago and some landowners were unhappy because the proposed line would split their fields, require new fencing to protect livestock, etc.
Would it be easier to force a half a dozen farmers/landowners to give up part of their property than thousands of house/apartment dwellers? Of course. But its still forcing people and that results in lawsuits which will drag things out.
I can't help but be skeptical when it costs from 3-7 times more than the bus and is only 20-30 minutes shorter of a trip. Of course, it's far more comfortable, but for a 3-5 hr trip, the trade-off is well worth it.
If you book in advance, the Northeast Regional is $50 each way from NYC to DC, versus about $30 for the Washington Deluxe bus. And a far more comfortable trip that's less likely to get stuck in traffic.
Oh, I honestly didn't know the train was that cheap when you book in advance. I've only ever booked the week of.
Still, it's like an hour longer of a trip than the bus. (I guess I haven't had bad experiences with traffic--on my last trip to Boston, we arrived early.)
Another reason to be skeptical is that the longest existing HST line is still only 1/2 as long as the distance from NYC to Los Angeles. If the USA is really interested in country wide coverage like that, it would be unprecedented.
You're right, Amtrak in the Northeast isn't very good. It's the only mode of transportation that completely prohibits carriage of bicycles. And simultaneously one of the few major rail services in the world to do so.
This is a little ridiculous. High-speed rail in the United States will never be feasible on a country-wide scale. It's just not necessary or logical.
14 hours across the country—or even 8 hours at nearly twice the viable speed—will never match the speed or efficiency of air travel.
High speed rail can compete in the regional corridors where it makes sense. And it should. But the United States is huge. It's not Europe. We shouldn't expect this or desire it.
There's two physical factors working against air travel here:
1) it's much less cost-effective to make a flight comfortable, in terms of leg room, food, space to walk around, than a train trip; this difference only gets bigger as liquid fuel prices go up;
2) as a practical matter you can't build airports in the heart of urban centers.
If there was a 14 hour train between NYC and LA, with the leg room and amenities of existing Amtrak trains, I think it would give air travel a run for its money. Taking a cab to LaGuardia takes about half an hour from Manhattan, and it's like a $40 cab ride. Another half an hour in LA, probably $50+ cab ride. Add at least an hour at the airport for each scenario, plus the 4-5 hours of actual travel time, and you're looking at an 8 hour trip.
Would you rather have 8 hours of cab -> airport -> airplane -> airport -> cab, or 14 hours just walking from a downtown train station, onto a train, and out of another downtown train station? Would you rather spend 5 hours crammed into an economy class seat, or 14 hours in an Amtrak train with a ton of leg room, where you can get up, go to the snack car, etc?
On your point (1) : making a flight comfortable is just a case of paying more money for bigger seats and better food. However, people are extremely price-sensitive about travel, and will eagerly give up comfort for price savings. You can't argue against this : the market bears it out. A train service wanting to compete will reduce seat size and service to something approximating what air travel does now.
Bigger seats (more leg room) and better food costs weight. Weight is a much bigger problem for air travel than it is for train travel. It's just physics.
Actually, bigger seats reduce weight because it reduces the density of people (and luggage) that the plane carries.
It's a capacity issue. A plane can carry a fixed maximum amount of people, the same as a train. Finding the economic load for any transport solution is a function of what people are prepared to pay plus the fixed and variable costs for operation. So while weight is part of it, that's only as it pertains to the maximum lift of a plane. Weight isn't as important to a train, but a train still has limits to the amount of people and luggage it can carry.
Even ferries have different price/class levels, some using airplane seats, some allowing cabins. This is totally dependent on what the market pays over anything else.
I haven't sat down and done the math, but I'd imagine that if you compared the energy cost of counteracting lift induced drag, it would account for a larger portion of the operating cost of an airplane than the energy cost of counteracting rolling resistance does for a train. Assuming that is the case, it's less important for trains to minimize weight per passenger than it is for airplanes to do so, and space/comfort is directly related to weight per passenger.
This gets even more attractive when you factor in the possibility of taking an overnight train and booking a sleeper berth.
When I worked in Beijing, trips to Shanghai were greatly improved by booking sleeper berths and catching a night express trains out and back. The entire experience was generally extremely low stress and utterly reliable.
I want to agree with you, but as a practical matter, Amtrak currently offers service from NYC to Montreal that takes about 11 hours. Instead of offering it overnight, when it would add value by allowing passengers to sleep during most of their journey (and thus reduce their perceived travel time to less than that of a plane), they offer the service only during the day, in both directions! So you spent literally the entire day on the train. I would take a 14-hour train across the country, but not if it runs from 9 AM to 11 PM. And it would, because all those people in the middle don't want to have to join the train at their regional station in the middle of the night.
I'd take the 5 hours of moderate discomfort, because I only get 2-3 weeks of vacation, and I'd rather not burn basically 2-4 days of my vacation on transit. Not only that, but the plane is likely also much cheaper, and I'm not going to spend hundreds of dollars on greater comfort for a few hours.
