engineerboy6561

I mean, if you use the Milan method (modified cut and cover, where you build the walls first, then dig a trench down to about where you're putting the roof, then put your equipment in the trench, then pour the roof slab and deck over everything) you can probably get fairly low construction costs and fairly low disruption, and that's significantly better than boring anything if we can do it that way. According to Alon Levy the cost of a TBM scales roughly linearly with diameter, and so the extra cost for a 65' diameter tunnel (two tracks plus platforms) would only be about 1.5x the cost of a 45' diameter tunnel (able to carry two levels of two tracks), but would also save us the cost of drilling and blasting a cavern at every station. I don't actually have hard numbers for the cost per station cavern, and I'd be very interested to see what the breakeven station spacing is (i.e. how many stations per mile do you need to have for the cost of a 65' tunnel with platforms inside to come out even with or less than the cost of a 45' tunnel of a given length plus drilled and blasted platform spaces at each station). Furthermore, the "basement" can be used for an emergency walkway in the tunnel, and the "attic" can be used for machine rooms, offices, etc. near stations, and a place to put electrical/plumbing/ventilation equipment between stations.

So basically Eastern Parkway is a big enough thoroughfare for traffic going across Brooklyn in that area that trying to rip it up and rebuild the subway underneath it would provoke World War III with everyone who drives on Eastern (which sounds like a fair number of people, given that both Empire Blvd and Atlantic Ave are already full, and Empire Blvd just dumps out onto Flatbush Av at Prospect Park); just out of curiosity, what's the geological issue with the Parkway? I know that farther south in Brooklyn the soil gets sufficiently difficult to tunnel through that planned lines along Nostrand and Utica would need to surface somewhere around Avenue S or so.

Honestly I'd agree that it would be better to just rebuild Rogers as part of a flattening of Eastern Parkway west of Utica Av. I'm tempted to suggest that we leave Utica Av as a two-level station, add tail tracks to the lower level, set the lower level up with tail tracks, and add an additional interlocking to allow trains to either terminate on the lower level island platform or switch to and from the local tracks just west of the station. I personally would have Utica Av run off a rebuilt Jamaica trunk; the downside is that making that work would likely require six tracks between Essex St and Myrtle Av to handle combined loading (15tph to Metropolitan Av, 30 express tph to Jamaica and beyond, 15 local tph to Jamaica, 15 or 30 tph to Utica Av. The original Second System proposal had a crosstown line running along Worth St that would have taken 8 Av local trains and brought them over to join the South 4th St trunk; if you brought that back you could in theory send the down Utica Av to Kings Plaza. The downside is that the only way to do that is to rebuild Jamaica, and then build a Worth St line.

I'm not sure, especially because the relative widths of the platforms at some of the express stations (145 St, Tremont Av IIRC, and Fordham Rd) suggest that the original plan was to run four tracks; if I had to guess there wasn't money for four tracks across the entire first system, and because the Concourse line directly parallels Jerome the whole way from Bedford Park Blvd down to 161 St it was probably the place they figured that dropping the bidirectional express would do the least harm. It would be really interesting to get a solid list of pinch points like that in one place, especially if we could get a meaningful sense of what it would take to do something about them. For instance, looking at ENY yard on satellite view tells me that if you wanted to run 600' trains out of there you'd basically need to buy up the two or three blocks next to the yard (the area bounded by the tracks, Conway St, and Bushwick Av, plus the storage place there, and then build a new 600' car barn extending from where Stewart St is now through where the storage place is. If you did that you could probably extend about 10 or 15 of the storage tracks at ENY up to Conway St, and then bring the track that currently connects eastbound trains to the yard up and over the portal. You could do it, but it would basically entail completely rebuilding ENY to do it, and I'm not sure where you'd put all the trains that currently live there while you did that. Also Rogers Junction is a shitshow, and ideally would be rebuilt to the same design as the IND junctions just below Columbus Circle as a congestion management thing, but that would be interesting and quite possibly require the rebuilding of President St as well (President St looks like it ends around 300' south of where the platforms at Nostrand Av sit, and just west of Nostrand Av the northbound has to swing under the northbound to pop up in the correct spot by Franklin Av, and the distance between Franklin Av and Nostrand Av is only about 1500 feet. If we assume that the clearance for individual tracks to pass underneath each other is about 15', then I can take a rough whack at laying this out: Just east of Franklin Av, the two Nostrand Av tracks branch off and descend 15 feet over about 450 feet (so a 3.3% grade), as does the westbound track. The Nostrand Av and westbound tracks then level off and continue for about 450 feet, while the westbound track descends another 15 feet; at that point the Nostrand Av and westbound tracks swing south; the westbound track connects to the lower level of Nostrand Av, while the Nostrand Av tracks connect to President St; the westbound track slowly rises back up over the following 900' and joins the old tunnel lower level just past the end of Nostrand Av station. The downside is that making this happen would require the rebuilding of Franklin Av, Nostrand Av, and President St, which would likely take years because of the sheer amount of track that would be getting ripped up and relaid. Furthermore, while work on the junction is happening nothing can run through it; during that time you'd basically need to turn the at Bowling Green, turn the at either Franklin Av with a new crossover or Atlantic/Barclays using the central crossover there, and then turn the and upper section of the at Franklin Av using the crossover just to the north of there. The southern section of the would basically only run from New Lots to Utica Av at a fairly low frequency, and both Nostrand Av and Eastern Parkway between Utica Av and Franklin Av would need to be bustituted; in theory you could run a shuttle from President St to Flatbush Av, but unless you added a brand new track connection from President St to either level of the Eastern Parkway line east of Nostrand those trains would be trapped there without any way to maintain them (and if, as I suspect, making this work would require rebuilding President St then you can't do that either). You'd wind up doing the work in phases: Phase I would be a few months to a year of preparatory trackwork; we'd need to add switches so that the center track stub on the coming out of Utica connected to both local tracks west of Sutter Av (which wouldn't be too bad and could probably be done at night with shuttle buses replacing the between Utica and New Lots), add a crossover just west of Franklin Av to allow trains to turn there (which you could probably also do late nights), and add bidirectional signaling on all four tracks between Eastern Parkway and Franklin Av so Franklin Av can be used as a terminus). Phase II would be 3-5 years of absolute misery; and trains would turn at Franklin Av, trains would turn at Bowling Green all day, and trains would run in two sections (between 148 St and 42 St sharing trains from the , and between New Lots Av and Utica Av at a 10-minute headway using trains out of Livonia Yard). Shuttle buses would have to replace trains between Utica and Franklin Av, as well as between Flatbush Av and Franklin Av, so Nostrand Av would just completely lose subway service. Phase III could probably be done in a year or so and would offer partial relief but still be not great; express trains could run through Franklin Av while platforms are widened, elevators, added, etc. but local trains still wouldn't be able to. The would run to Eastern Parkway, empty out there, and reverse at the crossover without entering Franklin Av, and the would still run to 42 St. However, the would run to New Lots all times, skipping Franklin, Nostrand and Kingston Aves and the would run to Flatbush all times), so at least there would be some through service on the network.

