Line Spacing, and Intersections Without Transfers (Hoisted from Comments)
This post is inspired by two separate things. The first is my work on a fantasy subway map for Lagos; here is the current live version. There are twelve radial lines, all serving the western half of Lagos Island, converging on nine transfer stations. Under the principle that whenever two lines intersect there should be a transfer station, this greatly constrains the paths the lines can take. Result: the path between two CBD stations, Eko Bridge and Leventis, carries ten tracks underneath it. This is under a wide street, and it might be possible with a double-deck four-track-wide tunnel and two more tracks deep-boring around it, but it’s not easy to construct.
The second inspiration is a post by Brian Stokle about subway line spacing. Brian looks not at spacing between successive stops on one line, but at spacing between parallel lines, averaging a few North American examples. The average is a little higher than half a kilometer, narrower than the typical stop spacing. On Twitter, Joshua Mello notes Boston’s spacing was narrower; in comments, I add examples from New York and Paris, which are a bit narrower than Brian’s examples but wider than Boston (New York is one block in Midtown, so 280 meters, and Paris is 300 between Metro 3 and Metro 8 and 9).
These two examples together illustrate the tradeoff in subway construction. Most subways have a stop every 1-1.5 km; newer systems are at the high end of this range, mostly because of the demographic weight of China. It’s normal for stop spacing to tighten in the core, but not to a large extent. In Tokyo, the average stop spacing is 1.2 km, and in Central Tokyo it’s perhaps 800 meters. In London, the Tube lines seem to tighten from an average of 1.2-1.5 km to 600-800 meters in Central London.
At the same time, subway line spacing is necessarily short. The reason is that modern CBDs are geographically small. Midtown is maybe 4 km^2, from 30th to 60th Streets and from between 2nd and 3rd Avenues to between 8th and 9th. The Paris CBD, from just west of Les Halles to just east of Etoile, is also about 4 km^2 (see job density on PDF-p. 6 here). The Tokyo CBD, defined around Otemachi, Nihonbashi, Hibiya, Shimbashi, and increasingly Roppongi, is maybe 5-6 km^2, in a metro area of 38 million people.
Subway networks in such CBDs are necessarily crowded. The CBD is where people want to go. A subway line can get away with skirting it – Paris M4 does, and is in a near-tie with M1 for highest ridership per km. But avoiding it entirely is a ridership killer, except specifically for circumferential lines concentrating off-CBD travel: in Paris this suppresses M10 ridership, and in New York, it suppresses ridership on the J/Z (even though they serve Lower Manhattan) and the L (even though it serves Union Square). This means that the CBD of a large city will have many subway lines converging on a relatively small area. New York has its five north-south lines through Midtown.
Ensuring that every pair of intersecting subway lines has a transfer in this environment is difficult. Line spacing is usually narrower than station spacing, requiring kludges like the block-long walkways in New York, such as between Times Square at 42nd/7th and Port Authority at 42nd/8th. Paris managed to have an almost perfect network – before M14 was built, it only had one missed connection, between M9 and M12 (built by a competing private company) – but only by having very short station spacing, unusual even by the standards of the early 1900s, ruling out significant suburban extensions of the kind that are routine in London and Tokyo.
The situation in smaller cities is actually easier. The CBD is very small, often smaller than a square kilometer, but there are fewer lines, so it’s easier to make sure lines intersect properly. It’s also much easier to get line spacing right outside the CBD, where there’s less intense demand, allowing line spacing compatible with stop spacing on any intersecting or circumferential line.
The fundamental issue here is really about planning for the future. It’s not hard to gets lines 1, 2, and 3 to intersect nicely, or even lines 1-6. But beyond that, a city will often find itself in a situation where the best street alignment for line 7 happens to be right between two stations on line 1, spaced too far apart for a transfer. This is what happened to Tokyo. In New York, the three constituent systems (IRT, BMT, and IND) were each internally planned cohesively, so when two lines within the same system intersect, there’s a transfer, and, with difficulty, the IRT/BMT intersections have transfers as well. But the IND connects poorly to the other two systems, sometimes deliberately, and the IND’s layout made future extensions and service changes break transfers. My proposal to reduce reverse-branching in New York runs into the problem of breaking the transfers designed by the IND around a specific service plan.
