Department of Subways - Proposals/Ideas

[Caelestor]

The  should not be extended to College Point or Whitestone, which represent dead ends, as there is no justification to build an expensive subway tunnel to Throgs Neck. A northern extension would also permanently limit capacity on the eastward Northern Blvd extension to Broadway / Bayside, which is the denser corridor that needs to be served.

2 hours ago, RR503 said:

The way I see it, PW provides capacity for the now/short term, regional rail solutions in the medium term,  reliever line and  extension in the long term.

CityTicket should be extended to weekdays, and PW service should be increased off-peak, as well as during the peak after ESA opens. Queens needs every bit of capacity it can get.

Regional rail through NJT / LIRR consolidation would be nice, but it would only really open up trips to NJ. The real game changer would be an ESA extension to Union Square and Jersey City / Fulton St, which is even more long term than new SAS + Northern Blvd subway lines.

[BreeddekalbL]

How about making 74th on the  /  express can that be done? 

[bobtehpanda]

3 hours ago, RR503 said:

It’d be pretty easy to provide a QB-PW xfer in Elmhurst, but that’s beside the point.

The objection I lodge against putting extension/upzoning loads on the  is capacital. There’s already plenty of growth/upzoning along the rest of the line; dunno if it’s wise to add more fuel to the fire without providing relief. In a perfect world, you’d be able to harness transfers to provide said relief, but 74/QB is already shot and by the time you hit QBP you’re beyond the peak load point so 🤷‍♂️ (related ish: if you want to do development along the  in LIC you should really be increasing  service, as  capacity through, say, HPA is inversely proportionate to the number of people you can entice off at QBP). The way I see it, PW provides capacity for the now/short term, regional rail solutions in the medium term,  reliever line and  extension in the long term.

Indeed. For the short term, extension to Broadway LIRR should be sufficient to divert bus traffic away from Main St without causing too much density raising (the Kissena Blvd corridor is already moderately dense), and why my fantasy fourth big Queens line is a line going 86-Astoria-Kissena-Parsons to Jamaica.

[bobtehpanda]

5 minutes ago, BreeddekalbL said:

How about making 74th on the  /  express can that be done? 

I've never been a fan of this idea; 74th on the local splits crowds on the train pretty well. Roosevelt is way too congested to be adding more transferring passengers to the mix.

[T to Dyre Avenue]

Ever since moving to Queens in 2010, I’ve wanted to have a  extension to Springfield Blvd via Northern. I’ve also City Ticket for a PW branch with more frequent service and stations in Corona and Elmhurst. Hey MTA, what’s that saying about “when there’s a will...”?

Perhaps the combination of a  extension to Broadway LIRR, plus implementation of a weekday City Ticket on the PW could do wonders for congestion relief, in the short term, both on the  and on the streets of downtown Flushing, by keeping some of the bus routes from having to converge there to connect with the  now. Would be way better than the giant mess we have at Main Street-Flushing now.

Edited March 11, 2019 by T to Dyre Avenue

[RR503]

One thing that I’d be interested to know is what the practical capacity of  CBTC is (@Stephen Bauman can you speak to this?). Eking out even another 3-4 tph could mean a lot — provided that they’re properly scheduled. 

[Union Tpke]

3 hours ago, RR503 said:

One thing that I’d be interested to know is what the practical capacity of  CBTC is (@Stephen Bauman can you speak to this?). Eking out even another 3-4 tph could mean a lot — provided that they’re properly scheduled. 

Aaron Gordon found that the 7 ran 33 TPH during the 2000 World Series with close operations. You should tried to find more about that.

[RR503]

3 minutes ago, Union Tpke said:

Aaron Gordon found that the 7 ran 33 TPH during the 2000 World Series with close operations. You should tried to find more about that.

Interesting. There’s an old Stephen Bauman post on SubTalk that basically demonstrated that NYCT had, by the early 2000s, lost the competency necessary to operate 30tph. Was this 33tph sustained? 

