If You Were God/Goddess | Transit & Infrastructure Sandbox

[Delvin4519]

I'm pretty sure we would have no through-running routes -- everything would terminate at Lynn Central Sq. The 424 and 450 would get diverted via something like Washington St, and the 441, 442, and 455 would be truncated at Lynn. It is true that we need to deal with serving the areas between Lynn and Wonderland, but those should be separate routes, to keep the north-of-Lynn routes more reliable.

To your point, you have two corridors leftover: Common St/Western Ave/Salem Turnpike, and Lynnway/North Shore Road.

The eastern corridor sees pretty good ridership on both sides of the bridge, so yes, we'd probably have a shortened route pinging back and forth between Wonderland and Lynn Central Sq via Lynnway. That route will be more reliable if shorter, and can also have its frequency tuned to match the demand of that specific corridor, rather than having to fit the demand of the much longer and more varied corridor.

The western corridor sees solid ridership between Central Square and the West Lynn Garage, but it drops after the river. The lion's share of this corridor could be handled by increased frequencies on the 426, with riders oriented toward Central Square.

You do have some leftover pieces here, mostly on Western Ave -- between Washington and S Common St, and then the short stretch between Summer St and the West Lynn Garage. You could experiment with a short route that ping-pongs back-and-forth to take care of those. Alternatively, we could make an exception to "no through runs", and re-extend the 455 the extra 2.3 miles to cover those gaps.

~~~

If we truly want to play "Deity Mode", then let's assume that the River Works station becomes accessible from the street. You could then run the 424, 450, and maybe also an extended 455 into a secondary hub there.

~~~

But basically, even if you only curtail the Salem Turnpike segments of the 450, 455, and 424, that's still a significant increase. And even if it isn't strictly speaking "double the frequencies", it'll still be a massive improvement in reliability to have the hub at Lynn rather than Wonderland, even if you still need to cover some of the route-miles between the two. The reliability improvements coupled with at least some shorter routes will result in frequency increases. (Plus, providing rapid transit to the ~15,000 people who live in the 15-min walkshed of Lynn Station is nothing to sneeze at either.)

The 424 terminates at Eastern Ave., At that point, if the 424 were to be truncated at Central Sq. via Washington St., the 424 and the 456 could be eliminated altogether, and instead reroute the 435 via Eastern Ave. This frees up a bus or so to increase frequencies across other routes.

The elimination of the 456 is essentially unnoteworthy if the 450 is routed to Central Sq. instead of West Lynn Garage. In that case, one travelling along the Western Ave. corridor would now need to transfer to the 450 from whatever route originates at the Garage. Currently, it's a one seat ride on the 450 at present day from the Garage up Eastern Ave..

It's probably best to just extend 1 of the Lynn area bus routes from Central Sq., to West Lynn Garage. I'd opt for the 441/442 to extend to the West Lynn Garage., and the 455 terminates at Central Sq. It's better just to through run one of the routes, since it eliminates a transfer for at least 1 origin-destination pair in the Lynn area, and the last thing we need is an extra bus route competing for frequencies.

 

[Teban54]

I feel we should also consider the bus network redesign, which will be in effect way before BLX to Lynn opens:

Corridor	Current routes	Redesign routes
North Shore Rd	441/442	442, 455
Salem Turnpike	424, 450, 455	450

Also note the 424 is currently an AM peak-only service that is planned to be eliminated in the redesign. No bus routes will serve Eastern Ave in the redesign proposal.

The redesign has a 15-min-frequency corridor on North Shore Rd served by the 442 and 455, and the 450 on Western Ave/Salem Turnpike also has 30-min-or-better frequencies. In a post-BLX world, that's overkill for local demands on both corridors. You can probably have 30 min frequency on North Short Rd between Lynn and Wonderland, and 30 min frequency on Western Ave between Washington St and West Lynn Garage. Regardless of how you handle them operationally, that already saves you a lot of buses from not having to run to Wonderland (or in the case of 450 today, downtown).

(The stop at Salem Turnpike/Ballard St has its catchment overlap with the 426, so that shouldn't be a big concern.)

Operationally, I would do the following:

• 442: Reroute to West Lynn Garage via Washington St and Western Ave
• 450: Shortened to Central Sq Lynn via Washington St
• 455: Shortened to Central Sq Lynn
• New route between Lynn and Wonderland via North Shore Rd. In practice, it can be through-run with (some or all) 450 or 455 trips.

I'm not sure how important it is to maintain some one-seat rides to Wonderland to connect to buses there (T110, T116).

 

[Teban54]

Rather than start a new thread, I figured this would be a better option: Instead of "If you were God" description: If you were Maura Healey, what would your transit expansion priorities entail for your FIRST term? (Assuming the MBTA System is in a general state of good repair. I KNOW, that's CRAZY talk...........but let's assume.) Here is my list in order of my priority:

1. Blue Line Extension to Red Line at Charles Street
2. Blue Line Extension to Lynn
3. Green Line Extension/Spur from D Line to Needham Junction with new stations along Highland Avenue and the just announced Muzi Ford redevelopment.
4. Orange Line Extension to Route 128 or West Roxbury involving discontinued Needham line commuter rail. (Now covered by Green Line extension from Newton)
5. Green Line Extension from Union Square to Porter Square

That would be a nice vision in my opinion. I have a LOT more, but I think this list would be visionary but also accomplishable. (Meaning the funding and start could be accomplished in the first four-year term with a solid vision.)

What are your Maura Healey transit initiatives?

A bit late to this, but here's my list in no particular order. I agree with most suggestions before mine.

HRT/LRT projects to be started preferably ASAP:

• Red-Blue Connector
• Blue to Lynn
• Orange to Roslindale
  • This is #5 in the Go Boston 2030 survey that was released in 2017 (Boston only)
  • Phase 2 would be OL to W Roxbury and GL to Needham; I personally think both should be done together (to minimize impacts to Needham Line) and after Roslindale
• Green to Route 16
• Green to Porter

HRT/LRT projects to be studied, so that we can gain insights into projected ridership, engineering feasibility, cost-benefit analysis etc, but not necessarily starting them:

• Green to Hyde Square
  • Small extension that should be easy, but rarely mentioned. Ranked #18 in GB2030.
  • Borderline between this and "start ASAP" category.
• Urban Ring
  • I think the entire ring needs to be considered, not just a particular section. Grand Junction, service to Everett/Chelsea, best way and mode through LMA, connecting to Lechmere, etc.
  • In terms of actually building it, Grand Junction is the obvious priority, but probably not realistic for the short term
• Green to Nubian
  • The need is definitely there, but may be seen as less urgent because of SL4/5 being a thing. Also a few uncertainties in engineering, including (not limited to) Bay Village interchange and integration with future trunks.
  • "Silver Line tunnel connecting Washington St to S Station" ranked as #14 in the Go Boston 2030 survey. I'm sure residents who voted for it would be in favor of a GL branch even more.
• NSRL
  • The only reason why it's not higher is because I don't think it will happen any sooner. But at least undo the Baker sandbagging study.

