Joey Reply:
September 10th, 2011 at 8:07 pm

I was talking about the tracks themselves – who owns the land was not relevant to what I was saying. I have nothing against running freight and passenger trains on the same corridor. And of course, the arbitrarily large separations required by the FRA and UP are ridiculous, though given enough space, you probably do want a little bit of a buffer zone between heavy freight tracks and high-speed passenger tracks.

Beta Magellan Reply:
September 11th, 2011 at 9:11 am

Something Siemens suggested to the Midwest high-speed booster group was to use a sort of three-track blended plan for the tighter stretches of their suggested Chicago-Minneapolis line. Within Chicago and its inner suburbs a third track would be added to the Metra ROW and the corridor would be electrified—HSR would mainly use the new track, only going onto one of Metra’s tracks for the odd passing maneuver. Even so, they projected average speeds of around 125 mph for that segment.

The Metra line in question doesn’t carry much freight, though—I think I’ve seen a grand total of one freight train there in four years of Hiawatha-riding (and even then I’m not sure), and the corridor’s owned by Metra so you don’t have to deal with the private sector. Still, I could see such a solution working in corridors where there’s much less freight than capacity, passenger trains don’t need to go more than 150 mph, and you’re not expecting to run them at less than hourly frequencies.

Alon Levy Reply:
September 11th, 2011 at 3:29 pm

Fair enough. But I suspect that Berlin wouldn’t have a bypass anyway – the German concept of running HS emphasizes connections over nonstop service. Even terminals don’t get bypassed, or don’t get bypassed often – trains instead enter and reverse direction.

For another example, look at Japan. Instead of stub-ending at Osaka and having a bypass for trains going further west, JNR built Shin-Osaka. It also built no Tokyo bypass – it intended to through-run the Tohoku and Tokaido Shinkansen, though by the time the Tohoku Shinkansen got to Tokyo, JNR had split, and JR East and JR Central have incompatible ideas of how to run trains and aren’t interested in through-service. There are two Tokyo bypasses, neither used by HSR: the Yamanote Line, built long before Tokyo Station opened and through-service became possible, and the Musashino Line, built for freight trains (but now also used by passenger trains) after JNR made the Yamanote Line passenger-only.

Andrew Reply:
September 11th, 2011 at 8:35 pm

@Alon – Your Osaka example supports the opposite conclusion — they essentially bypassed Osaka and put the Osaka shinkansen station on the bypass. in any case Osaka wouldn’t be analogous to LA, whose original station doesn’t require a stub. The Tokyo analogy also does not apply here — much higher degree of economic centralization, no comparable degree of detour in order to reach Tokyo Sta., etc. Also, non-shinkansen lines make extensive use of bypassing, as you illustrate. The fact that shinkansen do not utilize the *same* bypass routes that commuter trains do is immaterial. Eg, why on earth would shinkansen use the musashino line as a bypass to Tohoku, when the present line is so much shorter? Also, the difference between going up the west and east sides of yamanote can’t be more than a few kilometers. No one would suggest an LA bypass if there were a line from San Diego up the coast to LA and straight up thru Tejon. The presentation of these clearly spurious analogies is baffling. When we offer up these exotic foreign examples, it should be for the purpose of presenting reasonably analogous precedents, not simply conveying the impression of worldliness.

Alon Levy Reply:
September 12th, 2011 at 8:55 am

“It’s a bypass. You’ve got to build bypasses.”

Alon Levy Reply:
September 12th, 2011 at 12:00 pm

By the law, less snarkily, you should check who I’m responding to – it’s Max, not you. In particular, the assumptions I’m contrasting the Japanese approach with are Max’s (coming from the TGV operating pattern) and not yours. In particular, I have no reason to try to start a pissing match about worldliness with someone who lives near Zurich and has been involved in citizen engagement in the public transit process there. In contrast, I have every reason to contrast the Japanese approach to HSR with the French approach; I’ve argued for years that the Japanese approach is both better in general and specifically more suited to the population distribution in California.

Now, look carefully again into why I’m bringing up Yamanote and Musashino. It’s not about the Shinkansen – it’s about why they exist as commuter rail bypasses. Specifically, it’s about the point that intercity bypasses of through-stations are rarely if ever built; that’s why I brought up Berlin first.

