Home View from Underground A few thoughts on subway ridership numbers

A few thoughts on subway ridership numbers

by Benjamin Kabak

A Saturday evening, Manhattan-bound Q train had no empty seats to spare. (Photo: Benjamin Kabak)

While working on a somewhat related piece tonight, I started collecting a series of numbers regarding the last few decades’ worth of subway ridership that I wanted to share. These numbers tell a story of a city that’s growing and a transit agency that’s going to struggle to keep up. They tell the story of planners potentially caught off guard and economics and construction timelines that are impossible to sustain. The numbers leave many questions up in the air, and I’m not quite sure what the next few decades will bring.

Let’s start in September of 1996. Right before MetroCard discounts were announced, the average daily subway ridership was 3.684 million. Four years later, in September of 2000, daily subway ridership hit 4.745 million. Last October, average daily subway ridership reached 5.974 million. So in the span of 20 years, the MTA saw, on average, 2.3 million more entries per day or an increase of nearly 66 percent. That is, simply put, remarkable growth. On an annual basis, in 1992, overall ridership was below 1 billion; in 2015, that total topped 1.762 billion.

On the other hand, in the intervening twenty years, the MTA has opened a new station, and that new station has been open only since September. The agency is currently constructing three more with the first substantial addition to the subway map in a generation set to open within the next seven months (give or take a few), but this seems like a woefully inadequate response to a system that would have felt downright empty in the early 1990s as compared with our packed trains at most hours of the day.

This is, in a nutshell, the capacity crisis that has gained recent headlines. As I wrote last week, there are few immediate solutions and most transportation proposals seem to be bespoke ones driven by outside interests. The Brooklyn-Queens Streetcar, for instance, is going to do diddly-squat to help a Bronx commuter find a few square inches of space or a Q train rider at 7th Ave. fit into a Manhattan-bound subway at 8:30 a.m. Bike share solves some of the city’s last-mile problems, and despite my annoyance with the attention on ferries, they can help around the margins. But when a good year for the ferry system means 1.2 million riders over the course of 365 days (or 20 percent of today’s total subway ridership), we’re really comparing apples to oranges.

Today, we’re living with the consequences of both deferred maintenance and a lack of foresight. At some point in the 1930s, for a variety of historical and economic reasons, New York City simply stopped expanding its subway, and a few decades later, the city stopped investing in regular upkeep. Thus, when the state took over, it had a backlog of maintenance and no money for expansion. Today, the subway still needs money for maintenance, but the MTA can’t expand fast enough or cost-effectively enough to meet demand. (In 2007, when the Second Ave. Subway broke ground, annual ridership was 1.56 billion — over 200 million less than it is now.)

So what happens? I’ve been beating the drum for open gangways for a long time, and it’s a solution the MTA needs to explore and adopt as soon as possible. It’s also imperative to find a way to build faster and cheaper. Many options are simply fingers in the dike of a flood of riders, and without a commitment to a high-volume, cost-effective expansion effort, the subways are going to be this crowded for decades to come. And what happens if ridership growth continues on its upward trajectory? That may just be a question without an obvious answer.

You may also like


John-2 May 9, 2016 - 2:06 am

Really, the maintenance problems had their seed nearly 100 years ago, when the IRT and BMT were contractually locked into the 5 cent fare, and post-World War I inflation started to cause costs to rise, at the same time city politicians decided to make preservation of the nickle fare a major selling point for themselves to voters.

The fare would endure at 5 cents for another 30 years, by which time New York was seeing the same problem WMATA’s facing with their system in 2016 — i.e., once a system hits the 30-40 year old mark, all those once shiny and cutting-edge new technology starts reaching the end of its useful life, and will break down far more often if it’s not upgraded or completely replaced. It was the initial aging infrastructure that in part led the city to cannibalize the 1950 Second Avenue Subway bonds to keep the older BMT and IRT systems going, and by the time you got to the late 1960s, the main portions of the IND were starting to hit the 35-year mark and it’s 1930s equipment was reaching the end of its lifespan.

So New York’s system has been playing catch-up in one form or another for going on 70 years now, and the politicians overseeing the MTA only turned things around from the absolute nadir when the total collapse in the early 1980s couldn’t be ignored any longer. That’s still fresh enough in enough people’s minds so that enough is done to avoid a return to the 1970s and early 80s, but there’s still a bit of a mindset that hasn’t come to grips with the boom in subway ridership over the past two decades, and has not planned for a system with two-thirds more riders, in terms of added rolling stock and new stations.

SEAN May 9, 2016 - 7:46 pm

What you are describing can be compared in real estate terms as “the thirty-year cycle.” Basicly it means it is cheaper to replace a structure than maintain it.

tacony May 10, 2016 - 2:24 pm

In 30 years it’s cheaper to replace the structure than maintain it, but if you continue to maintain it for 30 more years, it becomes “historic” and worth double the value.

SEAN May 10, 2016 - 2:46 pm

In the older cities like NYC, that concept is well understood. That is not the case in places like Dallas, Houston, Phoenix or Indianapolis where the real estate growth was post war suburban style & the donut whole took effect.

Eric May 11, 2016 - 4:00 am

The donut hole is because “criminal” population moves into a network and middle-class whites think they have to move out. Not because of the buildings themselves.

