Crazy Transit Pitches

[Brattle Loop]

Thanks for the info regarding the conditions of the existing tunnels/track at Copley Junction. I do not have the ridership data for people getting on/off from Brigham Circle to Heath St., but the stops between Heath St. and Brigham could be eliminated. E route would then terminate at Mattapan.

You can't eliminate Heath Street, not with the VA hospital there. (The T has tried, multiple times, and gets a ferocious outcry from the federal legislators, among others, every time.) Running light rail from Mattapan to Heath and/or Brigham Circle, without touching the subway, isn't likely going to pass a cost-benefit analysis because of the forced transfer to the E-branch. Running into the subway via Copley is probably impossible and certainly inadvisable given the Central Subway's capacity issues and the need for more cars to maintain headways on the ZE (especially if it's on the surface).

1. Increase transit options to Franklin park. This should reduce car traffic to the park. Park at Forest Hill and take the train. People that live along the CR route to Forest Hills may opt to taking the train, instead of driving. Similar for families coming from Milton areas or south. Park at Mattapan St. then train to Franklin.

Not an unworthy goal, but one that would benefit from numbers. How many people are going to the park, how many of them are driving, and how does that match up with LRV capacity versus, for instance, bus capacity.

2. Eliminate bus route #16.

I'm not too familiar with the bus system, but if I'm understanding the schedule it looks like the #16 runs on roughly 15-minute intervals. Even if all of those are articulated 60-foot buses, we're still talking significantly less capacity per run than LRT (I'd wager that an XDE60 at crush load can probably carry about half the people of a single-car Green Line LRV). If the #16 is over-capacity at present, it would be significantly cheaper to add more buses than install Green Line infrastructure along the entire route; even then, doubling the #16 schedule would represent roughly equivalent capacity to single-car LRT on the current #16 schedule).

1. Increase Transit along Blue Hill Ave.

Blue Hill, as a transit corridor, is a good idea, going through an area of mostly good density (apart from the parks) that's bracketed by two HRT lines but only served directly (on rails) by Fairmount (which has mode-related reasons that it can't be a proper rapid transit line, though there is room for improvement from its current operation). The problem is that, as @Riverside pointed out, there isn't a natural place for that transit to feed into. I question whether running a tunneled ZE into the Central Subway is feasible at all, between Copley Junction's foibles and the additional capacity that would be necessary to maintain headways on the long run down to Mattapan, but it would at least be more feasible than running a surface ZE onto the E-branch and then into Park via Copley, which would be operationally impossible. (Similar concerns would apply to non-tunneled routings Mattapan-Nubian-Park.) Improved bus transit feeding into Orange, Red, or, ideally, a Green Line extension to Nubian is probably the best-fit pitch for Blue Hill. (Maybe a surface LRT branch if Huntington was tunneled and connected to the Tremont tunnels per some of the Green Line Reconfiguration thread ideas, but that's a whole other conversation.)

2. One seat ride from Park St. to Franklin Park. That sounds awesome, IMO.

One-seat rides usually sound awesome, but agencies' obsession with them can be directly harmful to the goal of improved transit. There are only so many transit dollars to go around, and it may well be that a more efficient and overall-beneficial use is to use a portion to improve bus connections to Blue Hill (including the park) from the existing transit corridors, and using the remainder for other projects. Annoying, if your primary interest is getting to the park as quickly and easily as possible, but only those nodes with the highest demand merit the most attention for one-seat rides. How many bus corridors could get dedicated, protected lanes and improved facilities for the cost of one LRT line down Blue Hill? Ten people having to transfer so that a thousand get to work faster would be a tradeoff that any transit planner would take in a heartbeat.

 

[Riverside]

I second everything @Brattle Loop has said. Blue Hill Ave has all the trappings of a streetcar boulevard, forming the strong spine of a neighborhood, except that it's the wrong-way-round from a transit-planning perspective: at its northern end, the travel corridors branch out and become diffuse before joining up into a new set of strong spines as they traverse the Inner Belt region and continue on to downtown. In an "ideal" scenario, the branching and diffusion would occur at the outer end of the spine, not the inner.