1) Comfort is relatively speaking entirely irrelevant to the short air trips within the US.
2) As a practical matter, you don't need to. It takes you 30 extra minutes to get from the outskirts of cities to the centers, versus an extra 8 hours for train travel to get you to the city center. Perspective.
Sure, you have some good points. Trains might have something for overnight trips with beds. But that's only a fraction of the market, and frankly, only for the West-to-East direction where redeye flights are most prevalent due to the time zone loss. Going East-to-West, the time advantage is ridiculously good: get on a plane in the morning in NYC and arrive in the morning in LA. You just can't beat that.
> 1) Comfort is relatively speaking entirely irrelevant to the short air trips within the US.
It's not. There is a reason Amtrak is absolutely destroying air travel in the Northeast corridor, especially among business travelers. I can get work done on a NYC to DC train trip. I can't get any work done on the corresponding airplane trip, between the tight spaces, the cabin pressure changes, etc.
> 2) As a practical matter, you don't need to. It takes you 30 extra minutes to get from the outskirts of cities to the centers, versus an extra 8 hours for train travel to get you to the city center. Perspective.
But not all trips are NYC to LA. The shorter the distance, the more time spent getting from an airport into the city impacts the overall travel time. HSR would get you from downtown NYC to downtown DC in 95 minutes. Getting from downtown to LaGuardia is 30 minutes and from Reagan to downtown is another 15 minutes best case, possibly a lot more with weekday traffic. More importantly to a business traveler, that's 45-60 minutes they're probably not getting any work done, while the train traveler is getting work done the extra 25 minutes the train is actually traveling versus the flight.
HSR is supposed to do Boston to DC in 3 hours, versus 90 minutes for a plane trip. Add in airport time and time spent in the airport, the total travel time is probably a wash, but again much of the travel time for air travel is cabbing and waiting in line at the airport during which you can't get any work done. So the HSR still wins.
DC to Boston is about 450 miles. There are a lot of major North American cities within ~500 miles of each other. E.g. Chicago to Toronto, Minneapolis, St. Louis, Detroit, Pittsburgh, Columbus, Milwaukee, etc. There's tens of millions of people in a ~500 mile radius of Chicago, and HSR would beat air travel for those trips.
Which is why the entire point of my post was entirely in agreement with what you're saying here. I wasn't against HSR at all, just the idea that it needs to span the country.
It should connect large urban areas as needed, to bring the megalopolises closer together and easier to access, and in a way that's advantageous over air travel. It should not attempt to compete with cross-country flights, since it's not logical given those conditions. If you change the conditions, then sure, it makes perfect sense!
You're absolutely correct about all the routes you mentioned. Those are perfect. I want HSR on those routes. It handily beats air travel in every way.
But NYC to LA? Seattle to DC? Nope. Sure you could do that, but it would make more sense to fly, and that's simply a logical combination of transit for this large country. The map doesn't show that—it shows routes all the way across the country in ways that make no sense. We don't need to connect the coasts with high speed rail, because it's not logical to do so. That was my point, and I'm sure you would generally agree.
Totally in agreement. In fact I think the NYC to LA stuff drags the HSR debate into the mud more than anything else. We should instead be focusing on the U.S. mega-regions and pitching how HSR could help revitalize cities like Philadelphia or Detroit by connecting them to their thriving neighbors within their mega-regions.
It accommodates the state of the track (in most cases). To go faster would require wholesale replacement of most existing track, which is thought to be too expensive (but is the approach that was used in Japan and France).
> That's at least 12 hours to get from New York to Los Angeles.
I think many people would accept that, if the price was reasonable and the cars had suitable accommodations.
> It accommodates the state of the track (in most cases). To go faster would require wholesale replacement of most existing track, which is thought to be too expensive (but is the approach that was used in Japan and France).
It does not accomodate current tracks. No rail lines in the US are designed for speeds over 150 mph.
I should have said that a comparatively modest roadbed upgrade would allow this speed -- but only compared to the effort required for the higher speeds possible elsewhere.
It's a nice map, but it elides over the difficult parts of high-speed rail in the US. Running track through rural areas is easy, modulo getting across the Rockies and the Sierras. The hard part is routing through urban areas. Look at the Los Angeles and New York "stations". They cover the entire metro areas. Same with San Francisco. No one wants to be forced to sell their house to the government for high-speed rail, but without new dedicated track carved deep into the hearts of our urban areas, it will never compete with air travel.
I think that the high-speed rail map that the Obama Administration put up on the American Recovery and Reinvestment Act (ARRA) website is a little more close to a tenable future.
It highlights the boston-washington route (NEC on the map or BosWash some like to say[1]), which seems to be easily the most do-able route in the near future, in terms of people served.
I like it. The idea, I mean. But its hard to get people to come around to spending money on anything remotely like this seems. I think all societies should have very lofty goals published and promoted often[2], but I think at the same time we need to always provide concrete reality-scale plans when trying to make the argument to non-believers, which is why I prefer the ARRA map, or something even smaller.