Eric May 11, 2016 - 4:01 am

“neighborhood” not “network”

Raycee May 9, 2016 - 8:36 am

Open gangway question. Historically they were on articulated units (triplets & multies). Can Non-articulated trains also have open gangways?

John-2 May 9, 2016 - 10:55 am

They can, but then it comes down to the sharpness of the curves.

The MTA always shut the end doors on the 75-foot trains (and earlier, on the 67-foot BMT standards) because some of the curves, like the one between Cortlandt and City Hall on the R, cause too much separation between cars at that length. They did allow through passage on the 60-foot cars until recently, while the original multis the BMT had were only 37-foot sections, so the separation between the sections on curves, even on a route through the Montague tunnel, would be limited.

What they have to decide now with the R-211 test order is if today’s more stringent safety standards will allow open gangways on 60-foot cars on something like the S-curve on the R, or if the open gangways will end up with the same type of restrictions as the 75-foot cars have had, and will be banned from certain lines where the curves are deemed to be too much of a potential liability to allow riders to move between cars during those times.

Alon Levy May 9, 2016 - 2:10 pm

Yes and yes.

And in the first link, it’s on a line whose tightest curve has radius 40 meters, where New York’s tightest are 45.

DF May 9, 2016 - 8:48 am

Is the increase in overall ridership really the most relevant number in considering the extent of a capacity crisis? It seems to me that it would be more relevant to consider the increase in the number of people riding at peak times/places (or who would ride if not for problems caused by too much crowding). Increases in demand midday and on weekends do not really imply a need for new stations or lines (although of course better headways at those times would be nice).

While only a proxy for what we really want to know, according to the “Hub-bound travel” reports, from 1996-2014, the change in traffic entering the CBD was as follows.

8-9AM: 367k -> 372k
7-10AM: 799k -> 851K
24h: 1690k -> 2141K

Larry Littlefield May 9, 2016 - 8:52 am

In fairness, the growth over many of the years didn’t require more stations. Much of it was off peak, filling what had been empty space, a net benefit.

The crisis has only hit in the past few years, aside from lines such as the Lex where it was all along.

Moreover, the city did continue to expand the subway until the state took over. Consider the IND connection to the 60th Street tunnel, the Christie connection, the 6th Avenue express.

Deferred maintenance started in the Wagner Administration, which allow cut a deal for a state law preventing the city from requiring city workers to live within its borders. The political/union class then decamped to the suburbs and a few suburban areas of the city, and has been against public services at a fair price for the serfs ever since.

GregK May 10, 2016 - 10:40 am

The growth is off-peak because that’s the only time it can be. Peak’s already at capacity in many places. We have no idea what growth would look like at peak times if there was the capacity there to serve it.

paulb May 9, 2016 - 9:04 am

Like open gangways, because more capacity on subways we have. (CBTC important for this, too?) But we need new service where there isn’t any, don’t we? It seems to me elements of the IND second system are still worthwhile. Houston St out Metropolitan Ave and Union Tpke to far eastern Queens, maybe even into Nassau County. Plus there’s Utica Ave extension (or maybe it would be better to run that from a new Queens line?) And Triboro RX–can that really be done, I read somewhere that parts of the ROW are used to move a lot of freight. And Rockaway branch revival. Also 7 train extension at Flushing end. The Bronx has good coverage but some of the lines are so slow. Could those be sped up, some way? Seventy years of postponement and it makes a long list. The money’s impossible, I imagine.

Jeff May 9, 2016 - 1:34 pm

Money and time.

Brooklynite May 9, 2016 - 7:38 pm

Regarding an extension from Flushing: the line is already at capacity, which will only worsen as the city’s population increases. Extending a line that can’t fit its existing passengers is not a great idea.

Some of the others, especially Triboro RX, need to be built.

Eric May 10, 2016 - 3:42 am

Well, the first thing to be done is to increase frequency and equalize/integrate fares on the Port Washington line, to essentially make it into another subway parallel to the 7. This would have the same effect as extending the 7 to the east, and take some of the riders off the 7 at the same time.

After that is done, the 7 could be extended north to College Point or Beechhurst.

Panthers May 11, 2016 - 1:27 pm

That’s true. An extension of the Flushing line solves nothing. Running the 7 to Bayside and College Point and Whitestone at Ft. Totten will only make crowding worse. However, if you reactivate the Worlds Fair line and make it possible to change at Citifield for IND service, you may alleviate some of the crowding at 74th/Roosevelt. If somehow you can tie the unused Rockaway branch from LIC to the 7, creating a true, rapid express into LIC, then you will see business boom further in LIC. If you tie the LIC terminal into say, Courthouse Square, or the F or G or E, M, R, you have only redistributed people. People would surely leave the 7, hop on the express, and board a connection later down the line rather than sooner.

Supposing now you make the J or Z a true express past Myrtle. Take the Z.
Myrtle Ave
Marcy Ave
Essex St.
City Hall
Fulton St
Broad St

Bypass Bowery and Canal using the two sets of unused tracks that are walled off.
J Train:
Myrtle Avenue
Marcy Avenue
Essex St.
Canal St.
City Hall
Fulton St.
Broad St.

M remains a local.