This is what I was thinking of when I referenced a better way to do TBM construction if we want to build four-track trunks that way in the future; this first image is a cross-section of the tunnel used for Barcelona Line 9; they used a 40' outer diameter tunnel to fit two tracks, one on top of the other, against one edge of the tunnel and the platforms on one side of the tunnel next to the tracks. That way, the platform itself resides inside the tunnel, so the only extra mining/drilling/blasting you have to do is to build the entrance/exit box, which can be significantly smaller than what we built for Phase I with the full length mezzanine. That said, a quad-track bored tunnel that's big enough to include stations has a 65' outer diameter (60' inner diameter inside the tunnel lining), which is pretty huge. A single-level cut-and-cover tunnel would be 60' wide x 15' tall between stations, and then at stations widen to about 100-110' wide at platform level if all tracks are on the same level; a two-level cut-and-cover tunnel constructed the same way as above would probably be a 30'x30' square cross-section between stations, widening out to about 55' wide at stations. You'd move a lot less earth and get away with a lot less complexity if you just built the tunnel cut-and-cover or Milan-style. Also apparently BART is trying a large stacked tunnel like this with only two tracks and somehow they're getting costs of $1.5B per mile, which is just ridiculous. The reason I bring up the stacked structure is because while Northern Blvd, Merrick Blvd, and 2 Av/3 Av in Manhattan are all wide enough to comfortably take four tracks (95-100 feet between buildings) 3 Av in the Bronx can get a bit tight (closer to 80 feet between buildings), and Broadway and Jamaica Av are both in the 70-75' wide range, so I'm not sure if it would be possible to fit a four-tracked single-level line underneath them without hitting everyone's basements along the way. On the other hand, 55'-60' maximum width is narrow enough that it can basically go under any street and be OK.

