A brief Los Angeles transit history lesson.

50 Replies to “Sunday Open Thread: Why isn’t the Expo Line a subway?”

  1. Wow! Lots of lessons here! Still, it’s a little simplistic to conclude how things in LA evolved in a sentence or two.

    I would mention that US rail systems built after 1985 are mostly surface light rail. That approach, encouraged by FTA scoring for grants, has dominated the rail expansion mindset for 30 years. It wasn’t a choice unilaterally made by LAMTA. Of course, there are many painfully slow surface rail segments around the US because of it.

  2. This really make me appreciate what we’re building in the Seattle area. While there are things that we wish were different/better, we’re effectively building a subway in much of the urban core which is something that no other US cities are doing now. This leads to a much higher cost per mile, but also a system that is truly a game changer for mobility in the region.

    1. Yes exactly, that’s what I’ve been saying for a long time. The cost of a rail system is directly related to the benefit. If we had gone with the original low-budget approach Link would be on I-5 or Eastlake between downtown and Northgate, and on the surface between Mt Baker and SeaTac. The tens of thousands of people on Capitol Hill would not down the steep hill to the station; they’d just continue suffering with the 43 and 49. And the other tens of thousands of people going to UW would have to walk or bus from I-5 and enjoy the freeway atmosphere around the station. Luckily we had already built the DSTT in the 1980s so there was no consideration of surface downtown (which Portland, San Jose, San Diego, Dallas, and others have).

      However, the video is right that you have to balance getting it right vs getting it at all or sooner. Every year of delay means a year of people’s time wasted on slower, less frequent, less reliable buses, and a limitation on the number of activities they can get to in a week. We have to fight hard for tunnels in downtowns, grade-separated stations in the center of large urban villages, minimal surface segments with level crossings, and surface segments no worse than MLK.

    2. >> we’re effectively building a subway in much of the urban core which is something that no other US cities are doing now.

      Except that the subway skips important parts of the urban core, for the same reason that the Expo line is above ground: politics and money. ST feared the political repercussions of a cost overrun, if they ran into bad soil (something that didn’t stop Bertha, of course). After a couple of losses, they figured they needed to keep things under budget, and do what they said they could do (connect the UW with downtown). So they didn’t build a station there, and there are no plans to.

      Likewise, there are no plans for a subway station at Belltown. Unfortunately, this sort of omission is common, as we build things out of order. Fife before Belltown. West Seattle before the Central Area. Paine Field before a line connecting Ballard to the UW. Eventually we may build a subway in our urban core, but it will take a very long time, and we will build lots of other stuff first.

      In the meantime, L A. is actually building a subway in parts of their inner core, it is just that they are mixing in surface light rail (which has to wait at traffic lights) at the same time. I wonder, though, which will be easier to fix. I would guess that retrofitting a First Hill station would be very difficult (you would probably be better off simply making a First Hill station be part of a Metro 8 subway) but burying the Expo not as bad. On the other hand, as the video showed, they spent a huge amount of money on a train that gets stuck at red lights, which is really a bad idea. It probably should have been a bus line until they got enough money to bury it. Of course, like a lot of things, hindsight is 20-20.

      When I look around the country, I find very few new mass transit rail systems that are good values. D. C. seems like an exception. It covers just about all of the urban core, as well as the inner suburbs (Georgetown being the exception, and only because they didn’t want it). The result is a subway system that is extremely popular for a city its size. The successful design of that system may have been due to luck. Or it may have been due to federal government oversight and funding. Almost all cities go through what Seattle and L. A. have. Propose something, then only if people vote for it, build it. This leads to them making vague promises and focusing on what will be popular amongst the public, but what transportation experts would think is silly.

      The average person thinks of public transportation as they would a freeway. Unfortunately, the analogy breaks down almost immediately. A freeway that includes downtown and Capitol Hill would certainly be seen as serving First Hill, as it is only a short drive until you get to the on-ramp. But it is too far to walk quickly, and messy from a bus intercept standpoint, and thus a big omission. A freeway to Tacoma is great (much faster than a highway with traffic lights) but a subway there is very slow (an hour and 15 minutes just to the outskirts of town). Likewise, while it doesn’t seem bad to “go around” from the UW to Ballard (and you certainly would if there was a an uncongested freeway doing that) it just doesn’t make sense from a transit standpoint. It takes too long, as the improved speed of a subway doesn’t make up for all of the extra stops and extra distance.

      The result is making plans that look great on paper — that sound great when proposed to the general public — but would probably be laughed at in a board full of transportation experts. I’m afraid that while Seattle is unusual in that it is spending a huge amount of money on public transportation, it is ordinary when it comes to getting good value from the money spent. I think that is because it runs into the same political problems that most of those other cities have.

      I wonder if D. C. did things differently, or just got lucky.