Would that help in redistributing people especially when the L goes out?

Stephen Bauman May 9, 2016 - 9:11 am

Your analysis falls short because you are considering only the 24 hour or yearly turnstile counts. The subway ridership growth has not been uniform over the 24 hour period. The growth has occurred during the non-peak hours. Peak hour use is relatively unchanged from 1994. Peak hour use is lower than it was in 1985, when non-peak use started to increase.

Ignoring this trend leads to concentrating resources to solving a problem that doesn’t exist – increasing peak hour service capacity.

Tower18 May 9, 2016 - 10:36 am

Your contention is there is no capacity issue at peak periods?

Stephen Bauman May 9, 2016 - 11:10 am

That is correct for most lines. This specifically includes the Lex, which is the poster child for more UES service (aka SAS).

The figures for peak hour passenger loads for the local and express are:

1984 – 36,250(exp); 21,340(loc)
1994 – 29,797(exp); 20,249(loc)
2014 – 27,244(exp); 23,772(loc)

The 24 hour passenger loads are:

1984 – 126,400(exp); 96,960(loc)
1994 – 142,858(exp); 142,880(loc)
2014 – 190,614(exp); 168,749(loc)

Tower18 May 9, 2016 - 11:34 am

Interesting–how is this possible (peak hour ridership down 11% since 1984) for a line which is now and has been over capacity in both passenger and track capacity measures? Was there roughly-equivalent more *trains* in 1984? Surely it’s not because we were 11% thinner 🙂

Stephen Bauman May 9, 2016 - 12:57 pm

The 1970’s NYC financial crisis and the elimination of split shifts led to the reduction of peak hour service. Prior to that, peak hour service tried to squeeze as many trains as would fit. The 1971 Hub Bound Report shows they ran 547 peak hour rail cars on the Lex (local and expresss). That got reduced to 410 peak hour rail cars in 1984, 450 in 1994 and 490 in 2014.

tacony May 10, 2016 - 2:47 pm

The MTA ran 547 peak hour rail cars in 1971 yet can only manage 490 in 2014? Why are the media allowing the MTA to constantly lie and blame peak crowding on the constraints of the physical infrastructure?

The idea that we can’t run as many peak hour trains today as we could in 1971 (what a banner year for the NYC subway!) runs very much counter to the narrative most New Yorkers would believe and if that’s true this should be a scandal that the MTA should fess up to.

johndmuller May 10, 2016 - 10:31 pm

Maybe the trains were going faster and/or driven closer together and/or other operational features that may now be considered too dangerous or too litigation-provoking or has otherwise become against the work rules for better or for worse. Rumors have it that the Moscow operators run more trains, faster, etc than we do here, so perhaps it’s an operational issue.

tacony May 11, 2016 - 9:09 am

If we can’t run them as fast/close together as they could in 1971 for some safety/work rule/litigation reason, the MTA should fess up to that fact instead of continually lying.

Jeff May 9, 2016 - 1:37 pm

Very interesting. This basically throws conventional wisdom out the door.

Capacity isn’t the main issue, operational funding is.

Stephen Bauman May 9, 2016 - 2:16 pm

Follow the money.

Where capacity is an issue, it’s best to know what constrains it.

Eric May 10, 2016 - 3:44 am

And operational competence. When the Lexington can only run 25 trains per hour, but comparable lines in Europe can run 30 to 40, something is wrong.

Larry Littlefield May 9, 2016 - 10:39 am

So why has non-peak service not increased?

The answer can’t be physical capacity. It must be financial capacity.

Stephen Bauman May 9, 2016 - 11:16 am

I totally agree one reason non-peak service has not increased to meet demand is financial. The other reason is NYCT’s inability to track and react to changing traffic patterns in a timely manner.

AlexB May 9, 2016 - 9:12 am

Also more station entrances. Another bandage on the wound, sure, but it would help load-balancing, cut travel time for some riders, and is useful in emergency situations. The L has this problem severely, but many if not all lines have significantly uneven loads car by car.

I ride the 2/3 often on the weekends, when it is easier to see the load-balancing issues. I see trains with near-crush loads in cars 4-7 but ~10 open seats in cars 1-2 and 9-10. That’s convenient for me since I want to be first/last car at my usual stations but a sub-optimal use of train capacity.

Maybe the MTA could engage some engineering students to identify the most valuable new station entrances. Could be a great team project.

Rob May 9, 2016 - 9:12 am

“consequences of both deferred maintenance and a lack of foresight”. And the elephant in the room: huge, unbridled immigration, legal and illegal. Perhaps a reason why NY just chose Trump in the primary?

PS He was not my choice.

VLM May 9, 2016 - 9:27 am

This comment needs one giant [citation needed] if not a racism warning. You’ve made it many times in this forum before, and it is baseless. NYC’s population boom and 70% increase in subway ridership over 25 years isn’t being driven by illegal immigration. If you truly believe that, I’ll just assume you’re another ignorant, xenophobic racist.

Tower18 May 9, 2016 - 10:39 am

As the spouse of a legal immigrant and soon to be naturalized citizen who happens to be college educated, middle class, and white, I can assure you immigration is not easy or unbridled, no matter what self-interested professional provocateurs may like you to believe.

