Green Line Reconfiguration

[The EGE]

Yes, those parcels were intentionally acquired by the MBTA as a future ROW. I have to wonder if it was even before the Urban Ring - the West Village buildings were constructed around 2000, before the Urban Ring was in full planning.

 

[Riverside]

Oooh that's very interesting. Yes, assuming you could thread a sidewalk through those trees (which, I agree, looks totally doable and quite pretty), then you can free up the 6 feet of width in that 50-foot current footprint, which can easily fit two travel lanes and two bus lanes:

Still might need to do some fussing on the Huntington-Parker block to the west, but that seems a little more doable.

So, yes -- if we can get bus lanes on Ruggles, then by all means, funnel everything through there. It's definitely better overall than Forsyth is.

 

[Teban54]

Not sure I explained this earlier, but part of the LMA Transitway concept includes converting Forsyth St into a transit mall (combined with some modifications to Ruggles Station to enable through-running at surface level).

Forsyth has relatively few connections to the external street network. If Northeastern really threw a fit, we could probably figure out a compromise where limited local traffic is also allowed, but it seems generally reasonable to reallocate that streetspace to transit.

I'm not familiar with Northeastern campus, but based on my experience of other college campuses, Forsyth St looks like the kind of campus streets that are inundated with students crossing the street pretty much everywhere possible. Google Streets View also shows tons of on-street parking and bike lanes near junctions, which further reduce the width available to buses.

So yeah, the Ruggles St alignment assuming realigned sidewalks seems like a clear better option.

 

[Downburst]

I'm not familiar with Northeastern campus, but based on my experience of other college campuses, Forsyth St looks like the kind of campus streets that are inundated with students crossing the street pretty much everywhere possible.

As a double Husky (retired) I can attest that this is absolutely the case, and that cars (or LMA shuttles, for that matter) barely move on Forsyth during inter-class periods. For all the reasons mentioned above and more, Ruggles St. is preferable.

 

[Stlin]

Still might need to do some fussing on the Huntington-Parker block to the west, but that seems a little more doable.

Take it with a large grain of salt, since I can't verify it, but I was once told that the Wentworth Field property line on the North side of Ruggles Street doesn't extend to the sidewalk, but instead only to the line of its fence. If true, that gives you/the city ~6ft of extra space to relocate the sidewalk.

Also, unrelatedly, the discontinuity of Ruggles St across Tremont is actually quite annoying now that I'm looking at it on a map. During any future rebuild, can we choose just one name for this segment that, depending on the block, is Louis Prang - Ruggles - Whittier - Cabot?

 

[Riverside]

Take it with a large grain of salt, since I can't verify it, but I was once told that the Wentworth Field property line on the North side of Ruggles Street doesn't extend to the sidewalk, but instead only to the line of its fence. If true, that gives you/the city ~6ft of extra space to relocate the sidewalk.

Yeah I had my eye on that fence; even if the city doesn't own the strip that would be needed for a relocated sidewalk, the need seems modest enough that something could probably be worked out with Wentworth.

And good feedback all around from everyone on Ruggles vs Forsyth! Consider me converted; I've adjusted my map.

@The EGE I'm still thinking about your point about frequency vs signal prioritization, in the context of running the Grand Junction service (my J) to Ruggles (now with a slightly more direct route). I'm thinking in terms of "full" vs "half" frequencies -- let's say "full = 15 tph = 4min freq" and "half = 7.5 tph = 8min freq".

By branching the Seaport (Magenta) service at MFA, the burden on the LMA Transitway is reduced to half frequencies. (Let's set aside the Silver Line for the moment.) My Gold Line already operates at "half frequencies" because it shares the Grand Junction with the G branch of the Green Line. So if we extended it back down through the Transitway to Ruggles, the Transitway would only need to accommodate ~4 minute headways (cumulatively a single full frequency service).

Do you think that's a modest enough frequency to quell concerns about signal priority?

 

[The EGE]

That's still a train every 2 minutes through any given signal; depending on the cycle length, that's a train every 1-2 cycles. That means you can't be very aggressive with TSP, probably just a few seconds of early or extended green. Something like full preemption is right now, since that takes several signal cycles to return back to the normal coordination.

