Photo by joshuadf

A valid year-on-year measurement, unlike for Link:

The end of the year data shows continued ridership growth on the South Lake Union Streetcar.  There were over half a million riders in 2010, a 15 percent increase over 2009, and 25 percent greater than ridership in 2008, the first full year of operation.   The gains were driven largely by increased weekday trips.  Average weekday ridership was over 1,800,  peaking at over 2,200 in August 2010.  While seasonal peaks continue to be evident, employment growth in South Lake Union led to sustained ridership growth.  In fact, the month with the highest increase over 2009 was November with an increase of 128 percent.

For a fun but totally unfair comparison, at 1.3 miles it’s exactly 12 times shorter than Central Link (15.6 mi). The streetcar would have 21,600 daily weekday boardings for 15.6 mi at that rate of boardings per mile. Full 2010 Central Link ridership is not yet available, but is likely to be not too far from there.

72 Replies to “Seattle Streetcar Ridership up 15% Year-on-Year”

  1. Hopefully the streetcar line can be expanded northward to Eastlake Avenue, once the Fairview bridge can be retrofitted. There would be a much greater patronage of the line if the residents around Fairview/Eastlake could avail themselves of the streetcar. As it stands now, the odd situation with the streetcar is that during peak commuting hours, the flow of passengers is AWAY from downtown in the AM, toward downtown in the PM. It would be nice to even that out somewhat with commuters heading downtown in the morning from the Eastlake apartments and condos. Right now, very few residents actually LIVE along the line.

    1. Those are pretty respectable numbers considering how short the line is currently. Has anyone heard any recent information on the timeframe for continuing up to Eastlake and the U-District?

    2. The problem with an extension up Eastlake is there are two bridges with issues. The Fairview bridge needs to be rebuilt and I think the University Bridge needs to be upgraded; plus there would be more special work at that bridge to coexist with the ETB wires. I think the Fremont streetcar is more likely in the near term, for that reason.

      I saw on Twitter that the McGinster has formed a committee to start working on the the long-promised rail proposal. I hope some good ideas come out of that.

      Any significant streetcar capital project should include ORCA readers, functional ticket machines, and an extension of the SLUT down Stewart/Olive to 1st.

      1. SDOT has just hired an engineer to do final engineering for the replacement of the Fairview bridge with one that will support the streetcar extension. Actual construction might start in 2011 or 2012.

      2. SDOT posted an RFQ for engineering services for the Fairview bridge replacement in December, so that’s a start.

    3. If they extend it they’ll really have to do something about the slowness on the existing route. Maybe eliminate the 7th and Mercer stops. I don’t remember exactly how long it takes from Westlake-Lake Union but it’s like 15-20 minutes. That’s one thing if it’s the entire line, but another thing if it’s a third of the line.

      1. No, it’s only 10 minutes from Westlake to Fairview, giving the operators about 5 minutes to change to the other end of the train. The 70 is actually slightly slower over the same stretch, depending on traffic.

  2. One other thing I keep wondering about, the streetcars still don’t have Orca card readers. I’ve asked the supervisors what the holdup was, and they didn’t really seem to know. I don’t imagine it’s a technical problem, it doesn’t seem like it’d be any harder to install the readers on a streetcar than on a bus. I’m guessing it’s an inter-agency issue of some sort?

    1. Considering that the vast majority of riders have transit passes (UW, Group Health, Amazon, etc) ORCA readers are a pretty low priority for the SLU line–not much benefit for the cost. I’m guessing their working on it for the First Hill line and will install in SLU after that’s done.

  3. If the streetcar line is approximately 1 mile long, and it takes 11 minutes to go from end to end, that works out to about 5.6 miles per hour. The average human walks a little under 3 mph.

    1. When you add in the stop and wait time for an “average human” at a reasonable number of intersections, where the light won’t be green when you arrive as you walk, you would then have something resembling a valid comparison.

      1. And in practice [BA]’s analysis is right on – if the trolley leaves at the same time I do walking, there’s no way that I keep up.

        But that’s if it’s waiting there ready to leave. Assuming they come once every 15 minutes, that adds an average of 7.5 minutes to the trolley trip. The trolley still wins, but not by a huge amount.

        That said, I’m sure they’re run them more frequently as demand increases, and if demand increases enough they’ll have a good argument for timed lights and even it’s own right of way.