But wait, what does this have to do with the subways again?

smotri May 9, 2016 - 10:43 am

Can’t agree more with VLM on the illegal alien nonsense.
As a user of the Lexington Avenue line, I see a huge increase in use of that line, but also of other lines, after the evening rush hour and on the weekends. Many, many tourists use the subways, whereas 20, 30 years ago, they did not. Plus, even Manhattanites use the subway on ‘off hours’ to go to places like Williamsburg, Flushing, etc. There’s just a general increase in the use of subways overall, in my experience.

Rob May 10, 2016 - 8:59 am

Great. When you have no case, you play the race card.

Any questioning of immigration – and its impact on transit – is obviously a real sore spot with you, as I believe I mentioned it just once before, but that’s enough to get you in a lather.

Bolwerk May 10, 2016 - 10:23 am

Just think about this. Do you think any significant number of people in NYC are that neurotic about immigration? Even neoliberal law-‘n-order twats like Bloomberg don’t care. Hell, even Giuliani, practically a literal out-of-the-closet fascist nowadays, may not care (if he cares now, he certainly didn’t care when he was in office). Transit doesn’t care about the immigration status of its users either.

Think immigrants as a percentage of the population in NYC have been decreasing anyway. Crowding’s probably being driven be domestic in-migration.

Rob May 10, 2016 - 10:32 am

It’s little more effort to get the facts than shout ‘racist’, so I figured I’d present a few.

22.6% NYS foreign born
48% Queens foreign born
5.7 million NYC metro foreign born

So stop bleating that immigration has no effect on subway ridership & crowding. That’s what this discussion was abt, remember?

AG May 10, 2016 - 9:36 pm

The comments make no sense. If not for immigrants revitalizing neighborhoods – much of NYC would look like 1995. Just ask places like Baltimore and Detroit.
In any event – since it was named New Amsterdam – this place has thrived off of migrants. Your comment just doesn’t add up in the grand scheme. Subways are crowded because the powers that be didn’t take the right steps. Simple.

Eric May 10, 2016 - 4:00 am

Given that the parts of NY state where immigrants move to are also the parts where very few people vote Republican, I would guess this factor is pretty near irrelevant.

Avi May 9, 2016 - 9:23 am

How much of the growth has been at peak times? Off peak growth requires running more trains(increasing costs) and making maintenance schedules more difficult, but it can be solved. The real issue is growth during peak hours. What are the plans to address that? SAS will help a little on the UES, but won’t help beyond that. Do any of the east river crossings have capacity for more trains? If not CBTC needs to be deployed faster.

Nick Ober May 9, 2016 - 10:08 am

A few East River tunnels have spare capacity: the 63rd Street Tunnel (carrying the F), the Rutgers Street Tunnel (carrying the F), and the Montague Tunnel (carrying the R). The trouble is matching that capacity with space on the Manhattan trunk lines.

Currently, only the Montague Tunnel’s spare capacity could be taken advantage of. The MTA could either use it for an extension of the J/Z south to Brooklyn (essentially what we had prior to the 2010 cuts with rush hour Nassau Line M trains) or a southern extension of the upcoming return of the W train on the Broadway Line.

Before the M train was rerouted to the 6th Avenue Line, the express tracks of the Culver Line could have been used for an additional 6th Avenue service to Brooklyn (taking advantage of the Rutgers Tunnel), but to do that now would require robbing Myrtle Avenue Line riders of their one seat ride to Midtown.

Perhaps if these riders were assured of plenty 6th Avenue trains to transfer to at Delancey/Essex and a Bowery/Grand Street transfer to the B/D, you could pull off the switch. It would also help alleviate capacity issues on Queens Boulevard since you could run 10 car trains again. But the M’s rerouting has contributed to a lot of growth along the Broadway Line in Brooklyn, so it’s still a tricky proposition.

Tower18 May 9, 2016 - 10:41 am

The current M service is a net-gain in overall usefulness and likely in ridership as well, vs. using that space for a Culver Express.

John-2 May 9, 2016 - 11:00 am

The Culver express tracks won’t have a potential new service option at least until Phase III of the SAS, from 63rd to Houston Street, is built, and a connection could be made between that line and the Rutgers Tunnel. So it’s going to be a long, long time before Rutgers or the trackage south of Jay Street can be utilized to anywhere near full capacity.

Stephen Bauman May 9, 2016 - 10:57 am

Neither CBTC nor any signal system (collision avoidance) affects service level capacity (trains per hour) to any significant extent. Safety is assured by keeping trains further apart than their maximum emergency braking distance. All signal systems do is assure that this distance is maintained by all trains at all times.

The distance between a leader and follower on a single track with stations is a minimum when the follower is approaching a station that the leader has just left. This intermediate station service level capacity is determined by: the follower’s emergency braking distance; the leader’s dwell time within the station; the leader’s acceleration rate from the station; the follower’s service braking rate into the station and the signal system’s reaction time. The final item is usually an order of magnitude less than any of the other items.

The intermediate station service capacity, assuming a 30 second station dwell time, is a nominal 40 trains per hour, for rapid transit equipment that has been used for over a century. The Moscow Metro operates at 40+ tph using a standard block system. The 30 second station dwell time provides a 17% cushion for crowded stations.