That's not the end of the world depending on how your signal system is set up. If you have consistent headways entering the transitway, consistent dwell times, that makes it easier to set up signal progressions that allow trains to proceed through with few delays. There some very good surface light rail systems out there with very consistent running times because every single train moves exactly X blocks per cycle.

At the end of the day, the question becomes what tradeoffs are you willing to make? There is no free lunch with signal timing - you have 3600 seconds per hour, and giving some of them to one mode or one route inevitably reduces the amount you can give to others. Two of the main tradeoffs you end up with on surface transitways like here are:

• It's extremely difficult to have coordination on crossing streets. Even if they're on the same system (ie all the traffic signals are synced to have the same cycle length), there are going to be conflicts. The more time/priority you give to transitway trains at Audubon Circle, the less control you have over signal progression on Beacon. Same with Brookline and Huntington. If the cross street is on a different system (more likely) then it will essentially be crossing the transitway at a random point in the cycle, without a timed progression to minimize its wait.
• Similarly, there's a direct tradeoff between pedestrian crossings and transit priority. TSP is limited by the length of pedestrian cycles: you can curtail the white hand and skip to the countdown for TSP, but you can't shorten the countdown. (In most cases, the green time for minor side streets is determined by the pedestrian countdown length, not the green needed for traffic.) The fewer pedestrian crossings you have, and the longer green you have for the trains, the better they will perform. But the fewer pedestrian crossings you have, and the longer they have to wait, the more you discourage walking (and encourage jaywalking).

That's why my crayon design has grade-separated ROW across the LMA, even though it would come at massive cost. Better a few years of disruptive construction than have to make those tradeoffs forever.

 

[Teban54]

That's why my crayon design has grade-separated ROW across the LMA, even though it would come at massive cost. Better a few years of disruptive construction than have to make those tradeoffs forever.

How feasible is that, though? IIRC, in previous discussions of Urban Ring, F-Line insisted that any subways for the SW quad have to go through Kenmore and that digging under a route like BU-Park Dr would be infeasible. Can someone still find those posts?

 

[Brattle Loop]

How feasible is that, though? IIRC, in previous discussions of Urban Ring, F-Line insisted that any subways for the SW quad have to go through Kenmore and that digging under a route like BU-Park Dr would be infeasible. Can someone still find those posts?

This looks like it might be one of them, and another in the discussion on Page 10 of this thread, and an older reference here.

F-Line's thinking may have hardened on this over time, I don't know, but the impression I got was that the Park Drive route had serious physical blockers (and flood risk if it had a portal around Fenway. I don't recall if F-Line was of the opinion that all southwestern quadrant routes from BU were impossible to tunnel, or just so outrageously expensive that they'd auto-fail a cost-benefit analysis in a state still somewhat traumatized by the expense of the Big Dig.

 

[F-Line to Dudley]

This looks like it might be one of them, and another in the discussion on Page 10 of this thread, and an older reference here.

F-Line's thinking may have hardened on this over time, I don't know, but the impression I got was that the Park Drive route had serious physical blockers (and flood risk if it had a portal around Fenway. I don't recall if F-Line was of the opinion that all southwestern quadrant routes from BU were impossible to tunnel, or just so outrageously expensive that they'd auto-fail a cost-benefit analysis in a state still somewhat traumatized by the expense of the Big Dig.

These were the same neighborhoods that turfed the I-695 Inner Belt tunnel 50 years ago, so there's ample precedent that it's going to be hard to nearly impossible getting through Brookline. The street grid and Muddy River flood zone are enemies at getting through LMA. It would be maximally expensive...probably so maximally expensive that you'd have to work on a very generous installment plan. And that's going to mean an Urban Ring that initially must interface with Kenmore (where you can at least do it easily by burying the B reservation from BU Bridge).

 

[bigeman312]

Oooh that's very interesting. Yes, assuming you could thread a sidewalk through those trees (which, I agree, looks totally doable and quite pretty)

Not only doable to "thread a sidewalk," but in-fact that's the exact location of the proposed Rox-Fens Connector multi-use trail.

The Roxbury to Fenway Connector is a proposed Green Link. It would include a path between the existing double row of trees on Ruggles Street, between Ruggles Station and Parker Street.