I would love to someday see a Park-Mattapan via Blue Hill LRT, but I'm very skeptical it could ever work. Even back in the streetcar heyday, BERy (or its predecessors) had stopped running streetcar routes from beyond Dudley (Nubian) into the subway by 1920 -- it was just so much faster to transfer to the El. Such a route would be 6.5 miles long -- we'll call it 6 miles on the surface -- and, assuming you kept SL4/SL5 stop spacing of ~1,000 feet, would need to have over 30 stops. (Twice as many as today's B and C branches. Even the Arborway Line, back in the day, only hit 23 stops, and some of those were notably less than 1,000 feet apart.)

And I do think you'd need to maintain the 1,000 ft stop spacing of the current bus routes. And I think it's interesting and worth noting that all of the stops you've placed on Blue Hill Ave are less than half a mile from a (potential) Indigo Line station. If you're looking for express service through Dorchester, 10-min-or-better frequencies on the Fairmount Line seem like a much better bang-for-buck. If you're going to go to the trouble to build LRT, it seems worth avoiding duplicating the existing infrastructure.

On a more positive note: I really do like the idea of recreating these old streetcar boulevards. I don't live in Dorchester and might feel differently about the idea if I did, but in the abstract it sounds like a nice placemaking venture, particularly if it can be used to knit together the fabric of the neighborhood.

If Dorchester said, "We want an LRT line, not because we want to get downtown faster but because we want our community to be centered on a light rail line," I would build it along Blue Hill Ave and Columbus Ave:

Wide boulevards that are already planned to be redesigned with dedicated transit lanes. Follow well-established travel corridors used by current bus routes. And you have options at the northern end: if Tremont St receives dedicated transit lanes, then you can head over to Huntington from there, and terminate at surface level around Brigham Circle -- essentially doing the same thing the proposed T22 will do, as a Longwood-focused service; you could alternatively extend further up to Columbus to hit Ruggles, with the option to jog over to Huntington from there (again, as a Longwood-focused service; downtown commuters would transfer to Orange as they do today, or would take Indigo). Non-revenue connections to Huntington would provide yard access to the rest of the LRT network.

^ This is actually pretty close to what you propose. The main difference is traveling via Egleston Sq rather than Nubian Sq. Is it unequivocally the most in-demand corridor? No. But it would be the easiest option to build, and if the goal is building an LRT line for the sake of building an LRT line (which I think is indeed a valid reason), then you would satisfy that goal and have a reasonably good piece of transit while you're at it.

Subsequently or alternatively, you could anchor an LRT line at the northern end to an infill Indigo Line station at Ceylon Park:

(In principle, you could also anchor it instead at Four Corners/Geneva, but there's more room for a surface station at Ceylon Park.)

In this iteration, the Blue Hill LRT is explicitly a local feeder service into high-frequency mainline service on the Indigo Line. If Mattapan Yard could be beefed up a bit, maaaaaaybe you could build a Mattapan-Ceylon Park LRT line without building a connection to the Green Line? Unclear.

I agree that Columbia Road could be nicely reimagined as a streetcar boulevard, as a Phase 2 or 3 building on the above:

And over time, you could continue to convert large roads into transit boulevards -- Morton St would probably be the next candidate, either toward Forest Hills, Ashmont, or both. I think it could be done. But in general, I think @Brattle Loop's fundamental point holds true: How many bus corridors could get dedicated, protected lanes and improved facilities for the cost of one LRT line down Blue Hill? Ten people having to transfer so that a thousand get to work faster would be a tradeoff that any transit planner would take in a heartbeat.

If the desire explicitly is for an LRT line for an LRT line's sake, then I think you're looking at Columbus-Blue Hill, potentially with a spur to Ceylon Park. (Which, incidentally, gets you that nice connection to Franklin Park, at three of its five corners.) But beyond that, I think you're looking at the combination of:

• As-close-to-full-build BRT infrastructure on all the proposed high-freq corridors (eg. T28, T22, etc) -- Blue Hill, Warren, Columbus, Morton, Talbot, Washington going into Ashmont, Columbia, with extension to LMA and Kenmore on the north end
• 10-min-or-better Indigo Line service on the Fairmount, with full fare integration as a de facto rapid transit service
• Extending the Green Line to Nubian (though I have an idea about this I may post later)
• Potentially converting the Mattapan Line to heavy rail

All of which (possibly except Mattapan HRT) reflect known, stable travel patterns, build on existing infrastructure, and can be built incrementally and evolve with the community's needs over time.