NYC operates 40 tph on certain lines but not on a sustained basis. The reason is there is not enough rolling stock (and support equipment) for sustained operation. The last documented instance of NYC operating 40 tph on a sustained basis was in 1949 on the Third Ave El (42 tph). The El wasn’t equipped with CBTC.

AMH May 9, 2016 - 11:18 am

Signaling systems do affect capacity because of their affect on throughput. If they keep trains farther apart than they actually need to be (because of the nature of fixed blocks), they will constrain throughput. An even larger problem is the fact that components fail regularly and replacements are not available. Signal problems and maintenance (on which we spend a fortune for custom-made components) are probably a leading cause of delays and capacity problems right now.

Stephen Bauman May 9, 2016 - 12:04 pm

Any poorly designed signal system can adversely affect maximum service levels.

Maximum service levels depend on block length for block systems. Block lengths can vary in length, being shorter at station approaches. This is where more accurate train location is critical because trains are closer together. NYCT’s block system uses this approach.

Communications based systems also have location errors. These errors are due to processing and communications delays. There’s a maximum allowable delay that’s permitted before an emergency stop is executed. If one translates this maximum allowable communications delay into distance, one discovers it’s equivalent to the block length used at NYCT station approaches. Thus, CBTC is not inherently better at train location than the current block at the critical station approaches.

Maintenance of custom-made components is expensive. This applies to CBTC as well as the conventional hard wired relay logic. One example of a critical custom made CBTC component on the Canarsie Line is its digital communications system (DCS) radio. Its manufacture was discontinued before CBTC became on the Canarsie Line. The difficulty and expense in procuring additional Canarsie compatible DCS radios is one reason Canarsie Line service levels will be capped at 22 tph after the additional substations are built. BMT peak operations were 24 tph on the 14th St Line.

Brooklynite May 9, 2016 - 3:33 pm

Quick note on that last point:
In 1954 capacity on the Canarsie line was believed to be 32tph, although only 24tph was operated. Have a look at this interesting map:


Stephen Bauman May 9, 2016 - 9:30 pm

The NYCTA subsequently removed switches that permitted alternate trains to terminate at 6th Ave. The MTA removed switches and structure that permitted trains Brooklyn bound trains to terminate at Atlantic Ave.

These two improvements have limited capacity to the 24 tph that the BMT operated.

Brooklynite May 9, 2016 - 3:31 pm

THIS. A system that considers ~60% of international best practice service to be capacity levels should not be allowed to spend 10x international construction costs to build a relief line (SAS in this case). Most of the reasons for the low capacity we have today are long dwells, low accel/decel rates, and slow timer signals. None of this is impossible to fix. As the Germans say: “organization before electronics before concrete.” In New York we seem to have fixed the first part, given up on installing the second on any sort of timely basis, and chosen an obscenely painful and expensive way of doing the third.

Regarding 40tph, I believe we do not operate such levels anywhere on NYCT (even on a non-sustained basis), with the possible exception of the stationless segment of Rogers Junction where the 2, 3, and 5 share a track.

JEG May 9, 2016 - 10:27 am

There will have to be a mix of transit and non-transit solutions. For instance, up-sizing housing stock in the East and West 20s, 30s, 40s, and 50s. so that more people are living closer to the urban core. Distributing some office space to Brooklyn, Queens, and the Bronx so that certain commuters are not traveling into Manhattan to work.

Nick Ober May 9, 2016 - 10:57 am

6h Avenue in the teens and twenties seems especially underzoned for housing. There are 6 tracks of mass transit underneath a wide avenue and yet recent construction seems woefully short. Real missed opportunity. To be fair, there are some landmarks Ladies’ Mile department stores in the twenties, but zoning could still allow a Hearst Tower situation where a tall residential or mixed use tower could rise above the land marked base.

TimK May 9, 2016 - 11:55 am

I don’t know all the details of the geology, but it’s my understanding that the reason Lower Manhattan and Midtown Manhattan have skyscrapers is that bedrock is near the surface in those areas. The reason the areas in between don’t have them is that bedrock is much farther from the surface there.

Mid-rise buildings taller than the existing ones might be an option there, but really tall ones probably aren’t.

JEG May 9, 2016 - 3:26 pm

There more myth than fact to that belief. The New York Times did a story about the geology of Manhattan a few years ago, and the deepest areas of bedrock are actually in the City Hall area, which already accommodates some fairly tall towers. Zoning law, rather than geology is primarily responsible for the height of buildings across Manhattan, and in recent year, two quite tall towers have gone up south of East 23rd Street.

TimK May 9, 2016 - 3:45 pm

Thanks for the correction! Good to know.

Adirondacker12800 May 9, 2016 - 4:31 pm

If it’s 1925 and your choices are tearing down 4 and 5 story tenements where the bedrock is close to the surface or tearing down 4 and 5 story tenements where it’s going to be a lot more expensive to build a foundation, where are you going to build?

Snackbar21 May 9, 2016 - 3:28 pm

Haven’t we moved past that issue? If Saudi Arabia and Dubai can build mile-high skyscrapers in the sand, can’t we go a little higher now in these sections of Manhattan?

Tower18 May 10, 2016 - 5:44 pm

Chicago was built on a drained swamp, and had no problem building skyscrapers.