      2. SDOT has experimented with running the line more often. Even 12 minute service would help.

      3. “Assuming they come once every 15 minutes, that adds an average of 7.5 minutes to the trolley trip. The trolley still wins, but not by a huge amount.”

        If you just miss a car, it’s probably faster to walk, even the whole length of the line.

      4. For me the time it take is more a matter of whether it’s nice enough to walk. I’ll do it in the summer, but if it’s as cold as it was today, or raining hard, I’d rather go back into Westlake Center for some coffee.

        And when I’ve got my kids with me, it’s no contest. A 2 year old can’t walk very fast, and is obsessed with riding the trolley anyway.

      1. Getting to the heart of Eastlake alone might justify it, at least during peak hours.

        (Not that U-District isn’t a worthy ultimate goal.)

      2. I think it would be good to make sure it directly replaces a bus route in future extensions so it’s not so much of an extra expense. Extending just to Lynn would still require the 70, while extending to the U District would get rid of it.

      3. In the abstract I agree, Alex. My only thought is that we have (as far as I know) no idea if or when the University Bridge will be ready for a streetcar, and there are myriad routing options and stakeholders to deal with in the U-District: it seems getting there could take awhile and cost more than a bit, to say the least.

        On the other hand, the Fairview Bridge is already being rebuilt and there’s only one feasible route to the heart of Eastlake. It’s a project that can be completed quickly and cheaply while significantly boosting ridership and increasing the route’s utility. And there’s no reason it would have to impede progress on a U-District and/or Fremont extension.

    2. Amazon’s consolidation and a summer with the new SLU park open will probably kick ridership up enough this year to really make this discussion take off.

      1. I think also all the people moving into the new units will help as well. Didn’t they just have a few auctions recently? Population must be increasing in SLU.

      2. No condo auctions in SLU that I know of. Perhaps you’re thinking of Olive8 which is near Convention Place? There were two 2008 condo buildings that converted to apts (Rollin St and Equinox).

    3. The Eastlake Community Council is very much against an Eastlake Avenue streetcar.

      If you go to a meeting and talk about the possible route, you will be corrected by the board members that the Streetcar is neither a Streetcar nor a Trolley, but “Rail”, and that it will be noisy, harm businesses by reducing on street parking and all the usual anti-streetcar arguments.

      I would like to see a scientific poll done of the neighborhood in order to gauge what the level of popular support is in the the neighborhood for an extended streetcar.

  5. The interesting thing about SLUT is that it basically covers the one (and only) route in the Puget Sound that has enough density to justify mass transit.

    1. ??? How do you figure? the uptown->downtown route sees a bus about every 2 minutes or so when you count the 1,2,2x,3,13,15,15x,18 and 18x (ignoring the 3 and 4!). Most of these are full or even packed during commute times. Even further routes like Ballard have around 7 minute frequency of packed busses – certainly enough to justify mass transit.

    2. John, you realize the density came after the streetcar? Or is that not convenient for you?

      1. Ben, do you realize that the density had absolutely nothing to do with the streetcar? Or is that not convenient for you?

      2. The streetcar got a private contribution in order to encourage future density. What’s wrong with that?

        In terms of comparing densities, it would seem prudent to wait until March or April and compare current densities, not 10 year old data or estimates.

      3. About half of the original capital outlay was private contribution and none of the continuing operating cost (ten years of O&M will equal the private contribution). In exchange Vulcan got zoning changes which resulted in a much greater increase in property value than what was contributed to the SLUT. Don’t forget that most of the Amazon jobs moving to SLU are leaving an empty shell on Beacon Hill (which Link conveniently bypassed).

      4. While you are correct that Beacon Hill and parts of downtown will be hurting for the loss of Amazon workers unless someone new shows up, a lot of the Amazon employees in SLU are new hires… “headcount has grown 44 percent over the past year”:
        http://www.techflash.com/seattle/2010/10/amazon-headcount-grows.html

        Apparently there are a lot of Kindle-related jobs out there. The SLU hq is also massively larger, SLU buildings are going to be nearly 2 million sq ft while PacMed is 260,000 sq ft according to this:
        http://www.cityofseattle.net/ban/development-amazon.htm

      5. Amazon also had a lot more offices at 605/705 Union Station. More square footage than PacMed, I believe. Also a couple of floors in the Columbia Tower.