Max Wyss Reply:
September 12th, 2011 at 1:34 pm

Alon: When we are talking about terminals in Germany, it is essentially Frankfurt/Main which applies. And there, Frankfurt Hbf IS bypassed by the ICs/ICEs from Mannheim towards the Rhineland. Well, thanks to the connections, the trains from (München-) Stuttgart and from Basel meet at the same platform in Mannheim. One of the trains continues towards Frankfurt Hbf and then on towards Fulda (-Hannover-Hamburg), and the other one towards Frankfurt Flughafen (-Köln and beyond). An almost similar connection is in Frankfurt Flughafen between the ICE towards Köln via Neubaustrecke, and the IC taking the Rhine valley route, allowing you to come from Mannheim and from Frankfurt Hbf. Actually, this works because the travel time between Mannheim and Frankfurt Flughafen is just a bit less than 30 minutes, fitting very well in the hourly fixed interval timetable.

There are a few other bypasses of terminals: a few trains (I knew them as ICE Sprinters) bypass Stuttgart, and there are also a few ICEs stopping at Köln Deutz instead of Köln Hbf, in order to avoid a reverse move. (at least that was the case; it is already quite some time since I travelled around Germany on business).

On the other hand, reverse movements are not as critical for the DB, because they routinely reverse ICEs within 5 minutes (for example in Frankfurt Hbf).

Alon Levy Reply:
September 12th, 2011 at 6:19 pm

Well, I was thinking of the ICE Sprinters as one example, but, these aren’t a very high-ridership part of the network. There are just a few per day, rather than 1-2 per hour as on the other lines. My understanding is that they’re marketed as premium TGV-style express runs.

My guess is that the California equivalent for them is nonstop SF-LA trains meeting the legal requirement of under-2:40 travel time, whereas in reality most express runs are still going to make multiple stops near SF and LA, including but not only SJ.

Do you know what the travel time difference is between trains that reverse at Frankfurt and trains that bypass it? Presumably it’s a little more than 5 minutes, because of the slow approach speeds into a terminal and the need to accelerate to speed. I’ve been trying to find equivalent information for Marseille – i.e. the difference between in-train travel time on TGV runs that bypass Marseille and on the sum of a run to Marseille and a run from Marseille to Toulon or Nice – but the schedules are so inscrutable I can’t find anything with otherwise-equivalent stopping patterns.

Max Wyss Reply:
September 13th, 2011 at 6:20 am

Alon: I agree, the ICE Sprinters are special; in fact, if I remember correctly, there is one in the morning and one in the evening (per direction) between München and Frankfurt. Time gained by bypassing Stuttgart is about 10 to 15 minutes, because the shortcut is very close to Stuttgart Hbf.

I agree with the SF-LA directs speculation; there may be one pair in the morning and one in the evening. There may also be something similar to Sacramento (whenver Sacramento gets connected to the network). Such trains are mainly business trains (but then, a stop in San Jose would make sense, in order to get the travellers into the business areas by Caltrain).

What I also could imagine as possible schedule models is a pair of trains, one running “local” until about the middle of the line, and then “express” to the other end, being met by another train running “express” until the middle, and then running “local”. Such a model would allow for a rather low-cost connection between the local towns, but still provide fast direct connections for the long distance. A condition for this to work would be non-reserved operation; you get whatever train you can…

Back to Germany. As far as I remember, there are no trains running (from somewhere to) Mannheim – Frankfurt Hbf – Frankfurt Flughafen (and to somewhere beyond). Well, maybe early morning and late night trains could. Travel times between Mannheim and Frankfurt Hbf and between Mannheim and Frankfurt Flughafen are about the same. So, the difference with the reverse in Frankfurt Hbf would be the transit time plus the reversing time, which would result in about 20 minutes difference. Now, considering the fixed interval timetable, these 20 minutes would not be useful, because the train would be “off beat”… which means that the difference had to be extended to 30 minutes.