In 2019, it was named as a priority by the Emerald Network group.

No apparent movement for the past three years though.

 

[Riverside]

Continuing to fiddle with this corner of the network. In creating a dual-trunk (Boylston + Huntington) LRT network, the connection between Brookline Village and Kenmore (which I'm going to call the "Fenway Branch", for lack of a better name) remains the odd-one-out. Several years ago, we considered its usefulness as part of a B-D Connection that could provide Urban Ring service in a zig-zag fashion -- something like Cambridgeport - BU - Kenmore - Brookline Village - Huntington. There were two key problems with that route: first, it required reconstruction at Kenmore to connect the B to the Kenmore Loop, and even then, it would create an extended single track segment through Kenmore, which would be a pretty major capacity constraint.

My LMA Transitway proposal was intended to provide a transit ROW that connects the northwest quadrant (Harvard, Allston, and ideally the Grand Junction) with Longwood. For a variety of historical and geological factors, tunneling between the B and E Lines would incur massive cost.

However, as I look at it more closely, this "orphan" ROW north of Brookline Village also provides a dedicated ROW that gets you halfway from the E Line (assuming a D-E Connector to Brookline Village).

Moreover, coming from a BU Bridge station, there is a path that gets you quite close to the Fenway portal: the Mass Pike. Half a mile of decking and/or elevated gets you to Beacon St (right across the street from Lansdowne, providing a Regional Rail connection with express service to downtown and Back Bay).

And then from there, it's 700 feet to the Fenway Portal, mostly through a parking lot, via surface and/or elevated running to meet the current ROW (with a Lansdowne Station centered somewhere in the vague area highlighted in white, and potentially also including a bit of space to the north, above the Pike):

From there, the Fenway Branch, instead of being a vestigial crossover in a dual trunk network, becomes the anchor of a semi-independent LRT system radiating out from Brookline Village and Longwood toward Harvard and Sullivan.