 

[Riverside]

On a slightly different note...

Extending the Green Line to Nubian has been a proposal at least since the closure of the El, if not earlier. It features on pretty much every fantasy map you come across out there.

Many crayon maps of the self-described "crazier" variety venture one step (or perhaps ten steps) further and extend the Green Line all the way down to Mattapan. Back in the day, I myself extended it to both Mattapan and Ashmont. As discussed above, I no longer think it's viable to run a service from Park Street to Mattapan.

One obstacle -- whether an LRT service or BRT bus lanes -- has been Warren St, in particular between Blue Hill Ave and Boston Latin. This windy half-mile segment is the only stretch between Nubian and Mattapan where the street narrows such that transit lanes would be incompatible with continued on-street residential parking. It's worth noting though that the rest of Warren St, north of Townsend/Quincy Sts, is plenty wide enough for transit lanes. It really is just that one segment.

But that one segment has made me skeptical even of more reasonable proposals to extend a Nubian Branch down to Franklin Park, which itself forms a logical transit hub at the convergence of most of Dorchester's major corridors.

However, our discussion above has made me look again at the rail transit coverage of these potential crayon maps. Assuming Green to Nubian and a maximally-rapid-transit-like Indigo Line, plotting out the very approximate 15-minute walksheds gives us something like this:

There is a large gap between Indigo and Orange, but the southern two thirds of that gap is almost entirely parks and cemeteries. Which means there's actually a relatively small pocket of Dorchester that would be without rail access.

So, what I'm considering: given the broad width of the upper section of Warren St, would a further extension of a Nubian Branch to MLK Boulevard/Boston Latin be worth exploring?

With such an extension (a little less than a mile long), the walkshed map gets almost completely filled in:

(And, for what it's worth, a Blue Hill-Columbus LRT line entirely fills the remaining gap; if run to LMA, I'd argue that forms a legitimate transit + commuting corridor.)

Looking at the satellite images, it's unclear whether there's a lot of available surface space on Warren near Boston Latin -- probably not enough to create a "Nubian South" bus transfer terminal. So probably you'd have a simple island platform for Green Line trains to terminate on, with bus routes from the south continuing up Warren alongside the LRT trains, to terminate at Nubian or carry on to Longwood etc.

What do you get out of this? I'm not sure. It does expand the reach of the OSR just a bit, and I think the gap I highlighted in the first walkshed map is a legitimate one. Warren is wide enough, and the extension short enough, that I don't think you'd kill reliability on the Washington St segment. In my most recent Green Line Reconfig pitch, I send two out of the Green Line's four southern branches to Nubian, one of which short-turns at Government Center; I'd probably extend only one of these down Warren St, given that it would still be sharing lane capacity with the T28 and T23.

On the other hand, this particular neighborhood hasn't had a one-seat ride to downtown via Nubian in over 100 years. Is there demand for it? Would there be interest in it? (There's a longer conversation to be had here about gentrification and the rising rents we are seeing in the wake of GLX.) I'm not sure.

But, at least at first glance, I think it could be done, and I'm definitely going to think about it further.

 

[Brattle Loop]

Looking at the satellite images, it's unclear whether there's a lot of available surface space on Warren near Boston Latin -- probably not enough to create a "Nubian South" bus transfer terminal. So probably you'd have a simple island platform for Green Line trains to terminate on, with bus routes from the south continuing up Warren alongside the LRT trains, to terminate at Nubian or carry on to Longwood etc.

What do you get out of this? I'm not sure. It does expand the reach of the OSR just a bit, and I think the gap I highlighted in the first walkshed map is a legitimate one. Warren is wide enough, and the extension short enough, that I don't think you'd kill reliability on the Washington St segment. In my most recent Green Line Reconfig pitch, I send two out of the Green Line's four southern branches to Nubian, one of which short-turns at Government Center; I'd probably extend only one of these down Warren St, given that it would still be sharing lane capacity with the T28 and T23.