Adirondacker12800 May 10, 2016 - 7:05 pm

It’s not a technical problem. It’s just that it was cheaper to buy tenements in the north of 34th Street and build skyscrapers than it was to build them south of 34th.
…..Today there are 147 historic districts.

Brooklynite May 9, 2016 - 3:38 pm

Considering we’re currently building Hudson Yards, where the new structures have TRAIN TRACKS going through the foundations, I would wager that foundation costs aren’t an issue anymore. The issue with buildings where bedrock is far down is the cost of getting to said bedrock, not the possibility of doing so.

kevdflb May 9, 2016 - 2:00 pm

When I worked just of that section of 6th ave, the only buildings going up were 20-35 stories.

AG May 9, 2016 - 10:15 pm

Over the past 20 years – job growth has been faster in the outer boroughs than in Manhattan. They all have record numbers of jobs. There are just simply more jobs and more people everywhere in the 5 boroughs.

In any event – more of an argument for the Triboro Rx.

JJJJ May 9, 2016 - 3:06 pm

Cutting construction costs is obviously important, but that wont give you a new subway line in 2 years.

What is needed is political interest. That means leadership from someone who doesn’t live in their limo.

In 10 years, Mexico City built 78 miles of real BRT. It now carries close to 1 million riders a day. Line 5, 10km with 18 stations was built in 8 months.

NYC could do the same. Its easy, with all those 6 lane avenues. Well, easy if the leadership wants it.

Instead it takes 3 years to get a bus lane that operates 4 hours a day (in theory) and in reality is used for truck parking and NYPD nap time.

JT May 10, 2016 - 9:10 am

JJJJ nailed it.

Political will and good, fairly quick solutions like real BRT.

MordyK May 10, 2016 - 5:29 pm

I honestly don’t see why Cut-and-Cover in the outer boroughs is totally ruled out. If a street can be torn up for the replacement of underground utilities like sewers, it can be done for a subway line.

The trick is that they need to get the process done fast and use prefab tunnel and station components, so that they minimize disturbances and the streets are back to normal in under a month.

The time it takes to add the rail and signals and all the other rail and station infrastructure outfitting would be done underground, so there is no real difference between a TBM and cut and cover, although the time it takes current projects to get those done is obscene.

How did they build it faster when it was all being done by hand?

Dan May 9, 2016 - 6:02 pm

I’m surprised no one in the comments here has mentioned the reverse branching in the system (aka interlining). If the MTA rerouted so as to remove a few of the really bad choke points – 60th st tunnel and the tracks between DeKalb and the Manhattan bridge come to mind – higher frequencies could be run. And if overall system capacity is the big issue, a few less-than-ideal transfers throughout the system would be a small trade-off.

Dan May 9, 2016 - 6:12 pm

I would suggest:
– B/D local and express on Brighton, up 6th ave to Concourse in the Bronx
– N/Q going to West End and Sea Beach and both to the Upper East Side.
– R from Bay Ridge to the Queens Blvd local
– W to Astoria (alone but with high frequency)
– A/C terminating at Cortland in downtown Manhattan, from Inwood
– E/new service from Fulton in Brooklyn, local up 6th ave and through the 63rd st tunnel to the Queens Blvd express
– F through the Rutgers tunnel, switching to 8th ave local, then through 53rd st to Queens blvd local
– Constructing track connections to allow the G to loop via Rutgers, 63rd st and 8th ave local

Brooklynite May 9, 2016 - 6:58 pm

On one hand you mention several excellent ideas for de-interlining, but making the G into a loop route kills most of that. Also, the Fulton service won’t be able to run via 6th Av without interfering with either the WTC or Rutgers trains.

Brooklynite May 9, 2016 - 7:06 pm

There are plenty of issues, like dwell time, that significantly constrain service. Case in point is the 6 line, which runs ~25tph and has no junctions with anything other than its own express variant. The proliferation of junctions is certainly an issue that will crop up eventually, though.

My suggestions:
3 to Flatbush, 5 to Utica/New Lots: requires two new switches, one on each level, just west of Nostrand station.
C express 145-Canal, D local 145-59: can be implemented today.
Q via West End, D via Brighton: probably requires that the R tracks at Dekalb be dropped one level and a platform be built over them, so trains via the bypass can stop at Dekalb.

There are others, like untangling QBL, but they require more significant elimination of transfers. That would not go over well with the local community.

Dan May 9, 2016 - 7:53 pm

Ah, you are correct in your earlier comment regarding the Fulton service, I was mistaken. I do agree with you on the 3/5/C/D changes, and perhaps also the Brooklyn B/D/N/Q routing and the G loop. But if that G loop were replaced with the V, then:
– A/C continuing from Inwood to Fulton.
– F/V via Rutgers and 8th ave local, through the 53rd st tunnel and onto QBL express.
– E alone from Chambers and 6th Ave local, then sharing with the R on QBL local. (obviously imperfect because of the lesser local service on 6th Ave, but that line only skips 2 stops vs. the express).

Regarding Dekalb, I don’t think the forced transfer at Atlantic is the end of the world if it means the frequency would be higher on both trains a person is transferring to/from. Also, I’m not certain what the real issue is at QBL. Don’t the locals there go right to the yards, with the express continuing?

