Green Line Reconfiguration

Van,

Those maps are an amazing resource, but you have to acknowledge their extremely limited predictive power. They can't tell you anything at all about who would ride the green line from Porter to Union, Northpoint, and North Station in a future in which that line exists. And not only exists, but HAS EXISTED long enough for people to make choices about where they live and work which are predicated upon its existence.

Those areas are just starting a massive ramp-up as employment centers. To fail to recognize the future demand for transit to those places is to guarantee under-performing service (Seaport is the prime example, but also Kendall, Longwood, and just about every secondary employment center in the Boston metro). This is the same intellectual laziness that F-Line was always guilty of. You feel strongly validated using data to support a position, but if the data is completely irrelevant then so is your conclusion.
 
Agreed, I had the exact same thoughts expressed by FatTony as I drooled over the maps. But because I do love them so much, where did you find them? Are they available for other ZIP codes?
 
Data tells me current demand but trying to figure out future demand is always a guess. You're right about new employment centers and the potential of improved transit (believe me I am a strong advocate for investing in transit as part of a development plan) so I'll admit this doesn't address the growth that could happen. That said I still think it's too much of a stretch to think that someone in Watertown is going to ride the Green Line through Cambridge and not transfer. That doesn't mean that there shouldn't be improvements and extensions, just that they should do it in another fashion; via the Grand Junction for instance (which I have problems with as well). I'm not saying extending the Green Line is a bad idea, just that this idea is a flawed idea.
 
HenryAlan said:
Agreed, I had the exact same thoughts expressed by FatTony as I drooled over the maps. But because I do love them so much, where did you find them? Are they available for other ZIP codes?
Click to expand...

I may be wrong, but I believe those are city-data.com maps which are available for everywhere in the country, along with a plethora of other statistics.
 
Re: Park Street loop and system capacity

https://dspace.mit.edu/handle/1721.1/111263 makes the argument that we're likely to lose the Park Street loop if the Green Line is modified to have a larger ruling radius (which is potentially useful for making the Type 10 order more off the shelf and capable of higher speeds on the D branch); if we do something about the four tightest curves, apparently that would involve getting rid of the current Lechmere yard (which is going to happen for the Green Line Extension), some minor adjustments at the Lake St yard near Boston College, rebuilding or replacing the Heath St loop (which would have to happen if we want to run longer trains on the E branch or extend the E branch to Hyde Sq anyway), and eliminating the Park Street loop.

There's some mention of the possibility of a crossover in or near Park Street, I think with the implication of continuing to allow some trains to turn around at Park St, but it wasn't obvious to me exactly where that would be or whether it would be compatible with 300' trains or whether it would mean turning trains around on the Park St platforms. And maybe that thesis explains it and I didn't find the explanation; I read parts of it and skimmed others.

The thesis at some point talks about assuming we're going to continue to see the Central Subway running 40 trains per hour in each direction for reasons that would be explained later, but in my first pass through it I don't remember identifying exactly where there was a clear explanation of that.

I'm wondering what the limits of the Government Center Green Line platforms are in trains per hour: if Copley and Arlington had 40 trains per hour in each direction, plus we had 10 or 20 trains per hour through the Pleasant St Incline in each direction, would the Government Center platforms be able to handle 50 or 60 trains per hour, with 40 of those continuing to Haymarket and the rest looping just north of the Government Center platforms?

I'm also curious whether there's some way of explaining whatever factors force a minimum curve radius vs top speed tradeoff that would fit in a few paragraphs and be readily understandable by most people.
 
Re: Green Line Accessibility

https://dspace.mit.edu/handle/1721.1/111263 has some discussion of future accessibility improvements, and it's great to see it recommending level boarding everywhere.

It didn't seem to discuss whether platforms located on curves are going to have problems with level boarding.

There's mention of 12" platform height for Green Line level boarding; is that a height that would work for a platform shared between Green Line and buses that would provide level boarding for both? What's the height of the Chelsea busway platforms and the Barr Everett bus platforms?

It didn't seem to be saying anything about moving platforms to be far side at intersections where possible for better transit signal priority.
 
Re: Green Line Transit Signal Priority and headway management

https://dspace.mit.edu/handle/1721.1/111263 has some discussion of the tendency of Green Line trains to develop uneven headways even if they manage to depart the terminal at the beginning of the route with even headways.

I suspect this means that for eastbound trains headed toward Kenmore / Symphony it would be useful to make Transit Signal Priority be conditional; perhaps during rush hour this would generally mean that if the previous train requested priority less than 5 minutes ago at the same location, a train requesting signal priority will have its request ignored.

