Green Line Reconfiguration

[F-Line to Dudley]

You aren't mixing Green Line trolleys and Blue Line HRT cars on the same tracks. A collision between the two would be nasty because of the wildly different centers of gravity. HRT car striking the trolley would kill/maim the trolley operator in the cabin, and the high-slung HRT car might jackknife on the very low-slung trolley. For the same reason (carbody too high...not enough of a cowcatcher effect) the HRT cars would also be banned on grade crossings, meaning any mixed-traffic branches are no-go. To say nothing of whether well over 50 stations could even feasibly be converted simultaneously to high platforms with no pedestrian track crossings.

You're going to have to standardize on nuthin' but LRT rolling stock if any of this grand scheme is to work.

 

[ritchiew]

went insane and made a diagram of my final solution to the green line
sorry its messy my brain was melting out of my ears and my organs were all vibrating at different speeds when i was making this
View attachment 55111

Wait, what's the deal with Castle Street Junction? What did that used to connect? When was that taken down?

 

[F-Line to Dudley]

Wait, what's the deal with Castle Street Junction? What did that used to connect? When was that taken down?

That's where the Atlantic Ave. El connected with the Washington St. El, the lead to the Chinatown Orange Line portal, and the former lead to the 1901-08 re-route through the Tremont St. tunnel. It was formally known as Tower D Junction. The last remnants of it came down in 1987. This is what it looked like in 1982 with just one thru route.

 

[Hooledeli]

You aren't mixing Green Line trolleys and Blue Line HRT cars on the same tracks. A collision between the two would be nasty because of the wildly different centers of gravity. HRT car striking the trolley would kill/maim the trolley operator in the cabin, and the high-slung HRT car might jackknife on the very low-slung trolley. For the same reason (carbody too high...not enough of a cowcatcher effect) the HRT cars would also be banned on grade crossings, meaning any mixed-traffic branches are no-go. To say nothing of whether well over 50 stations could even feasibly be converted simultaneously to high platforms with no pedestrian track crossings.

You're going to have to standardize on nuthin' but LRT rolling stock if any of this grand scheme is to work.

the red line in cleveland mixes heavy and light rail? at tower city center their light rail and heavy rail uses the same tracks but board on different parts of the platform that are highfloor and lowfloor respectively, but yeah ill concede itd be better if it didnt so additionally: if the B branch is gonna be HRT it cant possibly stay above ground yes, so it would need to be sunken. the existing B branch tracks to packards corner would continue on west to the A branch restoration. the approach to kenmore would need to be rebuilt that thatll be a mess but i dont think thats the obstacle with any of this considering the amount of new track im proposing

they made it work when they dug the orange line tunnel, it surely is hard but not impossible. the green line trains are like so much shorter then the platforms so you could make half of it high floor (probably more like 2/3) while running current rolling stock, then switch to short HRT trains while building the rest of the platform, before beginning full length HRT cars

 

[TallIsGood]

went insane and made a diagram of my final solution to the green line
sorry its messy my brain was melting out of my ears and my organs were all vibrating at different speeds when i was making this
View attachment 55111

Is the E to D connection in Brookline Village still physically possible?

 

[F-Line to Dudley]

Is the E to D connection in Brookline Village still physically possible?

Yes. Via street-running on Pearl St. Here's a diagram showing a multi-directional configuration (i.e. Huntington trains being thru-routeable onto the D, and South Huntington trains being routeable from Kenmore).

An E-to-D train would proceed straight from Huntington to Washington, turn onto River Rd. utilizing the existing pedestrian signal, then turn onto Pearl via a driveway easement, crossing Brookline Ave. at a new signal coordinated with the pedestrian signal. Trains would turn onto the D at a universal crossover and proceed into the existing Brookline Village platform. Same in the reverse direction.

