Green Line Reconfiguration

[737900er]

Beautiful work, as always! Quick question: with the revival of the A branch, is the 57 necessary? I could see a short-turned bus pinging between Watertown and Oak Sq (if the A terminates at Oak), but would it really be necessary to have a surface transitway on Comm for it? Not saying it's a bad thing to have the redundancy. I could definitely see a similar treatment being useful on Huntington for the 39, unless of course the branch all the way to Arborway is revived.

I think you have to at least study going back to Watertown instead of terminating at Oak Square. The challenge is just getting across Newton Corner and the Pike. The rest is fairly easy. The whole corridor is worthy of BRT or LRT service.

Watertown also comes with a yard unlike Oak Square. In the Green/Gold scenario, Reservoir and Lake Street don't have the capacity to support all the Green services. I don't think we want a lot of interaction between the divisions to staff the A from Brickbottom or Riverside.

 

[Teban54]

I think you have to at least study going back to Watertown instead of terminating at Oak Square. The challenge is just getting across Newton Corner and the Pike. The rest is fairly easy. The whole corridor is worthy of BRT or LRT service.

Watertown also comes with a yard unlike Oak Square. In the Green/Gold scenario, Reservoir and Lake Street don't have the capacity to support all the Green services. I don't think we want a lot of interaction between the divisions to staff the A from Brickbottom or Riverside.

These are some good thoughts, but the issues I suggested earlier about reliability and road width remain.

"The rest is fairly easy" -- easy if you're willing to have LRT/BRT (or "not BRT") running in mixed traffic, but not easy at all if you want dedicated transit lanes.

In fact, this analysis made me a lot less optimistic about A branch resurrection at all (even to Oak Square) than I was. The roads are constrained to 46' (east of Oak Sq) and 43' (west of Oak Sq) at the bottlenecks. Boston Transportation Department's requirements are 11' for bus lanes, 10' for travel lanes, and 5' for bike lanes (preferably 6'). These would add up to 52', and I don't see an easy way to cut it down to 46'. Not to mention platforms, and the need to eliminate all parking.

If yard space is the concern, there are other solutions, and just Watertown Yard alone won't be able to support the entire GLR fleet anyway (not to mention fleet required for the extra distance from Oak Sq to Watertown). My preliminary guess is that eminant domain may get you yards in the following places:

• Everett, along Eastern Route
  • I actually think this is necessary for any Urban Ring proposal, regardless of mode choice and regardless of interlining with GL. It's one of the few places along the route that can support a yard.
• More space in Inner Belt
• Needham Industrial Park
• Alewife, along Fitchburg ROW (with a D branch extension)
• Around Albany bus garage and/or Widett Circle, if the I-93 El is built?
• Riverside expansion

So yeah, I'm not too concerned about yard space. Put it another way, the fleet increase from GLR will likely exceed Watertown Yard's capacity, anyway.

 

[Teban54]

While we're on this topic: How many people think it would be a good idea to have a dedicated "Design Your Urban Ring" thread?

Right now, such discussions are most frequently in Crazy Transit Pitches, and sometimes spill over to this GLR thread (as is the case here). Similar to how the GLR thread got split from Crazy Transit Pitches (see Page 1 of this thread), I feel that separating this rather concrete idea into its own thread will not only help concentrate its discussions, but also free up Crazy Transit Pitches for other innovative proposals so that they don't get buried.

 

[TheRatmeister]

While we're on this topic: How many people think it would be a good idea to have a dedicated "Design Your Urban Ring" thread?

Right now, such discussions are most frequently in Crazy Transit Pitches, and sometimes spill over to this GLR thread (as is the case here). Similar to how the GLR thread got split from Crazy Transit Pitches (see Page 1 of this thread), I feel that separating this rather concrete idea into its own thread will not only help concentrate its discussions, but also free up Crazy Transit Pitches for other innovative proposals so that they don't get buried.

We could just use this one. (As a sidenote, I love how this forum already has a thread for every infrastructure project imaginable, it's great.)