When lines are designed together, it’s easier to avoid this problem. Paris M8 and M9 share a route through the center, as they were built simultaneously as the street is wide enough for four tracks. In contrast, building a line under or next to an existing line is much more difficult; New York did it anyway, under Sixth Avenue, but this led to cost overruns that doomed the IND’s early plans for further expansion. It is also difficult to build a new station under an existing transfer station, as it usually requires underpinning; in Paris, this problem means that transfer stations tend not to have closely-aligned platforms, requiring long walks between lines. When I’m proposing running multiple lines in the same tunnel in Lagos, this is from the point of view of assuming coordinated planning, with sequencing that allows entire streets to be dug up at once.
However, in reality, even coordinated design has its limitations. Subway networks take multiple decades to build, and in the interim, the city changes. Planners can attempt to use zoning to shape city development in a way that facilitates further expansion, but some tendencies are too uncontrollable. For instance, high-income neighborhoods tend to commercialize; I mentioned Roppongi as a growing part of the Tokyo CBD earlier, which is an example of this trend. The hottest new part of New York commercial development, the Meatpacking District, is really not a subway hub. This means that even if a city plans out lines 1-12 to share tunnels appropriately, it may not be able to control where there will arise the most demand for line 13. Coordinated long-term planning makes things easier, but it will not solve the basic problem of optimal subway spacing and CBD size.
Pedestrian Observations 
There are a few tricks that can help with this: one is planning tranfer point to have three lines crossing, which concentrates more transfers onto fewer stations and lets you route a new line to be the third line at a series of transfers, as was done for Moscow’s Line 10. This can also be accomplished by having parallel lines converge to the opposite ends of the crossing line’s platforms. But even then, sooner or later the complexity of the network catches up with you and you end up with missed connections. Ideally though, you’d still only require two transfers between any two stations in the network.
That works a couple times, but then runs out. Imagine four transfer stations on a square. You can build a diagonal, but unless there’s a stop in the middle, and quite often there won’t be, that blocks the other diagonal.
Six lines around and across a square, that is conveniently as wide as the subway platforms are long.
Not that anyplace would ever need that much subway. It would be four levels deep. Five with the huge mezzanines at each corner. … there’s gonna be long ranks of turnstiles and umpteen stairs/escalators to handle the pedestrians flow isn’t there?
Make the square half as wide as the subway platforms are long and there can be even more lines with more levels. Not that anyplace would need that.
No, I mean a square that’s as wide as the normal CBD subway station spacing. There’s a near-example in New York: Grand Central, Bryant Park, Rockefeller Center, 51st/Lex. There are indeed a fuckton of turnstiles and stairs to handle passenger circulation.
If the square is blocks and blocks wide there is plenty of length to dip the tunnels up down and sideways. Threading the IND from the Upper West Side out to Brooklyn for example. The Flushing line is under 41st St. … South of Chambers Street or downtown Brooklyn…
Well, Chatelet-les-Halles has 8 lines (M1, M4, M7, M11, M14, RER-A, -B, -D; nine if you consider M4 has two stations at both ends!) and yes, a real Escher plan of escalators, stairs and travelators. Which brings me to the issue of interconnection (correspondance) that, unless I am missing something, is the main topic? Obviously in the Paris system there are some long correspondances such as the notorious double-enfillade-travellators of Chatelet, that in fact I don’t mind; likewise at Montparnasse (linking 4 Metros to each other and the train station) and the long walk at Concorde (and probably others–the ones connecting RER-B and Metros at St Michel). It surely is a valid and effective means of connecting parallel lines. For example, in the map of Toronto in the article by Brian Stokle there is no indication of the city’s underground arcades that link a lot of the city; surely they also link the Metro lines?
There are also some really long walks in Barcelona. There’s quite a nasty block-long walk when transfering between the train and the metro at Passeig de Gracia. I actually prefer transfering twice than making that long boring walk through a narrow and crowded pedestrian tunnel.
Beijing will soon have a very regular subway grid across most of its metro area.
What should it build after that? (assuming the existing subways are crowded, and funds are available)
I think part of the answer is to build lines at a 45-degrees diagonal to the current grid, so that each stop will be a 3-line transfer point.
Current plans seem to be to build “express lines” which skip some of the interchanges, which has the potential of severely overloading some lines while underserving others.
Yes, exactly: in a situation like Beijing’s – a highly gridded city with subway lines that mostly follow the four compass points – diagonal lines are extremely beneficial. Even in Paris, this is the case. If you squint your eyes, you see that M2 and M6 form a ring; M10, M1, M3, M8, and M9 go east-west; M5, M7, M4, M12, and M13 go north-south; M11 goes center-northeast; and M14 goes southeast-northwest. Ridership per route-km goes M1 = M4 > M14 > M2 > M6 = M11 > rest.