      1. Ross,

        it is just that they are mixing in surface light rail (which has to wait at traffic lights)

        Again with the anti-LRT exaggerations. More than 80% of the length of the Expo Line runs on gate-protected private right of way. It’s only the stretch on Flower Street south of the MetroCenter Subway portal and a few blocks just north of the Santa Monica terminus where it’s running in the middle of a street where it has to stop for traffic lights. Where it uses the old Santa Monica Air Line ROW it slams across streets without even slowing down because it has gates. It doesn’t go quite as fast as Westside MAX — Expo’s speed limit is 45 or 50 depending on the frequency of grade crossings — rather than MAX’s 55. But it’s pretty darn fast.

      2. Washington DC’s Metro subway benefits from direct Federal support at many levels. It’s personally important for Congressional members who have to live in DC, as well as for the huge numbers of Federal employees responsible for our policies and funding at the Federal level, including the Federal Transit Administration. It keeps them from needing to obsess about capital-cost-cutting measures like everyone else, and a better awareness of best practices from seeing what happens in the rest of the country.

        On the other hand, the DC area is also much more politically willing to zone for 20-story apartment buildings than the Puget Sound region is. If Mercer Island or South Bellevue or even Judkins Park stations were in DC, there would be an active discussion of placing at least 10-story if not 20-story buildings there — both residential and non-residential. That’s not only true for the District, but also for Arlington, Alexandria, Silver Spring, Rockville and Fairfax County. The closest thing that we have are plans at Bellevue’s Spring District, which still won’t apparently have buildings over 12 floors.

      3. “Except that the subway skips important parts of the urban core”

        I’d argue that Link as-is is twice as transformative as a lower-budget Link would have been. It doesn’t eliminate the need for a 43-like route and a 7-like route, but it does get people from the center of Capitol Hill to UW quickly, from Columbia City to UW faster than the 48, and in the future from more suburban locations to UW and the center of Capitol Hill quickly. If you have to walk or transfer to get to 15th or Summit or 23rd, well that’s a tradeoff. It’s a much smaller tradeoff than what Link does, even if we can imagine a fully-urban line that would be more transformative (within the city).

        “Likewise, there are no plans for a subway station at Belltown.”

        Link in general suffers from a lot of “old thinking”. The lines were drawn up in the early 1990s when there was a more suburban and P&R-focused mindset, when the urban redensification was just starting and people probably assumed it would remain small and irrelevant, and when congestion was only half that of today. So the Northgate and 145th P&Rs were seen as good places for stations. And the “extra stations” at Summit, 15th, 23rd, 520, and 85th were so unimportant they didn’t even consider them. Belltown and SLU weren’t even on the radar. SLU was added at the last minute, but that’s because the huge increase in commuting there really couldn’t be ignored. It was like adding a downtown Bellevue to Seattle, and the planners really messed up by underestimating the impact of adding that many jobs and highrises.

      4. So, why were Alexandria and Arlington so pro-density and Pugetopolan cities aren’t? Why did commerce and transit convenience have so much clout in DC but NIMBYs had more power here? Because they could have put in the DC metro and left the station areas unchanged.

        I’ve heard that Alexandria was in decline at the time, so maybe it was a “we want jobs” and “we want to bring people back to the city”. Whereas in Mercer Island, south Bellevue, Judkins Park, Mt Baker, and north Seattle it’s more “We’re affluent or gentrifying SF homeowners in leafy green neighborhoods who don’t want change. Maybe Alexandria was just more of a commercial/industrial area so there weren’t many leafy neighborhoods around?

      5. Although DC got huge federal support for building out its Metro, the operations and maintenance of that system depends entirely on contributions from local governments in DC, MD, and VA. WMATA does not have a dedicated funding source and there are issues with the management culture which is a reason for the service cuts and safety problems of the last few years. That is a shame because they got so many things right.

      6. “If Mercer Island or South Bellevue or even Judkins Park stations were in DC, there would be an active discussion of placing at least 10-story if not 20-story buildings there — both residential and non-residential. ”

        You mean, like what Redmond and Bellevue are doing with 8 of their 10 East Link stations? The only East Link stations that aren’t getting upzones and robust mixed-use development are South Bellevue and SE Redmond, both of which happen to be adjacent to giant parks.

        Mike is on to something – it’s much better to build new stations in low density commercial areas and aim for redevelopment than to build stations in dense but primarily SF residential zones and hope for further development.

      7. I think that higher density gets discouraged for another simple reason that doesn’t dominate D.C.: views. I can’t imagine any buildings taller than 6-9 floors in many places around Pugetopolis because the million-dollar-view homeowners a half-mile away would be up in arms!

      8. Meanwhile, D.C. itself has strict height limits set by Congress to keep the city’s Paris-like character. I wonder how much of an effect that has on the demand for higher density development just outside D.C. in Roslyn and Silver Spring and Bethesda, etc..