In Marseille, the shortcut is very close to St.Charles, which means that we can assume a time difference similar to Stuttgart, therefore around 15 minutes. This is a guesstimate, as there may be a few connections not serving Marseille, but they are hard to find in the SNCF timetable chaos (well, maybe eventually they learn it too, and discover the advantages of fixed interval systems … maybe in 2040 or so… (end of polemics)).

Peter Reply:
September 13th, 2011 at 6:55 am

The ICE Sprinter runs up to seven trains a day, split between three different lines. That’s not round-trip.

Alon Levy Reply:
September 14th, 2011 at 2:17 am

Ironically, PACA recently adopted a fixed interval schedule for the TERs.

In California, the problem with the half-local, half-express model is that the stations in the middle are all in high-speed territory, whereas the stations near LA and SF are in lower-speed territory. This means that a half-local, half-express train is still going to lose too much time. In addition, having trains connect in the middle, i.e. in Fresno or maybe Bakersfield, requires the overtake station to have four platform tracks, making the urban aerials even more expensive than they already are.

The population distribution in California is fairly Japanese – i.e. nearly all people live along a small number of lines. Together with the speed pattern of fast in the middle and slow near the edges, this suggests a Japanese service plan: local trains make all stops, and express trains make stops in the LA Basin and the Bay Area but run nonstop from San Jose to Sylmar, skipping the stations in high-speed territory. Bakersfield and Fresno don’t lose too much from this arrangement because they still get reasonably fast connections to LA, SF, and later SD, and their trains only make 1 or 2 stops on the way in high-speed territory; Gilroy and Palmdale get 0 high-speed stops in the useful commuter direction, and 3 in the far less useful intercity one.

Max Wyss Reply:
September 14th, 2011 at 9:41 am

Alon: Yes, for TER services, they are gradually discovering fixed interval schedules, and the results are, as far as I read, very good.

I agree that the “middle station” would be bigger. In fact, it would require 6 tracks instead of 4 (two center tracks for high speed passing; just provide enough distance from the platforms; it is safer; plus two island platforms per direction. Overall, this station would need about 12 meters more in width than a 4-track station. This layout would also allow connections at the time when the network is Y-shaped (as opposed to the final (or hypothetical) X-shape. Depending on where the heavier end is, this “big” station could also serve for reversing not running through trains. So, there would be quite a few operational advantages for an (IMHO) marginally higher price.

Even if operation patterns are flexible, it would be worthwile to look at them at construction times. The pattern with a “strong” center station would definitely send a message the the Central Valley.

Of course, the strong center station connections would (have to) fit between two express services (in lower density scheduling, or set up so that the “intermediate express” passes the “regionals” while they exchange passengers in the middle station.

VBobier Reply:
September 11th, 2011 at 2:39 pm

There is another way to get from Victorville to Palm Springs, head out towards the Lucern Valley, then head south towards Landers and then off to Palm Springs, It can be done by car now, some might take that route when the pass is closed due to snow(Hwys 18-247-62), a full tank is a must though as It’s the long way home.

Joey Reply:
September 11th, 2011 at 3:09 pm

Not a terrible route, I guess. It’s about the same length as the Cajón route, and somewhat easier (no cities and fewer tunnels, though certainly not tunnel-free). You do miss the opportunity for a major junction in the vicinity of Colton though, which would allow SF-SD, LA-Palm Springs/Phoenix, SD-Palm Springs/Phoenix, etc, etc all using the same routes. You also miss the opportunity for a major station at San Bernardino.

thatbruce Reply:
September 13th, 2011 at 1:34 pm

And you’ve still got the problem of getting from high desert to low desert.

Alon Levy Reply:
September 11th, 2011 at 5:23 pm

I doubt speeds are really going to beat the LA Basin, apart from the smaller number of stations. A Cajon tunnel would have to involve steep grades, limiting speed. Conversely, unlike on the Peninsula, in the LA Basin the plan is to run trains at high speed, I believe 150 mph.

