Green Line Reconfiguration

[Riverside]

Does a Grand Junction GL brach really need to be fed from both ends?

and if both ends, does it need a Wye on the west? Does it need a flying junction ?

Sure, make sure these connections are possible, but are they a prerequisite for cost effective service?

would having/not a West Station change your answer?

My takes:

Sullivan - Grand Junction: Core requirement of the whole shebang - definitely needed

Sullivan - Union: I think this could be a valuable way to provide additional service along the northern tier of the system (e.g. Porter to Chelsea) without straining the Central Subway(s) - not vital but worthwhile

Sullivan - Lechmere: I go back and forth on this, but I think the ability to relieve the Orange Line at Sullivan and offer one-seat rides from Everett is compelling; Chelsea I think is a little more uncertain -- I feel somewhat that Chelsea would be better served by true BRT (I'm talking physically separated lanes and high platforms with pre-payment) on the 111's corridor - 2.8 miles vs 5.8 miles via Lechmere, and somewhat more directly serving the higher-density sections of Chelsea; I'm not sure how many riders would be interested in going the long way around to reach downtown. But I think Sullivan on its own is a pretty compelling destination, so I feel that a Sullivan-Lechmere connector is important

Lechmere - Grand Junction [in Cambridge]: In an ideal world, this service pattern would be unnecessary. However, I don't know if there's a feasible location for a transfer station between a circumferential Grand Junction Line and a radial Green Line; combine that absence with the walking transfer at Kendall, and I think that adds up to making the Grand Junction Line too isolated

Cambridgeport - West Station: Likewise at this point a core requirement

Cambridgeport - Kenmore: I also go back and forth on this; Kenmore itself would be a big destination for the entire Grand Junction corridor, from Sullivan through Cambridge, but I don't see any sense in running Grand Junction service past Kenmore as a wraparound: the 1 bus, the Red Line, or traveling via Lechmere will be faster in I think almost every scenario. Looping at Kenmore gets complicated since the Commonwealth portal does not have access to the Loop, though there are ways around that. (A couple of months ago, I put forward a Crazy Transit Pitch to add a second LRT tunnel to connect Kenmore with Back Bay and the Seaport, and I think that could be interesting.) However, if West Station isn't built -- or otherwise is not yet a connectivity hub -- then obviously Kenmore takes the cake.

I think where I land is that Sullivan - Grand Junction - West Station and Lechmere - Grand Junction - West Station need to be there on Day 1, with provisions for the other legs listed here safeguarded but unbuilt.

 

[Teban54]

One benefit of a Lechmere-Grand Junction connection is providing a direct link between North Station and Kendall for commuter rail riders (especially Lowell and Haverhill lines), basically replacing what EZRide does today. Another benefit is to connect GLX to Kendall, though T101 arguably does that.

 

[Charlie_mta]

Here's my try at snaking a, LRV line on the GJ through the McGrath Hwy area. Yellow is new track, green existing track. Some reconstruction of the EB GLX viaduct where it ramps up from under McGrath Hwy would be needed to fit a new WB tunnel portal under it for WB LRV line on the GJ.

This configuration would enable a Everett/Sullivan - West Station route, and a Lechmere - West Station route.

This would require a "diamond" crossing where the tracks cross:

 

[roy_mustang76]

Here's my try at snaking a, LRV line on the GJ through the McGrath Hwy area. Yellow is new track, green existing track. Some reconstruction of the EB GLX viaduct where it ramps up from under McGrath Hwy would be needed to fit a new WB tunnel portal under it for WB LRV line on the GJ.

This configuration would enable a Everett/Sullivan - West Station route, and a Lechmere - West Station route.

This would require a "diamond" crossing where the tracks cross:

Considering the bottleneck that is the considerably simpler Copley Junction, having that kind of at-grade junction here troubles me. I mean sure, it's not all funneling into one pipe, but I suspect that makes it more brittle. If there's room for a flying junction, that's probably the way to go even at a higher cost imo. But... there are better informed posters than me here, and I'm curious to hear their takes.

It's definitely an inspired take though!