There probably should be a few major stations on the D branch that get intentional holds in the stations when traveling eastbound to even out the headways, too.

I suspect westbound that it's probably generally OK to not worry too much about bunching west of Kenmore / Symphony if a relatively small fraction of the passengers traveling on those trains are boarding at the surface stops.
 
Re: Park Street ramp from southbound Red to Copley bound Green

I'm wondering whether it would make sense to build a ramp in Park Street Station, 10' or 20' or 30' wide, that would start on the southbound Red Line side platform just to the west of the staircase to the Copley bound Green Line platform, and continue up roughly parallel to the Green Line tracks with roughly a 12:1 slope plus appropriate landings until it reaches the level of the Copley bound platform. (It might need to start under the western most Green Line track next to the staircase and then turn to be under the Copley bound platform once it gets south of the staircase.)

The goal would be to both provide a wheelchair accessible path between the two platforms, and to provide more capacity for people to walk between the two platforms.

I fully expect that it would be more useful for whichever two branches use the south end of the Green Line platform than for the trains serving the north end of the platform, at least as long as the Green Line trains are only 150'.

Are there any detailed maps of Park St with precise enough dimensions (including relative elevation and slopes) that we could work out exactly where this would end up going to understand how feasible it would be?
 
Re: Charles River bridge for Needham branch

Has anyone done renderings of what a Charles River bridge for the Needham Green Line branch might look like? Might a contest to design the best bridge for that location be an effective way to raise awareness of the opportunity to build the branch?
 
Re: Park Street loop and system capacity

Joel N. Weber II said:
https://dspace.mit.edu/handle/1721.1/111263 makes the argument that we're likely to lose the Park Street loop if the Green Line is modified to have a larger ruling radius (which is potentially useful for making the Type 10 order more off the shelf and capable of higher speeds on the D branch); if we do something about the four tightest curves, apparently that would involve getting rid of the current Lechmere yard (which is going to happen for the Green Line Extension), some minor adjustments at the Lake St yard near Boston College, rebuilding or replacing the Heath St loop (which would have to happen if we want to run longer trains on the E branch or extend the E branch to Hyde Sq anyway), and eliminating the Park Street loop.

There's some mention of the possibility of a crossover in or near Park Street, I think with the implication of continuing to allow some trains to turn around at Park St, but it wasn't obvious to me exactly where that would be or whether it would be compatible with 300' trains or whether it would mean turning trains around on the Park St platforms. And maybe that thesis explains it and I didn't find the explanation; I read parts of it and skimmed others.

The thesis at some point talks about assuming we're going to continue to see the Central Subway running 40 trains per hour in each direction for reasons that would be explained later, but in my first pass through it I don't remember identifying exactly where there was a clear explanation of that.

I'm wondering what the limits of the Government Center Green Line platforms are in trains per hour: if Copley and Arlington had 40 trains per hour in each direction, plus we had 10 or 20 trains per hour through the Pleasant St Incline in each direction, would the Government Center platforms be able to handle 50 or 60 trains per hour, with 40 of those continuing to Haymarket and the rest looping just north of the Government Center platforms?

I'm also curious whether there's some way of explaining whatever factors force a minimum curve radius vs top speed tradeoff that would fit in a few paragraphs and be readily understandable by most people.
Click to expand...

I haven't read the thesis yet but just to play with what you are asking about, 60tph MIGHT be possible to Park St but not to Govt Center. Boylston to Park is the only active 4 track section of the subway, after Park it drops down to 2 tracks again so GC would be 30-40 tph.

Track loops are a dinosaur design from a time when trains only had one conductor and couldn't just pull into a track and reverse. New designs and multiple car trains eliminate the need for loops. So to say we could get rid of the Park St Loop is totally reasonable with the addition of new switches between Park and Boylston. The terminal capacity at Park is limited due to the fact that there are no tail tracks (nor space for them). I'm used to thinking in terms of 10 car NYCT trains so 2-3 car trolleys can probably turn a lot faster, thus increasing capacity. So theoretically 30tph is totally doable without a loop at Park.

Now lets say you combine the D/E branch into a new subway and connect it to the Green Line via the old Pleasant St Portal. Then you can run 30tph OR MORE as the outside tracks which run between Pleasant St and GC are separate from the inside tracks. Assuming there are flying junctions built in Newton and Somerville then there would be virtually no bottlenecks and with even branching each terminal wouldn't be constrained. That's the only way to really get as many trains as possible through.