Andrew Reply:
September 11th, 2011 at 7:24 pm

Yes, the whole thing is based on the assumption of the main line going thru Tehachapi, the way it is planned now. The Cajon bypass makes even more sense on the assumption that it would cut straight from Cajon NW towards somewhere north of Lancaster, skipping Palmdale (which would already be served by the main line). People going between San Diego, Phoenix, Inland Empire etc. and NorCal/CV would save on the order of 30-40 minutes by using the bypass, partly because of skipping several stops (again, this is assuming Tehachapi). Under this scenario, the bypass line would converge with DX and LA-Phoenix at S Bdno, an ideal transfer hub if its residents will consent to all the bullet train traffic.

Alon Levy Reply:
September 11th, 2011 at 8:01 pm

30-40? More like 15-20, and all for a market that’s already going to be served at high frequency by combining LA-SD and LA-SF trains. It just isn’t worth it.

Andrew Reply:
September 11th, 2011 at 9:30 pm

No, 30-40. We’re comparing a) Riverside – SBdno – Bako and b) Riverside – wherever – wehrever – LA – wherever – wherever – Palmdale – Bako. For reference, official site shows Palmdale-Riverside as 59 mins.

80 miles with one stop vs. 140 miles with six stops…even with no LA basin stops and going at the high urban speed somebody else mentioned of 150 mph, the extra 60 miles would add 24 minutes. With all six stops, you’re looking at twice that. Where did you get 15-20?

Joey Reply:
September 11th, 2011 at 10:08 pm

My very rough calculations say that you would probably save about 20-25 minutes based on that 59 minute number. This is based on 320 km/h from Palmdale to Cajón, 200 through Cajón, and 160 approaching San Bernardino and down to Riverside, including stops at Palmdale, San Bernardino, and Riverside. This is definitely less time savings than I originally anticipated (it usually is), though it still might be worth routing a few trains through Cajón if you were building it anyway. But truthfully I can’t imagine many reasons why you would build it anyway.

Joey Reply:
September 11th, 2011 at 10:12 pm

And a minor point – all of the alignments I’ve seen show more or less periodic curves through the LA basin that would limit speeds to much less that 150 mph. But it doesn’t matter for the purposes of this discussion as we got that 59 minute number from the Authority anyway.

Andrew Reply:
September 11th, 2011 at 10:25 pm

The thing is, the proposed bypass also bypasses Palmdale. This saves another stop and miles of detour. Try the calculations with (a) 80 miles with one stop vs. 140 miles with six stops, as an upper limit of time savings, and (b) 80 miles with one stop vs. 140 miles with three stops, as a lower limit. Stops add 4-5 minutes each. Minimum time difference is 32 minutes, even skipping Ontario, Industry and Burbank and using the searing 150 mph urban speed Alon was referencing. Just to make sure we’re on the same page, here’s the bypass map.

As for OTHER reasons to build Cajon, the most important one is connecting with Victorville/DX. Palmdale/Sylmar would be many miles out of DX’s way.

Alon Levy Reply:
September 11th, 2011 at 11:36 pm

First, it’s not 60 miles. Trying to cut a few more km by going through Lancaster won’t work – it’s a trivial time saving, a lot of grade separations, and more likely an unnecessary line since DX is going to build Palmdale-Victorville by itself. Using mildly realistic numbers for a bypass, it’s 126 km, vs. 201 according to the official alignment. 75 km. (67 is using Google Maps for the HSR route rather than the CAHSR trip planner).

Second, 75 km/240 km/h = 18.75 minutes.

Third, you must be the only person who thinks every station a train passes near has to be served. Even Synon sometimes acknowledges you can pass stations. I think at the end CAHSR isn’t going to be skipping stops in the LA Basin or in the Bay Area, but that’s out of lower demand for super-express trains than for high frequency to intermediate stops.

Fourth, extra stops are a feature, not a bug. This is especially true of LAUS, the only truly necessary stop on the route. The higher frequency coming from serving SF-SD trips on the same trains as the much larger SF-LA and LA-SD markets is a good thing.

Fifth, in contrast, the route you’ve drawn has a very sharp curve at San Bernardino, and can’t reach the officially proposed UC Riverside station location (or in general the 215 route to SD) without taking another sharp curve. Basically, you’ve locked in two places where trains have to stop because there’s little reason not to, which is one more than on the official route.