 

[jklo]

Sullivan - Grand Junction: Core requirement of the whole shebang - definitely needed

I can't see them expanding the scope to include something that goes all the way to Sullivan. They'll just have stay on for two more stops to North Station to transfer. I see the benefit but again out of scope.

Same goes with any connection to the B line.

Lechmere - Grand Junction [in Cambridge]: In an ideal world, this service pattern would be unnecessary. However, I don't know if there's a feasible location for a transfer station between a circumferential Grand Junction Line and a radial Green Line; combine that absence with the walking transfer at Kendall, and I think that adds up to making the Grand Junction Line too isolated

Not sure what you are saying but that's what you are going to get. They would find some way to connect the GJ to the Green Line before it hits Lechmere and that's it.

Cambridgeport - West Station: Likewise at this point a core requirement

I'd argue that the Allston project shouldn't even be a dependency on it. Let alone West Station. I say that despite that the purpose of doing this is really to make Harvard happy. You might get some ridership from NS-Kendall riders.

In practice they probably won't get serious until the Allston project gets approved.

I'm hopeful that the MBTA will make their feelings known about how they feel about losing the train moves. That might be asking too much.

 

[Brattle Loop]

I'm hopeful that the MBTA will make their feelings known about how they feel about losing the train moves. That might be asking too much.

I'd imagine that the CR/Keolis people wouldn't love losing the GJ, but, that said, F-Line has a good encapsulation of what it would take to mitigate the loss of the GJ to the CR network here.

 

[Tallguy]

I'd imagine that the CR/Keolis people wouldn't love losing the GJ, but, that said, F-Line has a good encapsulation of what it would take to mitigate the loss of the GJ to the CR network here.

So, the $300M+ the T is planning on spending @ Readville on a new repair facility should ease the pain. And now that CSX owns the Ayer line and the Ag Branch, maybe rebuilding the connection could cut a few minutes off the weekly switch

 

[ceo]

Yeah, if "rebuild the McGrath Highway bridge" becomes part of the equation, a whole lot of things suddenly become easier.

 

[Brattle Loop]

So, the $300M+ the T is planning on spending @ Readville on a new repair facility should ease the pain. And now that CSX owns the Ayer line and the Ag Branch, maybe rebuilding the connection could cut a few minutes off the weekly switch

Yeah, CSX probably isn't going to tolerate Pan Am-level track conditions that made that run such a slog. Removing GJ from the RR mode has prerequisites, but the T and CSX are already working on taking care of some of them (because they serve their own interests) so any MBTA CR claim that the GJ is irreplaceable would probably be more about annoyance at being inconvenienced than actually describing fatal blockers (CSX's freight rights over the GJ would have to be sunset, but given that they no longer use them and have since eaten their way into every other freight route into Boston, that might not be that difficult to accomplish).

 

[737900er]

Sullivan - Grand Junction: Core requirement of the whole shebang - definitely needed

Lechmere - Grand Junction [in Cambridge]: In an ideal world, this service pattern would be unnecessary. However, I don't know if there's a feasible location for a transfer station between a circumferential Grand Junction Line and a radial Green Line; combine that absence with the walking transfer at Kendall, and I think that adds up to making the Grand Junction Line too isolated

Right, you end up with a map that looks like this nonsense:

Because you can't run the UR service through Lechmere, if you run Sullivan-GJ service, there is no choice but to run Lechmere-GJ service too for anyone coming from GLX or the Central Subway. My concern here is that this would be overkill levels of service, particularly offpeak.

 

[Brattle Loop]

Because you can't run the UR service through Lechmere, if you run Sullivan-GJ service, there is no choice but to run Lechmere-GJ service too for anyone coming from GLX or the Central Subway. My concern here is that this would be overkill levels of service, particularly offpeak.