On the other hand, this particular neighborhood hasn't had a one-seat ride to downtown via Nubian in over 100 years. Is there demand for it? Would there be interest in it? (There's a longer conversation to be had here about gentrification and the rising rents we are seeing in the wake of GLX.) I'm not sure.

But, at least at first glance, I think it could be done, and I'm definitely going to think about it further.

I like this idea. It's unfortunate that the narrow stretch of Warren between MLK and Blue Hill probably makes Nubian-Blue Hill a non-starter. Aside from the parking issue, it'd have to be street running, which would risk clobbering the entire Nubian branch (or at least the south-of-Nubian segment), particularly if Tremont-Nubian was surface from the Pike to Nubian, so I like this idea that hits most of the density between Fairmount and Orange on the northern half of the Mattapan-Nubian corridor. Unless something happened to that McDonald's, the mall, or the Walgreens a little further south (not that I'm rooting for disappearing businesses, mind you) that made one of those parcels available, I agree that we'd be talking about a simple, small-scale terminus with at most them borrowing a bit of the parking lot for a tail track-slash-mini-yard. Bus service would need to remain robust to bridge the Blue Hill-upper Warren gap, particularly if Mattapan-LMA LRT wasn't independently built.

As for reliability on Washington, I do think that we might have different conceptions of how to handle it, but I'd envision a good number of short-turns at Nubian, sort of a reverse E-branch pre-1985 when the Heath-Lechmere runs were overlaid on top of the Park-Arborway service, except here with downtown-Nubian as the main service and wherever-Warren as the secondary pattern more readily sacrificed in the event of disruption (because, as noted, Warren doesn't have room for a Nubian-scale bus terminal).

 

[Aprehensive_Words]

View attachment 31822

When I see this, my immediate thought is: Should it be a branch of a southern-half Urban Ring via Nubian in one direction and LMA in another?

 

[Riverside]

Unless something happened to that McDonald's, the mall, or the Walgreens a little further south (not that I'm rooting for disappearing businesses, mind you) that made one of those parcels available, I agree that we'd be talking about a simple, small-scale terminus with at most them borrowing a bit of the parking lot for a tail track-slash-mini-yard

Yeah I was thinking about this a little bit more, and I think unfortunately the footprint would need to be a little bit more than just an island platform in the middle of the street -- the problem is that turning trains would block the lanes used by buses. (Plus an island platform won't work for the buses anyway.) So, something a little more elaborate would probably be necessary.

The City owns the land that Boston Latin sits on, so maybe something creative could be done there:

Alternatively, Martin Luther King, Jr Blvd is something like 85 feet wide not including sidewalks, so a simple terminal could fit there without blocking the bus routes. It shifts our walkshed north slightly, but probably wouldn't shift the cost-benefit analysis greatly.

But again, I'm not sure whether Green-to-Warren is a good idea -- right now I'm really just spitballing what could potentially be feasible, should anyone wish to build it.

As for reliability on Washington, I do think that we might have different conceptions of how to handle it, but I'd envision a good number of short-turns at Nubian, sort of a reverse E-branch pre-1985 when the Heath-Lechmere runs were overlaid on top of the Park-Arborway service, except here with downtown-Nubian as the main service and wherever-Warren as the secondary pattern more readily sacrificed in the event of disruption (because, as noted, Warren doesn't have room for a Nubian-scale bus terminal).

Yeah, I had originally been thinking to extend my downtown-terminating service, mirroring the Arborway-Park model:

• F: Warren - Government Center
• G: Nubian - Grand Junction

But I like your idea too:

• F: Nubian - Government Center
• G: Warren - Grand Junction
  • short-turns from either direction at Government Center as needed to maintain reliability

 

[Riverside]

When I see this, my immediate thought is: Should it be a branch of a southern-half Urban Ring via Nubian in one direction and LMA in another?

Not sure what you are describing here, can you elaborate?

 

[Tallguy]

I remain convinced that OL to Nubian is the way to go.

 

[Brattle Loop]

I remain convinced that OL to Nubian is the way to go.