Andrew Reply:
September 12th, 2011 at 6:20 am

Alon, numerous errors here. I’d prefer to use km too, but the official map gives stuff in miles and people are quoting speed figures in miles, so let’s stick with English this time. the offficial map shows 126 miles river-palm, then there’s another 11 roughly to the proposed junction, = 137 mi. The rough sketch proposed bypass is about 80 miles, so the difference is about 57 miles, not 47 or whatever you’re working with. Also not 60 as I estimated yesterday — saw palmdale station location today was farther N than I thought. No more time now, I’ll finish this evening….

Alon Levy Reply:
September 12th, 2011 at 8:45 am

Yes, it’s 126 miles, i.e. 201 km. And your route is not 57 miles shorter, because the time saving vs. making it just 75 km shorter is so small (4 minutes at 150 mph) that there’s no point in the extra capital cost. Remember, Victorville-Palmdale is essentially free – and if for some reason DX builds Cajon in the next 15 years, then the best connection point for SF-LV is Mojave-Barstow.

Andrew Reply:
September 12th, 2011 at 5:29 pm

Alon, continuing from where I left off this morning…

* Repeating, it’s 126+11=137 miles, not 126 miles. If you have a different alignment in mind than the one in the map, then please say so, so we can talk about one thing at a time. It would be great to have an alternative map so we could compare them and see what others think…..Also, please let us use one set of units at a time.

* I am taking Tehachapi for granted (since the main line has to go somewhere and Tehachapi is the current plan), but there is no reason to take Victorville-Palmdale for granted. I think it is safe to say that DX would not build that segment if they had the option of connecting with HSR in the Inland Empire, providing a shorter and faster route to Vegas for many more people. What I’m suggesting is that we get a future Phoenix-LA line and an SD-NorCal (LA bypass) line into the discussion, and thereby allow DX to refocus their attention (and hopefully resources) on a more rational Cajon alignment. I cannot accept your reasoning that we should take advantage of the fact that “Victorville-Palmdale is essentially free.” A wasteful and irrational alignment is not free, especially when you count the opportunity cost of not having that money put towards building through Cajon. Thus in addition to its own merits, the proposed bypass line offers the potential of forestalling the construction of an irrational line between Victorville and Palmdale (which, as you pointed out, would not be of value to NorCal either).

* Your accusation that I “must be the only person who thinks every station a train passes near has to be served” is odd, given that I had explicitly compared the speed of the bypass to that of a train passing through LA but skipping half the stations (3 out of 6). As I pointed out, even if it skipped ALL the stations, it would still take considerably longer. It’s even more odd that, after making the above accusation, you predicted that hsr is NOT actually going to end up skipping stations in the LA basin. Based on that scenario, the bypass line would be stopping at FIVE fewer stations (1 (SBdno) versus 6), in which case the time difference would not be 30-40 minutes, but well over 40.

* YOu mentioned “extra stops are a feature, not a bug”. This is precisely what I’m always telling people. Your contention was that adding stops would provide the ridership to support more frequent departures. You did not explicitly relate that to the bypass line, but I assume you meant to suggest that the frequency of departures would be jeopardized by the bypass line. There are several logical and factual problems with this suggestion. First, we need to remember that the LA-Palmdale segment will also be served by trains originating from Anaheim, if I am correct in assuming that these will not be terminating in LA. Also, under the overall scenario I am proposing, service on the Inland Empire-LA section would be supplemented by Vegas-LA trains (and eventually Arizona-LA trains). Therefore, the bypass line also would also serve as a relief line for *potential* future bottlenecks in the LA basin. Moreover, by providing much-improved service between many destinations on the system (eg, Fresno/Sac/SJose to IE/SDiego), the bypass would attract more ridership to the system and thus support a general INCREASE in frequency of departures. There is no reason to believe that it would decrease train frequency in the LA basin. Finally, we need to remember that the important issue of departure frequency is ultimately just a subcomponent of convenience. Having more trains is one way to increase convenience; having selected trains use faster bypass routes is another way to do so.