I think there are some larger societal issues that affect the value of things like this too. I wish Americans would (could) travel more, and see what other developed countries are like.
[2] "If you want to build a ship, don't drum up people together to collect wood and don't assign them tasks and work, but rather teach them to long for the endless immensity of the sea" -Antoine de Saint Exupéry
The problem with the Obama map is that it still suffers from the same problem as the one posted: topography.
It is more a map to appease voters, without bearing into mind the cost of building it. If you look at it and go "oh wow, I can go from LA to SF up the coast", you of course can't tell that it would be madly expensive to build.
I wish Americans would (could) travel more, and see what other developed countries are like.
If they did this, they would realize why high speed rail makes sense in other developed countries but not in the US. Most other developed countries are tiny.
Aomori-Kagoshima (top of Japan to bottom) - 1928 km.
London-Moscow - 2885km.
Paris, France to Mosul, Iraq - 4522km.
NY-LA - 4469km.
And unlike the NY-LA route, all of the routes I mentioned have a lot of destinations in between the endpoints. For example, on the Paris-Mosul route you find Luxembourg, Germany, Austria, Slovenia, Croatia, Bosnia, Serbia, Bulgaria, Turkey and Syria. Between NY and LA there are Omaha, Chicago, Pittsburg, Cleveland, Denver and Vegas.
Even the short London-Moscow route passes through Antwerp, Dusseldorf, Berlin, Warsaw and Minsk.
>Most other developed countries are tiny
and have less people, less potential funds, etc. That isn't a valid arguemnt. Transportation infrastructure is almost by definition scalable.
Your point about destinations is pretty valid though. Getting to california from other parts of the US is an effort. The east coast has a lot of different possibilities though, and even things like Dallas-houston are frequented enough, and the time to get to DFW, go through security, etc. would be compensated easily by just taking a train.
The map the WH offered for the ARRA takes advantage of that fact. When big hubs are built, having single links between them becomes reasonable.
Rail scales up, not down. You can put two trains on a track and it costs marginally more than one train. If you put one empty train on the track, the cost is the same as a full train.
Having less money can actually help the economics of a rail line - if people are poorer (e.g. Europe), they can't afford a car and are stuck taking the train. The limiting case of this is India where low speed rail is extremely popular.
Here is why high speed rail in the US and Canada is stupid.
NY to SF: 2500 miles.
London to Moscow: 1500.
People who live between NY and SF: 313 million.
People who live between London and Moscow: 731 million.
The economics of it are completely backward, and you can't take high speed rail from London all the way to Moscow.
We'd be much better off to start by improving I80 and just wait for driverless cars (busses) to go 250 mph.
Also, yes, high speed rail in the sprawl (Boswash) makes sense, but it doesn't really make much sense anywhere else except for maybe Seattle to LA. Linking the coasts with high speed rail is probably pretty silly unless you can start getting close to 350-450 mph.
You don't have to link NY to SF. High-speed rail in the U.S. makes the most sense for linking the mega-regions. Chicago to Montreal is less than twice the distance from Bordeaux to Strasbourg and hits much bigger cities along the way. Birmingham, AL to Raleigh, NC is about the same distance hits similar sized metro areas.
Also: the fact that cars are driverless doesn't change the physics of 250 mph cars...
Actually driverless cars experience much different physics... Driverless cars can squeeze together much tighter drastically reducing wind resistance. Even the lead car will experience less drag because of vacuum effect. 250 MPH cars also need much straighter roads and allow for greater traffic flow which changes the economic modelling for highways leading to shorter journeys. Driverless cars also allow for greater degrees of sharing meaning that you'd only need a few 250 MPH cars shared amongst many individuals.
Of course you don't have to link NY and SF but that's what the map shows, well to be fair the map shows, NY to LA.
The rolling resistance of a rubber tire at 250 miles an hour is comically ridiculous. The Bugatti Veyron gets about 3 miles per gallon at 250 miles per hour.
There is an extremely glaring problem with this map: topography.
The route from LA to Seattle, with the branch additional bay-area line, has to deal with the following:
1. The tehachapi mountains, where the current rail line can handle up to 40 freight trains per day. This line is at capacity and cannot handle anything more. Worse even, it is mostly single track with steep grades and sharp curves. No regular passenger trains have run over it since the late 70's. In order to put a high-speed line here, it would cost billions, requiring many long tunnels through solid rock, along with large bridges and fills.
2: The Sacramento River Canyon, north of Redding. This is the only workable pass up to the Klamath River drainage, and quite frankly is extremely tough to pass. It has many tight curves, has had plenty of washouts and is a very steep grade past Dunsmuir. The I-5 alignment goes up and down and up, while taking up the only side of the canyon which can handle a right-of-way. There is no other option to get north, so this is pretty much right out.