      9. AJ,

        You’re right about view retention. And that’s a very good reason to line the crest of all the north-south hills with checkerboarded (like Van BC) mid-to-highrises. You’d be creating thousands of new view properties without blocking others’ views. Oh sure, if you put a row on Phinney Ridge then the folks high on the ridge in Maple Leaf will lose some of their views of the southernmost part of the Olympics, but if there were a line of highrises along Roosevelt they’d be looking at the Cascades.

        The same is true even of the ridge east of Delridge in West Seattle. Go high enough and you’re looking over the West Seattle plateau.

      10. >> Again with the anti-LRT exaggerations. More than 80% of the length of the Expo Line runs on gate-protected private right of way.

        Fair enough, but I don’t remember making anti-LRT exaggerations. I don’t know the details of the Expo line, and why it isn’t buried where it would matter. Again, I have no problem with trains running on the surface. I have a problem with a train stopping at a traffic light. The video implied this was an issue (i. e. suggesting that they should have paid the money to bury the train).

        Bruce Nourish had an excellent post a while back (before the ST3 vote) pushing for surface running to Ballard along with cut and cover downtown. In the article he had this to say:

        Lots of people worry about the top speed of transit service, but it’s not very important for in-city services (say, typical trips of less than ten miles, stops about every half-mile), because even a fully grade-separated train spends much of its time accelerating or decelerating for stations; frequency and reliability matter most. Assuming any of these lines will be both very frequent and reasonably reliable, the most important factor to minimize trip time is to avoid extended periods of very low speeds, e.g. slogging at-grade through the city center.

        I agree completely. I really don’t care what the maximum speed is of the Expo, I’m concerned if it spends a lot of time stuck at a traffic light. If this isn’t the case — if the video greatly exaggerated the issue — then I stand corrected, and it probably isn’t worth the money to bury it (just as it isn’t worth the money to bury the train on Rainier Valley). But if the train is stuck for long periods, then it probably is, and is just another example of an agency cutting corners where they shouldn’t.

      11. >> I’d argue that Link as-is is twice as transformative as a lower-budget Link would have been.

        Well, it depends on what you spend the money on. If the alternative is a low budget streetcar style option (which spends a lot of its time stuck waiting) and also skips Capitol Hill, then I agree with you completely. What we built is much better.

        But we could have also built each piece as well as possible, even if it didn’t go out as far (initially). We could also build it in a logical order. I don’t think anyone would look at a census map like this (https://arcg.is/0fS9WO) and immediately conclude that the first thing we need to do is go south — very far south. Even before you are told that a major university (which has nearby office towers and will soon have many more) lies a few miles to the north, you are drawn there because of the density. You also want to include the density to the east, and next thing you know, you’ve built U-Link. Except that it has a station on First Hill or 23rd and Madison (or both) as well as a station at 520. It is a bit more expensive, but it is also a lot better.

        Or maybe you don’t worry too much about First Hill with the first line, but you start working on it with the next one. Include 23rd and Madison just because it works well for buses, and then wait for the Metro 8. Once those two are built you are very close to achieving what Andy was talking about (covering much of the urban core). Eventually you add in the missing pieces: A Ballard to UW subway. An extension north to Northgate, along with a solid terminus for the north end buses (likely a bit north, close to the freeway). Build the south end line pretty much as ST did it, but don’t worry about going all the way to the airport (just build a south end freeway terminus for the buses). Run a line to Bellevue and Ballard. For the latter run it through Belltown. If you really need to add a second line, then add value to it — send it up to First Hill. So, basically something like this: https://drive.google.com/open?id=1kgurKg8UNhW8qQQMwyc5sBKfU74&usp=sharing

        If we want to extend that to the airport or West Seattle, fine. But that should happen later, as neither is a high priority. My guess is the second downtown line goes along with a West Seattle line, in that they aren’t necessary until much later. Even without that second line downtown, it should be obvious that this covers the core of the city. There are still plenty of places that are left out — some of which have relatively high density. Parts of Queen Anne, for example. But the core of the city is covered, and covered fairly well.

        The problem is that our system won’t look anything like this in thirty years, despite spending more money on it. More to the point, I don’t see it ever looking like this. Not only are we building things way out or order, and building things that aren’t worth the money, but we simply have no long term vision for covering our core. To a lot of people, ST3 is it. But even if you add to ST3 (e. g. build the Ballard to UW subway) you still haven’t covered Belltown and the Central Area — areas that are some of the most densely populated places in the state. These are areas that have the proximity and density to be very successful, but we have no long term plan to cover them.

        I think all of these problems are due to the political makeup of Sound Transit. Again, I may be giving Washington DC too much credit, but I think the nature of their process simply served them well, as opposed to simply being lucky, and having good leaders. I think our process is common, and D. C.’s was unusual. I think what is unusual about our system is that we will spend a lot of money, but not get something that is great. Most cities spend little, and get little. We will get a bit more, but spend way more.