And that brings us back to the need for a rebuilt and expanded Jamaica line; a four-track trunk from Jamaica into downtown via Williamsburg would do wonders for decongesting QBL and improving transit access to southeastern Queens. The core issue there becomes what Manhattan trunks to feed it with; you wouldn't want to build something that size only to end it at Broad Street because that would significantly reduce utilization and would wind up either not drawing enough ridership to make a difference or horrendously overcrowding Delancey, Canal, and Fulton with pax transferring to Midtown. The original plan for this under the Second System was to add a six-track trunk from Hanover St to Myrtle Av, and then four-track trunks down Utica Av to Kings Highway and along Myrtle Av/Woodhaven Blvd/Conduit Av to Ozone Park, with two-track extensions on Linden Blvd and 120 Av, and then to feed the whole thing with a mix of trains from 8 Av, 6 Av, 2 Av, and the Jamaica Line. I don't necessarily think that it would make sense to complete that design as proposed back then, but I think there are a few different versions of a combination Broadway/Jamaica Av/Utica Av second trunk that could accomplish a bunch of different things at various costs that I figured I'd put out here to see what folks think, from cheapest to most expensive. Option 1: Four-track Jamaica line running through to a four-track lower SAS, two-track branch to Utica Av; this version would convert the to 2 Av trains, along with the . This version would allow for 45 through TPH to Jamaica (30tph express, 15tph local) if everything were set up correctly; you'd have stops at: Rivington/Essex Sts (express) Pitt/Columbia Sts (local) Wythe Av (local) Marcy Av (express) Union St (local, connect to ) Flushing Av (local) Myrtle Av (express, branches off here to serve Middle Village) Arlington Av/Jamaica Av (local) Cleveland St/Jamaica Av (local) Highland Blvd/Jamaica Av (local) Cypress Hills St/Jamaica Av (express) 75 St (local) 85 St (local) Woodhaven Blvd (express) 104 St (local) 111 St (local) 121 St (local) Sutphin Blvd (express, connect to ) Parsons Blvd (express, connect to ) Pros: This is the cheapest option for doing this; the only work I'm proposing here is a rebuild of the Jamaica line, which would be about 11 miles total, plus tail tracks on the end of Parsons-Archer (so 44 track miles). If we gave up on full mezzanines we could use a single TBM about 62-65' in diameter to bore out a four-track two-level tunnel with the platforms fitting almost entirely in the tunnel bore; we'd basically only need to excavate two station boxes about 75' x 75' in floor area above the tunnel diameter that we could then add escalators and elevators to; the Milan method would likely be even cheaper ($50-100M or so per km in Italy and Turkey; double that because MTA, but that would still only cost about $3-3.2B to build a four-track line from Parsons-Archer to 66th Street and 2nd Avenue) Cons: If we just build that segment we wind up with two or three trains having to turn at 59 St because there's nothing to receive them heading north beyond that point; we could turn some of the trains downtown at Fulton or Broad St, but the extra transfers required to make that work would likely reduce ridership and reduce effectiveness of the new line. We could probably turn 15tph at Broad St and 15 tph or so at Chambers St, and that way 59 St either doesn't have to turn trains (if the are the only trains on 2 Av) or it only has to turn 15tph (if the go up 2 Av together to decongest the Broadway local tracks above 57 St. This is basically taking the existing Jamaica trunk and improving it, but that's about it). Option 2: Four-track Jamaica line, tied to a four-track 2 Av line and a four-track 3 Av line in the Bronx, and a two-track Northern Blvd line in Queens; this version would again make the all 2 Av trains, and would accommodate Broadway express trains all sharing the same line. Stops: Gun Hill Rd extension (2 tracks, served by ) Bay Plaza (2 tracks) Bartow/Edson Avs (2 tracks) Gun Hill Rd/Arnow Av (2 tracks) Gun Hill Rd/Seymour Av (2 tracks, connect with ) Gun Hill Rd/Boston Rd (2 tracks) 3 Av trunk line (served by local, express) : Gun Hill Rd/White Plains Rd (start of four-track segment, express, connect to , terminates here as 3 Av local) Norwood-205 St (express, branches off) Bedford Park Blvd/Webster Av (local) Fordham Plaza (express, connect to MNR) 3 Av/183 St (local) 3 Av/180 St (local) 3 Av/E Tremont Av (express) 3 Av/E 174 St (local) 3 Av/Claremont Parkway (local) 3 Av/E 169 St (local) 3 Av/E 161-163 Sts (express, provision for Boston Rd branch) 3 Av/E 156 St (local) 3 Av/E 149 St (express, connect to 3 Av/E 143 St (local) 3 Av/E 138 St (express, connect to ) 125 St branch (2 tracks, served by ): 125 St/St Nicholas Av (connect to ) 125 St/Lenox Av (connect to ) 125 St/Lexington Av (connect to , MNR) Upper 2 Av trunk (served by ): 2 Av/E 125 St(express) 2 Av/E 116 St (local, connect to ) 2 Av/E 106 St (local) 2 Av/E 96 St (express) 2 Av/E 86 St (local) 2 Av/E 72 St upper level (express, connect to (J) (M) (Z) ) Northern Blvd line eastern extension (2 tracks, served by express) Northern Blvd/Bell Blvd Northern Blvd/207 St Northern Blvd/Francis Lewis Blvd Northern Blvd/Utopia Parkway Northern Blvd/162 St (connect to LIRR) Northern Blvd/150 St Northern Blvd main trunk (4 tracks, served by local, express) Northern Blvd/Main St (express, connect to LIRR, ) Northern Blvd/Shea Stadium (express) Northern Blvd/114 St (local) Northern Blvd/108 St (local) Northern Blvd/100 St (local) Northern Blvd/Junction Blvd (express) Northern Blvd/88 St (local) Northern Blvd/82 St (express) Northern Blvd/72 St (local) Northern Blvd/62 St (local) Northern Blvd/54 St (express, connect to ) Northern Blvd/48 St (local) Northern Blvd/36 Av (local) 36 Av/31 St (express, connect to ) 36 Av/21 St (local) 36 Av/Vernon Bl (local) Roosevelt Island North (local) 2 Av/72 St lower level (express, connect to (T)) Lower 2 Av trunk (4 tracks, served by (J)(M)(Z)(T) 3 Av/E 61 St (express, connect to ) 3 Av/ E 52 St (local, connect to ) 3 Av/E 42 St (express, connect to ) 2 Av/E 34 St (local) 2 Av/E 23 St (local) 2 Av/E 14 St (express, connect to ) 2 Av/St Mark's Pl (local) 2 Av/Houston St (express, connect to ) Brooklyn Trunk (4 tracks, served by (T)(M)(Z)) Rivington/Essex Sts (express) Pitt/Columbia Sts (local) Wythe Av (local) Marcy Av (express) Union St (local, connect to ) Flushing Av (local) Myrtle Av (express, branches off here to serve Middle Village) Arlington Av/Jamaica Av (local) Cleveland St/Jamaica Av (local) Highland Blvd/Jamaica Av (local) Cypress Hills St/Jamaica Av (express) 75 St (local) 85 St (local) Woodhaven Blvd (express) 104 St (local) 111 St (local) 121 St (local) Sutphin Blvd (express, connect to ) Parsons Blvd (express, connect to ) Liberty Av/Merrick Bl (local) Merrick Blvd/108 Av (local) Merrick Blvd/Linden Blvd (express) Merrick Blvd/120 Av (local) Merrick Blvd/Farmers Blvd (local) Merrick Blvd/Springfield Blvd (express, (J) terminates) Springfield Gardens extension (2 tracks, served by (T)(Z)): Merrick Blvd/225 St Merrick Blvd/Francis Lewis Blvd Merrick Blvd/243 St (terminal). Pros: This configuration successfully accomplishes several main goals; it should pull a majority of the ridership off the Lexington Avenue lines, as it offers a faster alternative from Jerome Av, WPR, and the Concourse into the core, both on the east side and on the west side (because of the Broadway connection via 63rd Street); it should also offer a ride into downtown and the core from Jamaica that is approximately as fast and as frequent as QBL offers, which should relieve a significant amount of the pressure on QBL. Furthermore, the extension under Merrick Blvd covers a corridor that currently sees four to five very high-frequency limited-stop bus routes, and should relieve congestion at Parsons/Archer (because people who currently take buses from the area along Merrick and Brewer Blvds into Jamaica can instead walk directly to a local subway station). The configuration of the (T) and (Z) as one-seat express rides into the city from SE Queens should also allow them to pull a significant load off the (E); similarly the configuration of the (J) (M) and (Z) should pull a ton of people off the (7). Cons: This is significantly more expensive than the initial proposal; we'd be constructing about 30 miles of four-track trunk ( 6.6 miles of four-track trunk in the Bronx, 6.4 miles of four-track trunk from 125 St-2 Av down to the Lower East Side, fourteen miles of four-track trunk from LES-2 Av to Springfield Blvd in SE Queens, 7 miles of four-track trunk in northern Queens, as well as 8.6 miles of two-track extensions (1.15 miles along 125 St, 2.35 miles in the Bronx, 1.45 miles in SE Queens, 3.65 miles in NE Queens,). If we assume a cost of $320M per mile for two-track lines and about $600M per mile for quad-track trunk, that would cost out to around $18B for the four-track trunk and around $2.9B for the various extensions, so around $18 billion for the whole project (this assumes that we move forward with a combination of the Milan method for most construction, and that bored sections of quad-track trunk are constructed from a 60' inner diameter tunnel so that the platforms fit inside the tunnel bore and we only have to add small station boxes (say 75' x 75' at the 1/4 and 3/4 points of the station; I can post a drawing later tonight showing what I'm describing, but the gist of that is that it should significantly reduce station costs over the 2 Av subway model (using a TBM to carry the tracks outside stations, and then drilling and blasting giant caverns at stations). That total cost of $18B would be compared to an expected cost of around $8.4B for just an upgraded Jamaica line and lower SAS using roughly the same rules of estimation (14 miles of quad-track trunk from upper Midtown to Jamaica, but no extensions beyond that). Even then, those numbers assume a significantly lower cost than SAS Phase I cost (which was like $2-2.5 billion per mile) because it's dropping expensive features and returning to more affordable construction methods like cut and cover where at all possible. Unfortunately I'm doing some guessing at what the TBM version costs; Barcelona's Line 9 uses a similar construction method to what I've described above (a single big tunnel large enough to hold all the tracks, platforms, and related infrastructure), and it's come in around $200M/mile over there, and the tunnel diameter I'm describing is about 1.5x the diameter of the Barcelona deep tunnel. If we assume TBM cost scales with cross-sectional area, then estimated costs are probably about $450M/mile for the four-track version that I'm proposing (plus 33% because hey, it's New York).