How? There's no immediately-identifiable geometry to somehow swing from the Southwest Corridor trench to Nubian, well, anywhere, really. I suppose you could theoretically branch it south of Chinatown where the current alignment meets the old elevated incline alignment, but there was never a flying junction there, so you're either doing some fairly-delicate surgery under a major hospital (...and it's only after I wrote that did I notice the pun) or accepting a permanent operational demerit in the form of a flat junction. On top of that, even at boosted frequencies through the Washington Street Tunnel, does Orange have the capacity to maintain acceptable headways on two branches splitting that close to downtown. Pre-pandemic, at least, Back Bay had the third-highest ridership on the OL, and the BBY-NS Orange Line is the only current direct connection between the two halves of the Commuter Rail network.

More to the point, while I recognize that this is Crazy Transit Pitches, the Orange Line obviously can't run on the surface to Nubian. I think the Maple Leafs would sooner win the Stanley Cup than the city would permit a new elevated to be built down Washington Street, meaning it'd have to be a tunnel, which would cost a fortune. (While there's debate here and in the Green Line Reconfiguration thread about the operational trade-offs of surface versus tunneled LRT in a Green Line F-branch to Nubian, Green at least has the low-cost option of surface running that Orange inherently lacks.) I for one find it hard to believe that Orange is a viable option outside of the God Mode thread.

 

[Tallguy]

How? There's no immediately-identifiable geometry to somehow swing from the Southwest Corridor trench to Nubian, well, anywhere, really. I suppose you could theoretically branch it south of Chinatown where the current alignment meets the old elevated incline alignment, but there was never a flying junction there, so you're either doing some fairly-delicate surgery under a major hospital (...and it's only after I wrote that did I notice the pun) or accepting a permanent operational demerit in the form of a flat junction. On top of that, even at boosted frequencies through the Washington Street Tunnel, does Orange have the capacity to maintain acceptable headways on two branches splitting that close to downtown. Pre-pandemic, at least, Back Bay had the third-highest ridership on the OL, and the BBY-NS Orange Line is the only current direct connection between the two halves of the Commuter Rail network.

More to the point, while I recognize that this is Crazy Transit Pitches, the Orange Line obviously can't run on the surface to Nubian. I think the Maple Leafs would sooner win the Stanley Cup than the city would permit a new elevated to be built down Washington Street, meaning it'd have to be a tunnel, which would cost a fortune. (While there's debate here and in the Green Line Reconfiguration thread about the operational trade-offs of surface versus tunneled LRT in a Green Line F-branch to Nubian, Green at least has the low-cost option of surface running that Orange inherently lacks.) I for one find it hard to believe that Orange is a viable option outside of the God Mode thread.

OL ridership is heavily bias to the north, and with development in Sullivan, Medford/Malden, and Everett, that bias will only increase. There would be spare capacity to the south if frequencies are boosted to , say 3 min to accommodate northern increased. I am suggesting that the split happen at Shawmut St under the NEC. The tunnel(yes, tunnel, because street running GL is not "equal or better") runs under Shawmut until Peters Park then cutts under to Washington

 

[Tallguy]

OL ridership is heavily bias to the north, and with development in Sullivan, Medford/Malden, and Everett, that bias will only increase. There would be spare capacity to the south if frequencies are boosted to , say 3 min to accommodate northern increased. I am suggesting that the split happen at Shawmut St under the NEC. The tunnel(yes, tunnel, because street running GL is not "equal or better") runs under Shawmut until Peters Park then cutts under to Washington

Sorry, cuts. This could be done after or during the NSRL disruption.

 

[Aprehensive_Words]

Not sure what you are describing here, can you elaborate?

Sorry! A couple reasons behind my question:

• As a feeder to the Orange Line, it's powerful. But that also puts a lot of pressure on the line right at that station because there's a lot of folks who will want to go elsewhere in the city.
• The opportunity is there, too. If Mayor Wu's team is any kind of smart, Tremont/Malcom X will be getting dedicated transit lanes somewhere in the next 5 years, anyway, to support BNRD. This opens up routings to Mass. Ave. or Melnea Cass via Nubian or Kenmore or West Station via Longwood, does it not?
• Other systems repeatedly show the benefits of network effects in boosting ridership. More connections to other lines = more riders.