A South Huntington/Arborway train being routed from Kenmore would take the universal crossover and turn out onto sidewalk-facing platforms on Pearl. The sidewalk platforms would be a very tight fit for a 2-car Type 10, but it looks feasible. Then turn onto Washington and proceed to South Huntington via a new wye leg. Ditto the reverse direction. This pattern could also support 'circuit' service Kenmore-Huntington-Park/GC if there were ever a need for it.

If you were only going to do one of these patterns, you could build only half the track to/from the universal crossover and omit the other half. So, for example, if you're not keen on roping in Kenmore to potential patterns you could omit the west half of the Pearl loop and sidewalk platforms, as well as everything on Washington between the Pearl signal and River Rd. signal.


As the street grid is never going to go away, all of this should be permanently feasible.



EDIT: I forgot that there are parking garage curb cuts on Pearl where the platforms are depicted. That's no-go for the platforms. Have to see if instead you can fit turnout platforms on the D Line side of Pearl right after the crossover (perhaps move the crossover further back?). 1-car Type 10 should easily fit...the trick is to see if you can get a 2-car platform back there.

 

[Riverside]

As the street grid is never going to go away, all of this should be permanently feasible.

Right, this is the key thing to bear in mind here: these tracks are really close to each other. Like, "1000-feet-as-the-crow-flies" really close. Even with the S curve F-Line drew here, it's still only 1500 feet of track. And yeah -- the only way the street grid goes away is if we get Urban Renewal 2.0 and completely level all of the buildings there, which seems extraordinarily unlikely.

Yes. Via street-running on Pearl St. Here's a diagram showing a multi-directional configuration (i.e. Huntington trains being thru-routeable onto the D, and South Huntington trains being routeable from Kenmore).

View attachment 55294

...

A South Huntington/Arborway train being routed from Kenmore would take the universal crossover and turn out onto sidewalk-facing platforms on Pearl. The sidewalk platforms would be a very tight fit for a 2-car Type 10, but it looks feasible. Then turn onto Washington and proceed to South Huntington via a new wye leg. Ditto the reverse direction. This pattern could also support 'circuit' service Kenmore-Huntington-Park/GC if there were ever a need for it.

...


EDIT: I forgot that there are parking garage curb cuts on Pearl where the platforms are depicted. That's no-go for the platforms. Have to see if instead you can fit turnout platforms on the D Line side of Pearl right after the crossover (perhaps move the crossover further back?). 1-car Type 10 should easily fit...the trick is to see if you can get a 2-car platform back there.

A platform for Kenmore <> Arborway services is squarely in the "would be nice" camp. I'm a huge fan of such a service, but there's no way that it replaces the 39, which means that JP residents will always have a transfer to Riverside <> Huntington LRT (the "Gold Line") available somewhere around Mission Park.