The caution is that building subway lines that miss the city’s grid is more expensive, because they have to be bored. But I think that Beijing already bores the subway, and at any rate you can’t really cut-and-cover a line that has to go under multiple other CBD lines.
“It’s normal for stop spacing to tighten in the core, but not to a large extent.”
Why should this be the case?
If stops are very close in the CBD, you gain twice (less crowding=less dwell times at CBD stations, and easy connections when future lines are built) while losing once (slower trips through the CBD). Given that relatively few people are traveling all the way through the CBD, isn’t this a net gain, at least for “non-express” lines?
You really gain 1.5 times, not twice. Spreading demand between multiple stations can help reduce dwells, but only up to a point. Evidently, BART has long dwells at Embarcadero and Montgomery Street, which are ~600 meters apart.
Conversely, the more stops you add, the more you lose due to long travel time. When you interpolate a 1.2 km interstation, you get one stop penalty, say 50 seconds. When you interpolate the two resulting 600-meter interstations, you get two stop penalties, in this case 1:40. Now repeat that for a longer CBD stretch. This also raises construction costs – CBD stations are likely to be under older CBD stations, which means they have to be mined, and then additional stations have nontrivial marginal costs.
Does this change at all if you’re using large diameter TBMs (thus reducing the cost of each station). You still have to drill access shafts of course but the incremental cost of each station is less.
It does, but then you need to pay large-diameter TBM costs, which are really high (compare Barcelona L9/10 with anything else Spain builds).
Or one takes the Chicago solution of building one long station.
Re Lagos, are you just “drawing lines on a map”, or are there activity centers you are consciously trying to hit besides the CBD?
There are some activity centers I’m trying to hit. The big one is the edge city the city is building, Eko Atlantic – the southeastern end of the Ota Line (the yellow one) is in the open sea, in an area that’s being reclaimed (the sandy part just west of it isn’t a beach, it’s the area just reclaimed in the last ~3 years). I also made an effort to hit certain neighborhood centers with transfers, like Ikotun, the airport, LASU, and the University of Lagos. Beyond that, the size and density of Lagos meant I was mostly aiming to have regular line spacing, putting most of the built-up area within 1 km of a subway station. At a few places I deviated to serve neighborhood centers, like Shomolu, but for the most part I was aiming at regular line spacing.
If making a connection between two lines is a lot more expensive, is it worth it to make it anyway? I guess it depends on the situation.
I have a particularly frustrating missed connection in mind : in Brussels, the circle line around the center (lines 2/6) tragically misses the North station, by about 500 m or so. It’s really a shame, for two reasons :
– People who live in the city proper have a reduced access to the large business district around the North station*. French Wikipedia claims it’s the 2nd largest business district in the city, after the European Quarter.
– Trips from the northern suburbs to some secondary job clusters in central parts of the city (mostly Madou in the northeast and Ribaucourt in the northwest) are more annoying and time-consuming than they should be, requiring two transfers instead of one
Granted, the North station alignment would have been more expensive to build than the Rogier alignment which was used instead : the Rogier alignment allows you to keep building under a superwide boulevard (I believe the construction was just cut-and-cover) while the North station alignment would have required deep boring. But I think policymakers didn’t even consider the possibility of the circle line hitting the North station, they probably thought it was good enough to have a radial subway connecting to it. Anyway, I looked for contemporary studies and discussions of the 1960s/1970s plans for the Brussels metro, but I couldn’t find much.
*unfortunately it’s all 1960s urban renewal hell…
This is also the situation here in Paris! M2 frustratingly just misses Gares Saint-Lazare, du Nord, and de l’Est – it was built under and over the northern half of the street delineating the pre-1860 boundary of the city, and at the time construction had to follow streets. It hits all the radial Metro lines (as does M6, which does hit Montparnasse), but then its connections to the RER aren’t great. In the east and west the RER A connects to it, but in the north, the RER B, D, and E don’t – it’s too close to Gares du Nord and de l’Est. In the south, the RER B does connect to M6, but the RER C misses it in the southeast, since it passes just between Gare d’Austerlitz and BNF.