[BreeddekalbL]

In relation to my proposal of elimnating hewes street and flushing ave on the would it even be feasable to relocate marcy ave closer to the bus terminal  and exits including ADA facilities will be at the plaza end  it would include making the express track through so the trains can run through

[Union Tpke]

3 hours ago, RR503 said:

Interesting. There’s an old Stephen Bauman post on SubTalk that basically demonstrated that NYCT had, by the early 2000s, lost the competency necessary to operate 30tph. Was this 33tph sustained? 

It was very strictly done. It was stated in the following podcast: 

Quote

AARON GORDON: Not just among motormen but also in what's called dispatchers, and dispatching is a really, really intricate skill because a lot of it is still managed by one person who just really knows the system super well. They manage the switches and they make sure that trains end up on the proper tracks and that the track switches get executed quickly so trains can move through switches quickly. It's a really intricate dance. There was a huge, huge turnover after these kind of like not forced retirements but a lot of people basically said, "Yeah, please put in your retirement papers" because it was clear that they needed to cut costs. There was all this institutional knowledge lost and now you have a bunch of super, super cautious dispatchers, motormen. They can't run the system as well as it used to be run.

The ultimate example of this, during the 2000 World Series, subway series, right? It was between the Yankees and the Mets. Shea Stadium is on the 7 line. Lots of people take the subway to get to Shea Stadium. They ramped up service a little bit but not much and they were able to run about 33 trains per hour on the 7 line, which is a lot of trains.

CHRIS HAYES: 33 an hour?

AARON GORDON: 33 an hour, for like a brief period during the game.

CHRIS HAYES: Right, but that's a lot.

AARON GORDON: Yeah, yeah.

CHRIS HAYES: I mean, there's only 60 minutes in an hour for folks that are doing the math in their head.

AARON GORDON: And they were able to do this because they had super qualified motormen who knew those signals so well that they barely even needed to look at the lights as they were approaching it because they knew it just ... Like they knew by feel.

CHRIS HAYES: I love this. I'm picturing in my head like the "Ocean's Eleven" montage where they like lay like ... Like the MTA person comes around to like pick out who are going to be the people driving the trains during the subway series.

AARON GORDON: Right.

CHRIS HAYES: Like who's like the top quality ... Like has the best feel, like we need our best people on the job to drive these trains during the subway.

AARON GORDON: I mean, this was what it was before computerized signaling, like driving train really was a skill. It was about how well you knew the track, it was about how well you knew the equipment, it was about how good of a feel you had for the speed you were going, and you know, just all that stuff.

CHRIS HAYES: Right.

AARON GORDON: They were able to achieve 33 trains per hour. Now the 7 line just got upgraded to computerized signaling. They're working on the roll out now. It's called CBTC. It's like what the basis for this plan to fix the subway is and once they get it figured out they will be able to run 29 trains per hour on the 7 line.

 

[KK 6 Ave Local]

SAS proposal:

, starts at 125 and Bway, 2 av until Hanover, then to Atlantic Exp

If you deinterline a lot of things like QBL and Dekalb then it makes having a lot of SAS extensions hard to put in without chain effects

 

[RR503]

1 hour ago, Union Tpke said:

It was very strictly done. It was stated in the following podcast: 

 

Interesting. Here's the relevant link to the SubTalk thread:

http://talk.nycsubway.org/perl/read?subtalk=516989

[Union Tpke]

19 minutes ago, RR503 said:

Interesting. Here's the relevant link to the SubTalk thread:

http://talk.nycsubway.org/perl/read?subtalk=516989

Thanks!

That was excellent!

Edited March 12, 2019 by Union Tpke

[Maxwell179]

3 hours ago, BreeddekalbL said:

In relation to my proposal of elimnating hewes street and flushing ave on the would it even be feasable to relocate marcy ave closer to the bus terminal  and exits including ADA facilities will be at the plaza end  it would include making the express track through so the trains can run through

glad I’m not the only person who considered that before 

[engineerboy6561]

On 3/10/2019 at 7:35 PM, bobtehpanda said:

Where would you shove an additional track? The el basically touches building walls as it is.