 

[Riverside]

We could just use this one. (As a sidenote, I love how this forum already has a thread for every infrastructure project imaginable, it's great.)

Good goddamn, that's a throwback. And worth noting that @vanshnookenraggen's original post lays out many of the key principles of the Green Line Reconfiguration and some of the specific key proposals... 16 years ago.

But yeah, I think that would be a great thread to revive. @Teban54, you could post that excellent diagram of all of the possible SW UR alignments -- I think that lays out one of the key areas for debate quite well. If you do post there, I'll probably save some of my comments for there. In the meantime, let me respond to some of the Green Line etc questions...

 

[Teban54]

@Teban54, you could post that excellent diagram of all of the possible SW UR alignments -- I think that lays out one of the key areas for debate quite well. If you do post there, I'll probably save some of my comments for there.

Done!

 

[Riverside]

Huh, I thought I'd posted that diagram before but maybe not! Fun to see all the interest.

@Riverside Very nice map, super interesting subject. If you don't mind me asking I'm curious what program you used. My thoughts...

• Comm ave subway is very attractive to me, but six tracks with BRT? Is this just to avoid interlining completely?
• BU could really use a less wide comm ave. The highest use of comm ave as designed is with a subway and a tree lined mall down the middle or on one side. If it was multiple tracks you could revive the A and take away the 57. They instantly get their campus back as well.
• Doubling back at kenmore over digging a new subway provides some really nice frequency to that b branch trunk (if only it were more dense) but it seems like it would probably add more than a few minutes to the trip?
• While it's a tight fit, if we were trying to go without or with only minor tunneling, Given that the pike may be decked there, there's room between mountfort st and the pike for some kind of crazy viaduct, or even move mountfort over the commuter rail tracks / take a row of brownstones and put the trains down Park drive. As long as we avoid dorms, we can and should be taking some land in brookline

I think @Teban54 responded to most of this, but a couple of other thoughts:

6 tracks: yeah, this is where I should probably distinguish between proposing and provisioning. Do I think there is an urgent need for 6 tracks + surface busway? No, definitely not. But the idea is to make sure it would be possible if needed, hence making sure there are provisions for it. Would it be needed? Well, I think there are indeed multiple corridors superimposed here -- somewhat forced into the same location by the physical geography of the area (particularly the Charles River):

1. A radial heavy metro corridor radiating out to the west, toward Allston and eventually Watertown (Blue)
2. A light metro "streetcar-subway" corridor, of a type now rare, but exemplified by today's B and C and any revived A (Green)
3. A circumferential service or two, from Kendall and/or Harvard (Bronze)
4. A semi-local surface service along the 57 corridor, terminating at Kenmore

Now, you definitely can consolidate those corridors into single services. For example, a robust LRT subway from Kenmore to West Station or Packard's Corner would address items 1, 2, and 4 well enough. But, in a "full build" these are all occupying different niches, so I want to keep the option in the mix

The other thing to keep in mind: from what I can tell, the grade changes that would be necessary for a radial Green service vs a circumferential Bronze service actually mean they'd have very little potential overlap. Bronze needs to reach a surface portal at the west and needs to dive down underneath the Mass Pike + Beacon St Subway at the east -- basically descending as it goes east. Green needs to dive underneath the Pike at the west end and needs to reach the relatively shallow Kenmore station at the east -- ascending as it goes west.

Returning Comm Ave to BU: honestly this is a really interesting point that hasn't really come up before, but I think you're right that this could be a good selling point

Kenmore doubleback and using the Pike + Mountfort: Teban54 covered most of my thoughts here -- I'll probably circle back later with a couple further points

 

[Riverside]

Beautiful work, as always! Quick question: with the revival of the A branch, is the 57 necessary? I could see a short-turned bus pinging between Watertown and Oak Sq (if the A terminates at Oak), but would it really be necessary to have a surface transitway on Comm for it? Not saying it's a bad thing to have the redundancy. I could definitely see a similar treatment being useful on Huntington for the 39, unless of course the branch all the way to Arborway is revived.