Commuter rail: Start electrifying feasible lines such as Fairmount and NEC, and/or start running 15-min Regional Rail services within 128 wherever feasible. Others have done a good job detailing this.

BRT and others:

• SL6 or SL3 extension: Chelsea and Everett to Sullivan and Kendall (already has momentum)
• Downtown transit corridor: North Station to South Station and Seaport (#10 in GB2030)
• Mass Ave (#4 in GB2030, surprisingly popular)
• Blue Hill Ave
• LMA transit hub and bus lanes

 

[Delvin4519]

I feel we should also consider the bus network redesign, which will be in effect way before BLX to Lynn opens:

Corridor	Current routes	Redesign routes
North Shore Rd	441/442	442, 455
Salem Turnpike	424, 450, 455	450

Also note the 424 is currently an AM peak-only service that is planned to be eliminated in the redesign. No bus routes will serve Eastern Ave in the redesign proposal.

The redesign has a 15-min-frequency corridor on North Shore Rd served by the 442 and 455, and the 450 on Western Ave/Salem Turnpike also has 30-min-or-better frequencies. In a post-BLX world, that's overkill for local demands on both corridors. You can probably have 30 min frequency on North Short Rd between Lynn and Wonderland, and 30 min frequency on Western Ave between Washington St and West Lynn Garage. Regardless of how you handle them operationally, that already saves you a lot of buses from not having to run to Wonderland (or in the case of 450 today, downtown).

(The stop at Salem Turnpike/Ballard St has its catchment overlap with the 426, so that shouldn't be a big concern.)

Operationally, I would do the following:

• 442: Reroute to West Lynn Garage via Washington St and Western Ave
• 450: Shortened to Central Sq Lynn via Washington St
• 455: Shortened to Central Sq Lynn
• New route between Lynn and Wonderland via North Shore Rd. In practice, it can be through-run with (some or all) 450 or 455 trips.

I'm not sure how important it is to maintain some one-seat rides to Wonderland to connect to buses there (T110, T116).

To be honest, there aren’t really that many major changes to the Lynn area bus routes with BNRD, they’re mostly minor with the express routes eliminated. The only service loss is Paradise Rd. and Eastern Ave., which ithe latter is easily fixed by sending the 435 down there if the 450 is being routed to Central Sq.

 

[Riverside]

Shifting gears to pure "Deity Mode" here: I'm noodling around with the LMA-Nubian stretch of the Urban Ring corridor and am trying to figure out what an "ideal" path for a subway would be, without regard for cost or engineering feasibility -- I want to have an "ideal" articulated and then see how close we could get to that.

So, first it should be acknowledged that the "purest" ideal for this corridor looks something like this:

Zig zag between the core of Longwood, over to Ruggles, over to Nubian and then east from there. (I usually draw it going northeast toward the Seaport, but there's of course an argument for going toward JFK/UMass or Andrew.)

In some ways this is a helpful ideal, but it seems so impractical as to be of limited use. Short of a major TBM project, you simply will have to make some compromises.

Which leaves you with a few options:

"Francis St"

\

Pros: straight shot, connects Longwood, Orange Line, and Nubian with shortest amount of tunneling.

Cons: misses Regional Rail transfer at Ruggles, less connectivity to northern LMA, unclear what the potential would be for cross-LMA extensions to the west in the future, significant tunneling under narrow-ish residential street

Mitigation: Surface BRT on Ruggles St could provide a strong link between Longwood and the Dorchester buses feeding in to Ruggles; a similar solution could be used to handle Dorchester <> Nubian <> Ruggles journeys, though it's unclear how many Ruggles <> Nubian trips might be canceled out if there is both a Green Line connection at Nubian and a subway connection to Huntington at Nubian as well.

"Ruggles St"

Pros: mostly straight shot, shortest amount of tunneling overall, hits Ruggles, provides transfer to Huntington LRT, Orange Line, Regional Rail, and Washington LRT, MBTA owns parcels along Ruggles that may simplify tunneling

Cons: misses Nubian entirely, originates at the far end of LMA away from the larger hospitals, requires longer Dorchester bus journeys to transfer at Ruggles (which may be unnecessary anyway depending on how many Dorchester buses are extended to Longwood), may need to interface/cross under with a Green subway under Washington

Mitigation: LMA access can be improved by interlining with Huntington LRT through Brigham Circle (although need be mindful of capacity in extended Huntington Subway)

"Ruggles St with Nubian Spur"

Pros: Pros: largely straight shot, hits Ruggles, provides transfer to Huntington LRT, Orange Line, Regional Rail, and Washington LRT, MBTA owns parcels along Ruggles that may simplify tunneling, connects Nubian to eastern quadrant (e.g. Seaport)

Cons: more complicated tunneling, including a junction around Melnea Cass, doesn't improve connectivity Nubian <> Ruggles or Nubian <> LMA connectivity

Mitigation: significant BRT along Malcolm X and Tremont, though at that point it's unclear what additional benefit an LRT subway really nets you

"Zig-Zag"

Pros: hits both Ruggles and Nubian, providing full connectivity to Huntington, Orange Line, Regional Rail, and Nubian, MBTA owns parcels along Ruggles that may simplify tunneling, enables bus transfer hubs at both Nubian and Roxbury Crossing

Cons: more roundabout path, with slowing curves and somewhat longer travel times, more tunneling than either "Ruggles" or "Francis" alts

Mitigation: nothing really

~~~

I think that the "Ruggles" and "Ruggles With Spur" alts are pretty much DOA for their lack of Nubian <> Longwood connection. I've crayoned a route using Ruggles <> Melnea Cass, but only when a direct connection to Logan was possible. (For the purposes of this thought exercise, I've arbitrarily decided that a LMA <> Nubian subway is in scope but a harbor crossing subway is not.)

What I've been noodling around with is a "Zig-Zag" surface BRT combined with a "Francis St" LRT subway:

My thinking:

One-seat BRT between Dorchester and LMA is a good thing. An Urban Ring will not change that. I've said before -- we should be thinking about radial service for Longwood, not just circumferential, and the BRT route I've sketched out above is essentially a radial trunk-and-branch network originating in Longwood.

The BNRD's short-term designs for Dorchester-LMA BRT via Francis St notwithstanding, Ruggles St remains a strong contender for BRT; improvements to Regional Rail (and certainly an NSRL) would arguably tip the scale back toward transfers at Ruggles rather than Roxbury Crossing. So the need for a Ruggles <> LMA connection (e.g. for Regional Rail <> LMA journeys) could be handled by frequent BRT buses (that would be able to travel deeper into LMA than an LRT subway could anyway).

As for circumferential LRT: if we're going to build it, there's an argument to be made to build it to be as good of a circumferential service as it can be -- without bogging it down with trying to be a portion of a Longwood radial network. So, focus it on connecting between the radial corridors (Huntington, Orange, Nubian) as straightforwardly as possible.

(And it's not like this is a bad LMA <> Nubian route -- you get transfers to everything except Regional Rail, you get access to the southern job-heavier half of LMA, and a subway along this route would positively zip along from Nubian to Brigham Circle.)