I think Broadway LIRR would be a better, slightly longer extension. Murray Hill is a quiet neighborhood and doesn't save much time over just going into Flushing itself (in fact, the Q15/A serves both); Broadway LIRR would provide more of a time savings, and its position at 162nd, Crocheron, and Northern means that there are lots of options for terminating buses.

I figured there would be room because the el runs with four tracks from 104 St to Willets, so I figured there would be room most of the way down, with maybe a couple of buildings coming down at station sites between 90 St and Woodside if things turned out to be a bit too tight. South of there everything's on a median until you get to QBP, which you'd need to reconstruct but could probably build with express tracks on a third, even higher level, and so four tracks wouldn't be that hard. South of QBP you'd run two stacked els, with the express tracks on the top and a stop at Court Sq; you'd need to get rid of the two layup tracks next to the existing  portal so you could drop the express tracks in next to the local tracks before expanding the line to four tracks out through Manhattan. Since the  is already the deepest layer of the subway system it would be fairly feasible to just build a second two-track line alongside it just to the north; shift the southbound local track over, and the current platforms at GCT, 5 Av, and Times Sq become the eastbound platforms and new westbound platforms are built to the north.

If this proves to be unfeasible, then instead of widening the el to four tracks rebuild the portion north of 46 St as a stacked 2+2 el with the express tracks up top and express stops at 61 St, 74 St, Junction Blvd, and Willets Pt before Flushing. East of Flushing just extend the whole thing out to Douglaston Pkwy, with new stops at Parsons Bl (local), 150 St (local), 162 St (express), Utopia Pkwy (local), Francis Lewis Blvd (local), Clearview Expwy (local), Bell Blvd(express), Springfield Blvd (local), Douglaston Pkwy (local), and Little Neck Pkwy (express).

If a full 60tph out past 162nd St seems a bit excessive, then it might make sense to send one of the express trains down to Jamaica Av, with stops at Utopia Pkwy/Hollis Ct Blvd, Hollis Ct Blvd/50 Av, Hollis Ct Blvd/Francis Lewis Blvd, Francis Lewis Blvd/73 Av, Francis Lewis Blvd/Union Turnpike, Francis Lewis Blvd/Hillside Av, and Francis Lewis Blvd/Jamaica Av. 

 

[bobtehpanda]

16 hours ago, engineerboy6561 said:

I figured there would be room because the el runs with four tracks from 104 St to Willets, so I figured there would be room most of the way down, with maybe a couple of buildings coming down at station sites between 90 St and Woodside if things turned out to be a bit too tight. South of there everything's on a median until you get to QBP, which you'd need to reconstruct but could probably build with express tracks on a third, even higher level, and so four tracks wouldn't be that hard. South of QBP you'd run two stacked els, with the express tracks on the top and a stop at Court Sq; you'd need to get rid of the two layup tracks next to the existing  portal so you could drop the express tracks in next to the local tracks before expanding the line to four tracks out through Manhattan. Since the  is already the deepest layer of the subway system it would be fairly feasible to just build a second two-track line alongside it just to the north; shift the southbound local track over, and the current platforms at GCT, 5 Av, and Times Sq become the eastbound platforms and new westbound platforms are built to the north.

If this proves to be unfeasible, then instead of widening the el to four tracks rebuild the portion north of 46 St as a stacked 2+2 el with the express tracks up top and express stops at 61 St, 74 St, Junction Blvd, and Willets Pt before Flushing. East of Flushing just extend the whole thing out to Douglaston Pkwy, with new stops at Parsons Bl (local), 150 St (local), 162 St (express), Utopia Pkwy (local), Francis Lewis Blvd (local), Clearview Expwy (local), Bell Blvd(express), Springfield Blvd (local), Douglaston Pkwy (local), and Little Neck Pkwy (express).

If a full 60tph out past 162nd St seems a bit excessive, then it might make sense to send one of the express trains down to Jamaica Av, with stops at Utopia Pkwy/Hollis Ct Blvd, Hollis Ct Blvd/50 Av, Hollis Ct Blvd/Francis Lewis Blvd, Francis Lewis Blvd/73 Av, Francis Lewis Blvd/Union Turnpike, Francis Lewis Blvd/Hillside Av, and Francis Lewis Blvd/Jamaica Av. 