3: The Siskiyou mountains, or the cascade mountains, to make it to Eugene. Both of them have rail lines, again with many sharp curves and steep grades. The Siskiyou line is absolutely brutal, which caused the Southern Pacific (the original owner, bought by the Union Pacific in 1996) to sell it off to RailTex. It handles 1 through train daily and is prone to washouts and landslides. The Cascade line is also single track, rough and at capacity. No other passes which can accomodate a line with moderate grades and few curves exist, requiring another expensive series of tunnels.
So, that one is pretty much a wash, as you can tell.
The route to San Diego from LA is already very good with high speeds (90MPH), but for whatever reason they want to route it through San Bernadino. This is absolutely stupid, as the route would follow I-15 and is already extremely rough with many changes in elevation and no consistent routing. There was a previous Santa Fe rail line through here that was promptly abandoned once they built the Surfliner route along the coast.
The route from LA to Denver is a pipe dream, sadly, due to:
1: Cima Hill, which runs through Mojave National Preserve. The only way across is along the existing Union Pacific alignment, which has several nasty curves.
2: The Wasatch Mountains. There is no good pass here outside Soldier Summit, which is rather brutal. Look it up to see what I mean.
3. The Rocky Mountains. You want to put a railroad through that? Go north through echo canyon and across the continental divide in Wyoming. That is the only HST viable route there, wide open for the taking.
The route from New Mexico to Denver is mostly owned by the state of New Mexico and Amtrak, but suffers from the beast that is Raton Pass. 3% grades, very tight curves and lots of general unpleasantness south of Trinidad. The lead up to it is nice, flat and high-speed, but the pass itself is nasty.
I could keep going, but I think the point has been made. Maps are all well and good, but what rules HST design is topography vs. budget. In order to do like the Japanese and punch a line down rough land, you have to spend gobs of cash and years of effort. With tight budgets and a booming national debt, it isn't really an option.
However, the Northeast Corridor is an excellent place, and several other routes across the midwest are also great candidates, should there be traffic to support them. Chicago-NYC would be a very good one, as the route has high demand and can follow water-level routes with gentle grades and few curves.
What makes HST succeed is either great land for it (France) or $$$ (Japan, China) or both in spades (Germany). I want to see it happen, but we have to restrict it to where it makes sense. Those routes are:
1: Chicago/Great-Lakes to NYC
2: LA to San Diego (the surfliner would be packed if the trip was on an HST)
3: Boston-NYC-DC-Richmond. Replace the Acela with something faster on a grade-separated alignment.
It seems you assume that existing rail lines would be put to use for a high-speed train. I doubt that would be the case. Amtrak shares lines with freight but it's also limited to the speed of freight. I suspect the logistics of jostling trains to passing lines would eliminate the usefulness of a high-speed train.
Also, there's not a logistical or technical problem with punching through hills and mountains when needed. Definitely a financial problem, but if you're gonna spend trillions building this thing, what's an extra few million to bore holes in rock? ;-)
It isn't that I assume they'd use existing lines, it is just that most of the good mountain passes are already taken. Furthermore, the existing alignments show how difficult it is to cross those mountains in the first place. Hell, there are many places in the US where even the Interstate network couldn't find a better route - like along the Colorado River east of Grand Junction.
A lot of politicians and advocacy groups like to point to existing lines as a solution to crossing mountains and getting through cities. This is rather short-sighted as rail lines in the US are designed for slower moving heavy-tonnage and don't take advantage of the differences HSR alignments have.
When you want to run an HSR line, your requirements are quite different. Regular (read: freight) lines require easier grades and can take sharper curves due to their slower speeds and heavier tonnage. The busiest mountain passes typically have no more than 2-2.5% ruling grades, using curves to lessen the steepness. HSR can use much steeper grades (5%) but requires extremely gentle curves, which then demands a lot of expensive tunnel and bridge work.
When you want to take an HSR rail line across a series of mountains like the Tehachapi range, you run into a major problem. The change in elevation is extremely sudden and drainages are short, curvy and have steep walls. The end result is that you have to do many cuts, fills, bridges and tunnels to get up the grade. The expense is astronomical compared to a regular highway, yet most proposals don't actually take that into account.
However, the man who designed the map is at least a bit more realistic about the cost (several trillion). I do firmly believe that HSR is actually quite viable in the US in many places, it just isn't something we can draw on a map and say "here we go!"
We have to solve the problem of topography vs. budget. If we want a truly national network like shown in the map, we have to accept that the cost will be enormous - likely many times that of the interstate system (in today's dollars). If the people are willing to take that cost, then it should be built, but we should be careful what we propose to make sure it is truly viable.
Theres quite a bit of back and forth, but the main conclusion is that the market for HSR won't support one. It seems in most cases people would rather fly, or the distance is short enough that a HSR doesn't make any sense.
Outside the NE corridor, any significant amount of rail is a pipe dream. Why spent trillions to provide service that planes are already providing more than enough service to? At least I can take a plane from here to wherever I want in the United States. Rail? Oh, I'd have to travel to the nearest station (potentially hours away) and then hope it goes anywhere near where I want to go. And judging based on that map, it will go near maybe 5% of the locations I'd like to go.