Riverside and F-Line are both of the opinion that a resurrected A branch should only go as far as Oak Square, and I agree.

• The biggest reason is again reliability. Oak Square to BU Bridge already has 3.1 miles of street running (only 0.3 miles shorter than the B would be once BU Central is buried). Running to Watertown Yard adds another 1.5 miles, and involved tricky traffic conditions when crossing the Pike, at which point it really becomes too long for me. Remember, the old A branch was chosen to be terminated (when the D was launched and there was a shortage of cars) exactly because it was the least reliable and had too much street running.
• Demand patterns on the 57 also support it. Bus route profile for the 57 shows that while many people board at Watertown Yard and some at Newton Corner, the remaining stops along Tremont St and Galem St are relatively light until the bus hits Oak Square. It's not clear to me whether the Watertown and Newton Corner riders are taking it all the way to Kenmore and downtown, but even if they were, there may be better solutions for them (e.g. improving 504 and making it cheaper, commuter rail Newton Corner infill, improving 70 and 71 for Watertown).
• The roads may just be too narrow for dedicated transit lanes. Parts of the A's route east of Oak Square already concern me, where the road is often 46' wide and not enough to accommodate 4 vehicle lanes and 2 bike lanes. (I assume F-Line's A branch proposal may have been before dedicated bike lanes.) But it gets even worse on Tremont St further west, where it's often just 43' wide. (To be fair, that portion also doesn't have dedicated bike lanes which may make it easier, but there's no reason to believe it won't ever get bike lanes; the neighborhood also seems lower density and more car-dependent, so getting rid of parking will be a hard sell.)

If the A only goes to Oak Square, the need for the 57 bus obviously remains.

(Edit: I realized I completely missed the part of @BosMaineiac 's question about shortening the 57 to Watertown-Oak Sq, vs. Running it full-length to Kenmore. But I think it's not clear either, because I suspect there's a decent demand from Watertown/Newton Corner to east of Oak Square, such as Brighton, Allston and BU. A forced transfer isn't ideal. But yeah, definitely worth the discussion. )

In fact, this analysis made me a lot less optimistic about A branch resurrection at all (even to Oak Square) than I was. The roads are constrained to 46' (east of Oak Sq) and 43' (west of Oak Sq) at the bottlenecks. Boston Transportation Department's requirements are 11' for bus lanes, 10' for travel lanes, and 5' for bike lanes (preferably 6'). These would add up to 52', and I don't see an easy way to cut it down to 46'. Not to mention platforms, and the need to eliminate all parking.

Yeah so... yeah, I kinda am less of a fan of resurrecting the A (and the Arborway Line) than I used to be. But first, yes -- if A to Oak Sq is rebuilt, then I'd keep the 57 running Watertown Yard <> Kenmore, providing feeder service to a Regional Rail station at Newton Corner and to Kenmore, and running local service on Comm Ave above a limited stop subway. I'd argue in favor of bus lanes for this new iteration of the 57, but I can see benefits to turning that space back over to the public.

But yeah -- the A Line is a high ridership corridor in a higher density area of Boston. It definitely merits some sort of light metro service (LRT or BRT), but that doesn't mean it needs to be a branch of the Green Line. What does it mean to be a branch of the Green Line? It means you get a one-seat ride from a surface transit station (available every few blocks) directly into downtown.

And... well, the fact that the B and C still offer this particular boutique service is a weird one. The fact that one of Greater Boston's most affluent communities has maintained its maximally convenient transit connection to downtown while almost every other similar service (57, 22, 23, 28, 16, 71, 73, 77, 116/117) has been cut back to a Kenmore-like transfer station... yeah, that feels pretty damn weird to me.

The A will be hard to resurrect in a modern reliable form. The A will introduce extra complexity into the LRT subway system. And the A would double back down on a transit model that will almost certainly remain unique to Allston/Brighton/Brookline (and maybe Chelsea/Everett, oddly enough), a model that is difficult not to see as inequitable. So... yeah, I'm not the biggest fan.