As for Riverside -> Kenmore journeys, there are numerous ways to offer that connection, including simply with a Riverside <> Kenmore "dinky" overlay; alternatively, the existing Highland Branch station could be expanded northward into the parking lot, to create a Brookline Village <> Kenmore shuttle platform.

~~~

That all being said, the real feasibility question about D-E is determined by when/whether/how/where a Huntington subway gets extended. The surface alt works with the existing surface-running, but there's some ideal future where the "Gold Line" runs in a tunnel all the way out Huntington, through an underground D-E Connector, and then surfaces somewhere west of Brookline Village. And then there are a full range of scenarios in between those extremes, including something like what @The EGE has proposed in the past -- a D-E connector via Longwood Medical Area (on Longwood Ave or Francis St). Those are the versions of this that would require more study.

 

[Riverside]

EDIT: I forgot that there are parking garage curb cuts on Pearl where the platforms are depicted. That's no-go for the platforms. Have to see if instead you can fit turnout platforms on the D Line side of Pearl right after the crossover (perhaps move the crossover further back?). 1-car Type 10 should easily fit...the trick is to see if you can get a 2-car platform back there.

Looks to me like it should be doable to put platforms in parallel east of a wye:

The existing Highland Branch ROW would need a little bit of realignment to accommodate full tangent tracks, but I don't think it would be earth-shattering. This alignment would also avoid interlining at the crossing point (although there would still be interlining to the east of here as the tracks go under the Jamaicaway).

 

[TheRatmeister]

Looks to me like it should be doable to put platforms in parallel east of a wye:

View attachment 55302

The existing Highland Branch ROW would need a little bit of realignment to accommodate full tangent tracks, but I don't think it would be earth-shattering. This alignment would also avoid interlining at the crossing point (although there would still be interlining to the east of here as the tracks go under the Jamaicaway).

A (potential) problem with this arrangement is that it needs to run in mixed traffic due to the Brookline Place garage, which has entrances on NW/NE Pearl St. I think just converting the western segment of Pearl St to a full transitway is a better solution. Then rather than shifting the Brookline Village platforms east, shift them a bit further west instead.

 

[Riverside]

A (potential) problem with this arrangement is that it needs to run in mixed traffic due to the Brookline Place garage, which has entrances on NW/NE Pearl St. I think just converting the western segment of Pearl St to a full transitway is a better solution. Then rather than shifting the Brookline Village platforms east, shift them a bit further west instead.

Like this?

 

[The EGE]

I simply can't imagine a surface connector with tight curves and traffic signals being useful for anything but non-revenue moves, especially connecting to the surface tracks on Huntington. It would be a substantial downgrade from the grade-separated Highland Branch. Currently, Brookline Village to Copley is scheduled for 13 minutes on the D, while Riverway to Copley is 19 minutes (and that assumes minimal traffic west of Brigham Circle, so not liable to drop significantly with transit lanes). That differential would be a few minutes worse with the additional distance of the connector.

For revenue purposes, there's five main reasons to consider a D-E connector:

• Routing D trains to a new Back Bay subway to use the outer tracks of the Tremont Street Subway. That's a method to enable higher frequencies and higher capacity (and potentially more branches) as we've discussed a great deal in this thread. Given the expense to create such a high-quality system, it makes no sense to route D trains over a slow surface connector.
• Routing D trains to the Seaport. Inherently entwined with the previous bullet and has the same issues. There's also no way that you're running a substantially increased level of service on the surface on Huntington Avenue.
• Increase service to the Huntington Avenue corridor. This is desirable, but the downsides of longer D running time and issues with high-frequency surface operations don't justify rerouting the D just for this. Type 10s will be able to handle a lot of additional demand on Huntington in the short/middle term.
• Enabling circumferential or loop service. Loop service isn't happening unless a Back Bay subway opens up more subway slots - otherwise, every loop service is a slot that doesn't serve the outer D or E. Except for a very limited Kenmore–Ruggles (or Nubian) service - which would be slower than buses due to the routing - any circumferential service would require either Kenmore rebuild + B subway or a completely new alignment to reach the Grand Junction. Again, a slow surface connector and difficult high-frequency service operations don't make sense.