~~~

It's possible this thought exercise is too abstract and arbitrary to be that useful -- I don't think I would prioritize digging up Francis St to build a subway anytime soon. But at least for me, it's helped me put a finger on something that is dissatisfying about the Zig-Zag alignment: one of the pressures contorting that route is that it tries to partially replace what should be a one-seat ride between LMA and Dorchester. Identifying that as a separate need from circumferential transit is useful.

 

[Teban54]

Shifting gears to pure "Deity Mode" here: I'm noodling around with the LMA-Nubian stretch of the Urban Ring corridor and am trying to figure out what an "ideal" path for a subway would be, without regard for cost or engineering feasibility -- I want to have an "ideal" articulated and then see how close we could get to that.

So, first it should be acknowledged that the "purest" ideal for this corridor looks something like this:

View attachment 32610

Zig zag between the core of Longwood, over to Ruggles, over to Nubian and then east from there. (I usually draw it going northeast toward the Seaport, but there's of course an argument for going toward JFK/UMass or Andrew.)

In some ways this is a helpful ideal, but it seems so impractical as to be of limited use. Short of a major TBM project, you simply will have to make some compromises.

Which leaves you with a few options:

"Francis St"

View attachment 32611\

Pros: straight shot, connects Longwood, Orange Line, and Nubian with shortest amount of tunneling.

Cons: misses Regional Rail transfer at Ruggles, less connectivity to northern LMA, unclear what the potential would be for cross-LMA extensions to the west in the future, significant tunneling under narrow-ish residential street

Mitigation: Surface BRT on Ruggles St could provide a strong link between Longwood and the Dorchester buses feeding in to Ruggles; a similar solution could be used to handle Dorchester <> Nubian <> Ruggles journeys, though it's unclear how many Ruggles <> Nubian trips might be canceled out if there is both a Green Line connection at Nubian and a subway connection to Huntington at Nubian as well.

"Ruggles St"

View attachment 32616

Pros: mostly straight shot, shortest amount of tunneling overall, hits Ruggles, provides transfer to Huntington LRT, Orange Line, Regional Rail, and Washington LRT, MBTA owns parcels along Ruggles that may simplify tunneling

Cons: misses Nubian entirely, originates at the far end of LMA away from the larger hospitals, requires longer Dorchester bus journeys to transfer at Ruggles (which may be unnecessary anyway depending on how many Dorchester buses are extended to Longwood), may need to interface/cross under with a Green subway under Washington

Mitigation: LMA access can be improved by interlining with Huntington LRT through Brigham Circle (although need be mindful of capacity in extended Huntington Subway)

"Ruggles St with Nubian Spur"

View attachment 32618

Pros: Pros: largely straight shot, hits Ruggles, provides transfer to Huntington LRT, Orange Line, Regional Rail, and Washington LRT, MBTA owns parcels along Ruggles that may simplify tunneling, connects Nubian to eastern quadrant (e.g. Seaport)

Cons: more complicated tunneling, including a junction around Melnea Cass, doesn't improve connectivity Nubian <> Ruggles or Nubian <> LMA connectivity

Mitigation: significant BRT along Malcolm X and Tremont, though at that point it's unclear what additional benefit an LRT subway really nets you

"Zig-Zag"

View attachment 32619

Pros: hits both Ruggles and Nubian, providing full connectivity to Huntington, Orange Line, Regional Rail, and Nubian, MBTA owns parcels along Ruggles that may simplify tunneling, enables bus transfer hubs at both Nubian and Roxbury Crossing

Cons: more roundabout path, with slowing curves and somewhat longer travel times, more tunneling than either "Ruggles" or "Francis" alts

Mitigation: nothing really

~~~

I think that the "Ruggles" and "Ruggles With Spur" alts are pretty much DOA for their lack of Nubian <> Longwood connection. I've crayoned a route using Ruggles <> Melnea Cass, but only when a direct connection to Logan was possible. (For the purposes of this thought exercise, I've arbitrarily decided that a LMA <> Nubian subway is in scope but a harbor crossing subway is not.)

What I've been noodling around with is a "Zig-Zag" surface BRT combined with a "Francis St" LRT subway:

View attachment 32623

My thinking:

One-seat BRT between Dorchester and LMA is a good thing. An Urban Ring will not change that. I've said before -- we should be thinking about radial service for Longwood, not just circumferential, and the BRT route I've sketched out above is essentially a radial trunk-and-branch network originating in Longwood.

The BNRD's short-term designs for Dorchester-LMA BRT via Francis St notwithstanding, Ruggles St remains a strong contender for BRT; improvements to Regional Rail (and certainly an NSRL) would arguably tip the scale back toward transfers at Ruggles rather than Roxbury Crossing. So the need for a Ruggles <> LMA connection (e.g. for Regional Rail <> LMA journeys) could be handled by frequent BRT buses (that would be able to travel deeper into LMA than an LRT subway could anyway).

As for circumferential LRT: if we're going to build it, there's an argument to be made to build it to be as good of a circumferential service as it can be -- without bogging it down with trying to be a portion of a Longwood radial network. So, focus it on connecting between the radial corridors (Huntington, Orange, Nubian) as straightforwardly as possible.

(And it's not like this is a bad LMA <> Nubian route -- you get transfers to everything except Regional Rail, you get access to the southern job-heavier half of LMA, and a subway along this route would positively zip along from Nubian to Brigham Circle.)

~~~

It's possible this thought exercise is too abstract and arbitrary to be that useful -- I don't think I would prioritize digging up Francis St to build a subway anytime soon. But at least for me, it's helped me put a finger on something that is dissatisfying about the Zig-Zag alignment: one of the pressures contorting that route is that it tries to partially replace what should be a one-seat ride between LMA and Dorchester. Identifying that as a separate need from circumferential transit is useful.

Tremont St, between Brigham Circle and Roxbury Crossing, still seems narrow with enough buildings around that building a subway underneath it can be a concern. So even if we didn't care about west of Huntington and connection to Regional Rail, from an engineering perspective, you pretty much have to choose one of the Ruggles St alignments for a subway.