Agreed; in an ideal world you could send the 125 St line via Ditmars Blvd, Grand Central Parkway, and Main St down to Jamaica, capturing the M60 and the southern half of the Q44 in one go. NJ actually is dense enough for a subway, at least along Bergenline Av and JFK Blvd south of Fairview, but that would probably be better left to a massively expanded PATH.

TBH I would love a 125 St crosstown, but I really think it would either be better as a shuttle or as Phase I of an -type line to Queens or NJ than as the primary endpoint for SAS; the Bronx needs the trains more, and it makes more sense to have crosstown be a separate shuttle with provisions for extensions to Randall's Island and Queens than to have it be part of 2 Av.

That's a good sign in some ways (that they're at least trying to keep the ball rolling), and in others (namely the single-minded focus on turning under 125 St) not so great.

I don't have a problem with having the track connection there in case CPW falls apart, but I don't think it makes sense to try to fit a third service onto the Concourse as a way of boosting service to the Bronx; we'd be better off running 125 St as a shuttle and having 2 Av trains all continue north via 3 Av.

I'd argue it's complicated because the subway desert in southeastern Queens means that QBL expresses are running 30tph packed to the gills during rush, and so there isn't really enough peak capacity on QBL to handle those loads coming in. There are a variety of ways we could fix that over the longer term; LIRR fare integration would allow people going from SE Queens to the midtown core only paying $3-$4 and get a transfer to subway and local buses, but the total available capacity on the LIRR isn't necessarily high enough to put a huge dent in that load, especially during rush when LIRR trains are likely coming into Jamaica already SRO. As far as I'm aware, the maximum theoretical bidirectional throughput of the LIRR into Manhattan is about 80tph (30tph through ESA, then 60 total TPH into Penn via the East River tubes, but subtract about 10 slots an hour for NJT moves to and from Sunnyside and Amtrak slots); the mainline is quad-tracked east of Woodside, which would allow about 60 total tph coming in from Jamaica. Jamaica in turn sees trains from nine branches (including Montauk and Oyster Bay) that it can allocate slots to. In a world where rolling stock issues are resolved and LIRR has a fleet of dual-mode MUs that can fit into GCM and keep pace with the M7s and M9s you could just treat all runs evenly, and that means you could allot each branch a slot into Manhattan every 9 minutes (the Port Washington branch is its own thing because of the way the tracks are laid out, but that could run every 5 minutes during rush to take a load off the if we wanted, assuming we double-tracked the line from Manhasset to Port Washington and turned every other train at Great Neck. Realistically speaking, though, making that happen would likely take 20-30 years at least and take a degree of focus and self-discipline the MTA currently doesn't have, and I don't have a good sense of how much pressure it would actually wind up relieving on the QBL. Honestly the only way I could see a sweeping reorganization of the MTA actually happen is as the "stick" half of a bailout bill. You could tie a reorganization of the MTA board into a locally elected body to the next bailout bill to come through Albany; the idea would be that counties that get MTA service get board seats apportioned by population (so Putnam County gets 1, Dutchess gets 3, Orange gets 3, Rockland gets 4, Westchester gets 10, Nassau gets 14, Suffolk gets 15, Fairfield gets 10, New Haven gets 9, the Bronx gets 14, Manhattan gets 16, Queens gets 23, Brooklyn gets 26, Staten Island gets 5). If you added NJ into the mix, Hudson County gets 7, Bergen County gets 10, Essex County gets 9, Union County gets 6, Morris gets 5, Middlesex gets 8, Monmouth gets 6, Ocean gets 6, and Mercer gets 4. All told, your board size would be 214 members, of which New York City proper would have 84. On the one hand that's about the size of the NY state legislature; on the other hand you'd wind up with a properly regional institution with meaningful local accountability. Now it's also possible with a board that size that you'd be trading a few appointed fools and wastrels for a lot of elected fools and firebrands, but an organization set up that way would at least be somewhat accountable to the people they're supposed to be serving (which is better than what we have now). If need be you could add a layer of management more concerned with short-term execution and day-to-day management that the big board would be responsible for hiring and firing, as well as setting the long-term agenda for, but layers of management are expensive and don't necessarily add a ton of value. The idea would be that a combination of fares and service taxes on county residents and people who work in NYC would cover most of the costs of running and expanding the system, and that a reorganization of this nature would refocus board members on providing quality transit rather than hanging out in sinecures or rewarding their buddies.