* Just in terms of norms of discussion, if you’re making a counterproposal, then say so; don’t just use language that implies that someone’s estimations are simply wrong. Ie, rather than saying “30-40? More like 15-20″, you should say, “It’s not cost-effective to build the line straight past Palmdale and Lancaster as you drew it, so the proposed bypass line would have to go thru Palmdale, in which case the time savings would only be ….” Two people cannot carry on a discussion if they are talking about two different things. The rest of us can only assume that you realized your own number was off and therefore changed the basis of your calculation mid-discussion (“Oh, I didn’t mean the line on your map; I meant one going through Palmdale”).

* You mentioned that the route I drew would force trains to stop at San Bernardino and Riverside. My earlier comments already made clear that I fully intended S Bdno to be a stop, and I was counting it in all time comparisons, except the one that had no stops along EITHER route. I referred to SB as a future transfer hub. As for Riverside, I was using that as the starting line for the time comparisons, assuming trains would be stopping there either way. If express trains thru LA WOULD be skipping Riverside AND there is no way to configure a bypass line such that it could also skip Riverside, then the bypass line’s advantage in fewer stops would be reduced by one. But assuming all trains stop at Riverside, we’re comparing 1 stop on the proposed bypass line versus 6 on the main line (minus any stops skipped on the Ontario-Palmdale segment).

On a more conciliatory note, I know this exchange has distracted you from more important and expert discussions with other contributors, so I propose that after your next reply we call this one a day.

Alon Levy Reply:
September 12th, 2011 at 6:04 pm

Everybody keeps telling me I have a tl;dr problem on my blog, and now I’m being told to write more and draw maps. Sigh.

Anyway:

- You’re completely ignoring the sharp curves in your alignment, and the sharp curve that is not in your alignment but should be if it’s modified at the southern end to actually be able to reach the proposed UC Riverside stop and San Diego. That’s why I said your alignment forces those two extra stops. It’s not a question of service plan; it’s a question of those curves being so slow trains lose little from stopping. For example, to turn from the 91 Line to the 215 requires takings, and if you don’t want to start leveling residential areas you’re limited to about 80 mph. If you want to avoid takings you have have to go on 215 instead of the railroad, but then the turn is sharper – about 1,500 feet, which corresponds to 62 mph.

- If you want to talk about stupid alignments, then don’t be selective. For one, Pacheco is stupid, especially in the presence of the second tube you advocate. (SF-LA via a second tube is 10 minutes faster than via Pacheco, and the cost difference is much smaller than that of Cajon vs. not-Cajon). The current plan for SoCal is also pretty bad compared to LA-OC-SD and LA-IE on a spur; LA-SD directly would save 47 miles, going by a highway alignment, and reduce the amount of frequency splitting rather than increasing it as you want to.

- Marginal markets, i.e. any market that doesn’t involve LA or a connection from the Upper CV to the Bay Area, don’t actually increase ridership all that much – certainly not enough to justify adding frequency. (And remember, the less important frequency is, the stupider Pacheco is.) What you’re saying about Palmdale is neither here nor there – we’re discussing the fastest trains from the Bay Area to SoCal, not local high-speed commuter runs.

- If DX connects to LA via Cajon, then the optimal way to connect SF and LV is Mojave-Barstow. Going by highway alignments again, the difference is 70 vs. 208 miles via LA and Cajon, and the route is flat and unpopulated. Alternatively, there’s Victorville-Palmdale, the current plan. Think very long and hard whether the savings of a few minutes in the high desert justify building an entirely new line. I don’t think they do; basically, instead of using the line you did use, I was using the slightly less unworkable line you should’ve used using the same concept. So sue me. I was also assuming you can actually connect to San Diego, which requires modifying the southernmost few miles on your map.

- Having more trains is convenient; it’s also costlier, in terms of track and terminal capacity (remember how bad Transbay is going to be?) and to some extent operating costs.

Alon Levy Reply:
September 11th, 2011 at 11:40 pm

Okay, so you’re traveling from Riverside to Bakersfield. The alternative option is to drive for about 3 hours. What’s more useful to you: a train that comes every hour and takes 1:20, or a train that comes every 15 minutes and takes 1:40?

Andrew Reply:
September 12th, 2011 at 6:19 am

Alon, I suggested every third train, so the comparison of every 15 mins vs every hour is way off. Also, the difference is 30-40 mins, not 20 mins. See other reply.