General thought about terminating at Kenmore: probably more of my comments here will go into the Redesign the Urban Ring thread, but here I will note that a lot of the reasons to short-turn services at Kenmore would also point to short-turning them at BU instead. A Blue extension to Kenmore would not be hugely difficult to extend further to BU. The new subway could be constructed to have turnbacks built in. And the existing reservation at street level (including the current turnback at Blandford St) could short-turn surface streetcar lines. And, putting the center of gravity at BU instead of Kenmore also solves the problem of having a circumferential service divert all the way to Kenmore -- a transfer hub at BU would be much better positioned.

 

[Riverside]

Briefly doing what might be the oldest necro quote in board history, of Van's original post in the Redesign the Urban Ring thread, which, like I said, really does lay out a lot of the principles of the GLR concept -- although with much much more aggressive capital projects:

All of these new lines running through the central subway would put tremendous pressure on an already crowded system. The most important part of this whole rethinking is to rebuild the central subway.


This is probably a good idea regardless of whether any ring lines are added. What you see here is a track diagram of what the central subway could look like (click here to see what it looks like today).

Briefly commenting piece by piece:

Here's what would happen:
We would take pressure off the Boylston St tunnel by redirecting D and E branch traffic by one of two ways:
A - Build a new tunnel under the current Boylston St tunnel, eliminating the Copley Junction which is the major choke point in the Green Line (all B and C trains need to stop so an E can pass first). There would be new platforms under Copley station but no new platforms at Arlington.

We still propose eliminating Copley Junction, but with an approach more like Option D below. But I think this does draw attention to a feature of the current GLR concept that we take for granted: other than signal and platform improvements, and potentially aside from some crossover installations at Park St, the existing Central Subway does not get touched at all. There are no rebuilds, no underpinning, no modifications to existing tunnels. For a project that is geared in part at increasing capacity, I think that's actually pretty notable.

OR

D - Eliminate the Copley Junction by building a new tunnel under Stuart St through the Back Bay. This new subway would have two new stations and connect to the central subway using the abandoned tunnels south of Boylston St station.

This is closer to the current proposal, and leans on what was, at the time, the more common proposal to build a subway under Stuart St. But, again interestingly, this proposal would require modifications to the (now disused) Tremont Street Subway, to add an additional flying junction. The current proposal takes advantage of the existing flying junction at the Pleasant St Portal, once again eliminating the need to modify existing infrastructure.

Both of these plans envisions the D branch connecting to the Huntington Ave subway which itself would be extended under Huntington Ave to Brookline.

The D-E Connector has been key since the beginning, and unites virtually all GLR iterations, from cheapest to most exorbitant.

All of this hinges on the most important element of the plan,
B - Completely rebuild the Boylston St junction and station. Build two tracks that bypass the station and go directly to Park St (lines in black). Also take the two tracks that are there now and drive them under the current station, creating a new platform. This tunnel would then continue to South Station and on to South Boston, there by connecting the hotels, offices, and convention center of the new SBW with the hotels, offices, and convention center of the Back Bay (through the Financial District!) A new Boylston St Upper station would then serve all traffic either from Newton and JP (via the rerouted D and E branches) and/or a new tunnel under Washington St to Dudley Sq (and beyond!)

The last part of the plan is to correct a problem that the Green Line has suffered from since it's inception.
C - Build a second, two track tunnel from Park St to Government Center. Rebuild GC with two parallel platforms and a new loop for the outer tracks, while the inner tracks continue to Haymarket. The Brattle Loop would be sunk and a new platform for the Blue Line would allow passengers from the Blue Line to connect directly northbound Green Line service, which will undoubtedly pick up after the West Medford extension is completed.

I mean, I gotta say, even 16 years later, this plan slaps. It would be wildly expensive and even with fantastic sums of money would probably be impossible in places, but, just, I mean, damn. Grade separating the Brattle Loop is something that I haven't thought about in a long time, and is piquing my curiosity. I don't rely on the Brattle Loop as much as I used to in my designs, and I'm not sure that grade separation actually buys you that much, but it's still interesting to consider.