That's fair; I'd honestly aim for the high end of that estimate (I believe my current proposal is for 8 stations on that stretch, including 156 St); you and I both don't mind walking 7-10 blocks to get to the train, but for older folks and disabled folks it's better to space things closer, and in really dense areas where you're likely to get very high ridership close station spacing lets you save some money per station because you can shrink mezzanines and related structures without having to worry about overcrowding. I think we're basically in agreement about 3 Av and Gun Hill, and don't worry about the junctions; you'll learn with time

I mean, I think it's worth pursuing fare integration, through-running, and related services in order to convert MNR/LIRR/NJT from commuter rail into high-frequency regional rail that serves as an express/super-express transport layer within the city. Unfortunately, making that happen would require NY, NJ, and CT to jointly agree to merge their services into a single giant RTA that would cover north and central Jersey rail and bus ops, PATH, NYC subway, NYC buses, NYC subway, MNR, and LIRR. Furthermore, if you kept a similar internal structure to the current MTA and NJT for the new agency you'd get a place where all three governors appoint their friends to the board, raid the cash box, get into jurisdictional food fights, and ram through ribbon cuttings for poorly conceived pet projects in election years and the trains still wouldn't be integrated or run on time. Making integrated regional transit work for NYC in the way you're describing would require a regional agency where nearly all the board members were elected and NYC has a clear majority in its own right, and getting NY, NJ, and CT politicos to cooperatively establish a structure through which a lot of money will flow without being subject to their influence is going to be an absolutely massive uphill battle. I mean, high-frequency regional rail is a thing, and would honestly be a thing worth pursuing (4tph per branch off-peak, 8tph per branch peak for LIRR, additional tph for Metro-North where possible, and through-running plus a new tunnel from Atlantic Terminal to Hoboken via Fulton St and West 4th St would be amazing to have. That said, making that possible would require a lot of wild and interesting political things to happen that I'm not sure we could swing.

About Phase III: I think it could go either way, and I'm also somewhat biased when it comes to station spacing; my personal belief is that for maximal integration with the bus network you want a stop on basically every major cross street where there's bus service, because otherwise you either wind up missing a lot of connections (which significantly reduce ridership) or you wind up distorting the bus network rather significantly to make connections possible. My thoughts here primarily come from living in the Boston area since 2018, where the subway station spacing is fairly far (14 stations over 11.5 miles of track on the Ashmont branch of the Red Line, or 20 stations over 11 miles of track on the Orange Line), and so the bus network is very hub-and-spoke around train stations. That sort of network works really well for getting people to and from the subway for bus-subway trips, but a lot of trips that would be one-seat rides on a well-designed grid-based network get turned into bus-bus trips (which don't work well in Boston because a lot of our bus frequencies are piss poor) or three-legged bus-subway-bus trips (which cost two fares) as a result. Making good connections on 3 Av in the Bronx means at a minimum you'd need stops at 138 St ( transfer to the ), 149 St (transfers to the ), 163 St (transfer to the Bx6, Bx13, and Bx21), 169 St (transfer to the Bx35), Claremont Parkway (transfer to the Bx11), Tremont Av (transfer to the Bx36), 180 St (transfer to the Bx40 and Bx42), Fordham Plaza (transfer to the Bx9, Bx12, Bx17, Bx22, Bx34, and Bx41), Bedford Park Blvd (transfer to the Bx25 and Bx26), and Norwood-205 St (transfer to the , but also the Bx10, Bx16, Bx28, Bx34, and Bx38). Adding those stops massively expands the catchment area of the new subway; at this point basically anyone on 3 Av can walk easily to one or two of the stations. Furthermore, doing it that way would make the new line the default choice for most people in the box bounded by Fordham Rd, Southern Blvd, Webster Av, and E 180 St, as well as everyone heading to the east side northwest of Westchester Av and Southern Blvd during rush when the skips all those stops. From there, I added 183 St because St. Barnabas Hospital is a decent-sized trip generator and employer that's right on the way, and then 156 St and 144 St because I'm using a guideline for local service of about 3-4 stops per mile in dense areas, and that inner bit of Mott Haven and lower Melrose is dense as f**k. By the time you add those stops in, you're probably looking at travel times on the being about on par with travel times on the . You'd still see some time savings because WPR has a few tight curves that slow things down, while being underground the new 3 Av line I'm proposing could have those curves smoothed out somewhat to allow for higher running speeds. Still, if you want to really pull folks off Jerome and WPR you need to offer meaningfully faster travel times (which is where express train service comes in). Express trains would basically only stop at a few key transfer points (Norwood, Fordham Plaza, East Tremont, 161-163, 149, and 138), and would offer all-day time savings of around 10-15 minutes over WPR and Jerome, which would be a huge draw (and would likely pull enough demand to justify the service). Phase IV: I don't have a problem with a Boston Rd/Amtrak ROW branch, and I suggested leaving bellmouths and provisions for a lower level to 161-163 Sts to accommodate such a branch. However, I'd argue that the Gun Hill Rd crosstown and Boston Rd/Amtrak ROW branch would serve two very different purposes, and it would make sense to do Gun Hill Rd first. Because Manhattan is so narrow compared to the Bronx, the subway network mostly enters the Bronx from the southwest corner and then fans out from there carving the Bronx into slices (ignoring Broadway). The western slice (ignoring Broadway and the ) is served by the , and , the central slice is served by the up to 180 St, and then the above that; the northeastern slice is served by the above 180 St, and the southeastern slice is served by the . By building the 3 Av line and extending the out to Bay Plaza you add a new central-west slice, and connect up all the slices on the north end, and so you're able to offer improved travel times to everyone across the northern Bronx fairly easily. By contrast, the Boston Rd/Amtrak ROW version misses the East 180 St connection to the during rush, so it's slightly worse on that front. On the other hand, that line would also take a load off the and focus more heavily on pulling pax off the lower WPR, so it might be worth adding later on and I definitely wouldn't foreclose it. DeKalb: That I don't know; I have a larger plan for pulling the out of DeKalb, but that's a longer-term thing that would come with the Jamaica Av rebuilding proposal I'm playing with. The idea would be to swap the and on Brighton, then instead of elevating the tracks north of Park Pl, send them into a tunnel under the existing ROW and under Franklin Av before connecting with the just southeast of Bedford/Nostrand Avs. From there, the and would run together as far as Hewes St station, then the would run under Harrison Av to Marcy Av where it would join a rebuilt underground Jamaica line and eventually rejoin 6 Av on the express tracks at Lower East Side-2 Av. There are pros and cons to doing that; the biggest one being the merges, and the need to run six tracks under the East River to accommodate the to avoid burning capacity for services to Jamaica and Middle Village. That said, doing that would leave the only merge at DeKalb being between the switching from 4 Av to Canal St lower level. That's still not great, but it's better than having the all rearranging themselves there, because it makes it far less likely that a single issue somewhere at the interlocking will propagate into a self-sustaining cycle of chaos and delays. It's probably also worth upgrading the interlocking to allow trains making diverging moves to operate at 20-25mph, and in general thinning out the timers around the interlocking.