Alon Levy Reply:
September 12th, 2011 at 8:42 am

There’s not enough of a market for every third train. Even it if makes every single stop along the route it may avoid too many large markets to fill a train (LA-Bako, LA-Fresno, LA-SF, LA-SD), and if you’re stopping in the Central Valley you lose all time advantage coming from a bypass.

Max Wyss Reply:
September 14th, 2011 at 12:25 am

Alon (and ericmarseille): All the options presented at the current public hearings round contain a new underground run-through station in St.Charles (snoop around a bit on the lgvpaca.fr website to see the various “scenarios”). Also, depending on the scenario, Toulon would get a “downtown” underground station, or a station in the east or west. Not being very familiar with Toulon, I can not say what’s bringing most advantages. BTW, project costs for this new line, being around 240 km long with 150 to 210 km new high speed line, and 80 to 90 km of tunnels, and a connection to the Italian border vary around EUR 18 bn (in 2008 Euros)…

ericmarseille Reply:
September 14th, 2011 at 1:35 am

To me, a Parisian for 39 years berfore moving to Marseille and now close to Toulon, the costliest solution is the only one : underground station under St Charles (for a lot of reasons, not the least respecting the shortest time between Paris and Nice), and underground station under Toulon centertown (for geographic and urbanistic reasons) ; costly ; enormously costly ; but key to the success and development of a whole region.

ericmarseille Reply:
September 14th, 2011 at 1:41 am

I forgot to answer your main topic Max : PACA region, though already urbanized, has still an enormous development potential ; Intra PACA HSR is absolutely key to the realization of that potential ; 2 goals must be achieved in parallel : ensuring the fastest link to Paris from any of the southern cities, and ensuring a very fluid (read : fast) mobility between those cities, which is absolutely not the case to this day, and I know that firsthand.

Alon Levy Reply:
September 14th, 2011 at 2:04 am

Yeah, I realized that there was a tunnel option too late. It’s good for frequency if they make St. Charles a through-station, obviously. But, how much cheaper would it be to build something like a Shin-Marseille station? That is, a through-station just outside downtown, for example at Blancarde, which is still in an urban location rather than out in the fields as the stations on the LGVs are. It would achieve similar travel time to the underground option, or perhaps slightly worse because of speed restrictions on legacy track in Marseille (but even that might be upgraded as in Lille), but cost much less.

A Shin-Marseille station would also allow, under current LGV infrastructure, to serve Marseille and Toulon on the same line, boosting frequency to Toulon.

ericmarseille Reply:
September 14th, 2011 at 7:47 am

A shin-Marseille station isn’t the solution because the billions that have been poured into redeveloping Marseille since the completion of the TGV Mediterranée line (saaaaay….) are centered on a quadrangle Joliette -Saint Barnabé-Timone-Borély ; the aim is to do what was done to Paris in the 50′s, to push the poor away from centertown (project euromeditérranée) ; in this design, Saint Charles becomes a transportation hub ; to leave it out of the PACA line is, in my opinion, incomprehensible (except for cost matters, but hey, once you’ve chose a solution, you must stick to it, not water it down).

Max Wyss Reply:
September 14th, 2011 at 10:04 am

Eric: Indeed, there is a big transportation potential along the Cote d’Azur, and the current schedules do show that; bi-levels at rather short intervals, making me believe that you can get around a lot without depending on a car (that’s I think mainly between Cannes and Nice/Menton). So, yeah, the LGV is definitely a way to provide the longer-distance transportation. (I wish they had such service at my end of the Paca…)

Alon: shin-Marseille would actually be Aix-en-Provence TGV, with accordingly good connections towards Marseille center. But as Eric states, considering the developments around St.Charles, it would not make thaaat much sense. The question is also whether the underground station would be used for high speed only, or whether some TER services would use it as well (similar to the Zürich Löwenstrasse station under construction right now). That could allow some through running Miramas–Aubagne, for example.

The price of St.Charles souterrain (and other underground city center stations) is higher, but if there is the political will to go for them, they indeed have the advantages over the “shin-city” stations.