Once again, I'm frustrated at how the streets are so narrow in LMA that it's hard to even put bus lanes through it, much less a subway. Such a shame for a major employment center. Brookline Ave is the only one that's wide enough, but it comes at the wrong orientation.

I really don't like the Brookline Village zigzag, but that seems like the most feasible scenario. Although I do think if we have the financial and political means to build just one subway (other than Red-Blue), LMA is pretty much #1 on that list. One of those whose benefits may actually justify the high cost.

 

[Brattle Loop]

Tremont St, between Brigham Circle and Roxbury Crossing, still seems narrow with enough buildings around that building a subway underneath it can be a concern. So even if we didn't care about west of Huntington and connection to Regional Rail, from an engineering perspective, you pretty much have to choose one of the Ruggles St alignments for a subway.

That section of Tremont is quite a bit wider than Boylston/Essex where Silver Line Phase III was supposed to go. That was at least theoretically possible, albeit expensive and painful tunneling (with a lot of the expense coming from underpinning the ancient GL/OL stations that thing would have passed under). Not saying it would be easy or cheap, but it's definitely wide enough for an LRT subway.

Once again, I'm frustrated at how the streets are so narrow in LMA that it's hard to even put bus lanes through it, much less a subway. Such a shame for a major employment center. Brookline Ave is the only one that's wide enough, but it comes at the wrong orientation.

Anything wider than Essex is at least theoretically wide enough to shiv an LRT tunnel through it. Outside of God Mode, it's likely that a lot of them are infeasible on cost grounds, but they're at least possible.

 

[Aprehensive_Words]

This brings up an assumption I've always had, but it feels worth double-checking with all you smart people.

If you were going to use a TBM for the Urban Ring (the only way you'll get @Riverside 's first, idealized map), you'd have to maneuver around all kinds of foundation pilings for the hospital and residential towers around there.

But while things like the Wentworth dorm tower and the newish BIDMC buildings obviously have deep foundations, how deep do older buildings' pilings etc. go? Would the Harvard Med School campus, for example, represent a veritable forest underground? Is there an easy way to guess at these underground obstacles by building age?

I sort of figured that the Alice Taylor Apartments and the Mission Main apartments would give any TBM enough maneuvering room to line up on Longwood Ave. or Tremont St., but maybe the 40s-era parts of Alice Taylor would put the kibosh on that?

 

[Riverside]

Wow, I just realized I said "Francis St" but in fact Francis stops at Huntington, the other side is Tremont. Goodness gracious, this would mean another "Tremont Street Subway".

@Teban54 yeah I mean the actual feasibility of any of these proposals is a whole kettle of worms. And of course if a subway via Ruggles proposal began to get serious traction, I would be very much in favor -- no interest in letting the perfect become the enemy of the good here.

That said, I think @Brattle Loop is correct -- while tricky and expensive, the width itself I don't think is an immediate dealbreaker. There are stretches of, for example, Court Street and Tremont Street (downtown) that are significantly narrower. (That said, @Brattle Loop wasn't Essex Street going to be vertically stacked rather than side-by-side?)

Parenthetically about the Brookline Village zig-zag: it may not be super clear in my map, but here I'm assuming all Urban Ring services terminate at Brookline Village -- e.g. three branches originating there and fanning out to Allston/Harvard, Cambridge/Sullivan/beyond, and Nubian/Seaport. Running "round the bend" doesn't make much sense to me.

@Aprehensive_Words I really don't know, but it's an interesting question!

 

[Brattle Loop]

That said, @Brattle Loop wasn't Essex Street going to be vertically stacked rather than side-by-side?

Uh...I didn't know that, but I think the answer's yes...now that I've looked it up. Though that was BRT, and you're correct that parts of the existing Tremont subway (especially between Park and GC in the stretch between the Suffolk building and the Parker House), are nearly as narrow as Essex (and narrower than the part of Tremont near Brigham Circle we've been discussing.

 

[Teban54]

That section of Tremont is quite a bit wider than Boylston/Essex where Silver Line Phase III was supposed to go. That was at least theoretically possible, albeit expensive and painful tunneling (with a lot of the expense coming from underpinning the ancient GL/OL stations that thing would have passed under). Not saying it would be easy or cheap, but it's definitely wide enough for an LRT subway.



Anything wider than Essex is at least theoretically wide enough to shiv an LRT tunnel through it. Outside of God Mode, it's likely that a lot of them are infeasible on cost grounds, but they're at least possible.

Point taken about feasibility.

This got me wondering about Silver Line Phase III though, since I wasn't around back then. Was it canceled solely because of cost? Or were other concerns equally as detrimental, such as impacts to the Central Burying Ground and concerns from Bay Village residents? Or was it not a serious effort from the start?

If it's the former, seems like it would kill the prospects of many expensive potential tunneling projects (such as the LMA that we're discussing here), even if the benefits may justify the cost. Hopefully these won't be as expensive due to not having to underpin old stations, though.

I continue to be amazed that tunneling under Essex St actually turned out to be a serious proposal that almost materialized, not something in Crazy Transit Pitches. That seems like the kind of project that would draw criticism even in the Crazy Transit Pitches thread.

Parenthetically about the Brookline Village zig-zag: it may not be super clear in my map, but here I'm assuming all Urban Ring services terminate at Brookline Village -- e.g. three branches originating there and fanning out to Allston/Harvard, Cambridge/Sullivan/beyond, and Nubian/Seaport. Running "round the bend" doesn't make much sense to me.

Hmm. I can understand why this is done, especially if both branches won't be able to get into the heart of LMA and will require walking anyway. But this would be a big disappointment for travel patterns across the west and south quadrants of the Urban Ring, basically between (Harvard, West Station, Cambridge/Kendall, BU) and (Ruggles, Nubian, BUMC, somewhere on Red Line south, Seaport).

I'm not sure how much demand is there for such patterns (and that alone will probably not justify tunneling through LMA), but I can imagine some of them are valuable, especially in a Regional Rail world.

Some of them can be mitigated with a Huntington-Seaport subway and improved bus infrastructure (e.g. T1), but not all.

To be fair, without expensive tunneling through LMA or aggressive implementation of bus lanes for a BRT southside Urban Ring, what you suggested seems to be the best compromise anyway. Even if a through-run service via Brookline Village exists, it will likely not be competitive in timing. Having BRT routes that terminate at Longwood station on the Highland Branch is probably the best option.

 

[Brattle Loop]

This got me wondering about Silver Line Phase III though, since I wasn't around back then. Was it canceled solely because of cost? Or were other concerns equally as detrimental, such as impacts to the Central Burying Ground and concerns from Bay Village residents? Or was it not a serious effort from the start?

