North-South Rail Link Discussion

Talk may be cheap, but these influential big guns would not be taking up the cause if they didn't believe the benefits would be worth the cost. With the T's financial "challenges" the Commonwealth obviously cannot afford to fund it, but if these guys can convince Washington that the project has regional and interstate benefits (which it obviously does) perhaps they can get the Feds the provide most of the $$$. It would open up the possibilities of direct Amtrak service to Portland and beyond and eventually Montreal. Of course the benefits to rail commuters would be enormous, being able to easily access jobs on opposite ends of the city. Some of the basic foundation work has already been installed by the construction of the Big Dig tunnels so I say "go for it".

It would also likely eliminate the need for the expansion of South Station, as far less stub tracks would be required.

Doesn't sound likely however based upon the Governor's comments in Thursday's Boston Globe.

ohalloranchris wrote:It would also likely eliminate the need for the expansion of South Station, as far less stub tracks would be required.

Doesn't sound likely however based upon the Governor's comments in Thursday's Boston Globe.

No, it wouldn't eliminate the need for SSX at all. For one, this comes so many years after SSX any which way that there's more immediate needs for expanding the station. Amtrak drives a lot of that because it's NEC congestion and deadhead moves to Southampton/Widett that are the main traffic clogs, and increased intercity service is only going to make that worse. NEC trains need to have more middle platforms they can fan out across conflict-free, and MBTA trains need to be able to scoot in/out to the layover conflict-free so they don't have to hog the platforms for extended periods as a de facto layover. Also...the Post Office side caps all the Old Colony platforms at 6 cars. Buzzards Bay is the furthest you can extend the Middleboro Line on a full daily schedule within that current SS platform capacity cap, and the BB build requires all 6 cars in every peak-hour M'boro/BB set to be 100% bi-level...any intermixed flats = sardine-can overcrowding. For now...future growth on the M'boro/BB Line puts that route in a real tough spot with overcrowding as it grows. Moreso if the extension exceeds growth projections like Middleboro station itself has exceeded projections. And forget about running a future full rush hour schedule to Hyannis without being able to do 7 or 8 cars. Even if a regular Hyannis peak schedule runs skip-stop on the mainland with just Middleboro, Brockton, and maybe one other intermediate...the growth demands of M'boro and Brockton at 8:00am and 5:00pm will overwhelm the capacity of a six-pack. So they can't let the USPS-side platforms be length-constrained forever, either.


Where SSX is stuck between a rock and hard place is that the Federal Gov't is so gridlocked USPS has its hands tied on agreeing to the relocation. There are costs that can be deferred from that project, such as delaying the Dorchester Ave.-side headhouse until later date and just building the new tracks with bare ugly "temporary" shelter into the main waiting room. That slices enormous amount of money off the project cost. But they're going to do it. For one thing, infilling Dot Ave. with new storefronts and mid-height buildings is an enormous revenue infusion into the city and the only way to integrate SS/Financial District contiguously with the Seaport and any future air rights development over the north/unobstructed end of Cabot Yard and downwind at Widett Circle. The city needs contiguous development all the way down the street to the Pike ventilation building so the "triple-junction" of neighborhoods just north of Broadway has no development gaps. That's going to push SSX along as a non-optional priority once they solve the USPS relocation. It's just too valuable a slab of land and connecting corridor to leave as-is. And if they're going to do it for Dot Ave., they're going to re-jigger the finances to at least do the full track layout.


The thing people have to wrap brain around for the N-S Link is that it's NOT a replacement for the surface terminals like SEPTA Center City was. It's all about new capacity and allowing for mainline service densities the likes of which we've never seen before. Such as rush-hour headways all day on every mainline out to 495, real 15-minute Indigo headways on all lines out to 128, and full-blown Amtrak 2040 Superduper HSR headways. To do that the Link needs to be rationed to the highest-priority matching pairs that carry all types of traffic. Having every branchline run thru to every branchline ends up limiting capacity in the long run because it's impossible to coordinate that many schedules. So the Rockports and Greenbushes and Plymouths and Foxboros still are going to be majority-surface turnbacks except for whatever couple highest-demand rush hour slots have enough reverse-commute ridership to poke to an opposite end of 128. Likewise, the capacity-crimped lines like Reading and Needham (if it hasn't been punted over to the rapid transit system) are close to capacity now because of mainline...not terminal...capacity caps and gain very little by going through the Link. Those need to stay on the surface too most of the time.