And the linked article doesn't seem to take into account the massive operating subsidies that rail requires. It's a money loser nearly everywhere (except, sometimes, in the NE corridor).
"In 1970, the year before Amtrak took over the nation's passenger trains, average rail fares were about one-third less than average airfares—about 18 cents (in today's pennies) versus 27 cents per passenger mile. Four decades of Amtrak management have reversed this ratio and more: by 2011, average rail fares were 110 percent greater than airfares—about 28.5 cents versus 13.8 cents
per passenger mile (see Figure 1)." [1]
It would be excellent to have high-speed rail in the southeast, in the Atlanta-Raleigh ("I-85") crescent. This region will eventually be a major urban agglomeration, and improved transit would help its future immensely, as well as the development of the secondary cities along the route. Raleigh, Charlotte, and Atlanta are doing fine; but the Upstate (South Carolina) and Piedmont Triad (North Carolina) could use some help, and being more integrated with the overall region would make a lot of sense.
74 comments
[ 5.3 ms ] story [ 138 ms ] threadAnd Amtrak isn't even very good! Years of neglect and bad regulation have left the service with old cars (except Acela), crumbling infrastructure, etc. But it beats the hell out of air travel, and the differential will only get more dramatic as fuel prices continue to increase.
Sounds good to NYC though—I will consider it for sure.
Also, you highlight an interesting deficiency of the Acela: it's only 30 minutes difference in time. On the BOS->DC route it's only a 70 minute advantage on a ~6.5 hour trip. Sort of makes HSR seem unnecessary when compared to "slow" standard regional train service. It's just not that much better.
They need true grade-separated modern HSR on this route. It would be excellent.
If France can have high speed train from Paris (about the size of Chicago) to Marsaille (about the size of Milwaukee) why can't we have high speed rail between Chicago and Milwaukee (which are much closer together)? Or for that matter Chicago and Detroit, the latter of which is bigger than the 2-4'th largest French metro areas put together?
A train from Detroit-Chicago-Milwaukee-Minneapolis would in a single 700 mile line cover 1/3 the population of France.
Detroit, Chicago, and Minneapolis and their metro areas together have $900 billion in GDP (about 1/3 the whole GDP of France) and several dozen Fortune 500 companies. There aren't three cities together anywhere in France that have comparable economic importance.[1] Yet, there are trains between a whole bunch of these French cities.
[1] France's second largest city, Lyon, has about the same GDP as Milwaukee or Kansas City.
GDP is a better thing to look at then current demand because it gives you an idea of the economic capacity of the region to create demand for travel, on the presumption that demand for travel is intrinsically roughly proportional to economic activity. It's a counter-point to the common refrain that American cities are insufficiently dense and near to support high speed rail. If I can get a high speed train from Frankfurt (GDP: $220 billion) to Paris (GDP: $560 billion), distance 460 miles, why can't I get a high speed train from Detroit (GDP: $250 billion) to Chicago (GDP: $570 billion), distance 280 miles? Why would the demand for travel between the latter pair of cities be intrinsically lower than the demand for travel between the former pair?
The larger point is that while the U.S. may be too big to support HSR criss-crossing from NYC to LA or Atlanta to Seattle, it is divided into several mega-regions, each of which have economic density not that different from that of European countries.
Boston-NYC-PHL-DC, Vancouver-Seattle-Portland, Detroit-Chicago-Minneapolis, Birmingham-Atlanta-Raleigh, even Houston-Dallas-Oklahoma City all link population centers that are roughly the same size and same distance apart as major European cities.
Taking "the northeast" to be New England + New York + Pennsylvania + New Jersey, it covers ~135,000 square miles and has a population of ~56 million people.
New Mexico alone has an area of ~121,000 square miles and has only ~2 million people.
This matters. A lot.
It's harder for a train to go 200mph through densely populated areas where NIMBYs don't want anything built. Cows, on the other hand, don't care. The train also doesn't have to stop very often -- a train can stop in Albequerque and Santa Fe, and zip north to Colorado Springs and Denver.
The biggest challenge to high-speed trains in the US is not population density or technology; the problem is political.
Those rural areas are also largely divided into private property holdings, which have to be resumed meaning compensation for all the property owners.
You're right, the problem is political, in that the politics of the country is set up in such a way that a centralised body cannot just go and throw railways everywhere. But that is a design feature, not a bug. The same governmental processes that stop people ending up with a nuclear waste dump next to their town also apply to railway lines. You can't just plan this stuff from a central office in Washington and make it so.
The real problem is economic, not political.
A NY-SF coach-class train ticket cost $65 in 1870. That may not sound like much, but that was 2.8 ounces of gold -- figure $4700 at current prices.
A first-class ticket with sleeping compartment cost $136 -- almost $10,000 in current dollars.