Clearly, M2 here still works. There isn’t any east-west service to Gare du Nord, but that’s okay, since there are two north-south Metro lines plus the RER. It’s also not a huge destination by itself. Saint-Lazare is just outside the CBD and is a huge destination, but is served from multiple directions. The M2/M6 ring is at the right radius for circumferential service, so they get good ridership. I think in Brussels the situation is different, in that M2 is too small (it’s 10 km, M2/6 are 26 km combined), and there are fewer radii, so there’s less need for that circumferential service. Other cities with that network length tend to do 3-4 radials and no circumferentials at all, like Lyon.
Yes, I agree that the radius of the Brussels 2/6 line is too small. If I were to build it from scratch today, I’d make it go 1 km east so that it connects to Schuman, at the heart of the European Quarter. And of course I’d make it hit the damn North station. The line diameter would be a little over 4 km, maybe on the small side, but it would be going through densely populated neighborhoods and connecting together 3 of the top 4 Brussels destinations (South station, Schuman, North station).
The M2 in Paris has a perfectly functional underground passageway from the metro station La Chappelle to the RER station at Gare du Nord. It’s a fairly short walk to the northern end of the B/D platforms, a bit further to the RER E platforms at Magenta, and a bit further still, but still manageable, to the mainline platforms at the southern end of Gare du Nord.
Plenty of Parisians take the M2 to Gare du Nord every day.
It’s several hundred meters away. The M2/M4 transfer is one stop north on M4 than Gare du Nord.
When I was creating a fantasy map for Seattle, I never got that in the CDB. While my biggest junction station, Stewart (Roughly on the site of OTL Westlake) had 10 tracks, it was One East-West line (My Line 1, The Flaunteroy Line), over Four North-South lines (Line 2 (Overlake Line), Line 5 (Aurora Line), Line 6 (Interurban Line), and Line 17 (Valley Line). In this version I was going crazy with the crayon, but was allergic to branching. Still, I wound up with 9 routes through the core (Plus 2 Mainline rail alignments). In later versions I tried to be more realistic with branching (While still building a substantial network), but that only reduced Stewart to 8 tracks.
And my fantasy network is a very different animal from Seattles current expansion plans, as they were more an attempt to build an ideal network regardless of what was already in the ground.
Was your Seattle map fantasy (i.e. for the future) or retro-fantasy (i.e. what if it had built something different from its current network)?
It’s Retro Fantasy, but a Retro Fantasy that was grounded in a knowledge of past plans for Rapid Transit, along with some level of how the engineering would work. I’ve just scanned the maps that I made a bit after my craziest era, and from the ones I was talking about last night. They’re abstract looking, and my writing is barely legible, so I’m going to have to explain some things. I’ll go line by line, you might want to have to the map for King county Metro and Sound Transit up, because I’m going to be talking about them in the context of KCM and ST routes.
Line 1 (Flaunteroy Line, Red) In the 2007 map, It basically follows KCM routes 2 and 11 East of Downtown, and 56, C, 120 South of downtown, crossing through downtown itself on Olive way and 1st Ave.
Line 2 (Overlake Line, Blue) In the 2007 map, it basically following KCM routes D, 13, 40 and 75 north of downtown, and ST 550, and KCM B East of Downtown. In other analogies, the Seattle-Ballard segment is a modified version of Option D from the Ballard Link studies, the Seattle-Redmond segment is a modified version of East Link, but extended further into Redmond. In Downtown it follows Atlantic Ave (I-90 doesn’t exist here), before curving north to 3rd Ave.
Line 3 (Rainer Line, Green) In the 2007 map, this line starts in the Magnolia neighborhood, curves south to Mercer St, turns onto Fairview, and then Boren, and then Rainier, before curving west to Hwy 99 all the way to Tacoma. The portion of this line south of the core resembles Central and South Link, and like them, serves the Airport.
Line 4 (Duvall Lin, Yellow) In the 2007 map, this line starts out at Colman Dock, before swerving onto James st, and then Broadway to serve First and Capitol Hill. It then swings west to hit Eastlake, before swinging east through the University District (And UW proper) to Childrens hospital. From there it swings north under 35th Ave NE, eventually shifting onto Lake City Way/Hwy 522, going through Lake Forest Park, Kenmore and Bothell before winding up in Woodinville. AFAIK, I haven’t seen anyone come up with anything remotely like this.
Line 5 (Aurora Line, Violet) North of Downtown, this line is simple: Between Everett and N 65th in Seattle, it basically follows Hwy 99 to the letter. South of there it drifts off, but still shadows it closely. In downtown, it uses 5th Ave, Spring St, and Union St as it’s route through the core, terminating at Union 34th.