The run-thrus are going to be the usual-suspect mega mains carrying lion's share of the traffic: NEC to NH Main (probably Providence matched to Lowell/NH and Taunton/South Coast/Cape-via-Taunton/Middleboro matched to Haverhill), Worcester to Eastern Route (Portsmouth if that's online, and ample amounts of 128 short-turns). Short-turns and 'tweener slots where the headways skip a beat would probably go surface. And then the rest of the lines--Franklin, Old Colony, Fitchburg--are probably time-specific calculations for the tunnel where it's a demand spike on certain slots putting them in the tunnel once every X runs but the demand-neutral default being a turn at the surface or something like that. The pairings, save for the ones like the NEC and NH Main that would pool the lion's share of users, have to be planned. But otherwise to have the commuter rail and intercity service levels of our dreams we need all of the above: the Link, SSX, and probably NSX with a reinstated Draw #3 and a few more platforms on the ex-Spaulding side of the station. It's transformative because it's a kitchen-sink system, not a replacement for the terminals.

Thank you F Line, I had no idea it was so multi-faceted.

I didn't realize this was a four-track tunnel! Still need to fill the tracks with lots and lots of trains to make this thing work.

BandA wrote:I didn't realize this was a four-track tunnel! Still need to fill the tracks with lots and lots of trains to make this thing work.

4 tracks of tunnel, but South Station was only going to have 8 platforms, Central Station 6, and North Station 6 or 8. So that limits the number of the trains that can short-turn in the tunnel and makes it so most trains have to follow a plausible destination pair to keep pushing through. Otherwise on-platform turnbacks make that capacity a lot less than it seems. This is why the surface terminals, especially South Station, are still needed for service that ends in the CBD and the minor branchlines running on more irregular schedules.

Considering each stub platform at the surface terminals sees approx 3 trains per hour, turning at the Tunnel stations would be very detrimental. Running through would get you over 20 tph, per tunnel track & with 2 RR tracks and 2 Heavy Transit tracks that was thought out by F-Line, you would still need surface tunnels. If the RR had a 4 track tunnel with throats reaching every branch, you could effectively make the two surface terminals obsolete for Commuter Rail needs. In the 20 minutes to turn a train at a surface terminal it would already be say 'halfway' to an Rt. 128 station on the other half of the system.

F-line to Dudley via Park wrote:

BandA wrote:I didn't realize this was a four-track tunnel! Still need to fill the tracks with lots and lots of trains to make this thing work.

4 tracks of tunnel, but South Station was only going to have 8 platforms, Central Station 6, and North Station 6 or 8. So that limits the number of the trains that can short-turn in the tunnel and makes it so most trains have to follow a plausible destination pair to keep pushing through. Otherwise on-platform turnbacks make that capacity a lot less than it seems. This is why the surface terminals, especially South Station, are still needed for service that ends in the CBD and the minor branchlines running on more irregular schedules.

16 Tracks (8 platforms) under a station that only has 13 on the surface! Was this proposed by an engineer or a crack-head? To build that the cost would approach the entire federal budget! To put six platforms under the State St. area, not as deep as at South Station, would require the taking of landmarked buildings - ain't gonna happen! That would be a 240' wide tunnel, 850' long, under a 160' wide tunnel, with lobbies and passageways, with the Blue Line tunnel 90 feet down, crossing the mess with a 300 foot platform length. Thus Central Station would be almost as wide as Aquarium Station is long, unless the platform width approximated an express station on the Brighton Line of the BMT!

A four track tunnel with two platforms at each end is plausible, but still expensive and awkward. Anything more is pure fantasy. Dream on.

Gerry6309 wrote:

F-line to Dudley via Park wrote:

BandA wrote:I didn't realize this was a four-track tunnel! Still need to fill the tracks with lots and lots of trains to make this thing work.

4 tracks of tunnel, but South Station was only going to have 8 platforms, Central Station 6, and North Station 6 or 8. So that limits the number of the trains that can short-turn in the tunnel and makes it so most trains have to follow a plausible destination pair to keep pushing through. Otherwise on-platform turnbacks make that capacity a lot less than it seems. This is why the surface terminals, especially South Station, are still needed for service that ends in the CBD and the minor branchlines running on more irregular schedules.