Right. Because they don't have vehicles, chickens or wildlife in any of the other nations with high-speed rail networks. These are clearly unsolvable problems.
Sometimes it seems like people are just reaching to find any possible excuse to dismiss the idea based on ideology alone, doesn't it?
As far as the problem with wildlife and trucks and rural roads—you just need a grade-separated track with passages above and below. Expensive but necessary and plausible.
There was talk of a new rail line to the south of Rochester, MN a number of years ago and some landowners were unhappy because the proposed line would split their fields, require new fencing to protect livestock, etc.
Here in Oregon, other landowners have gotten letters from the railways demanding payment for driveways across their lines just so they can get to their property. See http://www.kval.com/news/local/Neighbors-on-edge-over-new-ra...
Would it be easier to force a half a dozen farmers/landowners to give up part of their property than thousands of house/apartment dwellers? Of course. But its still forcing people and that results in lawsuits which will drag things out.
Still, it's like an hour longer of a trip than the bus. (I guess I haven't had bad experiences with traffic--on my last trip to Boston, we arrived early.)
14 hours across the country—or even 8 hours at nearly twice the viable speed—will never match the speed or efficiency of air travel.
High speed rail can compete in the regional corridors where it makes sense. And it should. But the United States is huge. It's not Europe. We shouldn't expect this or desire it.
1) it's much less cost-effective to make a flight comfortable, in terms of leg room, food, space to walk around, than a train trip; this difference only gets bigger as liquid fuel prices go up;
2) as a practical matter you can't build airports in the heart of urban centers.
If there was a 14 hour train between NYC and LA, with the leg room and amenities of existing Amtrak trains, I think it would give air travel a run for its money. Taking a cab to LaGuardia takes about half an hour from Manhattan, and it's like a $40 cab ride. Another half an hour in LA, probably $50+ cab ride. Add at least an hour at the airport for each scenario, plus the 4-5 hours of actual travel time, and you're looking at an 8 hour trip.
Would you rather have 8 hours of cab -> airport -> airplane -> airport -> cab, or 14 hours just walking from a downtown train station, onto a train, and out of another downtown train station? Would you rather spend 5 hours crammed into an economy class seat, or 14 hours in an Amtrak train with a ton of leg room, where you can get up, go to the snack car, etc?
It's a capacity issue. A plane can carry a fixed maximum amount of people, the same as a train. Finding the economic load for any transport solution is a function of what people are prepared to pay plus the fixed and variable costs for operation. So while weight is part of it, that's only as it pertains to the maximum lift of a plane. Weight isn't as important to a train, but a train still has limits to the amount of people and luggage it can carry.
Even ferries have different price/class levels, some using airplane seats, some allowing cabins. This is totally dependent on what the market pays over anything else.
When I worked in Beijing, trips to Shanghai were greatly improved by booking sleeper berths and catching a night express trains out and back. The entire experience was generally extremely low stress and utterly reliable.
2) As a practical matter, you don't need to. It takes you 30 extra minutes to get from the outskirts of cities to the centers, versus an extra 8 hours for train travel to get you to the city center. Perspective.
Sure, you have some good points. Trains might have something for overnight trips with beds. But that's only a fraction of the market, and frankly, only for the West-to-East direction where redeye flights are most prevalent due to the time zone loss. Going East-to-West, the time advantage is ridiculously good: get on a plane in the morning in NYC and arrive in the morning in LA. You just can't beat that.
It's not. There is a reason Amtrak is absolutely destroying air travel in the Northeast corridor, especially among business travelers. I can get work done on a NYC to DC train trip. I can't get any work done on the corresponding airplane trip, between the tight spaces, the cabin pressure changes, etc.
> 2) As a practical matter, you don't need to. It takes you 30 extra minutes to get from the outskirts of cities to the centers, versus an extra 8 hours for train travel to get you to the city center. Perspective.
But not all trips are NYC to LA. The shorter the distance, the more time spent getting from an airport into the city impacts the overall travel time. HSR would get you from downtown NYC to downtown DC in 95 minutes. Getting from downtown to LaGuardia is 30 minutes and from Reagan to downtown is another 15 minutes best case, possibly a lot more with weekday traffic. More importantly to a business traveler, that's 45-60 minutes they're probably not getting any work done, while the train traveler is getting work done the extra 25 minutes the train is actually traveling versus the flight.
HSR is supposed to do Boston to DC in 3 hours, versus 90 minutes for a plane trip. Add in airport time and time spent in the airport, the total travel time is probably a wash, but again much of the travel time for air travel is cabbing and waiting in line at the airport during which you can't get any work done. So the HSR still wins.
DC to Boston is about 450 miles. There are a lot of major North American cities within ~500 miles of each other. E.g. Chicago to Toronto, Minneapolis, St. Louis, Detroit, Pittsburgh, Columbus, Milwaukee, etc. There's tens of millions of people in a ~500 mile radius of Chicago, and HSR would beat air travel for those trips.