 

I didn't know we were holding auditions to become the new @Wallyhorse!

• Building a second pair of tracks under 42nd St is not trivial because buildings have foundations and exist below the street level
• Good luck asking the (7)'s neighbors and the local councilman to accept the demolition of "a couple of buildings" and a wider or stacked el impacting the neighborhoods even more
• The day Douglaston is dense enough for a subway is the day Staten Island starts the Revolution and everybody is wearing Che Guevara t-shirts
• Hollis Court and Francis Lewis as a subway alignment is wild. You're talking about one street which is so lightly populated the bus on it doesn't even run on weekdays between 9am and 3pm, and another whose catchment is half taken up by Cunningham Park.

[engineerboy6561]

37 minutes ago, bobtehpanda said:

I didn't know we were holding auditions to become the new @Wallyhorse!

• Building a second pair of tracks under 42nd St is not trivial because buildings have foundations and exist below the street level
• Good luck asking the (7)'s neighbors and the local councilman to accept the demolition of "a couple of buildings" and a wider or stacked el impacting the neighborhoods even more
• The day Douglaston is dense enough for a subway is the day Staten Island starts the Revolution and everybody is wearing Che Guevara t-shirts
• Hollis Court and Francis Lewis as a subway alignment is wild. You're talking about one street which is so lightly populated the bus on it doesn't even run on weekdays between 9am and 3pm, and another whose catchment is half taken up by Cunningham Park.

You're right about Hollis Ct Blvd; I saw the boulevard width on Google Maps and assumed it had some development on it. My bad.

Douglaston is probably overkill; you could probably turn trains at Bell Blvd and be fine. I tend to propose pushing builds a little bit too far out rather than right-sizing them to account for potential future demand.

As far as the el width and Manhattan extension I'd like to talk a bit more about my thoughts; the issue here is that 40 IRT TPH is not enough to serve as a trunk for most of northern Queens the way the   currently does. You could four-track the main Roosevelt Av el, but you run into the issues fitting it that you brought up (that I missed because it looked like the platforms on the el had ~10' of clearance from building walls, and so adding a fourth track would reduce that clearance to ~4' (which is tighter but still theoretically feasible), and even if you four-track all the way to QBP the extra 20TPH has to go somewhere that's not the current track pair under 42nd St, and I figured that the current  is deep enough that it would already be under most foundations and so you could expand it to four tracks more cleanly. If not we still need to figure out a place on the western end to turn those trains (unless we want to run locals only to QBP and then have all the trains from midtown run as  trains, which isn't a particularly good solution either.

So given that you can either double up on the existing track pair, send the new  trains through a new tunnel to 59 St, or leave the  alone and build something out with four tracks, either along Northern from QP or along Astoria Blvd from the UES. The alternative to expanding or burying the  (which to be clear I'm talking about doing when the el needs to come down and/or get renovated in large chunks) would be to build something new (which would add fairly large costs to the system in addition to the work that would already happen, bury the  (which makes the most sense, and then when burying it you can make it four tracks to improve capacity), with the core cost being that the el would need to stay up a few extra years during construction to avoid the massive disruption that would come with trying to tie the el to the new subway midway through. I'd personally prefer to quad-track the  underground, and then figure out what we'd have to do to have 60-80tph worth of capacity feeding into the new underground Flushing Line; I figured replacing the el with a four-tracked structure would do the thing for less money and be more palatable.

If you don't mind my asking, @bobtehpanda, what are your thoughts about what should be done?

Edited March 12, 2019 by engineerboy6561

[bobtehpanda]

55 minutes ago, engineerboy6561 said:

You're right about Hollis Ct Blvd; I saw the boulevard width on Google Maps and assumed it had some development on it. My bad.

Douglaston is probably overkill; you could probably turn trains at Bell Blvd and be fine. I tend to propose pushing builds a little bit too far out rather than right-sizing them to account for potential future demand.