      12. Ross,

        OK, caveat accepted. The section north of USC runs alongside — not in the middle like Link on Martin Luther King — Flower Street. There are about a dozen intersections that the train crosses before it descends to enter the MetroCenter subway. Most of those streets are less than important so the train gets the same sort of “wave” priority that Link gets on MLK. The operators makes a request for signal clearance leaving the USC station and gets a mostly set-up run, albeit at a low speed. I expect it’s 25 mph but don’t know the actual speed limit.

        However, there are a couple of major streets that don’t fit the wave every time delaying the train and often causing cascading delays until the next station. This happens on MLK with Link as well, though less frequently. There is also a level crossing at Flower and Washington where the Expo joins the Blue Line. I think that sometimes causes delays that get trains out of the wave as well.

        The upshot is that not every train — I’d expect not even very many trains — gets delayed by traffic signals, but enough do that it’s an irritant.

      13. @Al — Your first paragraph was what I was getting at. Not only the funding, but the planning. We are going to spend a huge amount of money on our system, but I don’t think it will achieve anything close to what D. C. has achieved. We’ve ignored best practices, and I think that is due to the nature of our system. Subarea equity along with trying to serve too large of a region with a rail line (i. e. the spine from Tacoma to Everett) was a big mistake. Everett and Tacoma should definitely have a seat at the table, just as Baltimore should be able to weigh in on choices made with the D. C. Metro, but the goal should not be to connect the cities with a very long subway line.

        I disagree a bit with your second paragraph, though. D. C. proper population peaked in 1950, at 800,000. Now it is a bit less than 700,000. As mentioned, heights are limited there and frankly I don’t see it being that big of an issue either way. Even with the relatively static nature (and small size) of D. C. proper, every line goes through the city, and trains spend the bulk of their time servicing the area. I don’t think it was really designed for what might eventually happen, but was simply built for what exists when they started planning.

        But you are right in that the close in suburbs are remarkably different. Alexandria and Arlington (which would be part of D. C. if it was as big as Seattle) have grown considerably over the years. That is one area where they planned to grow around the station, and did so. But Northgate and Roosevelt have grown over the years and they are roughly as far away.

        From what I can tell, the farther out you go, the more the stations look very suburban. I’m sure there are exceptions, but they look remarkably like what we are building. Stations close to freeway with very low density and big park and ride lots.

        Which is not to say we shouldn’t grow around the stations, but for the most part, when you build next to the freeway, you can’t get very big. The freeway itself just takes a big chunk of the available space. You can grow outside the walking distance, but that is true for the entire region (not just places like Shoreline, but also Seattle). Much of the growth in Northgate will (or has) occurred well away from the station (no longer within typical walking distance) but that doesn’t mean growth will occur up on Maple Leaf, despite being closer.

      14. The DC height limit is 130 feet, or 110 feet along wide streets. That’s notably taller than the 85 feet or 65 feet limit that applies to most of Seattle.

      15. “D.C. itself has strict height limits set by Congress to keep the city’s Paris-like character. I wonder how much of an effect that has on the demand for higher density development just outside D.C”

        DC also has several good urban station areas. That shows that the problem isn’t an 8-story height limit per se but that a lower limit must necessarily be extended over a larger area to reach the population target. Paris, Boston, Edinburgh, Amsterdam, San Francisco, second-ring Vancouver (Kitsilano/Broadway area), and Chicago’s north side show that you can fit a lot of people in a 3-10 story district or even a 2-4 story district if you make the district large enough and banish excessive “open space” and parking. The real problem is the fact that the majority of Seattle’s and Pugetopolis’ land is locked into single-family zoning, and it’s not the row houses or small-lot bungalows predominent in Boston and other northeastern cities. In fact the ideal “house-density” development is now illegal due to minimum lot sizes, minimum parking, and maximum FAR. If we can get everything zoned to lowrise and eliminate the space-wasting regulations, we won’t have to worry about highrises as much. Highrises are intrinsically expensive because they have to be reinforced more heavily to stand up over time, and most urbanites live and prefer to live in a 6 or 12-story neighborhood rather than a 40-story neighborhood. Most of New York City and even Manhattan is 6-12 stories. The skyscrapers are a minority of buildings, most of them are non-residential, and the residential ones are the most expensive housing in the city outside house-sized apartments.

      16. DC has basically the same issues as Seattle, with lots of SF zones that don’t allow for the missing middle:

        DC may have more midrise neighborhoods to make up for the lack of high rise neighborhoods, but it still has plenty of room for growth, just like Seattle.

        Seattle is able to grow at 3% (which is crazy fast by any modern, developed country standard) b/c it’s upzoning at all three levels – buildings high rises in SLU/First Hill, building midrises in Ballard, QA, Roosevelt, etc. and building low rise elsewhere (all those 4 for 1 townhome developments). And the city needs more of all three – highrises in U District and Northgate, more midrise in the hub villages, and more ‘missing middle’ elsewhere.