      6. Amazon will still be occupying space in the Pac Med building even after the move to SLU. Quite a bit of the space in Pac Med is still medical offices, so it’s hardly an empty shell. Most of the space that Amazon is vacating is Vulcan owned space in the ID.

      7. “Vulcan got zoning changes which resulted in a much greater increase in property value than what was contributed to the SLUT.”

        Really, Bernie?

    3. John, I thought Capitol Hill had the highest density in the western US? In 2 years, the First Hill Streetcar will be rolling down Broadway and covering that area.

      1. Michael, I think that should be “in the PNW.” I’m pretty sure a few neighborhoods in San Francisco trump Capitol Hill.

      3. I think Chinatown in San Fran is the most densely populated neighborhood in the Western U.S. followed by Koreatown in Los Angeles.

      4. As of the 2000 census, here are the densest census tracts in Washington State, and their densities (in people/km^2):

        74 (west slope of CH): 17,035 people/km^2
        53.01 (U-District): 16,498
        85 (First Hill): 12,825
        80.02 (Belltown): 12,381
        83 (First Hill): 10,436
        84 (CH/First Hill): 10,145
        70 (LQA): 8,668

        The densest census tract in Kent is 294.08, with 3,284 people/km^2.

        Admittedly, this data is 10 years old, but forgive me if I’m skeptical that any part of Kent became 6x as dense in 10 years.

      5. As far as California goes, let’s not kid ourselves — there are 56 census tracts between SF and LA which are denser than any part of Seattle. I didn’t check other California cities because the simple FactFinder interface can’t handle that much data.) The densest parts of SF/LA are almost twice as dense as the western slope of Capitol Hill.

  6. Get it to First Avenue and usage will skyrocket.

    Get it to Eastlake around Boston St., and usage will skyrocket regardless of whether it goes to First Avenue.

    1. One of the advantages of Eastlake as a corridor is that getting to the U-District would allow Metro to eliminate route 70 and use those service hours for the streetcar (and probably take down the trolley wire on Eastlake, too). If you don’t get to the U-District (or at least to the bridge), you can’t really do that.

      Metro will probably want to rearrange all the 7x busses to avoid downtown once U-Link goes in, so if you can get the streetcar to the U-District, you can replace five local busses in that corridor (+ maybe 66/77) with one streetcar that should have better trip times and higher capacity.

  7. It cost over $50 million, and averages 1,800 boardings per weekday? And it is basically free to ride if you have an ORCA card.

    Any way you look at it, it is an utter waste of money.

    1. Yes, a waste of money just like all transit and bike projects. If only we’d spent the $54m on a dozen inches of bored tunnel instead.

    2. Most of the cost was picked up by SLU land owners. Are you a SLU land owner, Norman? If they thought it was an utter waste of money they wouldn’t have voted to tax themselves to pay for it.

  8. Can the cars used on the SLUT (which will presumably be similar to the ones used on any extension) be hitched up to run in pairs?

      1. Their website leaves a little to be desired. They say that a custom model was developed for the “city of washington, U.S.A.”.

  9. A couple of observations:

    I’d second the utility of it running down to first avenue. I’d love to be able to get all the way to the waterfront on it especially during lunch and I can imagine alot of tourist uptick. I’d take that over an extension to UW which duplicates to some extent LINK service and the existing 70 buses.

    That said it never seems close to full yet whenever I ride it and I’d just invest more money in the 70’s over it, which do seem to be chronically extremely crowded especially the 76.

    Metro also really needs to start up more express bus service that ends in SLU in the long run as well if employment and density continue to rise in the area. Routing everything through the DTTT adds at least 10-15 minutes to the connection.

    For comparison: by car from my house in NE Seattle SLU is 10 minutes off peak and maybe 20 minutes during rush hour. I tend to bike most of the time which takes about 25 minutes. The bus connection is closer to 45 minutes. I’d be happy with even a consistent 30 minute connection via transit.

    Ben

    1. Just UWMC to SLU takes me 30 mins by bike, but I’m slow. :)

      There’s currently a SLU/QA mini mobility study being done that will propose adding SLU stops for freeway express buses–probably using Mercer express lanes instead of Stewart, then stopping at Fairvew and Harrison or so.