I'd be quite interested to see what you're thinking; I have similar ideas of my own I'd love to share.

Exactly; I honestly think that the long term goal would be to pull enough ridership that express tracks are required, and the tracks above 96 St and in the Bronx should be built in a way that makes express tracks and extensions east to Bay Plaza easy to add later. Here's my proposal: Phase II would be four tracks to 116 St followed by a Columbus Circle-style junction that would enable express or local trains to turn to serve 125 St, followed by two tracks to St Nicholas Av (or even Broadway if desired, but St. Nicholas Av works). Ideally add track connections that would allow trains coming from 135 St/CPW to turn onto the 125 St crosstown line; that would mean than an issue on CPW could be routed around by sending D trains via 2 Av local, 63 St, and then the old switches for the to the 6 Av express tracks; if that's too much extra work that would be understandable but redundancy is nice. At that point, trains get extended to 9 Av and Bay Ridge, trains get truncated to Whitehall St and 9 Av, and the run together to 125 St/St. Nicholas Av. Add an east/west station shell under 125 St-2 Av, but don't worry about populating it right away. Phase III would be local tracks via 3 Av up to Norwood-205 St, with express trackways provided but not necessarily finished. I'd put stops at 138 St/3 Av (express), 144 St (local), 149 St/3 Av (express), 156 St (local), 161-163 Sts (express), 169 St (local) Claremont Parkway (local), 174 St (local), Tremont Av (express), 180 St (local), 183 St (local), Fordham Plaza (express), Bedford Park Blvd/Botanical Garden (local), and Norwood-205 St (express); I'd also add provisions for a junction and lower level at 161-163 Sts to allow for a future extension under Boston Rd and onto the Amtrak ROW. At this point, the would run local from Norwood to Coney Island, and the would stay on 125 St. I'd estimate that the in this configuration would be faster than the and local , and maybe about even with the in terms of travel time and travel speed, so you'd likely pull a decent number of passengers off both the and , but not quite enough to start crowding 2 Av yet. Phase IV would be an extension of the and to Bay Plaza via Gun Hill Rd, with stops at White Plains Rd/Gun Hill Rd , Bronxwood Av, Boston Rd, Seymour Av/Gun Hill Rd, Bartow/Edson Avs, and Bay Plaza. That would then start heavily relieving WPR and Jerome passengers (for trips into the eastern core the would be about on par timewise with the , and have similar frequency, while the would be much faster than the for trips into Midtown); this would likely unbalance the load a bit farther in favor of the and , especially for passengers coming from the far northern and northeastern Bronx. If it turns out that 2 Av is now starting to overload, you have a fair number of options. The simplest thing to do would be to finish out that station shell under 125 St-2 Av, and convert the 125 St segment into a shuttle. Run the both into the Bronx via 3 Av; now you have 30 IND tph on the 3 Av corridor, which is good for about 60k pax/hr comfortably, 75k pax/hr crushloaded. That also leaves you a fair amount of freedom for what to do with the bottom half of 2 Av; my personal dream would be to see four-track trunks under Northern Blvd ( relief) and Broadway/Jamaica Av (QBL express relief) followed by the addition of express tracks under the 72 St-96 St portion of 2 Av, and in the new express trackways you built through the Bronx back in Phase III, but that's just one of many possible options.