      17. There is no “missing middle” construction in Seattle. That’s the gap between the existing allowable houses and the existing feasible apartments. New buildings in the 1950s had a wider range of sizes and styles, like inexpensive 4-8 unit apartments and dingbats. Those are what’s not being built any longer, those and infill houses, and SRO hotels (the earlier counterpart to micro-apartments). This means there’s an ever-shrinking availability of those kind of units as they’re torn down or the population increases relative to the number of units. It’s both a housing-type issue and a size issue. (1) The types of units working-class and lower-middle-class people can most afford aren’t being built. (2) Much lowrise-zoned land remains single-family because the unused capacity isn’t enough to make replacing the existing building cost-effective. (3) There’s such a severe restriction on where new units can be built that it doesn’t saturate the upper end of the market, so the remaining upper-income people go into lower-class housing and the lower-income people are shut out. If we built enough housing to saturate the upper-income market and go beyond it, then the supply/demand pressure will ease. When the developers can’t find any more rich people to lease their luxury units, they’ll start building more downmarket things for other people. They’ll do that because that will be the only way to make any more profit. And if they say no and leave (mostly Wall Street-backed investors), then smaller local developers will step up to do it.

  3. I really hope the EIS process is smashed and torched. Environmental impact now isn’t about the environment, its about preserving the dysfunctional status quo… ballooning a transit project’s budget by adding car lanes and minimizing auto wait times while making the packed high occupancy trains wait at lights.

    The same can be said for constraints imposed on LINK in the Rainier Valley, the Mercer Slough and 99 in Federal way.

    1. The California state environmental process is much more complex than the one in Washington state or at the Federal level. California also has approved many referenda that govern environmental impacts, so voters would have to weigh in on reforms there.

      I would also say that the grade separations are good for this line. Traffic on La Cienega and other streets that cross this line are much heavier than a typical Seattle arterial. My last visit to LA included a 30-minute midday traffic delay on northbound La Cienega starting 2 miles south of the rail station! Exposition is a major cross street with lots of turning vehicles too. Finally, it’s safer for train riders to not be crossing tracks when trains arrive at a station.

      1. You also have the density to justify the extra cost of grade separation.

        Frankly, I find it weird that agencies so often feel like they have a perfect “middle”, where light rail, stopping at a stop light, makes sense. Just to be clear I have no problem with surface running (like Link does on Rainier Valley). Yes, there are stop lights, but the train rarely stops. That is often well worth the money when compared to ground level and even elevated transit (which has political issues as well). But if you are OK with a long train stopping a long time at a stop light, chances are you don’t need a train. A bus will do just fine. Or, to put it the other way, if you really need a big train, isn’t it worth the extra money to make sure it can run reasonably fast?

        Again, I’m sure there are instances where this isn’t the case. They have just barely enough volume to justify a train, but not enough to justify grade separation (or even much signal priority). But I think that is rare. More often, light rail is built mainly for cosmetic reasons (i. e. it looks like a superior system) even though it isn’t the right tool for the job.

      2. There is a difference between no transit priority anywhere, and minimizing stoplights to a very few difficult areas that have too many physical constraints or competing needs. Europe doesn’t build mixed-traffic streetcars routinely, but it does allow segments in a very few cases where the constraints are especially severe, such as a road that can’t be widened because historic buildings are right next to it on both sides.

      3. You also have the density to justify the extra cost of grade separation.

        In the case of the LA Expo line, there was also a long history of freight service that justified street separation before light rail was added.

        You don’t need the density if the infrastructure to do grade separation is already there.

        Seattle has no such right of way. The only real right of way that might be useful for future light rail is the segment of the NP freight line that served SLU from Interbay along Westlake, and it has no existing structures that give it grade separation.

      4. A good example of the “middle” might be East Link? It’s generally grade separate but at grade, but is a “subway” in downtown Bellevue. In Bel-Red, it runs at-grades with a handful of at-grade crossings of minor streets, but invests in grade separated crossing of major streets like 124th.

      5. @AJ — Yes, that is a good example. I think much of Central Link (Rainier Valley on the surface, tunnel through downtown) is similar. I have no problem with that, just as I would have no problem with running Ballard Link on the surface (as long as it goes underground downtown).

        What bothers me is when a train basically stops when it gets downtown. Apparently the Green Link, in the Twin Cities does that. The result is a line that is not much faster than the bus it replaced (http://www.thetransportpolitic.com/2014/07/21/the-value-of-fast-transit/). That is the part that doesn’t make sense to me. Seattle was way smarter in the initial order that it built things. First make the buses faster (by building a transit tunnel) then replace it with rail. As much as I complain about all the mistakes we’ve made, I am very thankful we don’t have light rail trains running through the surface downtown, waiting for the light to turn green.

      6. Hopefully the next round of investments in these cities is fixing the core to allow for better service in the whole network. This is basically what Seattle is doing with the 2nd downtown tunnel and what LA is doing with the regional connector.