Whether you terminate at Brookline Village, turn west toward Reservoir, or east toward Huntington, is left as an exercise for the reader. (In reality, this will depend on a number of factors, and likely would evolve over time as different parts of the network are expanded.)

~~~

Is this perfect? Far from it.

• Access to Longwood is still somewhat lacking
  • This could be ameliorated by wraparound-running on to Huntington, and/or a cross-LMA BRT
  • (My idea for a Transitway on Fenway notwithstanding, I don't see a way around limiting/restricting private vehicles on Longwood Ave in the long-term -- it's just too central to the whole shebang)
• Poor connection to the C Line
  • A 5-min walk between St Mary's St and Fenway station isn't great
  • A beefed-up version of today's 65 and 66 could provide partial mitigation
  • I suppose if the flat junction in the subway under Beacon St were sacrificed, it may be possible to add a subway station on the C at Lansdowne?
• Requires a Blue extension all the way to BU rather than just Kenmore
  • This is annoying in some ways but a blessing in disguise in others; the jobs are at Boston University, not Kenmore, so Blue-to-BU has merits on its own

But I think there's a fair amount of merit here:

• Provides a "landing zone" for the LRT/northern half of the Urban Ring with partial connectivity to Longwood and strong connection to a beefed up D-E Line (which I continue to contend will be a monster)
• Provides a dedicated ROW through the troublesome "B-E Zone" that requires minimal landtaking, no tunneling, and a maximum of 5 grade crossings (Comm Ave, Mountfort, St Mary's, Beacon, Miner -- and all of those except probably for Miner could be eliminated with grade separation)
• Requires no modification to Kenmore and no impact on capacity in the Boylston St Subway
• Reduces "zig-zagging":
  • as the crow flies from a Cambridgeport station (i.e. at Waverly St) to the intersection of Huntington & S Huntington, it's 1.64 miles
  • the route via Lansdowne is 2.45 miles -- 150%, while the route via Kenmore is 3.1 miles -- a full 200%, and includes over 1,000 ft of unidirectional single tracking
• By detaching the Fenway Branch from the potentially both the Boylston and Huntington Subways, you free it up to be used at full capacity, rather than as an awkward vestigial route that briefly pretends to be circumferential, but isn't

EDIT: I should note that this idea I'm pretty sure is not original. @davem proposed something similar 8 years ago: https://archboston.com/community/threads/crazy-transit-pitches.3664/page-102#post-200449

 

[Brattle Loop]

And then from there, it's 700 feet to the Fenway Portal, mostly through a parking lot, via surface and/or elevated running to meet the current ROW (with a Lansdowne Station centered somewhere in the vague area highlighted in white, and potentially also including a bit of space to the north, above the Pike):

Umm, doesn't Children's plan to put up a big building where that parking lot is? That parcel's moving quickly towards where it'd take God Mode territory to get a station there, and Crazy Transit Pitches territory to even shiv an LRT line through there to Fenway (at least without street-running via some not-super-wide streets).

 

[Riverside]

Umm, doesn't Children's plan to put up a big building where that parking lot is? That parcel's moving quickly towards where it'd take God Mode territory to get a station there, and Crazy Transit Pitches territory to even shiv an LRT line through there to Fenway (at least without street-running via some not-super-wide streets).

Sigh.

 

[Riverside]

Umm, doesn't Children's plan to put up a big building where that parking lot is? That parcel's moving quickly towards where it'd take God Mode territory to get a station there, and Crazy Transit Pitches territory to even shiv an LRT line through there to Fenway (at least without street-running via some not-super-wide streets).

Yup, looks like it: https://www.bostonplans.org/projects/development-projects/819-beacon-street

Well, to the circular filing bin for that one.

EDIT: We could still look at creating a dedicated ROW on Park Dr between the Mass Pike and Fenway station. Only one signalized intersection, and we'd probably see only modestly high frequencies (e.g. 4 minutes). You lose the Regional Rail connection, but you'd get closer to a connection with the C. And we'd still be able to "land" our Urban Ring on the Fenway Branch.

Gosh golly though, it's almost like this kind of planning takes professional skill and experience! [*insert self-deprecating facepalm here*]

 

[Brattle Loop]

Well, to the circular filing bin for that one.

EDIT: We could still look at creating a dedicated ROW on Park Dr between the Mass Pike and Fenway station. Only one signalized intersection, and we'd probably see only modestly high frequencies (e.g. 4 minutes). You lose the Regional Rail connection, but you'd get closer to a connection with the C. And we'd still be able to "land" our Urban Ring on the Fenway Branch.

Gosh golly though, it's almost like this kind of planning takes professional skill and experience! [*insert self-deprecating facepalm here*]

I agree there's still a way to make the track connection, which is at least something.

Sufficiently long exposure to the Crazy Transit Pitches thread makes me reach instinctively for the Mass GIS property map and the Google machine whenever something comes up involving a tantalizingly-open lot or ROW, to see what mines are lurking to turn a slam dunk into a boondoggle

 

[Riverside]

I agree there's still a way to make the track connection, which is at least something.