16 Tracks (8 platforms) under a station that only has 13 on the surface! Was this proposed by an engineer or a crack-head? To build that the cost would approach the entire federal budget! To put six platforms under the State St. area, not as deep as at South Station, would require the taking of landmarked buildings - ain't gonna happen! That would be a 240' wide tunnel, 850' long, under a 160' wide tunnel, with lobbies and passageways, with the Blue Line tunnel 90 feet down, crossing the mess with a 300 foot platform length. Thus Central Station would be almost as wide as Aquarium Station is long, unless the platform width approximated an express station on the Brighton Line of the BMT!

A four track tunnel with two platforms at each end is plausible, but still expensive and awkward. Anything more is pure fantasy. Dream on.

No land-takings are required for SS Under because it's so very very far underground. Well below Red and Silver level and all building pilings because that's the only way it fits with the criscrossing tunnels through there. It requires 1-mile long lead tunnels to a Washington St. portal on the NEC and portals flanking both sides of the Amtrak building at Southampton for the Fairmount and Old Colony approaches. If Central Station gets built it's underneath I-93, almost as far under Aquarium as Aquarium is under the surface. North Station would be considerably shallower and cross the Charles no deeper than the Orange Line tunnel, but that station fits under the footprint of the Garden's back service lot, the Storrow Drive ramps right before they go subterranean, and a slice of the first couple surface platforms so cut-and-cover works there.

There won't even be a surface breach near SS other than maybe a construction shaft or two. This all has to be deep-bore and horizontal-dig because there's no other way to do it. The only tough part for impacts is how to get from SS Under to the insertion point under I-93. Probably going to require 1 block under Dot Ave., 1 block under Atlantic Wharf, and a gentle S-curve under open water and Seaport Blvd. to slip under 93 at Northern Ave. That's going to be pure pain. But less so than the other alternative of plowing under Dewey Square which is a near certain no-go because of the Federal Reserve building. The rest of the tunnel stays neatly under the 93 space or squarely under T land and 4 or more commuter rail surface tracks inside the north and south terminal districts. There'll be a couple years of disruptions for portal/approach construction as surface tracks get temporarily displaced and realigned for staging. But the same thing happened during the Big Dig without unduly affecting southside schedules, so that's not anything that hasn't been seen before.

nomis wrote:Considering each stub platform at the surface terminals sees approx 3 trains per hour, turning at the Tunnel stations would be very detrimental. Running through would get you over 20 tph, per tunnel track & with 2 RR tracks and 2 Heavy Transit tracks that was thought out by F-Line, you would still need surface tunnels. If the RR had a 4 track tunnel with throats reaching every branch, you could effectively make the two surface terminals obsolete for Commuter Rail needs. In the 20 minutes to turn a train at a surface terminal it would already be say 'halfway' to an Rt. 128 station on the other half of the system.

The problem with that is that too few mainlines are equipped for run-thrus that turn at 128 because of lack of terminal-quality layover space. Anderson RTC definitely has the land to take most of the Amtrak schedule and healthy amount of NEC thru-running with the open industrial land immediately abutting. You could easily build a BET II out there if needed. However...

-- How do you set up a near-Readville/terminal-quality size layover at Westwood/128 with the Neponset River running a few hundred feet to the side and not much track capacity left to take from the industrial park after all that recent redevelopment? Next available opportunity is a reactivated East Junction layover, which won't fly for the distance the northside would have to run.

-- ...or Dedham Corporate on the Franklin with its adjacent wetlands.

-- ...or Riverside behind the Green Line facility.

-- ...or the Fitchburg Line with its would-be Waltham/128 stop's adjacent wetlands that allow for a good-sized station facility by the Biogen campus but not a fan-out of extra tracks.

-- ...or on the Eastern Route before it branches? North St. Yard and Beverly MoW are tiny; deadheading through the single-track tunnel to Castle Hill a capacity killer. The branches are branches. And even if you turned out south of the portal to Castle Hill, I'm not even sure Castle Hill won't be redeveloped as downtown mixed-use density before this thing gets built.

-- ...or on the Old Colony anywhere whatsoever before it branches.

-- ...or the Western Route, because it simply can't take that many more trains through Somerville and Medford single-tracking to handle anything more than a singular Indigo pairing.