        What could be great for Minneapolis would be a regional connector like project. Same for Portland: http://www.wweek.com/news/2017/05/24/portlands-fantasy-transit-map-what-if-we-spent-billions-to-fix-the-morning-commute-with-something-other-than-cars/

    2. Maybe an overhaul of the EIS criteria would be better than throwing away the whole process. It does do what it says: disclosing and making transparent and quantifying what the effects of a project will be and allowing public input. An EIS itself doesn’t prohibit anything or mandate mitigation, it simply states the facts and gives a baseline for debate. In Robert Moses’ day transportation departments could simply build whatever they wanted with no public input: was that better? Those lamenting I-5 fhrough diowntown, First Hill, and SLU would say no. We need to look more at why Germany and Canada succeed so well in transit design. It is more top down like Moses but I assume there’s more public input in the alignment and goals. A lot of it is public acceptance and desire of an urban-focused transit network. But maybe they also handle EIS-like issues better.

      So the problem seems to be in the EIS criteria. Maybe it needs to weigh more heavily that high-capacity transit is a great public good rather than simply a negative impact.

    3. FWIW as of 2014, California no longer considers changes to road traffic Level of Service (LOS) “an environmental impact” to be mitigated under state law but in retrospect the grade separations were for the better. On the other hand, the part that they did not grade separate near downtown LA has become a bottleneck affecting speed and reliability on both the Blue and Expo Lines.

      1. Oran, yes, they need to put the northbound tracks of both the Blue and Expo lines in a trench — just extend the east track of the existing MetroCenter subway — so that southbound Blue Line trains don’t foul northbound Expo’s and vice versa.

        And then improve the signal priority between the junction and USC.

    4. Like a lot of things passed during that era, it probably needs to be amended. LBJ, for example, passed a lot of Great Society programs knowing they were flawed, but figuring they could be fixed later (by making minor changes). I would imagine that NEPA, passed under Nixon, was probably passed with the same mindset.

      Unfortunately, like so many things, i wouldn’t trust this Congress with it. I would trust Nixon and the Congress of that era to make fixes that allow us to continue to build things that we should build (without wrecking the environment) but I fear that this president and this congress would just throw the baby out with the bathwater. They are trying really hard to roll back generations of environmental protection as we speak.

  4. You so often see LA Metro held up as a model agency these days. Clearly some very real budgeting and scheduling issues in their recent past, though.

    1. The same can be said of Sound Transit. Its early years were quite rough but both agencies managed to bounce back and get two massive transit expansion packages approved by voters in 2008 and 2016.

      1. As someone who formerly lived in Portland, I can say Sound Transit is much better run than TriMet. Link light rail has frequent fare inspection, lots of security and spotless trains and stations. TriMet trains and stations are filthy, a serious crime problem aboard the trains and stations and zero fare checks or roving security/police.

  5. The presentation did not make clear the primary reason that the Expo Line is surface rail: there was a recently active and almost intact Pacific Electric right of way available for use for 2/3 of its length. Since rail had always had “railroad crossing” protections at the cross-streets, it was easy to move to a gated light rail line.

    I’ve ridden the Expo, grant, just as far as the Culver City Phase 1 terminal, but I’ll say up-front, it books! west of USC. Like Westside MAX it slams across streets without ever pausing for stop lights. It has. preemption everywhere west of the University. No bus line ever has full preemption and none ever will.

    There are four or five historic LRT lines in the eastern part of the country and in San Francisco which were spared bustitution because they had significant reserved right of way and/or tunnel sections for which buses were simply not suited. Boston’s Green Line, the Norristown High Speed Line west of Philadelphia, Pittsburgh’s West Hills system, Cleveland’s Shaker Heights Line, and Muni’s Market Street system because of the Twin Peaks and Buena Vista tunnels.

    And thank Dios for them. Surface rail operating in reserved right-of-way with railroad crossing protection and preemption is very nearly as good as elevated or subway alignments, if the right of way can be secured. Do not think “surface rail” has to mean Interstate or Burnside MAX. There are a few places in the US and hundreds worldwide where it works very well and attains speeds in the 50-60 mph range.

    The sainted Orange Line takes roughly 35-40% depending on the time of day longer to transit end-to-end than it would if it had gates. Buses simply would come too frequently to allow gates; north-south traffic in the San Fernando Valley would be snarled. So the busway operates on the same light cycle as the closest parallel arterial.

    Grant, it’s great not to wait long for a bus; they come quickly most times of day or evening. But a rider more than loses the reduced average wait time versus a train sitting on the bus waiting for traffic signals.

    1. P.S.

      The same thing is true for the Blue Line. The old Pacific Electric Long Beach main line was still in freight service and could be re-purposed from a couple of miles south of downtown all the way to Long Beach.