It should connect large urban areas as needed, to bring the megalopolises closer together and easier to access, and in a way that's advantageous over air travel. It should not attempt to compete with cross-country flights, since it's not logical given those conditions. If you change the conditions, then sure, it makes perfect sense!
You're absolutely correct about all the routes you mentioned. Those are perfect. I want HSR on those routes. It handily beats air travel in every way.
But NYC to LA? Seattle to DC? Nope. Sure you could do that, but it would make more sense to fly, and that's simply a logical combination of transit for this large country. The map doesn't show that—it shows routes all the way across the country in ways that make no sense. We don't need to connect the coasts with high speed rail, because it's not logical to do so. That was my point, and I'm sure you would generally agree.
It accommodates the state of the track (in most cases). To go faster would require wholesale replacement of most existing track, which is thought to be too expensive (but is the approach that was used in Japan and France).
> That's at least 12 hours to get from New York to Los Angeles.
I think many people would accept that, if the price was reasonable and the cars had suitable accommodations.
It does not accomodate current tracks. No rail lines in the US are designed for speeds over 150 mph.
Yes, and welded tracks, apparently a universal way to support high speed operations.
http://upload.wikimedia.org/wikipedia/commons/6/6f/High_Spee...
It highlights the boston-washington route (NEC on the map or BosWash some like to say[1]), which seems to be easily the most do-able route in the near future, in terms of people served.
I like it. The idea, I mean. But its hard to get people to come around to spending money on anything remotely like this seems. I think all societies should have very lofty goals published and promoted often[2], but I think at the same time we need to always provide concrete reality-scale plans when trying to make the argument to non-believers, which is why I prefer the ARRA map, or something even smaller.
I think there are some larger societal issues that affect the value of things like this too. I wish Americans would (could) travel more, and see what other developed countries are like.
[1] http://en.wikipedia.org/wiki/Boswash
[2] "If you want to build a ship, don't drum up people together to collect wood and don't assign them tasks and work, but rather teach them to long for the endless immensity of the sea" -Antoine de Saint Exupéry
It is more a map to appease voters, without bearing into mind the cost of building it. If you look at it and go "oh wow, I can go from LA to SF up the coast", you of course can't tell that it would be madly expensive to build.
If they did this, they would realize why high speed rail makes sense in other developed countries but not in the US. Most other developed countries are tiny.
Aomori-Kagoshima (top of Japan to bottom) - 1928 km.
London-Moscow - 2885km.
Paris, France to Mosul, Iraq - 4522km.
NY-LA - 4469km.
And unlike the NY-LA route, all of the routes I mentioned have a lot of destinations in between the endpoints. For example, on the Paris-Mosul route you find Luxembourg, Germany, Austria, Slovenia, Croatia, Bosnia, Serbia, Bulgaria, Turkey and Syria. Between NY and LA there are Omaha, Chicago, Pittsburg, Cleveland, Denver and Vegas.
Even the short London-Moscow route passes through Antwerp, Dusseldorf, Berlin, Warsaw and Minsk.
Your point about destinations is pretty valid though. Getting to california from other parts of the US is an effort. The east coast has a lot of different possibilities though, and even things like Dallas-houston are frequented enough, and the time to get to DFW, go through security, etc. would be compensated easily by just taking a train.
The map the WH offered for the ARRA takes advantage of that fact. When big hubs are built, having single links between them becomes reasonable.
Having less money can actually help the economics of a rail line - if people are poorer (e.g. Europe), they can't afford a car and are stuck taking the train. The limiting case of this is India where low speed rail is extremely popular.
We'd be much better off to start by improving I80 and just wait for driverless cars (busses) to go 250 mph.
Also, yes, high speed rail in the sprawl (Boswash) makes sense, but it doesn't really make much sense anywhere else except for maybe Seattle to LA. Linking the coasts with high speed rail is probably pretty silly unless you can start getting close to 350-450 mph.
Also: the fact that cars are driverless doesn't change the physics of 250 mph cars...
Of course you don't have to link NY and SF but that's what the map shows, well to be fair the map shows, NY to LA.
And the map shows a lot more than just NYC to LA.
The route from LA to Seattle, with the branch additional bay-area line, has to deal with the following:
1. The tehachapi mountains, where the current rail line can handle up to 40 freight trains per day. This line is at capacity and cannot handle anything more. Worse even, it is mostly single track with steep grades and sharp curves. No regular passenger trains have run over it since the late 70's. In order to put a high-speed line here, it would cost billions, requiring many long tunnels through solid rock, along with large bridges and fills.
2: The Sacramento River Canyon, north of Redding. This is the only workable pass up to the Klamath River drainage, and quite frankly is extremely tough to pass. It has many tight curves, has had plenty of washouts and is a very steep grade past Dunsmuir. The I-5 alignment goes up and down and up, while taking up the only side of the canyon which can handle a right-of-way. There is no other option to get north, so this is pretty much right out.