• Providing better access to the LMA. While Huntington is somewhat closer to the center-of-gravity of the LMA than Longwood station, the slightly shorter walking time would be offset by the longer ride time. It's also better to have stations on both sides of the LMA than just one. I remain convinced that the best design for a D-E connector is a Longwood Avenue subway - expensive, but provides great value.

 

[TheRatmeister]

Like this?
View attachment 55355

Yes, exactly. This would also allow for a much easier connection to the Huntington Ave Subway, just built a single portal on Pearl St or on the north side of Washington St between Pearl St and Brookline Ave.

 

[Teban54]

I simply can't imagine a surface connector with tight curves and traffic signals being useful for anything but non-revenue moves, especially connecting to the surface tracks on Huntington. It would be a substantial downgrade from the grade-separated Highland Branch. Currently, Brookline Village to Copley is scheduled for 13 minutes on the D, while Riverway to Copley is 19 minutes (and that assumes minimal traffic west of Brigham Circle, so not liable to drop significantly with transit lanes). That differential would be a few minutes worse with the additional distance of the connector.

For revenue purposes, there's five main reasons to consider a D-E connector:

• Routing D trains to a new Back Bay subway to use the outer tracks of the Tremont Street Subway. That's a method to enable higher frequencies and higher capacity (and potentially more branches) as we've discussed a great deal in this thread. Given the expense to create such a high-quality system, it makes no sense to route D trains over a slow surface connector.
• Routing D trains to the Seaport. Inherently entwined with the previous bullet and has the same issues. There's also no way that you're running a substantially increased level of service on the surface on Huntington Avenue.
• Increase service to the Huntington Avenue corridor. This is desirable, but the downsides of longer D running time and issues with high-frequency surface operations don't justify rerouting the D just for this. Type 10s will be able to handle a lot of additional demand on Huntington in the short/middle term.
• Enabling circumferential or loop service. Loop service isn't happening unless a Back Bay subway opens up more subway slots - otherwise, every loop service is a slot that doesn't serve the outer D or E. Except for a very limited Kenmore–Ruggles (or Nubian) service - which would be slower than buses due to the routing - any circumferential service would require either Kenmore rebuild + B subway or a completely new alignment to reach the Grand Junction. Again, a slow surface connector and difficult high-frequency service operations don't make sense.
• Providing better access to the LMA. While Huntington is somewhat closer to the center-of-gravity of the LMA than Longwood station, the slightly shorter walking time would be offset by the longer ride time. It's also better to have stations on both sides of the LMA than just one. I remain convinced that the best design for a D-E connector is a Longwood Avenue subway - expensive, but provides great value.

Fully agreed.

The only way I think a D-E surface connector makes sense for regular service is if we want to convert Needham to LRT in the short-to-mid term, but under the following circumstances:

• It's decided that Needham trains should (and can) be additions to existing service, not replacing any Riverside trains. In particular, service between Newton Highlands and Brookline Village will be "doubled" (albeit with great differences in travel time). This constraint removes the option of simply diverting some D trains from Riverside to Needham.
• The Pleasant St Incline - Back Bay subway has already been built, enabling the E to use Boylston Outer. Otherwise, you can't add more trains into Central Subway (or at least it's undesirable).
• However, the Huntington Ave subway has not been built yet.

These can only happen concurrently in a very specific time frame and under very specific political scenes.

As for a ring route that goes Kenmore-BV-Huntington, I have long been of the opinion that unless branding (rail vs. bus) is the only thing people care about, it's simply too slow and winding to be worth it, even with grade-separated ROWs. I can see such a pattern being useful for sending Heath/Hyde Park/Arborway trains to Kenmore via BV and Highland Branch, though. (Such a pattern may be relevant for various reasons, including but not limited to minimizing street-running and reliability concerns on GLR/Huntington.)

 

[TheRatmeister]

• The Pleasant St Incline - Back Bay subway has already been built, enabling the E to use Boylston Outer. Otherwise, you can't add more trains into Central Subway (or at least it's undesirable).
• However, the Huntington Ave subway has not been built yet.

Which would be really weird since the Huntington Ave subway is the GLRC segment that provides by far the most significant immediate benefit without any other projects. (And therefore the easiest portion to sell, you don't really need to explain to anyone remotely familiar with the E branch why the Huntington Ave subway would be a good idea.)

 