      It’s true that the much longer mid-street approach to MetroCenter delays the Blue Line as it does the Expo, but south of Slauson most of the way to Long Beach it operates like the Expo Line with gate protections and preemption.

      1. Same for the Gold line, or at least much of the Gold line extension, and the Green line extension. They are all leveraging existing ROW, which LA Metro tends to own.

        Seems like he did a good job tracking the political context, but failing to point out that Metro’s BRT & light rail lines tends to leverage existing freight ROW (or freeway ROW), while the subways are creating ROW, seems like a pretty big oversight.

      2. Hey AJ, video author here:

        Obviously, I’m aware of the great Pacific Electric and freight ROW that Metro has widely reused. The point was, during the Red Line subway era, reusing the Pacific Electric corridor to get to the Westside was seen as unnecessary (politically), because the Westside was going to get an extension of the full-subway Red Line. It actually took quite a bit of community effort for Metro’s predecessor to save the ROW at all from private development during the Red Line era.

        It was as a result of the subway’s failure that the ‘Transit to the Westside’ line was moved to the Expo Corridor. Politically both the Wilshire subway and the Pacific Electric ROW were equivalent ways to get to to the beach, but with a subway off the table Metro was forced to use the cheap ROW.

        There’s a lot more history of early LA Metro I was only able to skim through. There was a huge early divide between classic forces who wanted to build Full Subways, and forces who wanted to quickly expand through reuse of the ROW and light rail. This is a struggle that occurs everywhere, but LA is an interesting case because we actually did _both_ at the same time, and the county voters clearly preferred the latter.

        But for the sake of the Expo subway vs lrt argument, “Expo Line Subway” is actually just “The Purple Line Extension”. There would never have been a subway under Expo.

    2. Let’s keep it in perpective. LA does have a subway in its highest-volume corridor, from downtown to Hollywood. The light rails are in secondary corridors. Similarly, Seattle has a tunnel in its highest-volume downtown-UW-Northgate axis. SLU was a very recent phenomenon that occurred after the initial segment opened and University Link was under construction. We have RapidRide on secondary corridors: Aurora, Ballard, and West Seattle. They aren’t as effective as secondary light rails but they’re something and we’re a tenth the size of LA. And light rails are now approved for Ballard and West Seattle and will have a downtown tunnel. So that’s something too.

    3. I agree – I found the Expo line plenty swift west of USC. It really only slows down between USC and downtown, when there are many at grade crossings and the train doesn’t always get preemption. The Regional Connector will help a lot, but the Expo line will still be slower between USC and downtown than if it was truly grade separated.

      1. AJ,

        Yes, both the Blue and Expo could be greatly improved by improved signal timing with as much LR priority as the cross-streets can stand. However, the really big win is to depress the northbound track on the Expo from about 23rd to Venice and the westbound track on Washington west of Maple. This would mean that the Grand/LATTC station would have to have a lower level platform for inbound trains but it would also mean that all trains southbound on Flower could run through the junction without adverse movement.,

        Inbound trains could simply hold at the Grand/LATTC or LATTC/Ortho Institute station until the northbound junction is clear. Trains would move through the junction much more quickly than they do today.

        I know this is expensive but it would increase the possible capacity on the severely stressed Blue Line and of course provide future capacity in the Expo when needed.

    4. Very good points. Richard. It sounds like the video was a bit misleading. In general I would say that having a train stop at traffic light is a bad idea. It is too expensive for what it actually gives you. But there are a couple (if nor more) reasons why this makes sense, and it sounds like this line has both. As it turns out, Jarrett Walker mentions both in his critique of streetcars (http://humantransit.org/2009/07/streetcars-an-inconvenient-truth.html):

      1) Existing tracks. This changes the dynamic considerably. One advantage of an effective BRT system is that it leverages the existing infrastructure. Well, the same is true for an existing rail line.

      2) Capacity — Capacity issues are more complicated. There are areas (such as Paris) where surface rail just makes sense. Trying to do the same thing with lower capacity buses would be tough, as demand is such that buses have trouble carrying enough people, even if they run frequently. This is true for 99 B-Line in Vancouver. Even though the buses run every couple of minutes, they are still crowded. It would make sense to replace them with trains.

      But at that point, you have to question whether running on the surface makes sense. The long term plan for the 99-B is to replace it with the subway. I would think it would make sense to eventually do that with this line, but that would be a lot more expensive and harder to justify (there are simply less demand).

      Meanwhile, you have the signal priority issue. The video implied that this had very little of it. But it sounds like it does.

      Headways, signal priority and capacity go together. Obviously it depends a lot on the crossing streets, but I think it is telling that both the Madison BRT and Link have six minute headways. If Madison BRT gets really popular, and they run buses every two minutes, my guess is that they will lose some of their priority. Rail along the Rainier Valley is held to six minute headways for that reason. Madison is an area where surface rail would be a problem (the hill is too steep) but even if it was, it isn’t clear whether it would make sense unless ridership is actually high enough to justify the extra capacity.