F-Line has a thorough accounting of its demise...somewhere around here. (I'm not in the mood to engage with the search function at the moment, but I'll link to it if I come across it.)

As I recall from that, Phase III died because of the cost considerations. The tunneling under Essex (as BRT, requiring wider tunnels than LRT) was part, but not all of the cost blowout. They had an extremely difficult time trying to find an alignment to connect the Washington Street end to the tunnel (as a result of shotgun-marrying the elevated 'replacement' to the Seaport/Airport project). The preferred alignment when the project died called for a portal around the intersection of Charles and Tremont, then tunneling under Charles, Boylston, and Essex to the Transitway. I don't think proximity to the burying ground was a big concern, given they were going to be under the Green Line. The biggest cost blowouts came from having to underpin Boylston and Chinatown rather than specifically the narrowness on Essex (though that one wasn't necessarily easy or cheap, it just wasn't as fatal as underpinning two levels of 1897-built Boylston plus 1908-Chinatown).

I continue to be amazed that tunneling under Essex St actually turned out to be a serious proposal that almost materialized, not something in Crazy Transit Pitches. That seems like the kind of project that would draw criticism even in the Crazy Transit Pitches thread.

It absolutely would be (and has to a degree when it's been suggested in some proposals in the Green Line Reconfiguration thread). It was never an objectively good idea, just one that suited the people in power at the time, because it allowed them to pretend to replace the Elevated and do it on the cheap with their BRT fetish as part of the new-build Airport/Seaport transit project that was an offspring of the Big Dig. It died precisely because they tried to God Mode that shotgun marriage until the Crazy Transit Pitches-required real-world analysis killed it stone dead.

 

[Teban54]

F-Line has a thorough accounting of its demise...somewhere around here. (I'm not in the mood to engage with the search function at the moment, but I'll link to it if I come across it.)

As I recall from that, Phase III died because of the cost considerations. The tunneling under Essex (as BRT, requiring wider tunnels than LRT) was part, but not all of the cost blowout. They had an extremely difficult time trying to find an alignment to connect the Washington Street end to the tunnel (as a result of shotgun-marrying the elevated 'replacement' to the Seaport/Airport project). The preferred alignment when the project died called for a portal around the intersection of Charles and Tremont, then tunneling under Charles, Boylston, and Essex to the Transitway. I don't think proximity to the burying ground was a big concern, given they were going to be under the Green Line. The biggest cost blowouts came from having to underpin Boylston and Chinatown rather than specifically the narrowness on Essex (though that one wasn't necessarily easy or cheap, it just wasn't as fatal as underpinning two levels of 1897-built Boylston plus 1908-Chinatown).



It absolutely would be (and has to a degree when it's been suggested in some proposals in the Green Line Reconfiguration thread). It was never an objectively good idea, just one that suited the people in power at the time, because it allowed them to pretend to replace the Elevated and do it on the cheap with their BRT fetish as part of the new-build Airport/Seaport transit project that was an offspring of the Big Dig. It died precisely because they tried to God Mode that shotgun marriage until the Crazy Transit Pitches-required real-world analysis killed it stone dead.

Thanks. Upon a further reading of the Wikipedia article, it seems like even before the Washington St service was shotgunned into it, the Seaport Silver Line project already planned for underpinning of Chinatown and Boylston stations:

The MBTA conducted a feasibility study in 1987 and released a Draft Environmental Impact Report (DEIR) in 1989.[4]: 23 [41] The DEIR selected an underground "transitway" over alternatives including a surface light rail line, an elevated people mover, a commuter rail shuttle, and a relocation of the Red Line. The transitway was to use trolleybuses or dual-mode buses, rather than the light rail and people mover possibilities considered; it would connect with the Red Line at South Station, the Orange Line at Chinatown, and the Green Line at Boylston.

Here's the DEIR, where Alternative 4 was eventually selected:

This seems to be before the Washington St service was merged into Silver Line or even planned as BRT at all. In the same year:

In 1989, the MBTA announced that trolleybuses would be used on Washington Street, operating on 4-minute headways at peak hours.[8]: 3.1  By 1990, the MBTA expected service to begin in 1993, with an underground connection to Boylston station and the proposed South Boston Piers tunnel in a future phase.[35] After several more years of studies, the MBTA decided in 1996–97 to build the route as a bus rapid transit line using compressed natural gas (CNG) buses to avoid the visual impact of overhead wires.[8]: 3.2

So to me, it seems that what eventually caused the cost blowouts weren't because they shotgunned what should have been two separate projects together. Although it still appears that when a Seaport-Boylston line was planned, not enough consideration was given for feasibility and cost.

 

[Brattle Loop]

So to me, it seems that what eventually caused the cost blowouts weren't because they shotgunned what should have been two separate projects together. Although it still appears that when a Seaport-Boylston line was planned, not enough consideration was given for feasibility and cost.

Huh. Either the thought of a possible Washington Street connection was there at some level that early, or the project was even more monumentally stupid than I thought. There is no universe in which tunneling down Essex is unambiguously a good idea given how narrow it is, even if we assume for the sake of argument that they didn't realize how absurdly expensive underpinning those stations would be. It's an odd design choice to even go that way if you're not trying to reach Washington, because there's nothing of sufficient scale around Boylston to anchor the end of a transit line.

Kind of feels like they Crazy Transit Pitched their way to being able to say they had connections to the (non-Blue) HRT/LRT lines, without really investigating (at first, anyway) whether that all made sense. I feel like if they'd just gone forward with that project, they would have realized Essex was a bad choice between the narrowness, the underpinning, and the fact that there's nothing to anchor it at Boylston, but that at some point they shotgunned it to Washington, and that that's why it lingered around as a cruddy option, eventually crippling Nubian's transit corridor when it collapsed under its own weight.

So to me, it seems that what eventually caused the cost blowouts weren't because they shotgunned what should have been two separate projects together. Although it still appears that when a Seaport-Boylston line was planned, not enough consideration was given for feasibility and cost.

That DEIR doesn't seem to show what would have happened with the Silver Line Seaport tunnel once it hit Boylston. If it just ended underground it probably would have been a bit cheaper than the convoluted routing they had to try and find to connect to a portal for Washington wherever they could shiv one in once they'd shotgunned the two projects. But most of those costs would have still been there if the thing was just built Seaport-Boylston, you're right about that. I'd say in sum it's sort of a Schrodinger's cost blowout; anything down Essex was inevitably going to be costly, so in that sense shotgunning the two projects didn't cause that (though it added to it), but shotgunning did force them to swallow that alignment, triggering the latent blowout and killing the functionality of both projects.