As far as the el width and Manhattan extension I'd like to talk a bit more about my thoughts; the issue here is that 40 IRT TPH is not enough to serve as a trunk for most of northern Queens the way the   currently does. You could four-track the main Roosevelt Av el, but you run into the issues fitting it that you brought up (that I missed because it looked like the platforms on the el had ~10' of clearance from building walls, and so adding a fourth track would reduce that clearance to ~4' (which is tighter but still theoretically feasible), and even if you four-track all the way to QBP the extra 20TPH has to go somewhere that's not the current track pair under 42nd St, and I figured that the current  is deep enough that it would already be under most foundations and so you could expand it to four tracks more cleanly. If not we still need to figure out a place on the western end to turn those trains (unless we want to run locals only to QBP and then have all the trains from midtown run as  trains, which isn't a particularly good solution either.

So given that you can either double up on the existing track pair, send the new  trains through a new tunnel to 59 St, or leave the  alone and build something out with four tracks, either along Northern from QP or along Astoria Blvd from the UES. The alternative to expanding or burying the  (which to be clear I'm talking about doing when the el needs to come down and/or get renovated in large chunks) would be to build something new (which would add fairly large costs to the system in addition to the work that would already happen, bury the  (which makes the most sense, and then when burying it you can make it four tracks to improve capacity), with the core cost being that the el would need to stay up a few extra years during construction to avoid the massive disruption that would come with trying to tie the el to the new subway midway through. I'd personally prefer to quad-track the  underground, and then figure out what we'd have to do to have 60-80tph worth of capacity feeding into the new underground Flushing Line; I figured replacing the el with a four-tracked structure would do the thing for less money and be more palatable.

If you don't mind my asking, @bobtehpanda, what are your thoughts about what should be done?

This proposal only makes sense with a few assumptions:

• Els are feasible to build on narrow roads in modern day New York
• Roosevelt is the only place we can build new train relief for the  , or the best place we can do so.

I don't believe either is true. The problem with the  alignment is largely that it's a victim of its own success. Development around the line is tight. All traffic heads to this development or to transfers with the subway, so as a result the buses reaching the  are slow and congested, Jackson Heights-Roosevelt and QBP is severely overcrowded, etc.

You would be much better fanning out the crowds by doing one of the following:

• Relief line on Astoria (will draw some current bus passengers to a different subway line entirely)
• Relief line on Northern (same as above)
• Building an Elmhurst stop and one or two Corona stops on the PW (there's scope to improve capacity, either by increasing frequency or by rearranging seat layouts on trains, or even buying entirely new trains. It's worth noting that the platforms are 850 ft long to accommodate LIRR trains, so they're much higher capacity than even the extra-long Flushing Line)

Edited March 13, 2019 by bobtehpanda

[engineerboy6561]

3 hours ago, bobtehpanda said:

This proposal only makes sense with a few assumptions:

• Els are feasible to build on narrow roads in modern day New York
• Roosevelt is the only place we can build new train relief for the  , or the best place we can do so.

I don't believe either is true. The problem with the  alignment is largely that it's a victim of its own success. Development around the line is tight. All traffic heads to this development or to transfers with the subway, so as a result the buses reaching the  are slow and congested, Jackson Heights-Roosevelt and QBP is severely overcrowded, etc.

You would be much better off doing one of the following:

• Relief line on Astoria (will draw some current bus passengers to a different subway line entirely)
• Relief line on Northern (same as above)
• Building an Elmhurst stop and one or two Corona stops on the PW (there's scope to improve capacity, either by increasing frequency or by rearranging seat layouts on trains, or even buying entirely new trains. It's worth noting that the platforms are 850 ft long to accommodate LIRR trains, so they're much higher capacity than even the extra-long Flushing Line)

That would make a lot of sense; I personally think a relief line connecting Flushing to 2 Av via Astoria Blvd and then continuing out to Bayside via Northern would make more sense because of the ease of patching it into the Manhattan network. If you put an express lower level on 2 Av (or just expand it to four tracks), then you could send 2-4 tracks under the river north of 86 St and send them along Astoria Blvd; depending on what you're doing you could rope the Astoria el into the new line, since Astoria Blvd is 95-100 feet wide over the street you could fit a full size four-track el over the street, and the portion where Astoria Blvd is a service road to the GCP would make that much easier so that you don't have to build the station an extra 15-20' below street level.