Line 6 (Interurban Line, Orange) This line functions as an express overlay of Lines 3 and 5, but has a distinct route through the core that’s basically 5th Ave, 5th Ave S, and Hwy 99. There are huge parts of this line in Pierce and Snohomish County, but I’m going to ignore those largely because I think that they were overkill and inappropriate for their actual needs.
Line 7 (Ballard Line, Cyan) This line follows KCM route 44 on the West side of Lake Washington, and KCM route 271 East of it.
Line 8 (Tacoma Line, Pink) This is a Pierce County Line, while it’s off of important Pierce County corridors, I’ve come to think Metro is overkill here. So I’m ignoring it.
Line 9 (Kirkland Line, Chartreuse) This line evokes the old KCM 340 bus, sweeping in a wide arc Woodinville to Burien via Kirkland, Bellevue and Renton. If you notice, some parts of this were originally part of Line 1.
Line 10 (Federal Way Line, Brown) This line is in southern King and Pierce counties. Like Line 8, I do see it as sort of overkill at this point. Though a lower much scale version of what I’m getting at might have some worth as a project.
Line 11 (Madison-Delridge Line, Gold) This line follows KCM routes 11 and 12 north of downtown. The segment south of downtown, which hadn’t opened in the 2007 map would’ve followed the northern part of the 120, and would’ve also taken over Line 1’s route south of White Center (To set Line 1 up for an extension East to Rainier Valley).
Line 12 (University Line, Gray) This line basically combines KCM routes 111, 105, 106, 36, 48, 67, and 347 (from South to North). Looking at the older maps, you’ll see that the northern segment here was originally a part of Line 1, while the southern segment was a part of Line 5.
From here we get to lines that were more or less in my head.
Lines 13 and 14 wouldn’t have any new infrastructure, they’d just take the parts over of Line 3 and 5 in the deep suburbs (south of the Airport for Line 13, North of Lynwood for Line 14), in addition to the local segments of Line 6 in Tacoma and Everett respectively.
Line 15 Would’ve used NW 85th St, and NE 65th in Seattle before jumping across the lake to follow NE 85th St in Kirkland/Redmond, before arcing down the East side of Lake Sammamish to Issaquah.
Line 16 was going to be a relief line for 2 in Downtown, using Jackson St, 2nd Ave, and Elliot Ave W through the core, and taking over the segment of Line 2 north of Ballard.
Line 17 was going to be a Radial line that started in the SLU neighborhood. Going through the core under Westlake and 4th Ave, it would’ve followed the Sounder Commuter Rail Corridor south of Downtown to the suburb of Auburn.
Line 18 was going to be Redmond-Crossroads segment of Line 2 that was going to cut off from the rest of the line, and extended south into the Eastgate area.
Beyond this there were plans for extensions to new areas (Like Kitsap county, I was crazy), and other lines, that I’m not going to get into right now. Largely so I can ask you if you have any questions about this that you want answered.
Keep in mind, this wasn’t the latest version of this project. The later version is more realistic.
All of Washington State could move to New York City, make Staten Island a nature reserve, bulldoze the Rockaways and Coney Island and still have a farm or two in Queens. Metro Seattle could move to Brooklyn and Queens and still make the Coney Island and the Rockaways a nature preserve with farms in Bayside. In very very round numbers metro Seattle has half the population of New York City and five times the area. You fantasy doesn’t make any sense unless automobiles are banned.
Well, I did say it was crazy to begin with. But there are a few caveats. One, this is a very different metropolitan Seattle, one with about twice the metro population as OTL (roughly in line SF and DFW), and it’s one largely without I-5 and I-90. (Something that holds up in the later, more realistic versions. And this is in a version of the US that rather going to batshit building Highways to nowhere, it instead built Metro systems that are overkill.
People in Seattle like to eat. They need I-5 and I-90 to haul it in.
It comes in lovely packaging that needs to be hauled out.
There’s all that stuff coming into and going out of the port.
….Boeing needs parts… Starbucks ships things out…
Seattle might need that much infrastructure in a world with completely free immigration and no zoning.
Again, it’s a very different Seattle. The existence of the Metro would drastically shift development patterns, and the butterfly effect might mean that Starbucks won’t exist, and Boeing would using different sites.