 

[Teban54]

Shifting gears to pure "Deity Mode" here: I'm noodling around with the LMA-Nubian stretch of the Urban Ring corridor and am trying to figure out what an "ideal" path for a subway would be, without regard for cost or engineering feasibility -- I want to have an "ideal" articulated and then see how close we could get to that.

So, first it should be acknowledged that the "purest" ideal for this corridor looks something like this:

View attachment 32610

Zig zag between the core of Longwood, over to Ruggles, over to Nubian and then east from there. (I usually draw it going northeast toward the Seaport, but there's of course an argument for going toward JFK/UMass or Andrew.)

In some ways this is a helpful ideal, but it seems so impractical as to be of limited use. Short of a major TBM project, you simply will have to make some compromises.

Which leaves you with a few options:
[...]
It's possible this thought exercise is too abstract and arbitrary to be that useful -- I don't think I would prioritize digging up Francis St to build a subway anytime soon. But at least for me, it's helped me put a finger on something that is dissatisfying about the Zig-Zag alignment: one of the pressures contorting that route is that it tries to partially replace what should be a one-seat ride between LMA and Dorchester. Identifying that as a separate need from circumferential transit is useful.

This is a crazy proposal, and I'm sure people would give me 100 reasons for not doing this. But hey, we're technically still in the God Mode thread, right?

Much of the struggle here with Urban Ring alignments seems to be solely coming from serving both Nubian (existing bus hub) and Ruggles (Regional Rail, also a bus hub but with a significant number of routes from Nubian itself).

So this got me thinking: What if we move the bus hub at Nubian itself further north to meet the Urban Ring?

Assuming Urban Ring uses Ruggles and Melnea Cass, and a GL Nubian Branch goes down Washington St to Nubian, we can place a bus hub somewhere around Washington St/Melnea Cass Blvd, denoted "W/MC". Specifically, I think we may be able to take over either of these two parcels:

In practice, we may not even need the entire parcel.

Before I detail the route changes, I'll post a summary:

• Surprisingly, most post-BNRD Nubian routes already pass by W/MC and/or Ruggles, so they already have an Urban Ring connection!
  • Even if we can't build a W/MC bus hub, this is the most important takeaway. Having the Urban Ring station at W/MC instead of Nubian still gives you connection to most Nubian routes.
• The only significant change is to extend the 14, 38 and 44, which currently terminate at Nubian, further north to W/MC. If you want to, you can reroute the Malcolm X-Ruggles routes to W/MC and Melnea Cass.
• Every Nubian route can continue to serve Nubian. (The T1 and SL4/5 can still terminate there.)
• The only ones that can't also serve W/MC are the T28 and T66, and of these two, the only problematic connection would be T28 to UR east (towards Seaport).
• In the worst case, you always have the option of walking 5 minutes between Nubian and W/MC, or taking the GL for one stop.

Now, the detailed route changes:

Current/BNRD routing	BNRD routes	Proposed routing	Bus hubs served under proposed routing	Notes
Ruggles - Malcolm X - Nubian - South or east	T15, 19, T23, 45	EITHER: Unchanged OR: Rerouted to Ruggles - Melnea Cass - W/MC - Nubian - South or east	Ruggles, (W/MC), Nubian	(1)
Nubian - North	T1, SL4/5	Unchanged	W/MC, Nubian
Nubian - South, Southwest	14, 38, 44	Extend north to W/MC	W/MC, Nubian
Nubian - West	T66	Unchanged	Nubian	(2)
North or east - W/MC - Nubian - South or west	T12, 41, 42	Unchanged	W/MC, Nubian
West - Nubian - South	T28	Unchanged	Nubian	(3)
Ruggles - Melnea Cass - (skip Nubian) - East	10	Unchanged	Ruggles, W/MC	(4)
Ruggles - Melnea Cass - Nubian - East	None (currently 8, 47)	N/A	N/A	(4)

(1) Malcolm X routes: Staying on Malcolm X and switching to Melnea Cass and both have pros and cons.

• Both can still connect to UR at Ruggles and all other bus routes at Nubian.
• The Melnea Cass alignment (like current 19) provides a faster connection to Urban Ring east towards Seaport, and also has a faster maneuver into Ruggles lower busway, while avoiding the upper busway altogether.
• The Malcolm X alighment (status quo) maintains local service on Malcolm X, which would otherwise only be served by T12, T28 and T66 (thus losing access to Ruggles).
• The ideal way is probably to have some routes amended to Melnea Cass (e.g. T23, 45), while others remain on Malcolm X (e.g. T15, 19).

Yet another proposal regarding these routes: Some of them can be cut short to W/MC and no longer go to Ruggles, in a world where Nubian and W/MC gets quality transit (GL branch and UR).

(2) T66: I believe in a post-UR world, the T66's main role at Nubian will be to collect transfer passengers from other buses here, and bring them to LMA core, Brookline, Allston, etc. This role remains doable even without T66 itself having a W/MC connection. Much of its demand overlaps with the Urban Ring, anyway, so very few passengers will want to transfer between T66 and the Urban Ring around Nubian.

(3) T28: This is the only imperfect part of the proposal. Much like the T66, passengers from T28 will still have access to other buses at Nubian, Orange Line (at Roxbury Crossing), and LMA. The Urban Ring will likely intersect T28's route (Francis St-Brookline Ave) anyway, so transfers to UR west is also doable.

However, some passengers from the T28 may want to transfer to UR east towards Seaport, and there's really not a good solution for that. Going all the way to LMA for the UR east connection is impractical. They can also walk 5 minutes from Nubian to W/MC, but also not ideal.

If the connection between T28 and UR east is really important, the T28 itself can be rerouted to W/MC, Ruggles, then follow T47's route via Ruggles St and Longwood Ave (skipping Francis St).

(4) Currently, the 8, 19 and 47 use Melnea Cass between Ruggles and Nubian, with 8 and 47 continuing further east via Melnea Cass (thus making a detour specifically for Nubian). However, in BNRD, the 10 (which swaps route with the 8 in this section) skips Nubian entirely and use Melnea Cass straight to Ruggles, and the 47 is shortened to Ruggles, so having W/MC is a strict positive for the 10.

 

[Aprehensive_Words]

Unfortunately, both parcels have already been committed to affordable multifamily developments -- probably the least likely project type to suffer from the current credit crunch or any potential recession:

NUBA Residences | Boston Planning & Development Agency

Learn about the BPDA's development review process, view information about development projects in the BPDA approval pipeline, and learn how you can get involved.

www.bostonplans.org

Madison Tropical Parcel 10 | Bostonplans.org

Development Projects and Planned Development Areas (PDAs) that the Development Review division is coordinating.

www.bostonplans.org

(don't be confused by the render at the top of the second link -- someone at the BPDA seemingly forgot to update that)

 

[Teban54]

Unfortunately, both parcels have already been committed to affordable multifamily developments -- probably the least likely project type to suffer from the current credit crunch or any potential recession:

NUBA Residences | Boston Planning & Development Agency

Learn about the BPDA's development review process, view information about development projects in the BPDA approval pipeline, and learn how you can get involved.

www.bostonplans.org

Madison Tropical Parcel 10 | Bostonplans.org

Development Projects and Planned Development Areas (PDAs) that the Development Review division is coordinating.

www.bostonplans.org

(don't be confused by the render at the top of the second link -- someone at the BPDA seemingly forgot to update that)

Welp. There goes my late night crayoning lol.