Basically what I'd propose is two tracks from 86 St/2 Av branch off and swing under Hell Gate, come up between Hoyt Ave S and the Astoria Park running track, and swing over Hoyt Av S, with a stop at 21 St. I'd reconfigure 31 St/Astoria Blvd as a two-level station with the current  platform rotated so that it stretches from 32 St/Astoria Bl to 35 St over the center of the GCP and the 2 Av platform above it from 35 St to 31 St. East of the station the 2 Av tracks would come down to the  level, and the tracks would merge to form a four-track line (  providing local service, the 2 Av trains providing express service), and then local stops at Steinway St and Hazen St before a two-level express station at 82nd St (four tracks on the lower level, two above); the two upper level tracks would turn north and serve LGA directly with stops at 90 St/GCP and Central Terminal before terminating at Terminal D.

The mainline would continue along Astoria Blvd, with local stops at 88 and 94 Sts, an express stop at 31 Av, a local stop at Ditmars Blvd, an express stop at Shea Stadium North, and then a plunge underground before stops at Flushing-Main St (express, connect to ) Parsons Bl (local), 150 St (local), 162 St (express), Utopia Pkwy (local), Francis Lewis Blvd (local), Clearview Expwy (local) and Bell Blvd(express).

If you build this, and then add stops on the PW at 108 St (Corona) and Broadway (Elmhurst) then that should help a great deal; you'd basically be flanking the  with higher-speed services that run farther faster, and you'd also have one-seat access to both the east and west sides of Manhattan (which should decongest GCT and Times Sq by a reasonable amount).

Edited March 13, 2019 by engineerboy6561

[bobtehpanda]

1 hour ago, engineerboy6561 said:

That would make a lot of sense; I personally think a relief line connecting Flushing to 2 Av via Astoria Blvd and then continuing out to Bayside via Northern would make more sense because of the ease of patching it into the Manhattan network. If you put an express lower level on 2 Av (or just expand it to four tracks), then you could send 2-4 tracks under the river north of 86 St and send them along Astoria Blvd; depending on what you're doing you could rope the Astoria el into the new line, since Astoria Blvd is 95-100 feet wide over the street you could fit a full size four-track el over the street, and the portion where Astoria Blvd is a service road to the GCP would make that much easier so that you don't have to build the station an extra 15-20' below street level.

Elevated on Astoria is not possible because of the landing paths for LaGuardia. The runways end right before the GCP, so there wouldn't be enough vertical clearance.

I also think 2nd Av - Queens is a terrible idea, because 2nd Av has really bad transfers to the rest of the system. Better to have an 86th St line run across town and terminate at 72nd/Broadway, with future expansion down the FWS and eventually linking to the  . 

[engineerboy6561]

1 hour ago, bobtehpanda said:

Elevated on Astoria is not possible because of the landing paths for LaGuardia. The runways end right before the GCP, so there wouldn't be enough vertical clearance.

I also think 2nd Av - Queens is a terrible idea, because 2nd Av has really bad transfers to the rest of the system. Better to have an 86th St line run across town and terminate at 72nd/Broadway, with future expansion down the FWS and eventually linking to the  . 

The elevated on Astoria is fair; as far as 2 Av to Queens having terrible connections is concerned that's at least partially a function of how absolutely hilariously poor the planning for the 2 Av project is. Ideally 2 Av would have connections at 59 St (59/Lex  via the Broadway platform), 50 St (connection to 53rd St  via  platform), 42nd St( connection to  via a new station), 14 St (connection to  via the eastern end of the 3 Av station), Houston St (connection Lower East Side/2 Av  ), and Grand St (connection to  ).