Now there is the question of demand and this is getting back to what you were asking about running more trains to GC. Obviously what I've proposed works best in a vacuum. It's impossible to predict what might happen when the GLX opens in terms of ridership coming from the north but what we can see today is that most of the ridership drops after Park St. While it would seem to make the most sense to just run trains further it would cause delays as there is no more capacity.

The trick, then, is to evaluate the branches and see just how many TPH each one needs. It's very possible that the ridership isn't perfectly balanced, meaning that each branch doesn't need 15tph. Let's say the E to Needham needs much less than the D. This means that you could run 15tph Ds and just 8-9tph Es. That extra space would allow some C trains to be extended up to GC. This can be done for every branch to see the best use.

IDEALLY if you are running trains at peak TPH then you don't really need to extend past Park St as there would be a train every 2min or less so that transferring isn't all that much of a wait. That's the only trade off possible.
 
Re: Park Street ramp from southbound Red to Copley bound Green

Joel N. Weber II said:
I'm wondering whether it would make sense to build a ramp in Park Street Station, 10' or 20' or 30' wide, that would start on the southbound Red Line side platform just to the west of the staircase to the Copley bound Green Line platform, and continue up roughly parallel to the Green Line tracks with roughly a 12:1 slope plus appropriate landings until it reaches the level of the Copley bound platform. (It might need to start under the western most Green Line track next to the staircase and then turn to be under the Copley bound platform once it gets south of the staircase.)

The goal would be to both provide a wheelchair accessible path between the two platforms, and to provide more capacity for people to walk between the two platforms.

I fully expect that it would be more useful for whichever two branches use the south end of the Green Line platform than for the trains serving the north end of the platform, at least as long as the Green Line trains are only 150'.

Are there any detailed maps of Park St with precise enough dimensions (including relative elevation and slopes) that we could work out exactly where this would end up going to understand how feasible it would be?
Click to expand...

That's not a bad idea. It's been a long time since I've been to Park St, what's the elevator situation there like?
 
Re: Green Line Transit Signal Priority and headway management

Joel N. Weber II said:
https://dspace.mit.edu/handle/1721.1/111263 has some discussion of the tendency of Green Line trains to develop uneven headways even if they manage to depart the terminal at the beginning of the route with even headways.

I suspect this means that for eastbound trains headed toward Kenmore / Symphony it would be useful to make Transit Signal Priority be conditional; perhaps during rush hour this would generally mean that if the previous train requested priority less than 5 minutes ago at the same location, a train requesting signal priority will have its request ignored.

There probably should be a few major stations on the D branch that get intentional holds in the stations when traveling eastbound to even out the headways, too.

I suspect westbound that it's probably generally OK to not worry too much about bunching west of Kenmore / Symphony if a relatively small fraction of the passengers traveling on those trains are boarding at the surface stops.
Click to expand...

Streetcars bunch just like buses do. What's needed is signal priority on traffic lights and a better balance in branching. The way the GL works now it's impossible to get even spacing, hell it wasn't even design for that! So the only way to to group the branches, B/C along Boylston and D/E along Huntington to a new subway via the Pleasant St Portal. OBVIOUSLY signal priority is much cheaper but there is only so much you can do when the fundamental system is not designed for something.
 
Re: Park Street loop and system capacity

vanshnookenraggen said:
I haven't read the thesis yet but just to play with what you are asking about, 60tph MIGHT be possible to Park St but not to Govt Center. Boylston to Park is the only active 4 track section of the subway, after Park it drops down to 2 tracks again so GC would be 30-40 tph.
Click to expand...

If we're being pedantic, don't Kenmore and JFK/Umass also have four track sections (and perhaps Savin Hill counts as a four track section, with the Braintree branch there not having a platform)?

I'm also wondering if level boarding and extra doors that are extra wide might cut down dwell times and help a bit to increase the viable number of trains per hour.

Track loops are a dinosaur design from a time when trains only had one conductor and couldn't just pull into a track and reverse. New designs and multiple car trains eliminate the need for loops. So to say we could get rid of the Park St Loop is totally reasonable with the addition of new switches between Park and Boylston. The terminal capacity at Park is limited due to the fact that there are no tail tracks (nor space for them). I'm used to thinking in terms of 10 car NYCT trains so 2-3 car trolleys can probably turn a lot faster, thus increasing capacity. So theoretically 30tph is totally doable without a loop at Park.
Click to expand...

If you have operator fallback, is there any difference in how long it takes to reverse ends of the train on a 2 car vs 10 car train?