(Also, shoutout to calling it the "new South Boston Waterfront", before the "Seaport" branding took hold.)

Overall, I'd argue this original plan had three key pieces, two of which we address directly today:

1. Eliminating Copley Junction
2. Redirecting one or more services away from Park St to the Seaport
3. Expand capacity between Park St and Government Center

It's that last one that has proven to be immovable. Just nowhere to put a tunnel that doesn't blow a lot of stuff up. But we've tried to address Park <> GC capacity with other, more indirect measures:

• Reduce the needed capacity by terminating some services at Park
  • This has the additional benefit (we hope) of increasing the reliability of the through-running services, which effectively buys you a little more capacity
• Moving away from running "Ring" services through the Central Subway
• Extend the Blue Line to Kenmore to redirect transfers, on top of Blue-Red's role in reducing pressure
• More heavily emphasizing the Seaport as a load-bearing destination

I should also say: Van published multiple later iterations of this concept (including several that we have coalesced around today, IIRC), and I'm pretty sure he would not support the more extravagant elements of this plan today. This post isn't meant to criticize these early ideas (certainly not coming from me, since I've proposed some bonkers stuff in the past), but rather to link back a lineage of ideas, even if the details have changed.

 

[Teban54]

Here are my comments on both Van's initial proposal (in the UR thread but really talking about GLR more than UR), and Riverside's comments above.

Engineering and detailed route design

This is closer to the current proposal, and leans on what was, at the time, the more common proposal to build a subway under Stuart St. But, again interestingly, this proposal would require modifications to the (now disused) Tremont Street Subway, to add an additional flying junction. The current proposal takes advantage of the existing flying junction at the Pleasant St Portal, once again eliminating the need to modify existing infrastructure.

While the consensus has shifted towards a Marginal St subway now and for good reason, I don't think Stuart St has ever been fully off the table. We're speculating that Marginal will be easier to build, but Stuart will always be a backup option and probably Alt #2 in any real-world plans. Even F-Line in 2021 completely agreed with this.

A Stuart-Kneeland alignment from Back Bay all the way to Seaport will likely see less certain utility relocations (though Marginal isn't completely without utility either, they're just better mapped), and of course, has the problem of needing to modify the Tremont St tunnel. But Marginal has its own risk factors:

• Much narrower than Stuart, often with decently tall buildings. (That section has always concerned me.)
• Needing to interfere with OL/CR Back Bay station. While it offers a better transfer, it does bring engineering uncertainties, and possible political uncertainties dealing with MassDOT for digging under the Pike.
• Dealing with Hudson St and/or the Big Dig tunnel (Dewey Square tunnel), which I've been increasingly concerned about -- even though I do think multiple alternatives exist.

Some of these are reasons why I occasionally lean towards Stuart now, especially for the eastern half (Kneeland).

But I think this does draw attention to a feature of the current GLR concept that we take for granted: other than signal and platform improvements, and potentially aside from some crossover installations at Park St, the existing Central Subway does not get touched at all. There are no rebuilds, no underpinning, no modifications to existing tunnels. For a project that is geared in part at increasing capacity, I think that's actually pretty notable.

In particular, the magical moment is when you realize that (1) Boylston-Park has 4 tracks, but we only use two; (2) Pleasant St Incline, an entire 1200' tunnel with its own freaking flyover and a former portal, just exists right there. Honestly, these two reasons are why I'm utterly surprised with how attention to GLR is almost entirely limited to this forum, especially due to (1). I mean, has no one else looked at the track diagram and thought "maybe we could use the quad-tracked Boylston-Park somehow"?

(Though, to Reddit's credit, someone was talking about "Stuart St realignment" the other day, presumably regarding the same idea.)

I mean, I gotta say, even 16 years later, this plan slaps. It would be wildly expensive and even with fantastic sums of money would probably be impossible in places, but, just, I mean, damn. Grade separating the Brattle Loop is something that I haven't thought about in a long time, and is piquing my curiosity. I don't rely on the Brattle Loop as much as I used to in my designs, and I'm not sure that grade separation actually buys you that much, but it's still interesting to consider.