Sufficiently long exposure to the Crazy Transit Pitches thread makes me reach instinctively for the Mass GIS property map and the Google machine whenever something comes up involving a tantalizingly-open lot or ROW, to see what mines are lurking to turn a slam dunk into a boondoggle

You know, I'd heard of that GIS map for years but never actually looked at it until today. That's simultaneously a treasure trove and a month's worth of rabbit holes right there.

In any case, here's a sketch of what this might look like system-wise:

It guts me to let go of Seaport - South Station - Back Bay - Prudential - LMA - West Station - Harvard on one rapid transit service... and I suppose technically this plan wouldn't close the door to that at some point in the future. But this iteration does at least move us away from Crazy Transit Pitches, and in some ways is "back to basics" -- perhaps a more effective simplification of the Green Line Reconfiguration concept.

The design is relatively straightforward: hand the Fenway Branch over to the Gold Line, and run both the Harvard and Grand Junction branches into it. Simplify the Magenta, and have it pick up one of the Highland services at "half frequencies" (7.5 tph). Max out frequencies in both the Huntington and Boylston Subways with 30 tph each (22.5 of which stop at Brookline Village, meaning Gold transfers are quick and easy).

(The map nerd in me is tickled to have the H Line go to Harvard -- and that there's some reasonable alphabetical logic to it!)

I've also been tinkering with the BRT layers, and I'm intrigued by some of the possibilities, but I don't love the current state:

We know the southern half of the Urban Ring is going to have to be largely BRT, if for no other reason than the Ted Williams Tunnel. One of the things I've really been trying to more fully think out is what kind of possibilities open up with BRT that are not available with LRT. Chief among those are open BRT, where buses operate partially within BRT infrastructure before spreading out on conventional streets. Obviously this is the kind of thing that can be really effective, and can also be done really badly, so it's worth thinking through.

In general, I want the southern half of the Urban Ring to run from Longwood (and maybe Kenmore) to Ruggles & Nubian and then to the Seaport and then to the Airport -- that's the primary corridor, though obviously there are a number of branches that merit attention (e.g. JFK/UMass). Hence why in my previous map I drew a single line threading its way through all of these places, with overlapping short-turn services to improve reliability.

But the discussion in the Bus/BRT thread reminded me that, Urban Ring or not, it's still going to make sense to run at least some of the Dorchester buses (eg the T22 and T28) into Longwood. This means, one, they will be contributing to Ruggles/Nubian-LMA services, and two, they will require capacity. Conceptualizing the southern Urban Ring as a partially open BRT system makes it easier to incorporate the additional routes that provide benefit to part of but not the entire corridor.

In theory.

In practice (insofar as making maps is "practice"), it gets a bit messier. What I've sketched out is supposed to be something like this:

• SL2, SL3, and SL4 are the "closed BRT network" that follow the primary corridor I mentioned above, and are intended to have separated lanes for their entire route
  • Turning buses at Longwood Station is dicey, and may be a bridge too far -- a 5-min walking transfer to a turn location on the other side of the Muddy River may be the only actual option
  • Having the Gold Line avoid Kenmore increases the pressure for a transfer at some place like Longwood
• SL1 begins at Kenmore, uses bus lanes on Brookline Ave and Longwood Ave before picking up the T22 route (and maybe the T28 -- I dunno how complicated the branding/wayfinding can afford to be), using the Columbus Ave and hopefully Blue Hill Ave bus lanes
  • SL1 would also incorporate other Dorchester buses that extend to Kenmore, under the theory that multiple routes will add up to create a legible service along the corridor
• SL6 is the BRT corridor along Brookline Ave, which would incorporate today's 60 and 65 and maybe the T66, plus one or two buses that just ping back and forth along the corridor to keep reliability up
  • This is one of the ideas I feel less confident about, but is a prime example of open BRT with feeder routes being used to create a legible corridor
  • Rerouting the Fenway Branch away from Kenmore does reduce Kenmore-Longwood connectivity, so there's some need there
  • If we're going to make Brookline Village a major transfer hub, Brookline Ave BRT would provide relief to the Huntington Subway by serving the western half of the LMA
  • And Brookline Ave is wide enough for bus lanes, so I guess why not?
• SL5 is probably a terrible idea but is meant to be bidirectional bus lanes on Boylston (or paired one-way lanes on St James & Stuart, it doesn't really matter) that connect to lanes on Kneeland which connect to the Albany St - South Station lanes proposed upthread, in order to provide a two-seat journey for A, B, and C riders going to the Seaport
  • Like SL6, it's a combination of feeder routes and a few dedicated vehicles. In this case, it's an extension of the T39 and the 55
  • I really don't love this, but am playing around with whether it could be any better somehow
  • Maybe this would be better if it leveraged the T9's bus lanes through the South End and then cut over to the Seaport from there? Ugh, I don't know -- at least doing that wouldn't require bespoke infrastructure for a service that duplicates a subway line a quarter-mile to the south

Anyway, yeah. The BRT stuff is very much a WIP. As is the first map, of course, though I feel like I've thought through those ideas better.