Now there is the question of demand and this is getting back to what you were asking about running more trains to GC. Obviously what I've proposed works best in a vacuum. It's impossible to predict what might happen when the GLX opens in terms of ridership coming from the north but what we can see today is that most of the ridership drops after Park St. While it would seem to make the most sense to just run trains further it would cause delays as there is no more capacity.
Click to expand...

The thesis argues that ~220' trains at 6 minute headways (10 TPH) in conjunction with better dispatching will likely be sufficient, and that just running D trains through to Tufts in Medford is likely to be sufficient, but it argues that also extending C trains at rush hour to Medford would work if ridership ends up being especially strong. (It doesn't seem to get into much detail about how well turning a train around every 3 minutes at College Ave will work; I suspect it's probably doable but may not leave room for long layovers to help smooth out the inbound headways when the outbound headways get inconsistent.)

The trick, then, is to evaluate the branches and see just how many TPH each one needs. It's very possible that the ridership isn't perfectly balanced, meaning that each branch doesn't need 15tph. Let's say the E to Needham needs much less than the D. This means that you could run 15tph Ds and just 8-9tph Es. That extra space would allow some C trains to be extended up to GC. This can be done for every branch to see the best use.
Click to expand...

The satellite images leave me with the impression that there is much more around the Needham branch stops than the Riverside branch stops, and it might make sense for Needham to simply take over all of Riverside's slots through the Central Subway, and run one car trains from Riverside to Kenmore. Maybe we could convert the 128 / Highland Ave interchange to a Diverging Diamond Interchange, and use the land that is freed up by getting rid of the cloverleaf for parking for a Needham Branch equivalent of the Riverside park and ride.

https://pedestrianobservations.com/...-frequency-guidelines-are-the-wrong-approach/ argues

all lettered lines except the J/Z and the L should have the same frequency, and in addition the 2/3/4/5 should also have the same frequency
Click to expand...

because apparently mixing different frequencies on different routes that converge onto a common trunk line is problematic, as that blog post explains in more detail.

IDEALLY if you are running trains at peak TPH then you don't really need to extend past Park St as there would be a train every 2min or less so that transferring isn't all that much of a wait. That's the only trade off possible.
Click to expand...

Having some trains terminate at Park works as long as you have a viable way to turn them around without them getting in the way of the trains running through; my concern is that if we get rid of the loop so that we can run off the shelf trains, we might end up not being able to have capacity for Pleasant St Incline trains. But maybe using Park St's wall tracks for the trains running through to Government Center and turning trains around on the fence tracks by switching ends might work.
 
Re: Park Street loop and system capacity

The simplest thing to do about the Pleasant St Incline in a world with no Park St loop might be to simply continue running 40 TPH through Park St and Government Center in each direction, and make that 8 TPH per branch instead of 10 TPH per branch when adding Dudley service. If the trains get lengthened from ~220' to ~275' at the time the headways are made less frequent in that fashion, there should be no change in overall capacity on each branch.

The other option might be to put the Government Center through trains on Park St's wall tracks, and reverse the B branch trains (or whichever branch ends up reversing at Park St) on the fence tracks. If that's done, it might be desirable to allow passengers to cross the fence tracks at the north end in much the same way that's allowed for the northbound wall track. In theory, if everything goes smoothly, it might be possible to always use what is currently the Copley bound fence track to turn around the B branch trains and never use what is currently the Lechmere bound fence track, but if anything goes wrong to cause a delay in reversing a B branch train, that Lechmere bound fence track will be the place that the next B train should probably go to get out of the way of the Government Center bound trains, and if passengers trying to get on a train to Boston College then suddenly need to be on the northbound platform instead of the southbound platform, that at grade crossing at the north end of the fence would be useful...
 
Re: Park Street loop and system capacity

Joel N. Weber II said:
If we're being pedantic, don't Kenmore and JFK/Umass also have four track sections (and perhaps Savin Hill counts as a four track section, with the Braintree branch there not having a platform)?
Click to expand...

I meant 4 track trunks. Kenmore and JFK were designed with 4 tracks to ease merges. If C trains looped at Kenmore then you could consider it a trunk station as then B trains could be increased. JFK was just poor planning.

I'm also wondering if level boarding and extra doors that are extra wide might cut down dwell times and help a bit to increase the viable number of trains per hour.
Click to expand...

Abso-lutely.

If you have operator fallback, is there any difference in how long it takes to reverse ends of the train on a 2 car vs 10 car train?
Click to expand...

Probably a min or two faster and when you compound that throughout the day that is some seriously expanded capacity.