[Teban54]

Which would be really weird since the Huntington Ave subway is the GLRC segment that provides by far the most significant immediate benefit without any other projects. (And therefore the easiest portion to sell, you don't really need to explain to anyone remotely familiar with the E branch why the Huntington Ave subway would be a good idea.)

While that's true, the impression that people will get from a standalone Huntington subway project (without Bay Village) is not as clear-cut as you may think. I can easily think of several concerns:

• "We have so many transit deserts without any service. Why are we replacing a corridor that already has excellent service?" (Most other rail extensions clearly enlarge the transit footprint.)
• "A subway costs billions of dollars. Why spend them just so that some people can speed up their journey by merely a few minutes?"
• "Why not other Green Line branches? Why don't you bury the B which has higher ridership? Or the C? Why not a subway to Watertown and/or Nubian?"

The third question had actually been raised on Reddit when I presented my GLR map there. A few users were accusing me of preferential treatment towards the E. If transit fans think like this, the general public probably will, too.

In comparison, Bay Village's benefits may be less trivial, but easier to accept with some explanations. Every Green Line branch can benefit from increased TPH due to Boylston Outer. Its cost would also likely be lower than the full Huntington subway simply due to shorter distance.

 

[Delvin4519]

While that's true, the impression that people will get from a standalone Huntington subway project (without Bay Village) is not as clear-cut as you may think. I can easily think of several concerns:

• "We have so many transit deserts without any service. Why are we replacing a corridor that already has excellent service?" (Most other rail extensions clearly enlarge the transit footprint.)
• "A subway costs billions of dollars. Why spend them just so that some people can speed up their journey by merely a few minutes?"
• "Why not other Green Line branches? Why don't you bury the B which has higher ridership? Or the C? Why not a subway to Watertown and/or Nubian?"

The third question had actually been raised on Reddit when I presented my GLR map there. A few users were accusing me of preferential treatment towards the E. If transit fans think like this, the general public probably will, too.

In comparison, Bay Village's benefits may be less trivial, but easier to accept with some explanations. Every Green Line branch can benefit from increased TPH due to Boylston Outer. Its cost would also likely be lower than the full Huntington subway simply due to shorter distance.

Absolutely. I find Red-Blue to be a letdown and am somewhat disappointed with it. While it is definitely essential for fixing broken color transfers downtown, it doesn't really extend frequent service to new neighborhoods the same way GLX does. Red-Blue, being a huge cost-blowout to build, would still leave befind the same transit deserts coming out of the project, as with before the project. With only 45% (~765k) of streetcar suburbs accessible from at least 1 frequent route, 55% of our streetcar suburbs are still inaccessible on the frequent grid.

 

[F-Line to Dudley]

Absolutely. I find Red-Blue to be a letdown and am somewhat disappointed with it. While it is definitely essential for fixing broken color transfers downtown, it doesn't really extend frequent service to new neighborhoods the same way GLX does. Red-Blue, being a huge cost-blowout to build, would still leave befind the same transit deserts coming out of the project, as with before the project. With only 45% (~765k) of streetcar suburbs accessible from at least 1 frequent route, 55% of our streetcar suburbs are still inaccessible on the frequent grid.

Giving North Shore communities two- instead of three-seat access to Cambridge-based jobs is a letdown? It's worth a +6500 daily ridership increase on Blue, +2800 all-new daily transit riders who aren't taking any transit, and 5,800 daily transfers to reduce the loading on DTX and State so everybody else's Downtown commute especially on the Red Line isn't delayed by severe loading issues. That's not chopped liver. It's one of the highest-ridership transit expansion projects evaluated despite its short length and no numerical net gain in stations served.

There's more than one type of beneficial transit expansion project, you know? It's one-dimensional thinking to assume that an expansion project is "disappointing" solely because it isn't tapping new "streetcar suburbs". People need to get across Downtown to their destinations, too, especially when we're flushing more riders through Downtown than ever before. This includes a whole freaking lot of "streetcar suburb" residents presently riding the system (whether on a frequent originating route or not) and inconvenienced by the extra transfers and crush-loading in the center of the system. It's not a healthy system or a system ripe for more expansion to those very streetcar suburbs if we don't fix this problem.