I might be missing something, but as far as I can tell the only missing leg that doesn't have a fairly-easily-identifiable solution for its alignment is Sullivan->Lechmere. GJ<-->Lechmere and GJ<-->Sullivan have possible routes via the GLX inclines, the stub turnouts, and the carhouse leads. To me, it would make the most sense to find some way to fill in the Sullivan->Lechmere leg of the Urban Ring (and it's not as if they don't have options, they'd just take some more effort). Do that, and you get the opportunity for anything to run GJ (or Union, probably) to or from Sullivan or Lechmere, and anything from Sullivan to or from GJ or Lechmere as well, meaning you can fillet and adjust service patterns to meet the prevailing demand. (And if you properly use the Brattle Loop then you avoid mucking with the worst-congested part of the Central Subway to boot.) Offpeak, there'd be nothing to stop you short-turning at Lechmere (or Union, or wherever) so that the headways were appropriate to demand just alternating between GJ<-->Sullivan and GJ/Sullivan<-->Lechmere.

 

[Charlie_mta]

Yeah, if "rebuild the McGrath Highway bridge" becomes part of the equation, a whole lot of things suddenly become easier.

Based on its poor condition, it probably needs major rebuilding, if not outright replacement, anyway.

 

[Charlie_mta]

To me, it would make the most sense to find some way to fill in the Sullivan->Lechmere leg of the Urban Ring (and it's not as if they don't have options, they'd just take some more effort).

A flyover at the newly built GLX junction would work, IMO. Yellow is new track on new viaducts, green is existing track:

 

[Riverside]

I might be missing something, but as far as I can tell the only missing leg that doesn't have a fairly-easily-identifiable solution for its alignment is Sullivan->Lechmere. GJ<-->Lechmere and GJ<-->Sullivan have possible routes via the GLX inclines, the stub turnouts, and the carhouse leads. To me, it would make the most sense to find some way to fill in the Sullivan->Lechmere leg of the Urban Ring (and it's not as if they don't have options, they'd just take some more effort). Do that, and you get the opportunity for anything to run GJ (or Union, probably) to or from Sullivan or Lechmere, and anything from Sullivan to or from GJ or Lechmere as well, meaning you can fillet and adjust service patterns to meet the prevailing demand. (And if you properly use the Brattle Loop then you avoid mucking with the worst-congested part of the Central Subway to boot.) Offpeak, there'd be nothing to stop you short-turning at Lechmere (or Union, or wherever) so that the headways were appropriate to demand just alternating between GJ<-->Sullivan and GJ/Sullivan<-->Lechmere.

I think we are all saying the same thing — I think that @737900er’s point is that you can’t run GJ-Sullivan service via Lechmere. Which does indeed create the silliness illustrated effectively by the diagram in that post. The need for GJ-North Station service is because there’s no transfer point available between Cambridge St and Sullivan (except mayyyyyyyybe at the junction just west of McGrath, with a triangular platform, but…).

 

[Brattle Loop]

A flyover at the newly built GLX junction would work, IMO. Yellow is new track on new viaducts, green is existing track:

There are definitely options. I don't think it's likely, but I'd like if they built it where you indicated, because the views would be cool from a ramp over the existing high GLX ramp

I think we are all saying the same thing — I think that @737900er’s point is that you can’t run GJ-Sullivan service via Lechmere. Which does indeed create the silliness illustrated effectively by the diagram in that post. The need for GJ-North Station service is because there’s no transfer point available between Cambridge St and Sullivan (except mayyyyyyyybe at the junction just west of McGrath, with a triangular platform, but…).

Yeah, my post was partly me trying to get confirmation that I wasn't missing something and a little bit quibbling with the suggestion that multiple service patterns would overload the GJ. I'm having a hard time envisioning any kind of a transfer at the junction near McGrath (in part because I've always been leery about whether there's quite enough space to duck under the Fitchburg and get on-alignment to the GLX incline, with or without eating some of at least the parking lot of that building on the southwest corner of the GJ-Fitchburg junction.