Honestly you're right; I think they really shot themselves in the foot there, and I think they also didn't really bother to think through what decongesting the Lex would actually require (namely that you offer a time-competitive alternative that runs from the area where Lex users live and where they work/shop/etc.) As it stands now, SAS is useful for people who live on the UES and want to go into the core of Midtown, and for people downtown and in Brooklyn who have jobs and doctor's appointments at the hospitals on the UES and that's about it; the current setup eliminates the forced transfer at Atlantic-Barclays/Canal/Union Sq for people who are coming from neighborhoods on the Brighton Line, and adds 34 St-Herald Sq and 63 St-Lexington Avenue as possible transfer points for folks coming from other parts of South Brooklyn; extending the up with it provides a one-seat ride to those areas from Sea Beach and would basically give all of Brooklyn south of Atlantic/Pacific and west of Utica Av a cross-platform transfer to those areas, which is great from a network access perspective. That said, a huge portion of the crowding on the Lex is Jerome and WPR folks trying to get down to the east side, and the existing plans don't do anything to address that. It's possible I'm overestimating how much crowding that is; I only ever lived in the Bronx when I lived in NYC, so I know from years of riding experience that the are absolutely slammed in the mornings coming from the Bronx while I don't have similar experience with trains coming from Brooklyn, and a 125 St crosstown does absolutely nothing for those folks, while a 3 Av trunk in the Bronx (especially one offering express service on 3 Av with an extension on Gun Hill Rd) would be well-positioned to capture folks who currently take the or from the Bronx (South of East Tremont Av, 3 Av is about equidistant between Southern Blvd and the Concourse, so folks south of there who would normally take the into the core would see similar-ish travel times on the , folks along that run who currently use the would see better travel times on the 2 Av express, folks north of Allerton Av who normally take the would either walk down to Gun Hill for the or transfer at Williamsbridge, and folks north of Allerton who currently take the would either walk to Gun Hill for the 2 Av/3 Av express or transfer at Gun Hill Rd/Seymour Av. Add in the fact that other than Hunter College a lot of the bigger employers on the East Side are hospitals/research institutions/etc. over near the water, and the 2 Av line becomes a no-brainer for folks in the Bronx who work there. The down side is that building out a system that would effectively use the 2 Av subway and additional trunks to relieve crowding and congestion across the system would be a project about the size of the core of the Second System as proposed, and if we wanted it built sometime before the 2090s a lot of things at the MTA need fixing on an organizational level.

The sad thing is that the MTA isn't even the worst when it comes to this sort of thing; the parts of WMATA's subway system that basically function as commuter rail in from the MD and VA suburbs got built long before the Green Line (even though the Green Line was actually within DC and served a bunch of black working-class and poor neighborhoods, which are the people the subway system was supposed to be for in the first place) because DC city appointees only make up a fourth of WMATA's board (VA, MD, and the feds make up the other 3/4), and the only way construction eventually got started ahead of yet another suburban extension was when the DC mayor threatened to veto WMATA's budget if something wasn't done. I think that building the things I'd like to see built (four-track trunks on 2 Av/3 Av in Manhattan, Northern Blvd in northern Queens, 3 Av in the Bronx, Broadway/Jamaica Av/Merrick Blvd in Brooklyn and southeastern Queens, with two-track extensions on Gun Hill Rd to Bay Plaza, Northern Blvd from Flushing to Bell Blvd, Boston Rd/Amtrak ROW from Bay Plaza to 169 St/3 Av, Junction Blvd from Northern Blvd to LaGuardia Airport, Merrick Blvd from Springfield Blvd to Hook Creek Blvd and Malcolm X Blvd/Utica Av from Broadway to Kings Plaza, plus a funky rearrangement of the that would replace the Franklin Shuttle and pull it out of DeKalb Junction) would constitute a 30- to 50-year plan if the MTA was actually serious about doing proper design and construction in-house, but I think that building that stuff out is worth what it would buy us.

And that's currently hard to do more often than not because machine politics is hard to root out, and in a two-party system where one party is captured by a machine and the other party is captured by arsonists it's really hard. You'd basically need politicians who could speak directly to voters without going through the traditional channels (which means being able to get and keep virality on decentralized media without turning into a demagogue or degenerating into a grab bag of disjointed hot takes), and then people who are able to hold supporters through the transition from calling out all the broken parts of the system to actively trying to fix those parts. It's difficult and messy to try to do that; I personally feel like the progressive wing of the Democratic Party is the best option we have for that right now, but they still have an uphill battle in front of them.

The problem with that right now is that internally the MTA is badly disorganized and different departments tend to feud back and forth a lot. Also, the MTA is controlled by a state as the whole, and the LIRR suburbs are populous enough to include a lot of legislative districts, historically fairly swingy, and at least somewhat provincial while the city itself is rock-solid Democratic, which means the governor is heavily incentivized to focus on the needs of the suburbs rather than the city (at least when they're focused on making the MTA do things at all instead of appointing their friends and raiding the cash box, which doesn't happen all that often). To be honest, if we want the MTA to be a place where things like handing an in-use LIRR ROW to the subway actually happens we would likely need to reorganize the MTA the way I outlined a few pages back. Right now, the only way I see the current MTA being willing to give up that ROW would be if we could fund quad-tracking of the ROW via St. Albans, construction of an all-stop Laurelton station on the St. Albans branch to use as a transit center with connections to the new , a flyover somewhere along that ROW to get the Far Rock/Long Beach tracks on the outside of the West Hempstead tracks, and a widening/rearrangement of the Main Line ROW between Hillside Yard and Jamaica Station to make room for those two new tracks. Personally I think it's worth costing that out alongside costing out 2- and 4-track subway extensions under Merrick Blvd out to Springfield Gardens and seeing what's most affordable and quickest to do; I'd argue that the subway extension might well be faster even if it's a bit more expensive to implement. This is also part of why I was arguing for a four-track rebuilding of the Jamaica line earlier; if you send the current to Springfield Gardens everyone's going to take the and crowding on QBL will get even worse, whereas a four-track Jamaica line would be able to provide QBL-comparable travel time to Midtown on the express service, and would thus pull a good chunk of passengers off QBL while simultaneously improving the lives of people in southeastern Queens significantly. That said, to make that viable you'd probably wind up doing a fair amount of reverse branching in order to offer fast one-seat rides to the relevant parts of Midtown; my preferred option would be to combine that with a four-track Second Avenue Subway, send one of the 2 Av expresses to Coney Island via West End, and then send the and the other 2 Av express to Springfield Gardens with stops at Marcy Av, Myrtle Av, Broadway Junction, 75 St, Woodhaven Blvd, Sutphin Blvd, Jamaica Center, and then local stops the rest of the way. Like ideally Metro-North and LIRR would be integrated with the city enough that we could use Metro-North and LIRR spare capacity to supplement for subways the way London uses the Overground network, but that would likely require a fairly heavy overhaul of the MTA in a way that runs counter to the incentives of most of the folks that choose MTA board members at the moment. I'm down to overhaul it in the way I talked about with the long-term goal of driving down construction costs and expanding total system capacity in a way that can comfortably accommodate the sort of long-term population growth NYC needs to have in order to mitigate the housing crisis; I just don't know if we can get that done.