 

[Riverside]

Sorry! A couple reasons behind my question:

• As a feeder to the Orange Line, it's powerful. But that also puts a lot of pressure on the line right at that station because there's a lot of folks who will want to go elsewhere in the city.
• The opportunity is there, too. If Mayor Wu's team is any kind of smart, Tremont/Malcom X will be getting dedicated transit lanes somewhere in the next 5 years, anyway, to support BNRD. This opens up routings to Mass. Ave. or Melnea Cass via Nubian or Kenmore or West Station via Longwood, does it not?
• Other systems repeatedly show the benefits of network effects in boosting ridership. More connections to other lines = more riders.

I have thoughts on this, but I actually was more asking about what you meant in terms of the actual route you were describing with “Should it be a branch of a southern-half Urban Ring via Nubian in one direction and LMA in another?” In particular, I’m not sure I follow what you mean about “Nubian in one direction and LMA in another”.

 

[Aprehensive_Words]

So here, in gold, is an example of what that radial routing might look like (crimping a bit from others' routings through Fenway/Brookline, here) from a sketch I did a while back. The Nubian routing assumes you don't have enough trains coming down an F branch that it wouldn't prevent you briefly interlining an Urban Ring service through there.

I'm too ignorant of rail ops to know how you would structure this with a branch heading down Columbia/Blue Hill.

• Does it start at West Station/Kenmore, and branch at the Roxbury Crossing into Mattapan and eastbound ring arms?
• Does it start at Mattapan and branch at Roxbury Crossing into eastbound and westbound arms?
• Start at the Seaport or somewhere else to the east, then branch into Mattapan and Kenmore/West Station arms?

 

[Riverside]

^ Gotcha gotcha, that is super helpful, thanks for clarifying! A few various thoughts incoming.

First, and this isn't directly related to your question, but on further consideration I would run this streetcar all the way to Ruggles, and not just Roxbury Crossing; if we are very lucky, in an NSRL future, Ruggles will have high-frequency express service to Back Bay, South Station, and North Station, and be a valuable alternative to the Orange Line. So, unless there happens to be a massive transit investment between Roxbury Crossing and Longwood, I would aim for Ruggles first.

Second: regardless of what kind of Urban Ring gets built, a Columbus-Blue Hill streetcar would basically have to have a track connection to the Huntington Line, in order to be able to interface with the rest of the LRT network for equipment moves etc. As we've discussed in the Green Line Reconfiguration thread, Ruggles St should be able to accommodate rapid transit lanes (initially used for buses, but could add in streetcar tracks potentially), so it would not be difficult to extend the extra one-quarter mile from Ruggles to a surface stop on Huntington.

Now, I'm taking for granted that there will be some sort of circumferential service along the southern half of the Urban Ring corridor, regardless of whether this Columbus-Blue Hill line (just gonna call it the "Franklin Park streetcar") gets built. If that service is LRT, then yes, that opens up options for thru-running. Do you need those thru-runs? I don't think so, at least not if the Franklin Park line is built specifically for the sake of centering those neighborhoods around a streetcar line; Longwood is a major job center and commuting destination, so a Longwood-Mattapan service is reasonably self-sufficient -- you're not going to overload the Orange Line any more than you do today, and as mentioned the NSRL would also help you spread out the passenger load.

You've asked an ops question about what is possible -- can you do those thru-runs? (Should you?) Rail ops is not an area I have particular expertise in, so take all this with a grain of salt, but my thinking:

In general, the Franklin Park streetcar is a looooong route. From Ruggles, it's over 5 miles of surface street-running with local stops. Even with dedicated lanes and priority signaling, that's still lots and lots of intersections and lots and lots of stops. (For comparison, look at the list of current US light rail systems -- sort it by system length and focus on the systems that have 1 line for a quick reference; most are shorter than 4 miles, and I believe most of the longer ones have private ROWs with rapid transit stop spacing.) To me, at most, I'd extend the service across Longwood to connect with the Green Line, either on the D Branch or maaaaybe at Kenmore; more realistically, I would hook over to Huntington and terminate at the current LMA stop: that gives you a strong destination, good transfers, and minimal interlining with other services.