Still, I think the point is that even if you can't have your Urban Ring route go to Nubian, most of the Nubian bus routes post-BNRD will have a UR connection anyway. The only ones that are hurt specifically by not building a W/MC hub are the 14, 38 and 44, none of which are important bus routes.

 

[F-Line to Dudley]

F-Line has a thorough accounting of its demise...somewhere around here. (I'm not in the mood to engage with the search function at the moment, but I'll link to it if I come across it.)

As I recall from that, Phase III died because of the cost considerations. The tunneling under Essex (as BRT, requiring wider tunnels than LRT) was part, but not all of the cost blowout. They had an extremely difficult time trying to find an alignment to connect the Washington Street end to the tunnel (as a result of shotgun-marrying the elevated 'replacement' to the Seaport/Airport project). The preferred alignment when the project died called for a portal around the intersection of Charles and Tremont, then tunneling under Charles, Boylston, and Essex to the Transitway. I don't think proximity to the burying ground was a big concern, given they were going to be under the Green Line. The biggest cost blowouts came from having to underpin Boylston and Chinatown rather than specifically the narrowness on Essex (though that one wasn't necessarily easy or cheap, it just wasn't as fatal as underpinning two levels of 1897-built Boylston plus 1908-Chinatown).

The project's fatal blow was in 2009, when the FTA changed its rating of the project to "Medium Low", rendering it ineligible for New Starts match funding. The immediate consequence of that was that there was no money to take it from 60% design to 100%. It had been on shaky ground with the Feds for 3 years prior to that over escalating cost projections and the preferred alignment controversies, but the last straw for them was when the budget had cleared $2 billion (with no end in sight). In their estimation at that level of cost leakage and with so much worsening of projected service levels through the tunnel as the design's compromises started to stack up, the project was no longer delivering max-leverage transit for its Federal contribution...and so it was to no longer be funded like a max-leverage project. The state was in no position to try to fund it alone, withdrew all its New Starts applications, and placed the project on indefinite hold. And that's where things have stood since at least Summer 2010.

There was no one failure point with the alignment. It was chewing costs everywhere. The underpins at Chinatown and Boylston (and the bad bus performance at the grades on those underpins limiting service levels). The underpins of the Green Line tunnel for getting to Charles St. South. The building mitigations on Essex. The building mitigations on Charles South. The historical impacts around the Common. It was all leaking into a torrent of overruns that crushed the FTA rating. Thus, to fix it you pretty much need to avoid every segment of the attempted alignment. Which thankfully you can do by reanimating at least a portion of the Tremont St. LRT tunnel, pair-matching the Transitway with the northbound Green Line instead of southbound Silver Line, picking a wider and more urban-renewal tamed South End street alignment (multiple possibilities), and choosing Tufts instead of Chinatown as an offset instead of stacked Orange Line transfer. So we're not out of options by a longshot. It's just going to have to be radically reimagined because nearly every tunnel foot on the alignment they originally chose was just hemmoraging cost to no end.

July 2002 - May 2009 covers the amount of time the project was under New Starts consideration, and when all of the major design work took place. The plot was lost at some point during that churn...but no a-ha! moments. It was a steady drip-drip-drip of obstructions here, opposition there, revised cost projections there coming from all sides...until the alignment was in the shakiest possible shape at $2.1B and rising with 40% of design still left to go. While it's possible they didn't know ahead of time it was going to be as extremely difficult going as it ended up being, critics had been warning just that for years that BRT on that alignment linking those neighborhoods in particular wasn't going to deliver on its promises. So it's fair to question who was doubling-down on it that hard for that long in the face of that many problems, and why that was their political choice of hill to die on. It definitely had a big-chill effect on major infrastructure projects, coming on the heels of the Big Dig and at the tip of a lot of major transit commitments. Us taking most of the 2010's off from promising new things in transit (and needing to be dragged kicking and screaming into something as comparatively basic as GLX) has definite roots in how badly SL Phase III belly-flopped.

 

[Teban54]

The project's fatal blow was in 2009, when the FTA changed its rating of the project to "Medium Low", rendering it ineligible for New Starts match funding. The immediate consequence of that was that there was no money to take it from 60% design to 100%. It had been on shaky ground with the Feds for 3 years prior to that over escalating cost projections and the preferred alignment controversies, but the last straw for them was when the budget had cleared $2 billion (with no end in sight). In their estimation at that level of cost leakage and with so much worsening of projected service levels through the tunnel as the design's compromises started to stack up, the project was no longer delivering max-leverage transit for its Federal contribution...and so it was to no longer be funded like a max-leverage project. The state was in no position to try to fund it alone, withdrew all its New Starts applications, and placed the project on indefinite hold. And that's where things have stood since at least Summer 2010.

There was no one failure point with the alignment. It was chewing costs everywhere. The underpins at Chinatown and Boylston (and the bad bus performance at the grades on those underpins limiting service levels). The underpins of the Green Line tunnel for getting to Charles St. South. The building mitigations on Essex. The building mitigations on Charles South. The historical impacts around the Common. It was all leaking into a torrent of overruns that crushed the FTA rating. Thus, to fix it you pretty much need to avoid every segment of the attempted alignment. Which thankfully you can do by reanimating at least a portion of the Tremont St. LRT tunnel, pair-matching the Transitway with the northbound Green Line instead of southbound Silver Line, picking a wider and more urban-renewal tamed South End street alignment (multiple possibilities), and choosing Tufts instead of Chinatown as an offset instead of stacked Orange Line transfer. So we're not out of options by a longshot. It's just going to have to be radically reimagined because nearly every tunnel foot on the alignment they originally chose was just hemmoraging cost to no end.