Going forward I would agree with most of this, except I'm not happy about having the serving as crosstowns to 125 St instead of serving the Bronx; 125 St would be served just fine with 15tph rather than the full 30, and I'd rather see 30tph go to the Bronx and 15 to 125 St. IRL the problem with this is that the MTA really painted themselves into a corner by connecting the tracks up the way they did; short-term it made a lot of sense but long-term they should have just built four tracks on the same level using the Milan method (which is basically a variant of cut-and-cover where you deck over the road as soon as you get your retaining wall sunk, so the road is closed to traffic for significantly less time and so you get the price tag of regular cut and cover but with much less disruption), or if they were truly insistent on using tunnel-boring to do the work they should have gone with a 60' inner diameter tunnel; that would have allowed for two levels of two tubes each constructed CPW-style with stations inside the tunnel bore. To elaborate on that second method, with a 60' tunnel inner diameter, assuming each level has 15' total clearance from floor of tunnel to roof of tunnel, with 2' thick concrete slabs between the two tunnel layers you could comfortably fit two 12' wide trackways and a 30' wide platform on each level. For local stations the platforms would be on the west side of 2 Av, while express stations would have island platforms to enable cross-platform transfers. If they'd done that and then used conventional station box mezzanines (instead of a full-length mezzanine just have two 75' x 75' mezzanines spaced at 1/4 and 3/4 the length of the platform raised about 10' above the top of the tunnel bore) they probably could have saved a pile of money. Because the tunnel would be that wide it would also be possible to have tail tracks and turnaround ramps situated on either tunnel level between stations to allow for operational flexibility, short turns, etc. That structure would also have connected to the 63 St tracks reasonably easily. As far as the 2 Av line being isolated I'd argue it's not great but a bit more complicated than that. I would like to see the downtown portion run under 3 Av from about 67 St to 37 St or so; that would allow for an easy connection at 59 St-Lex Av via the Broadway local platforms as well as connections to the at 53 St and the at Grand Central; south of there you'd have connections to the at 14 St, and the at Lower East Side-2 Av. That would basically resolve a lot of issues around people only being able to use the 2 Av line for service on the far east side; a 2 Av line with a routing along 3 Av in the upper core and 3 Av in the Bronx (especially one with express service) would actually be able to fulfill its stated purpose of decongesting the Lex because it would share basically all the same major transfer points as the Lex from Houston St to 59 St.

Honestly yeah; my personal thought would be via SAS, Forest Hills-Whitehall St or 9 Av (would depend on how many trains can turn at 9 Av), and Astoria-Bay Ridge or 9 Av (15tph , 15tph , 10tph , 20tph ) if we wanted to do this without adding any new tracks beyond SAS. The split on the between Bay Ridge and 9 Av, with every third train going to 9 Av, would be if Bay Ridge couldn't handle 20tph on its own, and would also (alongside the conversion of 9 Av from solely a work yard to a storage yard) give the direct storage yard access. Longer term, the SAS project kind of boxed itself into a corner by building only two tracks down the main corridor and then connecting them to Broadway express service; that basically either makes a impossible with streamlined operations (or would require doing something weird where the lower half of 2 Av is served by a completely different set of trains than the upper half). Part of why I proposed quad-tracking a bunch of things and building out new trunks in the Bronx, Queens, and Brooklyn was to basically turn 2 Av into a mirror of 8 Av/Central Park West, where some trains ran the whole length of the corridor, others serve only the lower half before doing their own thing in Queens, and others serve the upper half before branching off to serve a different Manhattan trunk. It's not as clean or easy to operate as a setup without merges, but it should split the difference reasonably well between ease of operation, frequency, and minimization of transfers.

Fair, I mean the important thing is to make sure there's the money to increase bus service frequency and speed on corridors that parallel overcrowded subway lines; ideally even if we can't give Northern Bl a subway or QBL a bypass we should seriously consider frequency increases, bus lanes, and limited-stop service on routes that parallel overused subway corridors. My thought would be to add bus lanes on Queens Blvd, Northern Blvd, and the Queensboro Bridge(the two boulevards and the bridge are more than wide enough to take it), boost both routes up to 12bph during the day, and offer 6bph limited-stop service along both from East 60 St.

That's fair; I could see the subway becoming the thing you do for situations where you need to be somewhere significantly more quickly than the buses, but then the buses become the default mode of transport for most people. Like you could totally do a commute from East Midtown up to North Riverdale on local buses and it wouldn't actually be that bad (M101 from East Midtown to 68 St, M98 from 68 St to 179th and Broadway, then Bx7 from 179th and Broadway up to North Riverdale; hell, back when I was living with my parents on Sedgwick and Van Cortlandt sometimes I'd take the M98 up to 179th and Amsterdam, then take the Bx3 most of the rest of the way home.

Some, but probably not as much as in the worst case scenario. The Q66 is scheduled to take 57 minutes to an hour to cover the same distance as the covers in 23 minutes and the covers in 17, the Q60 takes 75 minutes to cover what the plus walking covers in 48, and the B25 takes 45 minutes to an hour to do what the does in 13-15 minutes. You'll lose some folks to the buses, but I don't think there's going to be a huge number of people who are going to be willing to incur that sort of time penalty on longer trips.