      It really varies place by place as to whether the signal priority/capacity issue matters. Before you spend billions on converting the Orange Line to rail, for example, it probably makes sense to improve the intersections (possibly by building overpasses and underpasses). Once you do that, you are probably OK with BRT, as the capacity problems are reduced, and buses running every few minutes are just fine. I guess that is the plan (http://thesource.metro.net/2017/01/23/several-improvements-and-changes-in-the-works-for-the-orange-line/).

      It is easy to assume that it is all about the capacity. When you have big demand, go with grade separated rail. A little less and surface light rail makes sense. Less than that and BRT is the way to go. But it does get more complicated than that, and L. A. shows this.

      Another example of where it gets complicated is with a trunk and branch system. Quite often, the only place where you can justify the extra expense of grade separation is in the trunk. In that case, a BRT system is often the way to go. This is exactly the type of situation that Walker described in the “UPDATE:” section at the end of this blog post: http://humantransit.org/2009/11/brisbane-bus-rapid-transit-soars.html). You can put rail on the whole thing, but that gets very expensive, and probably isn’t justified on the various branches. You can just serve the trunk, but that means lots of transfers. You can build the whole thing but ignore grade separation where it matters most (in the trunk) but then speed suffers. Or you can build some sort of combination. My guess is it won’t be nearly as nice as a solid, well designed trunk and branch BRT line.

  6. The Expo line isn’t a subway because it doesn’t absolutely have to be. Main “decider” about tunneling, whose costs aren’t only unknown but unknowable.

    When the corridor requires it, it’ll be time to tunnel under it, and change its function to a feeder line.

    Bad traffic signal timing due to mistaken priority can be fixed with an order and a keystroke. Same with the MLK line. Same with First Hill streetcar on Broadway going by Swedish Hospital.

    Underground methane is a different matter. Lucky we don’t have any. Not only safer to tunnel. Fossil fuel carries worse politics than NIMBY-ism.


  7. Poncho, I’ve never lived in Portland, but for last thirty years or so have always enjoyed a train ride down there once every few months. Could say I watched MAX grow at same time I drove our trolleybuses, and was in on the start of the DSTT.

    Comparing Portland and Seattle, my impression has always been that Downtown Portland has always been a prettier place than the Seattle CBD, with some beautiful old parks.

    But also got the sense that many gracious things came from wealthy individual donors. In a city with a worse misdivision of wealth than Seattle. Which I think has been getting worse as our forest industries have contracted.

    Am I wrong to suspect that prolonged shortage of public money is responsible for both the lack of security and maintenance, and the condition of many passengers who in better times would have had the income to buy homes?

    And for whom there’s now mental hospital space to cure the results of that last fact?


  8. Really meant “NO mental hospital space.” Term “homeless” as it’s used in politics is a hundred percent evasion of the fact that our only real mental health agency is now our correctional system.

    Since we’re dealing with Oregon, could be fair to argue that Ken Kesey’s portrayal of a 1960’s mental hospital was partly to blame.

    The system was indeed in very bad shape. Intent of “de-institutionalization” was major increase in outpatient care. Intent of the Reagan Administration and its successors was to find another budget item that only hurt the helpless.

    Ken insisted that “One Flew Over The Cuckoo’s Nest” was really a metaphorical indictment of America’s self-inflicted mind control. Real “kicker” at the end is that all the pathetic patients whom the hero has been trying to rescue are there voluntarily.

    I’ve always wanted to get the story from the Chief Nurse’s point of view. Text refers to “Army nurses.” Villainess could’ve been at Chosin Reservoir in Korea. Now stuck with some chicken thief too lazy for the work farm. Latest budget cut probably drove her murderously lobotomous.

    If Max ever gets its own version of MUNI’s Forest Hill Station above a main hospital, hope they name it after Major Mildred Ratched. Because maybe then parents won’t be ashamed to name their little girls Mildred, as they did all those years before the book slandered her.


    1. The untreated mentally ill are a significant and most visible part of the homeless, but the fastest-growing part is people including families who can’t afford the rent spikes, can’t put enough together for first and last month and security deposit, lost their job, find themselves unemployable because they don’t have a job, become increasingly unemployable as the weeks go on and they can’t find a job, or have a major expense like a health problem, relative’s health problem, or their car conks out and they can’t afford to repair or replace it. None of this involves mental illness, and once you lose housing it can be difficult to get back into it, especially with the huge rent spikes, and especially because it’s hard to look for a job and be presentable when you don’t have a place to live or sleep.

  9. Woo Hoo!

    “Sound Transit contractors recently finished mining the last of the 23 cross passages between the northbound and southbound light rail tunnels under construction from the University of Washington to the Northgate neighborhood as part of the Northgate Link light rail extension opening in 2021. Here’s how they did it.”


Comments are closed.