The thesis argues that ~220' trains at 6 minute headways (10 TPH) in conjunction with better dispatching will likely be sufficient, and that just running D trains through to Tufts in Medford is likely to be sufficient, but it argues that also extending C trains at rush hour to Medford would work if ridership ends up being especially strong. (It doesn't seem to get into much detail about how well turning a train around every 3 minutes at College Ave will work; I suspect it's probably doable but may not leave room for long layovers to help smooth out the inbound headways when the outbound headways get inconsistent.)
Click to expand...

If you assume 10pth on each branch then it's possible to run 40tph BUT as you mentioned terminal capacity is something to consider. If College Ave is not being designed as a final terminal then the capacity will be constrained, though, probably to somewhere around 20tph so theoretically that wouldn't be an issue. This is all back of a napkin numbers and the final design will determine just what is possible.

The satellite images leave me with the impression that there is much more around the Needham branch stops than the Riverside branch stops, and it might make sense for Needham to simply take over all of Riverside's slots through the Central Subway, and run one car trains from Riverside to Kenmore. Maybe we could convert the 128 / Highland Ave interchange to a Diverging Diamond Interchange, and use the land that is freed up by getting rid of the cloverleaf for parking for a Needham Branch equivalent of the Riverside park and ride.
Click to expand...

Without a full study we won't know for sure. But it's entirely possible. Also the TOD potential is much higher in Needham than in Newton.

https://pedestrianobservations.com/...-frequency-guidelines-are-the-wrong-approach/ argues

because apparently mixing different frequencies on different routes that converge onto a common trunk line is problematic, as that blog post explains in more detail.
Click to expand...

I take everything Alon Levy writes with a grain of salt. He's not wrong but he thinks only in pure numbers and not how people actually want to ride transit. Everytime a rider needs to transfer their desire to use the system drops. The psychology of a one seat ride will always win out over a faster trip with transfers. Besides, both NYC and the Green Line were designed for interlining and it would be more expensive to redesign the systems to eliminate it than just dealing with the consequences.
 
In regards to interlining I believe that F-Line had said that adding more north side branches doesn't effect the green line negatively in most cases and if there isn't capacity to run more trains from the south to the north for a line it would be possible to restore the line to a four track trunk between Haymarket and Gov't Center and turn some trains from the north using the Gov't Center loop.
 
citylover94 said:
In regards to interlining I believe that F-Line had said that adding more north side branches doesn't effect the green line negatively in most cases and if there isn't capacity to run more trains from the south to the north for a line it would be possible to restore the line to a four track trunk between Haymarket and Gov't Center and turn some trains from the north using the Gov't Center loop.
Click to expand...

While this is technically true there are two things to consider:
1) demand
2) the loop track crosses at grade over the through track.

Per the second point, this creates a bottleneck and limits capacity. Per demand, I'm willing to bet that Park St will still be the #1 transfer spot so that any train looping will mean that passengers will still need to transfer to get to Park. This obviously doesn't hold true for games at the Garden but normal service will probably by 100% through.
 
Re: Park Street loop and system capacity

Joel N. Weber II said:
Having some trains terminate at Park works as long as you have a viable way to turn them around without them getting in the way of the trains running through; my concern is that if we get rid of the loop so that we can run off the shelf trains, we might end up not being able to have capacity for Pleasant St Incline trains. But maybe using Park St's wall tracks for the trains running through to Government Center and turning trains around on the fence tracks by switching ends might work.
Click to expand...

Didn't get to this before.

If the center loop is taken out and a double switch installed just south of Park St then the turning capacity is around 20-25tph. This might be more than enough. I don't know enough about the ridership to say which is best but the basic number we need to know is how many riders from the branches (past Copley and Kenmore) are going to and past Park St. Riders along the trunk don't matter in this case because a rider can just wait for the right train to take. But if we know what the branches bring in then we can better adjust the terminal points.

The more I think about the operations of the moves the more I realize that the loop is probably the best option. Park St is basically the length of two full trains (3 cars each?) and allows for different trains to berth at specific locations. This would need to change if the loop was removed meaning that the center tracks would work best with only one branch. Otherwise with two branches the dance of trains terminating and leaving becomes more complicated. Even in NYC each branch of each line has their own track space for terminating (Forest Hills-71 Av being the most complicated but there are tail tracks and yard leads past the station to help). There is no room past Park St to build terminal tracks which would help with multiple branches so ultimately keeping the status quo seems to be the best option.
 
Last edited:
You must log in or register to reply here.

Back
Top