~~~

ALSO. Something to draw attention to, again from @davem's map from 8 years ago (it's like they say, everything old is new again): an infill station at Riverway Island (which based on Google results makes me suspect is a name generated by realtors). A station here would provide surprisingly improved access to the southern end of LMA, including Brigham & Women's, the Yawkey Center for Cancer Care, and the Beth Israel West Campus. Courtesy of app.traveltime.com, you can compare the walksheds; while there is significant overlap, Longwood station itself barely reaches Children's in its 10-minute walkshed, which means that Riverway Island's additional access to Francis St is non-trivial:

 

[Teban54]

Continuing to fiddle with this corner of the network. In creating a dual-trunk (Boylston + Huntington) LRT network, the connection between Brookline Village and Kenmore (which I'm going to call the "Fenway Branch", for lack of a better name) remains the odd-one-out. Several years ago, we considered its usefulness as part of a B-D Connection that could provide Urban Ring service in a zig-zag fashion -- something like Cambridgeport - BU - Kenmore - Brookline Village - Huntington. There were two key problems with that route: first, it required reconstruction at Kenmore to connect the B to the Kenmore Loop, and even then, it would create an extended single track segment through Kenmore, which would be a pretty major capacity constraint.

My LMA Transitway proposal was intended to provide a transit ROW that connects the northwest quadrant (Harvard, Allston, and ideally the Grand Junction) with Longwood. For a variety of historical and geological factors, tunneling between the B and E Lines would incur massive cost.

However, as I look at it more closely, this "orphan" ROW north of Brookline Village also provides a dedicated ROW that gets you halfway from the E Line (assuming a D-E Connector to Brookline Village).

Moreover, coming from a BU Bridge station, there is a path that gets you quite close to the Fenway portal: the Mass Pike. Half a mile of decking and/or elevated gets you to Beacon St (right across the street from Lansdowne, providing a Regional Rail connection with express service to downtown and Back Bay).

And then from there, it's 700 feet to the Fenway Portal, mostly through a parking lot, via surface and/or elevated running to meet the current ROW (with a Lansdowne Station centered somewhere in the vague area highlighted in white, and potentially also including a bit of space to the north, above the Pike):

View attachment 31397

From there, the Fenway Branch, instead of being a vestigial crossover in a dual trunk network, becomes the anchor of a semi-independent LRT system radiating out from Brookline Village and Longwood toward Harvard and Sullivan.