Brooklynite May 9, 2016 - 9:09 pm

De-interlining, while important, isn’t a holy grail. London is planning to operate 32tph on their sub-surface line core, even though they have exclusively grade-crossing junctions (like 142nd/Lenox here), and they’re getting up to 36tph on their deep tube lines. On the other hand Moscow runs 40tph, albeit without interlining. Basically, operational practices and parameters like acceleration rates are more significant for the service than interlining or lack thereof.

The transfer at Atlantic can take five full minutes, given crowding and the up-and-down nature of the passageway. Passengers are unlikely to accept that for a frequency increase, especially if the reduction in waiting time will be a minute at most.

I like your A/C/E/F/V proposal. It has a couple of flaws, though: sending all QBL expresses through Lex/53rd causes crowding there, as it did before the 63rd St tunnel opened in 2001. The local stops on 6th Avenue might have some issues with the ~8tph that the E would end up running under your proposal, which would also underuse the 63rd St tunnel. More service on the E would be empty, because it wouldn’t offer QBL riders a proper Lex transfer, and would overload the 71st Av terminal, which has very low capacity because of NYCT’s insistence on checking that every train is empty before it relays.

I suppose the ultimate issue with high frequencies on NYCT is the dwell time, which is why they should be avoided in the CBD where possible. (See: QBL manages 30tph while Lex doesn’t.) That would imply that there have to be at least two services on the express, which entails interlining…

Dan May 9, 2016 - 10:24 pm

I don’t see those operational practices ever changing, the politics of being overly cautious are too entrenched in the city/country.

And yes, the transfer at Atlantic is rough, but you’re already losing 5 minutes to transfer because each of those trains only run at best 7-8 tph, so while annoying, it doesn’t cost any real time. Some people would get annoyed, vastly more people would get better service. De-interlining doesn’t solve everything (and in places like QBL interlining is probably a necessary), and I’m sure there are places in this proposal that could be improved upon, but if it any of it would mean eking out even an extra train per hour per service, that would be a huge capacity increase with almost no capital investment.

Dwell time is a very good point, and probably that the trunk lines wouldn’t be able to get 30+ tph. But the lexington line still gets a lot better than the joke we have now with most of the B division trunks running barely more than 15 tph. And the MTA claiming they’re running at capacity…

Chris May 9, 2016 - 10:05 pm

I doubt we’ll see any relief in the next 5-10 years. First, the Second Avenue subway doesn’t travel far enough downtown for it’s 2017 (scheduled) opening to help much. (I do not consider a switch over to the Broadway line an East Side downtown connection.) We need to extend the line up to 125th street and at the same time, build the line down to Hanover Square (or, as I prefer) to South Ferry/Whitehall Street (If Possible). Of course, this will require more money and manpower than NYC will have available over the next 20 years.

If we were to accept short term out of the box thinking, why not build Els on the East Side using modern technology, along with a new bridge linking them to lines (and yards) in the Bronx. We certainly wouldn’t need as much metal as in the old structures, and we could preserve more light on the ground if done right. No tunneling, and we’d get the lines built quickly. Is this a serious proposal? Not really – we’d never be able to sell the idea. But it would be much cheaper than tunnels, and we’d relieve the transit crunch in less than another 50 years…..

JT May 10, 2016 - 9:09 am

Els destroy neighborhoods.

Will May 10, 2016 - 12:07 pm

Els dont destroy communities. Please give me sources or your lying out your butt

Chas S May 10, 2016 - 1:49 pm

Long Island City and Astoria are pretty nice.

Will May 10, 2016 - 7:14 pm

Chicago is thriving with its Ls. Thailand just built couple of else. We should tear down the corona, Broadway, white planes Rd el

AG May 10, 2016 - 9:47 pm

I’m a little confused by the comment… When did Chicago build new El’s??? There’s is a legacy system – like ours… Well except we got rid of a whole lot.

Eric May 11, 2016 - 4:09 am

Newly build els can be extremely quiet, if they are made of concrete and constructed correctly. Only the old steel ones like in NYC are loud.

For example in Vienna: “the viaduct is flanked by sound-absorbing walls which cover at least half the train. For neighbours along the line, this is, of course, good as you can hardly hear the trains roll by”

will May 11, 2016 - 7:58 am

LA just built some el section for its expo line and Honolulu is building thier entire mass system as a el as well

AG May 11, 2016 - 10:06 pm

I’m not opposed to El’s completely – but Honolulu and LA are both light rail and those lines are in low-rise areas.
LA’s is not even grade separated in some portions (and has to stop for car traffic). Honolulu – well they are as bad as being overbudget and late as we are.
In the end though – NYC probably needs more subways – not El’s.. There just needs to be reining in of costs.

Will May 11, 2016 - 11:39 pm

She you have the money to build subways for cheap the you vs build subways but with budget issues and the cost of subways tremendously high, you go with the cheaper option

Jeff May 11, 2016 - 10:27 am

Many modern els are also built along highways as well, so the impact from additional disruption and “blight” to the neighborhood isn’t that much.

NYC has the JFK Airtrain as a model for how it can potentially be done.

Miles Bader May 14, 2016 - 9:36 am

Building along a highway makes rail transport much less useful, because land around a highway is typically unsuitable for dense development, as being near a highway is undesirable.