GJ-Sullivan through service obviously can't go via Lechmere, but if the GJ<-->Lechmere and Sullivan<-->Lechmere segments were completed you could effectively bootstrap cross-platform transfers at Lechmere as a Sullivan->GJ service option. (For instance, if there was some technical reason that made it more feasible to link the Sullivan->Lechmere viaduct directly to a GLX viaduct without a link to the carhouse leads. I don't know why they'd do that, but this is the MBTA we're talking about, some future governor could decide that's cheaper, consequences be damned.) Not being able to through-route wouldn't be ideal (and might put way too much stress on the Central Subway especially if they don't fix the Brattle Loop snafu), though you could still have the travel-pattern without the one-seat ride. My personal preference would be to have it built for all route options, but with plenty of signage and effort to encourage cross-platform transfers between the Ring quadrants at Lechmere, to train (no pun intended) the passengers into more experience with transfers and overlapping routes and to encourage more ridership by perceived greater convenience (no need to wait specifically for a Sullivan train if GJ-Lechmere-Sullivan's also an available option, even if that's not the most common or preferred service pattern, it demonstrates greater flexibility and utility...that and I know from experience that there are plenty of people who, however irrationally, get more annoyed waiting on a platform and will accept longer/more complex overall trip sequences if that means they can get going now. Any additional convenience that can be offered is, to me, a good thing.)

 

[737900er]

I think we are all saying the same thing — I think that @737900er’s point is that you can’t run GJ-Sullivan service via Lechmere. Which does indeed create the silliness illustrated effectively by the diagram in that post. The need for GJ-North Station service is because there’s no transfer point available between Cambridge St and Sullivan (except mayyyyyyyybe at the junction just west of McGrath, with a triangular platform, but…).

Sorry, I wasn't clear. My point was simply that if the GJ is operated as a circumferential route, it also needs to be operated as a radial route.
I'm not convinced that West-Kendall-Sullivan (without connection to Kenmore, which seems to be where the current study is focused) would have enough ridership to warrant its own service pattern, but obviously that's something to study.

 

[Brattle Loop]

I'm not convinced that West-Kendall-Sullivan (without connection to Kenmore, which seems to be where the current study is focused) would have enough ridership to warrant its own service pattern, but obviously that's something to study.

Definitely worth study, and to me that service pattern would merit more consideration as, basically, a short-turn if GJ was connected to Kenmore (and what's now the SL3 past Sullivan->Everett) as an option for boosting headways and filling gaps. I'd be curious as to what a study would show, because I agree it doesn't necessarily feel like a viable service pattern right now, especially with basically nothing but a void at Beacon Park and recurring haziness on what's going to be built there and when.

 

[Teban54]

Moving the following discussion here from @737900er 's post, as my comment has little to do with maps and colors:

Also, am I too tired, or are the service patterns could enable a Kenmore-Kendall OSR insane? It seems like there's no way to do it without running a big loop or putting a stub track at Kendall. If you have to run a weird spiral route (Nubian-Boylston-Lechmere-West-Kenmore-GC Loop or Medford-Lechmere-Kenmore-West-Sullivan) your chances of making a decent map seem low.

Maybe my fantasy of a Park-Kenmore-Kendall-Lechmere-Park loop service may not be so crazy, after all

In seriousness, if we assume Kenmore-GJ trains are sent to Sullivan (which is obviously a more logical choice than Lechmere), the terminus on the Kenmore side pretty much has to be Park St, unless Kenmore gets major modifications.

So if we're not running my fantasy loop service (or a slightly-less-insane GC-GJ-Park "C"-shaped service), the most realistic service patterns in a full build that maintain a Grand Junction-Kenmore connection would be as follows, assuming engineering at Brickbottom allows trains in all directions:

1. Chelsea - Sullivan - Kendall - Kenmore - Park St
2. Chelsea - Sullivan - Lechmere - Park St - (Nubian? Seaport? B/C?)
3. West Station - Kendall - Lechmere - Government Center ( - Park? Nubian? Seaport? B/C?)
4. West Station - Kendall - Sullivan - Chelsea

(1) is the only one that can connect Kendall-Kenmore, but its usefulness essentially ends there, aside from sharing the Sullivan-Kendall load with (2). If Sullivan-Lechmere is feasible, demand from beyond Sullivan will likely be easily handled with (2) to downtown and (4) to West Station, so (1) will essentially be a Sullivan-Kenmore shuttle that has to be extended on both ends for practical reasons. (1) may see more use in a world without a Sullivan-Lechmere connection. Unfortunately, I think this might be another reason or excuse for not building a Kenmore-Kendall connection.