That sort of imbalance, and improbability of solving because of too many wetlands immediately where each of these mains crosses 128, is what makes it impossible to run everything thru and turn it back soon enough before extreme schedule length becomes its own capacity limiter. You certainly aren't going to go 495-to-495 without having to slacken up the train spacing a hell of a lot, so the capacity increases get eaten back if Framingham, Walpole, East Junction/Attleboro, Ayer, and the like are the places you have to run a full route from the opposite half of the system in order to find enough yard space or big enough slab of land to build a mainline-capacity facility. That's going to enforce its own rationing. You can send most Providence trains and all Amtraks to Anderson. You can probably send most Worcester trains to Beverly or Peabody and divvy them up that way if B&A is the Eastern's only pairing. But everything else needs to pick-and-choose its spots and probably stay on the surface a lot of the time. Or cede tunnel slots to the Indigo pairings that are just ping-ponging from one 128 stop to another with short sets that turn back immediately with dwells brief enough that a pocket track serves the need.

Could any of the Readville sites be reused for yards or turnbacks? There are several in the area. The original locomotive and car shop properties straddling the Boston/Dedham line have had several development plans fall through. Although portions of the property bordering Sprague Street are now small industry sites, acres of the original "Readville" remain vacant today. A smaller yard on the southeastern side of the Midland Div. is being used by the MBTA for car storage or something. And how valuable is all that CSX yard between the Midland and the Corridor to their operations - it looks like there could be some surplus property there. The wide ROW near the site of the original Readville switch tower was at one time all freight yard and car storage, but much of that has been taken up by a commercial storage building now - but could that be taken back by eminent domain?

There's a giant yard called Beacon Park, sitting empty...

...a yard which is planned to have a modest sized layover facility built on it once the Pike is done being realigned.

F-line to Dudley via Park wrote:However...

-- How do you set up a near-Readville/terminal-quality size layover at Westwood/128 with the Neponset River running a few hundred feet to the side and not much track capacity left to take from the industrial park after all that recent redevelopment? Next available opportunity is a reactivated East Junction layover, which won't fly for the distance the northside would have to run.

-- ...or Dedham Corporate on the Franklin with its adjacent wetlands.

-- ...or Riverside behind the Green Line facility.

-- ...or the Fitchburg Line with its would-be Waltham/128 stop's adjacent wetlands that allow for a good-sized station facility by the Biogen campus but not a fan-out of extra tracks.

-- ...or on the Eastern Route before it branches? North St. Yard and Beverly MoW are tiny; deadheading through the single-track tunnel to Castle Hill a capacity killer. The branches are branches. And even if you turned out south of the portal to Castle Hill, I'm not even sure Castle Hill won't be redeveloped as downtown mixed-use density before this thing gets built.

-- ...or on the Old Colony anywhere whatsoever before it branches.

-- ...or the Western Route, because it simply can't take that many more trains through Somerville and Medford single-tracking to handle anything more than a singular Indigo pairing.

One answer would be to focus only on the lines that can be 100% dual-tracked within 128 and plan around 4-car EMU sets for the indigo pairings to keep physical yard sizes down, while pushing headways as tight as you can to maximize capacity. That leaves you with Riverside and Fairmount (to Dedham?) to the South, and Fitchburg and the Eastern to the north, which aren't a bad set of pairings in and of themselves. Add in through-running Providence (and maybe... one day... SCR?) <-> Lowell/Lawrence(via wildcat) service to spare SS some of it's most frequent platform-hogs, and because the ridership is honestly going to be there to support clock-facing push-pull service on those corridors, and it saves the T from having to clog up those already-busy ROW's with Indigo EMU's.

The only services that would get the shaft are 1: The Western, which should honestly just be turned into the OLX to Reading anyways; 2: The Needham, which should be carved up and turned into OLX and GLX projects; and 3: The OC lines, which honestly are so constrained by the whole 3-branches-into-a-single-track-bottleneck between SS and Braintree that it'll take billions in ROW improvements along that stretch to even BEGIN to get to the point where Indigo frequencies could even begin to enter the realm of possibility.

But back to the 4-car EMU plan, which would cut down significantly the size of the layover yards that the T would need to build at Salem, Bear Hill, Riverside, and Dedham. It would be a great idea in the short term, but I'd posit that the extra investment to get layover yards to handle 8-car EMU sets makes much, much more long-term sense.