While it should be very obvious that any proposal for quad-tracking Park-GC under Tremont St (50' building-to-building between 10-story-high buildings) is effectively a non-starter... I must say, the idea of it is still absolutely wild. And the benefits of quad tracking here are also pretty obvious: Making the Central Subway more like a "real" line, instead of a "stub" that terminates at Park St. The need to turn all of them at Park has really bothered me, especially when considering that GC (part of "Greater Haymarket") itself is a major employment center and tourist destination.

If grade-separating Brattle Loop is your concern, though: Can we bulldoze C&C under City Hall Plaza to straighten out the Haymarket-GC segment, like this?

Probably still very expensive and disruptive, but it has two main benefits:

• Easily set up 4 tracks towards GC that are much closer together than before: you can then have 2 inner tracks terminating at GC, resurrecting Brattle "Loop", while 2 outer tracks continue to Park
• Straightens the tight curves around Haymarket and GC: not just the non-revenue 50' Brattle Loop, but also the 60' revenue track southbound entering GC
  • May even allow far-future HRT conversion of GLR (depending on North Station-Science Park curves)

More abstract route design (GLR only)

This tunnel would then continue to South Station and on to South Boston, there by connecting the hotels, offices, and convention center of the new SBW with the hotels, offices, and convention center of the Back Bay (through the Financial District!)

Gotta be honest, there was a point recently where I was doubting the value of a Back Bay-South Station-Seaport connection. Especially when I was wondering if commuter rail riders needing a transfer to Back Bay may simply take another commuter rail train instead (especially with NSRL).

But this totally convinced me, and this way of explaining the importance of a Back Bay-Seaport route is understated. (I've certainly seen visitors living in Back Bay hotels who want to visit Seaport, but gave up because of how tedious the transit trip is.) And Van's comment didn't even begin to consider LMA.

Overall, I'd argue this original plan had three key pieces, two of which we address directly today:

1. Eliminating Copley Junction
2. Redirecting one or more services away from Park St to the Seaport
3. Expand capacity between Park St and Government Center

It's that last one that has proven to be immovable. Just nowhere to put a tunnel that doesn't blow a lot of stuff up. But we've tried to address Park <> GC capacity with other, more indirect measures:

I myself am actually not too concerned about capacity between Park-GC, if that's the origin-destination pair we're talking about. Orange Line almost does that, and Tremont St subway can at least see the exact same number of trains under GLR as it does today. Even now, I doubt GC-Park is the capacity bottleneck: I'd think the most severe overcrowding is west of Park St.

High-level philosophy
I realize that all I wanted to say here are really comments on Urban Ring and not GLR, so I've moved them to the Urban Ring thread.

 

[Riverside]

While the consensus has shifted towards a Marginal St subway now and for good reason, I don't think Stuart St has ever been fully off the table. We're speculating that Marginal will be easier to build, but Stuart will always be a backup option and probably Alt #2 in any real-world plans. Even F-Line in 2021 completely agreed with this.

Yeah I didn't mean to suggest that Stuart St is off the table now. More I just meant that, at the time, Stuart was really the only alignment that was discussed.

A Stuart-Kneeland alignment from Back Bay all the way to Seaport will likely see less certain utility relocations (though Marginal isn't completely without utility either, they're just better mapped), and of course, has the problem of needing to modify the Tremont St tunnel. But Marginal has its own risk factors:

• Much narrower than Stuart, often with decently tall buildings. (That section has always concerned me.)
• Needing to interfere with OL/CR Back Bay station. While it offers a better transfer, it does bring engineering uncertainties, and possible political uncertainties dealing with MassDOT for digging under the Pike.
• Dealing with Hudson St and/or the Big Dig tunnel (Dewey Square tunnel), which I've been increasingly concerned about -- even though I do think multiple alternatives exist.

Some of these are reasons why I occasionally lean towards Stuart now, especially for the eastern half (Kneeland).