The idea of a hook-around service that originates at Nubian or points east is an interesting one. I could be convinced by ridership analysis that a direct connection to Nubian would be valuable, though I suspect you wouldn't see a huge added benefit if you're already hitting Ruggles, and you'd be hit with the cost of the indirect route. I definitely wouldn't want to through-run from east of Nubian down to the Franklin Park line -- too long (as discussed above) and too roundabout.

Originating at South Station or the Seaport is intriguing; I wouldn't route it through Nubian, but potentially you could look at alternatives via Newmarket or JFK/UMass + Columbia Road. That being said, at that point you are almost directly duplicating the Indigo Line; I agree that it's worthwhile to have a local service running parallel to the Indigo's express service, and likewise think a path to the Seaport that bypasses South Station is valuable, but I don't think it's practical (or even necessarily desirable) to string it all into one service. The local service element can be addressed with a Blue Hill-Columbia-JFK/UMass route, and the Seaport access problem will be addressed one way or another by Urban Ring services.

That all being said: at the end of the day, it's relatively easy to reroute trains once the tracks are in place. So, if a Franklin Park streetcar is built, and there are connecting tracks that spread out to Longwood, Nubian, Kenmore, Seaport... the option will be there, and maybe changing conditions in the future will make it worth it! Either way, we know that we need stronger radial transit to Dorchester, and stronger circumferential service all around; how much interoperation there is between the two -- to some extent, only time can tell.

 

[Brattle Loop]

OL ridership is heavily bias to the north, and with development in Sullivan, Medford/Malden, and Everett, that bias will only increase. There would be spare capacity to the south if frequencies are boosted to , say 3 min to accommodate northern increased. I am suggesting that the split happen at Shawmut St under the NEC.

Even if the main trunk of the OL is running at 3 minute headways, you're not going to be able to get that on the branches. 50-50 split would be 6 minute headways to Forest Hills and Nubian, which is hardly terrible, but doesn't permit growth over what the OL is more-or-less capable of now on the southern end (they used to run close to 6 minute headways, at least at peak, on the Orange, before the line completely fell apart). At the very least you'd need to show data as to why permanently cutting Back Bay (and the rest of the Southwest Corridor) to branch frequencies isn't going to screw up the line. If you start getting dwell time penalties cascading through the system because every train coming in from Back Bay is overcrowded because of insufficient service, for example, that wouldn't help anybody. I don't know how feasible non-balanced branches would be from an operational standpoint, but all you'd be doing is taking from one branch to help another.

I am suggesting that the split happen at Shawmut St under the NEC. The tunnel(yes, tunnel, because street running GL is not "equal or better") runs under Shawmut until Peters Park then cutts under to Washington

Which is suggesting major surgery to a tunnel that wasn't designed to be expanded, which would be required to remain open during as much of the construction as possible. The geometry would be interesting, given the proximity of Tufts Medical Center station and the portal, not to mention the fact that that tunnel is mostly curving in the wrong direction for a Shawmut alignment the whole time. Either way, it'd be a fairly-complex project to build that junction, followed by the relatively-simpler tunneling under streets to Nubian at the going rate, which would cost a fortune. From a technical standpoint it's probably all feasible (though with some operational question marks), but from a Transit Pitches standpoint it's impossible. No, surface-running Green is not "equal or better" than the elevated in many respects; it's a damn sight better than the cruddy silver bus that exists now without so much as a single connection to the RT lines inside fare control. The state and the T were perfectly content in replacing the elevated with nothing, then hamstringing its replacement by choosing a lousy mode and stapling it to a completely different project (Seaport transit). Meaning the real-world decision isn't "Green or Orange". If this OL branch were officially proposed, even an honest tunneling cost estimate would look like the Big Dig compared to a Green Line branch on the surface, because the only new tunneling that's needed would be between the sealed portal and Washington Street. You'd need to show why the state would fund the "better" service in the face of an option that's a.) way cheaper and b.) way better than the existing service, in a historical context where they've demonstrated that they do not care, particularly, about Nubian's quality-of-service.

 

[Tallguy]

 

[Tallguy]

View attachment 31902

Pre-Covid data. OL northbound peaks at BB but reaches 75% of capacity, at 6 min frequencies

 

[Tallguy]

You could increase ridership by 30% without problem

 

[Tallguy]

Will it cost $1B? Yes, but Roxbury deserves the same nice things as Somerville