3: The Siskiyou mountains, or the cascade mountains, to make it to Eugene. Both of them have rail lines, again with many sharp curves and steep grades. The Siskiyou line is absolutely brutal, which caused the Southern Pacific (the original owner, bought by the Union Pacific in 1996) to sell it off to RailTex. It handles 1 through train daily and is prone to washouts and landslides. The Cascade line is also single track, rough and at capacity. No other passes which can accomodate a line with moderate grades and few curves exist, requiring another expensive series of tunnels.
So, that one is pretty much a wash, as you can tell.
The route to San Diego from LA is already very good with high speeds (90MPH), but for whatever reason they want to route it through San Bernadino. This is absolutely stupid, as the route would follow I-15 and is already extremely rough with many changes in elevation and no consistent routing. There was a previous Santa Fe rail line through here that was promptly abandoned once they built the Surfliner route along the coast.
The route from LA to Denver is a pipe dream, sadly, due to:
1: Cima Hill, which runs through Mojave National Preserve. The only way across is along the existing Union Pacific alignment, which has several nasty curves.
2: The Wasatch Mountains. There is no good pass here outside Soldier Summit, which is rather brutal. Look it up to see what I mean.
3. The Rocky Mountains. You want to put a railroad through that? Go north through echo canyon and across the continental divide in Wyoming. That is the only HST viable route there, wide open for the taking.
The route from New Mexico to Denver is mostly owned by the state of New Mexico and Amtrak, but suffers from the beast that is Raton Pass. 3% grades, very tight curves and lots of general unpleasantness south of Trinidad. The lead up to it is nice, flat and high-speed, but the pass itself is nasty.
I could keep going, but I think the point has been made. Maps are all well and good, but what rules HST design is topography vs. budget. In order to do like the Japanese and punch a line down rough land, you have to spend gobs of cash and years of effort. With tight budgets and a booming national debt, it isn't really an option.
However, the Northeast Corridor is an excellent place, and several other routes across the midwest are also great candidates, should there be traffic to support them. Chicago-NYC would be a very good one, as the route has high demand and can follow water-level routes with gentle grades and few curves.
What makes HST succeed is either great land for it (France) or $$$ (Japan, China) or both in spades (Germany). I want to see it happen, but we have to restrict it to where it makes sense. Those routes are:
1: Chicago/Great-Lakes to NYC
2: LA to San Diego (the surfliner would be packed if the trip was on an HST)
3: Boston-NYC-DC-Richmond. Replace the Acela with something faster on a grade-separated alignment.
4: Texas to Kansas City - this...
Also, there's not a logistical or technical problem with punching through hills and mountains when needed. Definitely a financial problem, but if you're gonna spend trillions building this thing, what's an extra few million to bore holes in rock? ;-)
A lot of politicians and advocacy groups like to point to existing lines as a solution to crossing mountains and getting through cities. This is rather short-sighted as rail lines in the US are designed for slower moving heavy-tonnage and don't take advantage of the differences HSR alignments have.
When you want to run an HSR line, your requirements are quite different. Regular (read: freight) lines require easier grades and can take sharper curves due to their slower speeds and heavier tonnage. The busiest mountain passes typically have no more than 2-2.5% ruling grades, using curves to lessen the steepness. HSR can use much steeper grades (5%) but requires extremely gentle curves, which then demands a lot of expensive tunnel and bridge work.
When you want to take an HSR rail line across a series of mountains like the Tehachapi range, you run into a major problem. The change in elevation is extremely sudden and drainages are short, curvy and have steep walls. The end result is that you have to do many cuts, fills, bridges and tunnels to get up the grade. The expense is astronomical compared to a regular highway, yet most proposals don't actually take that into account.
However, the man who designed the map is at least a bit more realistic about the cost (several trillion). I do firmly believe that HSR is actually quite viable in the US in many places, it just isn't something we can draw on a map and say "here we go!"
We have to solve the problem of topography vs. budget. If we want a truly national network like shown in the map, we have to accept that the cost will be enormous - likely many times that of the interstate system (in today's dollars). If the people are willing to take that cost, then it should be built, but we should be careful what we propose to make sure it is truly viable.
http://www.reddit.com/r/pics/comments/17nvfk/i_really_hope_t...
Theres quite a bit of back and forth, but the main conclusion is that the market for HSR won't support one. It seems in most cases people would rather fly, or the distance is short enough that a HSR doesn't make any sense.
And the linked article doesn't seem to take into account the massive operating subsidies that rail requires. It's a money loser nearly everywhere (except, sometimes, in the NE corridor).
"In 1970, the year before Amtrak took over the nation's passenger trains, average rail fares were about one-third less than average airfares—about 18 cents (in today's pennies) versus 27 cents per passenger mile. Four decades of Amtrak management have reversed this ratio and more: by 2011, average rail fares were 110 percent greater than airfares—about 28.5 cents versus 13.8 cents per passenger mile (see Figure 1)." [1]
[1]: www.cato.org/pubs/pas/PA712.pdf