The long-term impact of a well-executed link can't be underestimated. With proper indigo-ization of appropriate lines and upgraded headways for all of the others, the state is essentially getting a SECOND rapid transit system overlaid over the current T, one that covers major gaps in the existing T service area, while also being much, much faster and more modern. Look to the RER in Paris to see what I mean. The T would get to a point where the old subways are used for short-distance trips, while shunting long-distance travel into the new Indigo EMU system, which becomes even more of a possibility if the T were to focus on reasonable extensions of it's RT lines to tie into Indigo service (i.e. either a peoplemover to connect Wonderland to a Revere Indigo stop, or the BL Lynn extension, building an underground connection between Yawkey and Kenmore, etc) and make jumping from service to service that much easier.

Now, the RER has been such a runaway success that it now fills 8-car single and multilevel EMU sets at rush hour, even at peak headways, and there's no reason to expect that a post N-S link Indigo EMU service wouldn't do the same things. The Paris metropolitan area is only twice the size of Boston's, and since this indigo-ized inner CR service would net the T about half the number of the RER's lines, the per-line ridership numbers could easily equal out. So I guess what I'm saying is that IF the state were to ever build the link, it would be worth it to man up and spend the extra $$$ for the layover yards at the 128 termini of whichever lines end up Indigo-ized so that they could handle the 8-car EMU sets that could well be needed in the relatively near future.

highgreen215 wrote:Could any of the Readville sites be reused for yards or turnbacks? There are several in the area. The original locomotive and car shop properties straddling the Boston/Dedham line have had several development plans fall through. Although portions of the property bordering Sprague Street are now small industry sites, acres of the original "Readville" remain vacant today. A smaller yard on the southeastern side of the Midland Div. is being used by the MBTA for car storage or something. And how valuable is all that CSX yard between the Midland and the Corridor to their operations - it looks like there could be some surplus property there. The wide ROW near the site of the original Readville switch tower was at one time all freight yard and car storage, but much of that has been taken up by a commercial storage building now - but could that be taken back by eminent domain?

The Boston Revelopment Authority is trying again to get some of Yard 5 (not all of it, but subdividing of more acreage next to the school bus years) re-zoned as a "hi-tech incubator" or whatever their marketing speak is this week for a NIMBY-nonthreatening industrial park. But they've tried that before and potential developers have been scared away by the truck restrictions on the DCR parkways and Dedham city streets, so take with grain of salt. The weedy half of Yard 2 (commuter rail yard) next to the Neponset currently used for the recycling center is probably an easier reach for the T, since they can literally double the number of layup tracks by booting the recycling center. Not sure that's all that urgent a need, though, because they're getting a permanent yard easement at Beacon Park next to the straightened Pike that serves SSX-related needs (including the Indigos). And if that whole Olympics "Midtown" hullabaloo at Widett Circle gets revived in the future you would think there'd be an easier path for underwriting the decking costs for the 'bowl' the Food Market currently lives in by giving the T a permanent easement underneath for nuthin' but train + bus (to consolidate Cabot + Albany + Southampton garage storage) storage while the cover-over opens up the new street grid-interfacing development on top. That could serve most or all southside central storage needs in the Link era. The fact that Boston 2024 didn't think to breach this subject and expected a master developer to pay for a half-billion's worth of decking costs up-front was one of the prime reasons the bid imploded.

Personally, I'm not even sure they'd need Readville for anything more than Fairmount layovers if the Widett deck-over offered up that much space. They could sell Yard 2 and Albany + Southampton bus garages to the city/BRA, shift the Fairmount layover to the small unused 4-track yard sandwiched between the mainline and Yard 2 (forget which yard number that is), let the BRA build several hundred housing units on the Yard 2 property as a contiguous extension of Wolcott Sq. that does wonders to ease the city's affordable housing crunch, and still be totally fine for future expansion needs with the Beacon Park easement and the still significant unused rail acreage of Yard 5. That's what I'd pursue if I were Planning Czar, since all the land-swappy action of valuable real estate would make the Widett "downstairs" relocation cost-neutral. Possibly even slightly profitable for the T. But the Menino-coattails BRA marches to its own quixotic drummer, so these things never get talked about.