Still, I-405 does exist, as does Hwy 520. They’re taking on the role of I-5 and I-90 (Hwy 520 is actually extended west to Interbay TTL.). And that not even getting into the that there’s actually 3 times as many crossings of the Ship Canal (No, I’m not pulling that out of my ass.)
And that’s not even getting into Freight rail infrastructure (In the crazier scenario, Traffic is shifting to Everett, in saner scenario, Everett and Tacoma.)
How familiar are you with Seattle business history? Asking out of curiosity – a lot of its recent tech development comes from fully vested employees of Microsoft and Amazon starting their own businesses (e.g. Valve was started by Microsoft millionaires), but I don’t know why exactly all these companies were there in the first place. Bill Gates grew up in the area, but Jeff Bezos moved there to start Amazon; do you know if it was because of Microsoft? Already in the 1990s Seattle had a hip reputation…
Considering that Microsoft moved to Washington in 1979, Seattle’s reputation in the 90s probably didn’t have anything to do with it. It’s warm in the winter and cool in the summer? They couldn’t afford the rents in California or Massachusetts? And didn’t want to live in Texas?
Amazon? Someplace that is warm in winter, cool in the summer, has a lot of programmers and a low population so things shipped from there aren’t subject to sales tax in California and New York? Part of the initial business model was that the mail order shipper doesn’t have to collect sales tax. . . and the USPS charges the same price no matter where you are shipping the 4th class mail…
I don’t think Amazon would’ve had to pay sales taxes either way? I think online purchases were tax-free regardless of whether they were interstate or intrastate.
Intrastate, sales tax gets collected. There are assets the state can seize if you don’t. Hard to wiggle out of, the shipping company has records.
Yes, but in the 1990s, e-commerce was explicitly exempt from sales taxes.
No it wasn’t.
State income tax of zero. (Most of state budget funded by massively regressive sales taxes.)
I don’t know the details, but I do know that the boom Seattle had after World War II was led by Boeing, and then it had a huge bust at the end of 60’s. I know about this bust because it sunk a vote Seattle had to build Rapid Transit system, known as Forward Thrust. Forward Thrust served as the original jumping off point for the maps that you see above, with the first draft being almost identical to the map above, but with a different order of Lines. I then decided to move things back 25 years, and reordered the lines to account for that.
IIRC it really was Microsoft that played a role in developing the regions startup culture.
Having been to certain parts of Lagos, I can offer you some feedback on them.
-The Red line has no reason to go to the Wharf and East Wharf stops. Those are within a container shipyard and just won’t generate much traffic.
-The islands within and south of Badagry creek are not fit for human habitation. Sure people _do_ live there, but the island are very swampy and malarial. Without a very extensive land reclamation project, no one will ever be able to sink foundations there, so that area will never get the density needed to support subway lines. This affects the purple, dark orange, and teal lines.
-From my perpective, the prime destination is the western end of Victoria Island. The subway density (line spacing) there seems about right; something like 800m. Why so much denser in other parts of the map.
-This especially applies to the entire west/northwest/north parts of town, which are more or less just shantytowns. Why so many lines out there, when density isn’t so high and people will be more than willing to walk to a station.
Ooh, thanks for feedback. Responding based purely on what I’ve read in various places:
– Yeah, the circumferential wharf stops are questionable.
– Bagakry Creek and Lagos Harbour would need work, yeah… but way more work is being done for Eko Atlantic. Ultimately, a bigger, richer Lagos needs more land, even in areas that are currently less wealthy than Ikoyi and Victoria Island.
– The CBD is Lagos Island, not Victoria Island. But clearly the CBD is migrating east (just as the Paris CBD migrated west from Chatelet to the Opera and wants to migrate even further, to Champs-Elysees). Hence, the high density in these areas, rather than just lines terminating in Lagos Island.
– The shanties won’t stay shanties forever… and Lagos is big enough it might need all that capacity.
You mentioned lines that skirt the CBD and thus hinder their ridership. I have no clue why you mentioned the L. The L is one of the most heavily used lines in NYC. It is known for its overcrowding. The press coverage surrounding the Canarsie Tunnel shutdown shows how heavily the L is used. People said nothing about the Greenpoint Tube shutdown on the G.
The L is actually the third least used of the nine “colors” in New York, after the G and the J/Z. It’s overcrowded because it runs shorter trains (eight 18-meter cars rather than ten) and has lower frequency than the rest (around 16 tph if I’m not mistaken – they’ve upgraded signals to allow for more, but the electrical infrastructure can’t have so many trains at once on the line).