Things could get interesting if the initial build doesn't include Sullivan (not explicitly part of Cambridge's feasibility study). Then you would have to send (1) to Government Center or, well... Park. But again, Kenmore is also not explicitly part of the study, so it's possible that we only get (3).

Edit: Thinking about it again, would it be feasible to just run (4) and Park-Kendall-Park loop (or GC-Kendall-Park)?

 

[Brattle Loop]

If Sullivan-Lechmere is feasible, demand from beyond Sullivan will likely be easily handled with (2) to downtown and (4) to West Station

I find it very hard to believe that Sullivan->Lechmere would be outright impossible, though they absolutely didn't make it easy. The value proposition of the Urban Ring's northeast quadrant drops through the floor (especially as LRT) if it can't hit Lechmere because that's how the downtown connections get made. Hitting Kenmore would then become mandatory just to preserve the downtown and transfer access...though it'd make for an underwhelming choice of long way in green (via GJ) or long way in silver (via TWT/SL3).

Things could get interesting if the initial build doesn't include Sullivan (not explicitly part of Cambridge's feasibility study). Then you would have to send (1) to Government Center or, well... Park. But again, Kenmore is also not explicitly part of the study, so it's possible that we only get (3).

GJ<-->Kenmore might lose some of its (artificially inflated) ridership numbers in a Sullivan->Lechmere world by virtue of the distended loop not being necessary, but it'd still be very useful to have. I'll leave it to people with sufficient patience to have pored over the Pike throat plans to answer if it's even technically possible to run GJ->West, but (especially unless there's a reasonable yard), running even GC-Brattle->West->back would be less flexible (and possibly require more cars) than if you could choose-your-own-adventure Kenmore->BU->BC or Kenmore->BU->GJ.

Either way, if there's no GL connection on the B-branch side, you're hard-limited to whatever options exist on the northern end. GJ<-->Lechmere is easily the most feasible given that all you need to do is get from the GJ to the GLX Union Square branch, which would require some digging and maybe some light property takings (hopefully not incline mods, but cross that bridge when we get to it). Connecting GJ<-->Sullivan really isn't much harder, because thankfully the Union Square branch was pre-provisioned with the flyover stub and the access track to the carhouse pointing in the right way. The only one without an obvious and not-that-hard-for-this-thread-anyway solve is Sullivan->Lechmere, though there do appear to be options.

Edit: Thinking about it again, would it be feasible to just run (4) and Park-Kendall-Park loop (or GC-Kendall-Park)?

Chelsea/Everett/Sullivan gets the shaft, somewhat, if there's no Sullivan->Lechmere option, meaning they'd have to either take the distended loop (if there's a Kenmore connection) or backtrack on GJ to switch inbound to Lechmere and points south. Might get kind of brittle (and the ridership would look terrible on paper if it was broken out whole-line) compared to if you could feed the two Ring quadrants out of either each other or Lechmere (and GJ out of Kenmore too), but there's nothing to stop a (Kenmore?)->Kendall->Sullivan->Wherever service pattern (even in a full-build Ring, that pattern might be good for stiffening up service intervals).

If you're talking whether it'd be feasible to run Park Loop->Kenmore->Kendall->Lechmere->Park, that'd absolutely be feasible, though you'd want the Brattle Loop running to deal with the inevitable irregular operations. Can't run with Park as an endpoint without the Kenmore connection, though it would be feasible to run, say, West->Kendall->GC-Brattle (though why you would want to I do not know).

 

[Teban54]

One transfer access for Everett/Chelsea that should not be overlooked is taking the Orange Line at Sullivan. In a scenario where Sullivan->Lechmere can't be built and thus my pattern (2) can't be run, I wonder if Everett/Chelsea riders would rather transfer to OL (or BL from Chelsea) than going all the way to Kenmore and backtracking into the Central Subway on (1).