July 2002 - May 2009 covers the amount of time the project was under New Starts consideration, and when all of the major design work took place. The plot was lost at some point during that churn...but no a-ha! moments. It was a steady drip-drip-drip of obstructions here, opposition there, revised cost projections there coming from all sides...until the alignment was in the shakiest possible shape at $2.1B and rising with 40% of design still left to go. While it's possible they didn't know ahead of time it was going to be as extremely difficult going as it ended up being, critics had been warning just that for years that BRT on that alignment linking those neighborhoods in particular wasn't going to deliver on its promises. So it's fair to question who was doubling-down on it that hard for that long in the face of that many problems, and why that was their political choice of hill to die on. It definitely had a big-chill effect on major infrastructure projects, coming on the heels of the Big Dig and at the tip of a lot of major transit commitments. Us taking most of the 2010's off from promising new things in transit (and needing to be dragged kicking and screaming into something as comparatively basic as GLX) has definite roots in how badly SL Phase III belly-flopped.

Thanks, great post as usual. Sad to think that if only we opted for LRT instead of BRT from the beginning, maybe the alternative, more logical routing of integrating both fronts into GL branches would have come more naturally, instead of dying on the hill of the Essex St routing.

Speaking of which... If we were to put forward a proposal to "restart Phase III" and brand it as "extend both ends of SL to meet the Green Line by reusing the Tremont St subway, with branches going to Nubian and Seaport via Bay Village, and converting the mode to LRT with much greater capacity and reliability", which should hopefully result in much more manageable costs, what are the chances of this actually going anywhere?

 

[Delvin4519]

Went on a mind thought train and categorized transit routes into 2 categories.

A: All dedicated transitways, these can be rail tunnels, rail stub tracks, dedicated busway tunnels, or abandoned railways. Subway tunnels are differentiated from dedicated ground level rail lines that funnel from North Station, South Station, or the BRB&L terminals. This includes the stub track and the SL transitway in South Boston.
B: All street running routes and on street streetcar routes. This means the B, C, and E streetcar trolley lines are categorized with the buses, west of Kenmore, outside the tunnels.
Categorizing this way, this makes it easy to see where rail routes must define rail replacement shuttle bus routes, separate from bus routes and streetcar/trolley lines.

Inner boston core transitways and railways current/historical/abandoned, plus the rapid transit branches to Braintree via Quincy Center, and Riverside. I've also added the Watertown Branch extension to Waltham Center, even though it's outside of the inner core.

From this map, without using the North-South Rail Link, all rail lines on this map, are able to be fed directly into the subway tunnels past Downtown and through-run, on the other side, with the exception of the Fairmount line.

All northern routes are able to be fed into the Orange Line tunnel, with the exception of the Blue Line, and the B&M East Boston Line along Route 1A in East Boston. The Minuteman Branch can be fed into the Red Line tunnel. All branches that pass through the core of Somerville (E. Somerville/Union Sq) can also be fed into the Green Line tunnel.

All southern routes are able to be fed into the Orange Line tunnel, with the exception of routes that can be fed into the Red Line tunnel, and the Fairmount Line. The Highland Branch can be fed into the Green Line tunnel, but historically connected to the ROW that can feed into the Orange Line tunnel, so we'll say it can feed into both GL and OL tunnels.

Here are some of my thoughts on connecting existing transit and rail lines to form rapid transit lines, before adding new lines or digging new tunnels:

1. The Red Line tunnel will add a connection to Waverley. Braintree trains would be assigned to run to Waverley, as it could potentially be extended to Waltham Center, and Braintree trains serve Quincy Center. Ashmont trains will travel to Arlington Heights Busway. The Mattapan via Neponset line will be sent to terminate in the Seaport district via the Convention Center.
2. A rapid transit line would run from Waverley, to Davis, to East Somerville, then loop northwards to continue to Wellington to Linden Sq. This provides a one seat ride between Gilman Sq. and Davis. This is a crosstown connection line.
3. A rapid transit line from Newton Corner, then use the entire Grand Junction ROW. This is a crosstown connection line.
4. Aside from those two lines, Wellington Station would see the Medford Sq. Branch and Oak Grove Branch travel into downtown. The main corridor for Linden Sq. and the Chelsea-Revere line would travel to downtown via Everett. So 2 branches on each corridor.
5. For the Green Line, the Medford Branch would be extended to West Medford. There are now only 3 branches, as I removed street running service from the Green Line central subway tunnel due to reliability issues of street running traffic. Watertown Branch trains will be assigned to run through the Pleasant St. portal. West Medford trains will run through the Symphony portal (tunnel extension to cover the rest of the E branch). Arlington Heights trains will run via Porter and Union Sq., and continue through Kenmore (service continues on the D branch through Reservior).
6. On the south side, since the Medford Sq. and Oak Grove Branches are shorter, they will be assigned to run the longer routes to Dedham Mall and Wolcott Sq. Readville. Oak Grove trains will continue due south to Readville, Medford Sq. trains will curve westwards to Dedham Mall.
7. Linden Sq. via Everett trains will travel to Newton Corner, Chelsea-Revere trains will travel to West Roxbury VA Hospital.
8. The Orange Line tunnel is already responsible for now carrying 4 branches. However, there are currently no direct one seat service between Community College to any of the Somerville based rapid transit routes, even though there are 2 potential routings to reach Somerville from Community College. Charlestown riders must travel to Sullivan or North Station to transfer to a Somerville based route.

The Blue Line is mostly self contained, aside from extensions to Charles MGH, and potential extension past Revere/Revere St. to Lynn, which would continue to Salem Depot and Beverly. The B&M East Boston branch is mostly industrial along Route 1A, so unfortunately, I'm not sure if that line holds much value for connection. It does bring East Boston riders closer into the core portion of Revere, as opposed to Revere Beach, which is slightly further away from the core of Revere than the B&M East Boston branch. For now, it could be fed into the Blue Line tunnel.

The Fairmount Line is currently not connected and currently dead ends at South Station. The same is true for the Seaport Transitway.

Anyhow, these are my thoughts. I've left the map blank, since there's a lot of permutations on how to connect each line. The map here only extends to the extent of streetcar suburbs, as further out, the permutations for branches increases dramatically. Some ideas of note:

1. Use as many historical/existing lines as possible. There is no need to: use every junction in areas with more than 1 junction; or to use every dead end line in East Boston or South Boston.
2. Consider disconnecting the Green Line tunnel from all street running services, and only serve lines that are not street running. You may convert the Green Line tunnel to heavy rail if desired. This is due to reliability issues with street traffic, and the 4 street running branches decreasing capacity in the central tunnel.
3. Extra non-existant ROWs or tunnels that aren't current or abandoned are mostly not in scope of this. There are potential extensions of the Green Line subway tunnel from Symphony to Brigham Circle to Arborway, or Pleasant Street to Nubian to LMA/Forest Hills. Aside from that, the scope here is to reuse as many existing and former rail lines, and to minimize new tunnel/ROW digging. The idea here is the best way to optimize through running existing/historical ROWs. New metro tunnels could be dug under existing bus routes, so new tunnels would mostly be identifying Key Bus routes to dig subway tunnels underneath them.