Rail transport needs to go where the people are (or can be), not where it’s easy to build.

AG May 14, 2016 - 11:48 am

I hear you… Which is why I was surprised when I went to Chicago and saw rail lines and stations over “highways”. If I’m not mistaken the DC Metro does that in Virginia from time I spent out there. Certainly not ideal.

Panthers May 11, 2016 - 1:26 pm

I just came back from Bangkok. Downtown is thriving. Condos are being built all around BTS lines, especially Sukhumvit, which bisects all the Sois. Nana Plaza and Askoe especially are growing by leaps and bounds. The trains above had zero affect on the markets underneath them. In addition, there are tons of bars and shops on both sides of the Sukhumvit. It has boosted business.

Will May 11, 2016 - 11:42 pm

This allergic ideal of not building Els is ridiculous. It’s cheap and have the same capacity of a subway line

Eric May 12, 2016 - 5:09 am

To be fair, Thailand is not a first world country. Bangkok is an impoverished (by our standards) megacity that needs transportation capacity much more than beauty and comfort. Unlike Bangkok, Manhattan is a tourist destination in part because of how it looks (Midtown skyscrapers, SoHo cast iron midrises) and els would interfere with that. Also, it is difficult to build a transfer from an el to existing subway lines. In general, the more dense an area, the more people will suffer from the el’s shadows and noise, so the more desirable a subway is.

Bottom line, els are great for the outer boroughs, but not for Manhattan.

Will May 12, 2016 - 3:14 pm

Els could be made on the cheap and London has more density then us plus the docklands light rail which is an el, so please don’t come with this crap that els can’t be built here in Manhattan. We had freaking Els until real estate agents were declaring that els were bad for land values but city planers were not looking at the transportation value that els contribute to the city

Adirondacker12800 May 12, 2016 - 3:28 pm

Second and Third Avenue had low rise tenements along them. Like Fourth Ave did until the trains were put underground and the real estate developers had it renamed Park Avenue.

will May 12, 2016 - 4:18 pm

What does “blight” have to do with transportation need of the region. If you don’t have money to build subways then what not build other forms of transportation.

Adirondacker12800 May 12, 2016 - 5:24 pm

The people you are building the transportation for move away?

Miles Bader May 14, 2016 - 9:40 am

Els destroy neighborhoods.

No they don’t. Certain types of elevated lines are worse than others, but it’s perfectly possible to do it well, so that the impact isn’t huge, and is far outweighed by the positive impact of having the line.

JT May 10, 2016 - 9:08 am

I wish service was more frequent late nights and early , so there would be some incentive for commuters who can control their schedule to move their commutes to those times.

As it is now, if you get into a subway station at say, 6am, you often have to wait a long long time. And might not even get a seat. If more people could get fast service and seats at those time, it’d help a little.

I realize this won’t fix the core problems, but it’s something that could be done with “only” more money for staffing trains.

Tower18 May 12, 2016 - 11:41 am

Forget late nights, I wish I didn’t have to wait 10-12 minutes for a C train at 9:15am on a weekday. Ridiculous.

LLQBTT May 10, 2016 - 9:38 am

It seems that disinvestment in the subway is a generational thing because the ridership and safety situation is the exact opposite of the late ’70s, early ’80s.

Not sure how long the current circumstance will remain politically tenable.

Frank May 10, 2016 - 3:53 pm

A new system built by a new operator is needed, as Johannesburg did with the Gautrain because its Metrorail was so horrible. It’s two-tiered, but that’s par for the course in a highly unequal city.

Eric May 11, 2016 - 4:10 am

You mean like the IND? 🙂

Frank May 11, 2016 - 10:05 am

Like the IND in the sense that it would be a separate system, but not like the IND in that it would need to be placed on a firm financial footing from day one.

Bring in Hong Kong’s MTR corporation to build and operate it. They know how to turn a profit from transit properties.

Jeff May 11, 2016 - 10:24 am

The MTR is a real estate developer in disguise, one that operates in a city that previously still had plenty of undeveloped land and some of the most expensive real estate prices in the world. That model isn’t going to work in NY

Japan’s private operators might be a better model though.

Sam May 13, 2016 - 3:14 pm

New York City actually has the perfect conditions for an MTR approach, save the political will.

A large portion of the city is relatively underdeveloped, (think Staten Island and Canarsie) but we lack the political will to zone for greater density and the stomach for a public agency to operate as a developer.

Jim D May 12, 2016 - 12:13 pm

This may be a crazy question, but has any other subway system rebuilt coupled cars into open-gangway units? Even if the R211 test is successful, the large R142 and R160 fleets will be around for decades still.

will May 12, 2016 - 4:20 pm


wiseinfrastructure May 26, 2016 - 12:12 am

yes modern els are quiet and can look pretty (incorporate planters and you have “flower avenue”)

build a one way elevated down 1s Ave and uptown on 2nd Ave
-minimizes the visual impact

-use pop(proof of payment) instead of fare collection to reduce station sizes/impact
-have construction include a bike/walkway (the new east side highline) which will double was a wide emergency walk way and appease the locals.

include direct build/store front access and you have created a realeast mine.


Leave a Comment