Whether you terminate at Brookline Village, turn west toward Reservoir, or east toward Huntington, is left as an exercise for the reader. (In reality, this will depend on a number of factors, and likely would evolve over time as different parts of the network are expanded.)

~~~

Is this perfect? Far from it.

• Access to Longwood is still somewhat lacking
  • This could be ameliorated by wraparound-running on to Huntington, and/or a cross-LMA BRT
  • (My idea for a Transitway on Fenway notwithstanding, I don't see a way around limiting/restricting private vehicles on Longwood Ave in the long-term -- it's just too central to the whole shebang)
• Poor connection to the C Line
  • A 5-min walk between St Mary's St and Fenway station isn't great
  • A beefed-up version of today's 65 and 66 could provide partial mitigation
  • I suppose if the flat junction in the subway under Beacon St were sacrificed, it may be possible to add a subway station on the C at Lansdowne?
• Requires a Blue extension all the way to BU rather than just Kenmore
  • This is annoying in some ways but a blessing in disguise in others; the jobs are at Boston University, not Kenmore, so Blue-to-BU has merits on its own

But I think there's a fair amount of merit here:

• Provides a "landing zone" for the LRT/northern half of the Urban Ring with partial connectivity to Longwood and strong connection to a beefed up D-E Line (which I continue to contend will be a monster)
• Provides a dedicated ROW through the troublesome "B-E Zone" that requires minimal landtaking, no tunneling, and a maximum of 5 grade crossings (Comm Ave, Mountfort, St Mary's, Beacon, Miner -- and all of those except probably for Miner could be eliminated with grade separation)
• Requires no modification to Kenmore and no impact on capacity in the Boylston St Subway
• Reduces "zig-zagging":
  • as the crow flies from a Cambridgeport station (i.e. at Waverly St) to the intersection of Huntington & S Huntington, it's 1.64 miles
  • the route via Lansdowne is 2.45 miles -- 150%, while the route via Kenmore is 3.1 miles -- a full 200%, and includes over 1,000 ft of unidirectional single tracking
• By detaching the Fenway Branch from the potentially both the Boylston and Huntington Subways, you free it up to be used at full capacity, rather than as an awkward vestigial route that briefly pretends to be circumferential, but isn't

EDIT: I should note that this idea I'm pretty sure is not original. @davem proposed something similar 8 years ago: https://archboston.com/community/threads/crazy-transit-pitches.3664/page-102#post-200449

Inspired by the idea, would something like this be feasible?

• Urban Ring: Tunnel under the Pike and/or the Framingham ROW, then turn towards the existing D branch subway and take over the flat junction to Fenway station. Runs at the current C/D subway level just below the Pike. (No tunneling needed at the parking lot parcel)
• C branch: Rebuild the Beacon St subway to dig one level lower than the Urban Ring level/current flat junction level, to eliminate the flat junction.

So the layout would look like this:

• Street level: Beacon St, Mountfort St, Maitland St
• Basement 1: Mass Pike, Framingham Line
• Basement 2: Current C/D subways and flat junction, proposed Urban Ring subway
• Basement 3: Proposed C subway

 

[Brattle Loop]

• Urban Ring: Tunnel under the Pike and/or the Framingham ROW, then turn towards the existing D branch subway and take over the flat junction to Fenway station. Runs at the current C/D subway level just below the Pike. (No tunneling needed at the parking lot parcel)

It's worth noting that Google's accuracy on the location of (at least) the Green Line tunnels is notably poor in several areas. If the actual C/D split is further west (and OpenRailwayMap, for one, suggests that it is), you'd be tunneling a lot closer to those buildings at the northwest corner of the intersection, and/or needing a sharper (and therefore slower) curve to get on the insertion angle.

• C branch: Rebuild the Beacon St subway to dig one level lower than the Urban Ring level/current flat junction level, to eliminate the flat junction.

Wait, this would take out the connection between Fenway and Kenmore? (C's going to have to come up to Kenmore's existing level, and the choices would appear to be taking out the floor of the current C/D tunnel or widening it, if that's even possible.) I get Riverside's conception had the Fenway route given over to the UR Gold Line, but it seems like it might be a bad mistake to cut off the Highland Branch's access to Kenmore altogether. Basic operational flexibility reasons aside, it'd be a nightmare after Sox games (well, maybe not with the way FSG's been managing the roster lately) with Kenmore having to handle anything other than GJ/UR-direct traffic and no ability to feed cars into Kenmore from Reservoir/Riverside except via the surface branches.

You know, I'd heard of that GIS map for years but never actually looked at it until today. That's simultaneously a treasure trove and a month's worth of rabbit holes right there.

Sorry for any productivity you might have hoped to have this week

 

[Teban54]

It's worth noting that Google's accuracy on the location of (at least) the Green Line tunnels is notably poor in several areas. If the actual C/D split is further west (and OpenRailwayMap, for one, suggests that it is), you'd be tunneling a lot closer to those buildings at the northwest corner of the intersection, and/or needing a sharper (and therefore slower) curve to get on the insertion angle.

Welp, it is. I suppose an alignment like the following may still be possible, but it's ugly. Definitely slower, but probably still better than going to Kenmore and back (in terms of speed only):

Wait, this would take out the connection between Fenway and Kenmore? (C's going to have to come up to Kenmore's existing level, and the choices would appear to be taking out the floor of the current C/D tunnel or widening it, if that's even possible.) I get Riverside's conception had the Fenway route given over to the UR Gold Line, but it seems like it might be a bad mistake to cut off the Highland Branch's access to Kenmore altogether. Basic operational flexibility reasons aside, it'd be a nightmare after Sox games (well, maybe not with the way FSG's been managing the roster lately) with Kenmore having to handle anything other than GJ/UR-direct traffic and no ability to feed cars into Kenmore from Reservoir/Riverside except via the surface branches.

Good point. Technically, the part about burying the C below the flat junction is optional, as you can leave the C unchanged and keep the flat junction. Of course, that's operationally messy with the C and the Urban Ring interfering with each other (though at least better than present-day Copley, and any Kenmore solution has the same problem). The benefit is that it's more feasible in terms of engineering and cost.

An alternative is to widen the ROW and build new outer tracks for revenue service on the C that dig into Basement 3, while keeping the current inner tracks on Basement 2 and the flat junction intact for operational flexibility. I've drawn this in green above. The question is if Beacon St is wide enough for 4 tracks.

Btw, I attempted to fit an interchange station here, but that seems impossible with this alignment, especially with Type 10s in consideration.