----------

More to the point re: Link turnbacks, Readville's still sub-ideal because it requires deadheading from Westwood or Dedham Corporate. And the NEC in particular is going to be such an unending stream of trains every few minutes between Amtrak 2040 Superduper HSR and the Providence + Stoughton/South Coast Lines at Link-augmented frequencies that it might be an ops constriction despite the capacity available at Readville. At least at Anderson the surrounding ex-industrial/non-wetlands property makes it effortless to peel straight ahead out of the last station into a layover and swap sets faster than an on-platform turn. That's what ideally accommodates multiple mainlines' worth of turnbacks and a near-full Amtrak schedule's worth of turnbacks. I think you'd artfully have to mix-and-match between Westwood and Dedham Corporate termini to match that capacity well enough to turn northside run-thrus down there because of the low margin for error with on-platform turns. The Readville deadheads would have a very narrow window to turn the hell around and sprint any deadheads back to Readville before idle trains blocking a platform becomes the NEC's main north-of-Providence capacity limiter.

But at least you could plausibly do it down there if the ops were precise enough, and that would be enough to thread Lowell/Concord and Haverhill/Portland service down to both ends of 128. The problem is you still have zero such options on the Fitchburg, B&A, Old Colony, and Eastern (with inner Western capacity-limited to more or less just Indigo schedules by the single track). Those 128-crossing station candidates all have wetlands or space constraints. And so 100% run-thrus of Greater Boston everywhere and outright elimination of the surface terminals becomes impossible. You would have to find--hell or high water--an Anderson-like setup of varying size at every one of those 128 nodes to make tunnel-only satisfactorily handle the loads. 495 and on-branchline turnbacks just end up limiting the overall capacity by having to space trains wider inside 128 in order to funnel unlike-timed schedules from the outskirts. Whereas if the dispatch bench were shortened to just 128 turns once a line (any line) poked through the tunnel it's much easier to pack them ultra-dense one after the other.


It's a trade-off or priorities any which way. To do nothing but tunnel run-thrus and close + redevelop the surface terminals you either need to. . .

A) Have multi-platform small terminals with an immediate layover yard to turn out to (e.g. future Anderson) at every single mainline's 128 station. Which we can conclude right now today is 100.00% impossible at some of those 128-crossing locations, 100.00% impossible to rig up on every single line at once, and extremely unlikely to net the necessary space on majority of the lines.

B) Deal with extremely tight-timed on-platform turnbacks and hauling arse to the nearest close-enough layover 1-2 miles away (i.e. Westwood/Dedham Corporate and Readville). But that becomes overly dispatching-brittle if every line has to stage that same low-margin sprint. And that assumes there's buildable land within 1-2 miles on every single line for siting a 'close-enough' layover...something we can conclude today might work for some/most but is 100.00% impossible to swing for every single line at once.

C) Turn at 495 instead, ration the 128 turnbacks to the lines that can handle it (e.g. Anderson, Westwood/Dedham), ration the 128 turnbacks to the 128-to-128 Indigo schedules that are lowest-difficulty for dispatch to cram between other slots, and take the steep tradeoff capacity limitations on system-wide train spacing. For the cost of the tunnel this is treading water or only a slight increase over SSX's + NSX's future surface capacities, and makes it hard to justify the cost of building the tunnel and tearing out the surface stations. In short, the cost-benefit ratio changes to the point where no planner would rationally take the extra step of spending more on the project to outright decomission the surface terminals if that was the sum total difference between limiting capacity and retaining unlimited capacity. Therefore C ends up a 'placebo' alternative irrelevant to the real world.


Why bother? The surface terminals in their nearer-future expanded state are there, and it requires no additional capital investment to keep them there. 15-minute Indigos overlaid on every line. 30 minute all-day mainline service to 495, 45-50 minute all-day branchline service. Hourly regional intercity service to Springfield, Concord, Portland. Amtrak frequencies the likes we've never seen before. You get that whole package mixing and matching with tunnel slots skewed by a sliding scale of priority and amped-up surface terminal schedules on all else. Since Boston still is a CBD-centric travel market, and some lines just don't have screamingly obvious natural destination pairs. I can't fathom a rational reason to go full-blown SEPTA and get rid of the surface terminals when tunnel + SSX/NSX are the complementary pieces that get us real 100-year capacity increases and real Euro/SE Asia ops practices for those 100 years without having to make any radical changes to the rail network beyond the tunnel build itself and enough gradual electrification build-out to get an optimal majority-EMU / minority- dual-mode push-pull fleet mixture to cover every situation.