 

Edited March 13, 2019 by engineerboy6561

[bobtehpanda]

50 minutes ago, engineerboy6561 said:

The elevated on Astoria is fair; as far as 2 Av to Queens having terrible connections is concerned that's at least partially a function of how absolutely hilariously poorly this was planned. Ideally 2 Av would have connections at 59 St (59/Lex  via the Broadway platform), 50 St (connection to 53rd St  via  platform), 42nd St( connection to  via a new station), 14 St (connection to  via the eastern end of the 3 Av station), Houston St (connection Lower East Side/2 Av  ), and Grand St (connection to  ). Looping the  back around as an 86 St crosstown as part of the GCP to Flushing line makes sense; maybe send that on the LGA spur.

The concern isn't where the transfers are (the only one they're missing is 59th, which is fine because the SAS was never going to outcompete a transfer involving minimum walking) - the concern is that the transfers are all long. 2nd Avenue is at least one avenue block away from every single transfer.

I would prefer the 86 line lead exactly to one pair of tracks, because reverse branching is a sin that should be avoided.

Edited March 13, 2019 by bobtehpanda

[engineerboy6561]

50 minutes ago, bobtehpanda said:

The concern isn't where the transfers are (the only one they're missing is 59th, which is fine because the SAS was never going to outcompete a transfer involving minimum walking) - the concern is that the transfers are all long. 2nd Avenue is at least one avenue block away from every single transfer.

I would prefer the 86 line lead exactly to one pair of tracks, because reverse branching is a sin that should be avoided.

The best counterargument I could make is the transfer passage between TSQ and PABT, which is a full avenue block long but still sees a lot of heavy use. Also, most of the transfers are a lot shorter than an avenue block (53rd/Lex is about half an avenue block according to Google Maps, and 14 St/2 Av, Grand Av, and LES/2 Av are all literally right on top of 2 Av. The big issue would be the   at 42 St (which would ideally gain a stop between 2 Av and 1 Av to facilitate transfers, and would be a 900' walk otherwise).

The other argument I have is that to make 2 Av viable as a new trunk it needs connections to the Bronx, Queens, and Brooklyn. The Bronx is easy (four-track trunk up 3 Av to Fordham Plz, then up to Gun Hill Rd and swing across to Co-op), and Brooklyn's not bad either (replace the Jamaica el with a four-track line carrying a mix of 2 Av and Nassau St trains), but the best spot to put a Queens connection would be the Astoria Bl/86 St connection. It would make sense to not directly interline an 86 St crosstown with 2 Av trains for operational complexity reasons, but unless you want to run another line via Northern Bl/36 Av (which might not be a bad idea; it would leave all of NW and north central Queens within 1/2 mile on foot of a subway) then 2 Av should connect to Astoria Blvd.

The reverse branching thing is somewhat of a concern; I agree that interlining is nonideal because it means merge complexity and opportunities for delays to happen, but you're basically trading one-seat rides for some amount of operational sadness no matter how you do it. I believe that interlining and reverse branching should be kept fairly low, but I also think that unless we build out additional lines on Northern Blvd leaving the 2 Av line disconnected from Astoria Blvd is a mistake. The problem with my approach is there's going to be a merge somewhere unless all 2 Av expresses serve Queens (which is a poor tradeoff for other reasons), and so running the  to Astoria Blvd means trading off an  merge with some merging sadness on the 2 Av trunk. I'd rather have the merging sadness happen between the  and a 2 Av local train that would be continuing on as an Astoria Blvd local than have something like the 53 St  merge at 86 St/2 Av.

Edited March 13, 2019 by engineerboy6561

[Union Tpke]

6 hours ago, bobtehpanda said:

Elevated on Astoria is not possible because of the landing paths for LaGuardia. The runways end right before the GCP, so there wouldn't be enough vertical clearance.

I also think 2nd Av - Queens is a terrible idea, because 2nd Av has really bad transfers to the rest of the system. Better to have an 86th St line run across town and terminate at 72nd/Broadway, with future expansion down the FWS and eventually linking to the  . 

I have heard that because that area was landfill, groundwater could be a major issue.

[T to Dyre Avenue]

That seems like it would make Northern Blvd the better choice for the  relief line by default.

Either way, it should be its own self-contained line, so it can offer the maximum amount of trains per hour. It probably wouldn’t be able offer all that much relief as a branch off of the 2nd Ave Subway.

Edited March 13, 2019 by T to Dyre Avenue