I agree with you in general, but am confused about the first point. I agree with you that there are some buildings along Marginal that would be concerning, but (setting Hudson aside) I think the one at your link is the only one that really directly abuts? Plus what you'll need to do around Dartmouth St (see below).

Regarding Back Bay, from what I can tell the only interface with the Mass Pike would be building the spur off of the Huntington Ave Subway (where I think it would be necessary to extend the existing tangent track further northeast). There are some large buildings to maneuver around there, for sure, but a Stuart St alignment will have the same problems. As for Back Bay station itself, I would plunk the station down under/replacing the parking garage, build an enclosed connection over the Pike to the existing station and be done with it -- that seems relatively uncertain to me?

I wonder if an eastern Stuart Street alignment could be built off of the extant flying junction at Pleasant Street:

(Getting to Back Bay from Huntington is short but a real headache. The eastern half, in purple, shouldn't be that bad, as the distance between buildings is wide, and even the distance from the wall of the Mass Pike is wide. The western half, in blue, will require tunneling under the Pike, which will be harder. But the real pinch point I see is where I've marked the star, where it looks like there may be less than 40 feet between the wall of the Pike and the edge of the Westin.)

If grade-separating Brattle Loop is your concern, though: Can we bulldoze C&C under City Hall Plaza to straighten out the Haymarket-GC segment, like this?

Yeah, it's an interesting idea. Not very high on my priority list, but interesting all the same.

Probably still very expensive and disruptive, but it has two main benefits:

• Easily set up 4 tracks towards GC that are much closer together than before: you can then have 2 inner tracks terminating at GC, resurrecting Brattle "Loop", while 2 outer tracks continue to Park
• Straightens the tight curves around Haymarket and GC: not just the non-revenue 50' Brattle Loop, but also the 60' revenue track southbound entering GC
  • May even allow far-future HRT conversion of GLR (depending on North Station-Science Park curves)

While it wouldn't help with quad-tracking, what would prevent Government Center from being simplified like this?:

Eliminates the tight Haymarket turn entirely, but I think someone mentioned that there are foundation pillars or something like that, which would block it?

Doesn't solve the Brattle Loop problem, of course, but like I said, I'm less convinced of the need for that than I used to be. Though, looking at the above, perhaps you could squeeze three tracks in there, make the center one the terminating track in both directions:

I need to go back and look at the diagrams I found.

More abstract route design (GLR only)

This tunnel would then continue to South Station and on to South Boston, there by connecting the hotels, offices, and convention center of the new SBW with the hotels, offices, and convention center of the Back Bay (through the Financial District!)

Gotta be honest, there was a point recently where I was doubting the value of a Back Bay-South Station-Seaport connection. Especially when I was wondering if commuter rail riders needing a transfer to Back Bay may simply take another commuter rail train instead (especially with NSRL).

But this totally convinced me, and this way of explaining the importance of a Back Bay-Seaport route is understated. (I've certainly seen visitors living in Back Bay hotels who want to visit Seaport, but gave up because of how tedious the transit trip is.) And Van's comment didn't even begin to consider LMA.

Yeah, this has been my "thing" for a while, and is part of why I chose to "zoom in" on the Gold Line for the Transportation Dreams contest.

 

[Riverside]

Does anyone know what is currently in the bulge/gap between the Green Line tracks just north of Haymarket?

Here's what I have:

The Green Line tracks at the portal were relocated several times during the Big Dig, so I'm not 100% sure. My best guess is that of the original six-track wide portal, the current GL tracks are below the western two slots, the current OL tracks below the eastern two slots, and some portion of the incline remains in the center two track slots, which were last used to access the Causeway Street Elevated from 1997 to 2004.

 

[HenryAlan]

Like, South Station actually is pretty darn close to the Green Line -- about 2500 feet between South Station and Boylston. That's just not that far.

Yes, it's very walkable. I live on the Orange Line, and when I go to South Station, I usually walk from Chinatown, which is of course ridiculously close to Boylston. I find that often faster than going another stop to DTX, then potentially waiting for a Red Line train to go a single, short distance stop from there.