By your logic NSRL...the second-largest ridership generating expansion project ever evaluated in the modern era after the complete Urban Ring...is a "letdown" because it doesn't expand the map and doesn't as a capital construction project intrinsically increase frequencies unless paired with due-diligence Regional Rail service practices.

 

[Delvin4519]

Giving North Shore communities two- instead of three-seat access to Cambridge-based jobs is a letdown? It's worth a +6500 daily ridership increase on Blue, +2800 all-new daily transit riders who aren't taking any transit, and 5,800 daily transfers to reduce the loading on DTX and State so everybody else's Downtown commute especially on the Red Line isn't delayed by severe loading issues. That's not chopped liver. It's one of the highest-ridership transit expansion projects evaluated despite its short length and no numerical net gain in stations served.

There's more than one type of beneficial transit expansion project, you know? It's one-dimensional thinking to assume that an expansion project is "disappointing" solely because it isn't tapping new "streetcar suburbs". People need to get across Downtown to their destinations, too, especially when we're flushing more riders through Downtown than ever before. This includes a whole freaking lot of "streetcar suburb" residents presently riding the system (whether on a frequent originating route or not) and inconvenienced by the extra transfers and crush-loading in the center of the system. It's not a healthy system or a system ripe for more expansion to those very streetcar suburbs if we don't fix this problem.


By your logic NSRL...the second-largest ridership generating expansion project ever evaluated in the modern era after the complete Urban Ring...is a "letdown" because it doesn't expand the map and doesn't as a capital construction project intrinsically increase frequencies unless paired with due-diligence Regional Rail service practices.

The main issue of Red-Blue is that it's literally a full billion dollars just to extend the tunnel 1,000 feet down the road. Other countries outside the anglosphere can and have done this much cheaper. If Red-Blue were only a fraction of the cost, there'd be extra bandwidth to do both outward expansion and downtown capacity optimization. A full billion to extend the tunnel 1,000 feet further eats all the bandwidth away and there's no breathing room left over for anything else. It essentially pushes every other expansion proect, including NSRL, further away, and further back.

 

[TheRatmeister]

The main issue of Red-Blue is that it's literally a full billion dollars just to extend the tunnel 1,000 feet down the road. Other countries outside the anglosphere can and have done this much cheaper. If Red-Blue were only a fraction of the cost, there'd be extra bandwidth to do both outward expansion and downtown capacity optimization. A full billion to extend the tunnel 1,000 feet further eats all the bandwidth away and there's no breathing room left over for anything else. It essentially pushes every other expansion proect, including NSRL, further away, and further back.

It can cost too much compared to other countries and still be worth it, those aren't mutually exclusive.

 

[Teban54]

The main issue of Red-Blue is that it's literally a full billion dollars just to extend the tunnel 1,000 feet down the road. Other countries outside the anglosphere can and have done this much cheaper. If Red-Blue were only a fraction of the cost, there'd be extra bandwidth to do both outward expansion and downtown capacity optimization. A full billion to extend the tunnel 1,000 feet further eats all the bandwidth away and there's no breathing room left over for anything else. It essentially pushes every other expansion proect, including NSRL, further away, and further back.

Every transit project pushes other projects further back. By this argument, we should build nothing.

The inability to keep construction costs low is also irrelevant to the discussion for the same reasons. An expansion done at $1 billion per mile is better than no expansions at all, especially when it's not clear whether other projects of similar construction methods can be done at lower cost-to-benefit ratios (and thus may be just as expensive).

If anything, Red-Blue is one of the most equitable expansion projects. Most other extensions (BLX Lynn, OLX, RLX etc) only have localized benefits to specific cities and neighborhoods, but Red-Blue benefits everyone. Blue Line riders get connections to crucial destinations on the Red Line. Red Line riders get much-needed connections to the airport, East Boston, etc. Orange and Green Line riders also see less direct but still substantial benefits, in terms of fewer people using both lines as a tedious Red-Blue connector, fewer people clogging up the downtown transfer stations (including Cambridge riders going to GC and State), and an alternative to MGH.