In fact, this analysis made me a lot less optimistic about A branch resurrection at all (even to Oak Square) than I was.

I think realistically, an A-branch restoration as far as St Elizabeth's might be a good and feasible idea. But once you hit Washington St, it becomes much more complicated and ineffective for rail transit.

 

[TheRatmeister]

Does anyone know what is currently in the bulge/gap between the Green Line tracks just north of Haymarket?

Wikipedia claims that it's used for storage and electrical equipment, but I can't find a source for that.

 

[TheRatmeister]

"The rest is fairly easy" -- easy if you're willing to have LRT/BRT (or "not BRT") running in mixed traffic, but not easy at all if you want dedicated transit lanes.

In fact, this analysis made me a lot less optimistic about A branch resurrection at all (even to Oak Square) than I was. The roads are constrained to 46' (east of Oak Sq) and 43' (west of Oak Sq) at the bottlenecks. Boston Transportation Department's requirements are 11' for bus lanes, 10' for travel lanes, and 5' for bike lanes (preferably 6'). These would add up to 52', and I don't see an easy way to cut it down to 46'. Not to mention platforms, and the need to eliminate all parking.

I think realistically, an A-branch restoration as far as St Elizabeth's might be a good and feasible idea. But once you hit Washington St, it becomes much more complicated and ineffective for rail transit.

I don't think this is necessarily true. You could at least get to Brighton with a loop on Washington/Market/Henshaw. Beyond that is where it gets tricky. You could maybe do another loop along Langley Rd, mixed traffic for this part would probably be fine, but that would still likely require either widening the street or cutting back on about half the street parking. With both loops you could run <5 min headways even with single track between Langley/Market, assuming average speeds of 7MPH as seen on the C branch.

Another option would be to build Sparhawk/Arlington/Fanueil into a strong bike corridor, and providing bike lanes on streets that cross Washington wherever possible, potentially allowing you to skip bike lanes on Washington St itself which would leave room for 2 travel and 2 transit lanes. Alternatively Washington could have bike lanes but be made one-way only for cars, but for this Sparhawk/Arlington/Fanuiel would need to take some of the load, which would likely be quite unpopular given that the corridor is primarily residential.

In terms of platforms, the options would either be pick a solution that leaves enough leeway for platforms, cut into sidewalks a bit around platforms, or run the transit lanes along the edge, allowing for slightly raised platforms to do double duty as sidewalks. The last option would require things like fencing/barriers/green track to stop cars from accessing the transit lanes, which generally would not be necessary with median-running transit lanes.
Washington St would still lose basically all street parking with any of these plans though, so that remains a problem. In terms of changing public opinion, open street days remain a valuable tool for demonstrating that foot traffic alone can sustain businesses.

 

[RandomWalk]

That bulge is also where the original Haymarket platforms were. Until the superstation was built, they were mostly intact.

 

[The EGE]

The platforms are still there, with some storage use.

File:Old Haymarket station from northbound train, July 2019.webm - Wikimedia Commons

commons.wikimedia.org

 

[Riverside]

The platforms are still there, with some storage use.

File:Old Haymarket station from northbound train, July 2019.webm - Wikimedia Commons

commons.wikimedia.org

God I love the internet so much. "Hey what's currently located in this random stretch of semi-abandoned tunnel in the US's 25th largest city?" "Oh here's a video of that exact spot."

The next question I have is, how much of the stuff currently located there can be relocated? More to the point, could one or two turnback tracks be added there, potentially as a replacement for the Brattle Loop? I'm a little concerned about fitting a full-length supertrain but it looks vaguely doable.

 

[The EGE]

There's probably enough room for a single pocket track, but I doubt there's room for two with a crossover.

 

[TheRatmeister]

There's probably enough room for a single pocket track, but I doubt there's room for two with a crossover.

The whole line used to be quad-tracked there, I don't see why not. (Especially if you're willing to demolish the old platforms)