  10. A few of observations:

    I’d second the utility of it running down to first avenue. I’d love to be able to get all the way to the waterfront on it especially during lunch and I can imagine alot of tourist uptick. I’d take that over an extension to UW which duplicates to some extent LINK service and the existing 70 buses.

    That said it never seems close to full yet whenever I ride it and I’d just invest more money in the 70’s over it, which do seem to be chronically extremely crowded especially the 76.

    Metro also really needs to start up more express bus service that ends in SLU in the long run as well if employment and density continue to rise in the area. Routing everything through the DTTT adds at least 10-15 minutes to the connection.

    For comparison: by car from my house in NE Seattle SLU is 10 minutes off peak and maybe 20 minutes during rush hour. I tend to bike most of the time which takes about 25 minutes. The bus connection is closer to 45 minutes. I’d be happy with even a consistent 30 minute connection via transit.

    Ben

  11. For reference, Amazon Phase III should be occupied anytime now and Phase IV (the largest so far, two 11 story buildings) should open in April. Phase V has not even broken ground yet, and neither has UW SLU Phase III. There are also two multistory older buildings being fully renovated right now:
    http://thesouthlake.com/2011/01/08/slus-major-renovations-continue

    Unfortunately housing construction has been at a standstill. Hopefully a couple permitted apt projects on Dexter will begin in 2011.

    Also, the Hutch continues to expand, they just bought an empty Blume building:
    http://seattletimes.nwsource.com/html/businesstechnology/2013707333_hutch18.html

      2. Whoops, my other comment was supposed to be replying to yours. See below. :) Sorry!

        Another interesting data point is the people mover in the Seattle airport. According to Wikipedia, the system was designed to have a capacity of 14,400 people per hour. It cost about $48 million/mile (2011 USD). And in 1999, the Port spent about $186 million (2011 USD), or $109 million/mile, to refurbish the system.

        This tells us two things:

        – Even when the right-of-way is free (e.g. airports), these systems are expensive to build.
        – Transit spending has far outpaced inflation. The Morgantown PRT would be much more expensive to build today.

        If Link was entirely driverless, it very well might have cost twice as much to build, though operating expenses would be far lower. (It’s unclear how much of the capital cost could have been recouped by having smaller trains/stations.) Would I have voted for that? In a heartbeat. Would that measure have passed? I don’t know.

  13. Grade-separated transit with a guaranteed wait time of less than 5 minutes will always attract higher ridership than less-compelling options.

    What’s notable about Morgantown is primarily that it’s automated. By eliminating the need for a driver, they were able to provide far higher frequency service than you would expect from a city its size.

    This is a general truth about automated systems. See Jarrett Walker’s excellent article about SkyTrain. In Vancouver, the SkyTrain runs every 2-4 minutes, 20 hours a day, 7 days a week. The Miami Metromover is the same. And many airports have people movers which have a far higher frequency than any transit service in the nearby city.

    But there’s a catch. Automated systems need to be fully grade-separated (for obvious reasons). This means more expense. Morgantown’s PRT cost $60 million/mile (in 2011 USD), despite having a maximum capacity that is on the order of 10% that of Link’s maximum. And acquiring ROW there was surely much cheaper than it was here.

    Now, there is one way in which the pint-sized nature of this system does make a difference. Yes, each car can fit 20 people, but if the cars run every 15 seconds (which they apparently could), that’s a capacity of almost 100,000 per hour. Smaller cars are cheaper (though you need more of them), and smaller stations (especially underground) are *much* cheaper.

    So maybe the real question is: how much money could we save by building rail systems with much smaller vehicles and stations? And are the savings enough to pay for grade separation and full automation?

    (In a city like Seattle, where grade separation generally means tunneling, the answer is probably no. But I’d love to be proven wrong. And if building an elevated system would make it cheap enough to be viable, that’s worth noting too.)

    1. The reason the Morgantown PRT was so much more expensive than it should have been was political rather than technical. The wikipedia article you cited provides some background for this, but a synopsis is that Nixonian meddling caused design and management changes to the point where that system was being redesigned almost as it was being built. The Morgantown “phase 1” guideways were built to handle the largest vehicle that might have been used on them, which was the size and weight of a full-sized bus. This was rather dramatically overbuilt compared to the van-sized vehicles that were eventually deployed. The “phase 2” expansion used substantially lighter and less expensive guideways.

      Your point about transit spending outpacing inflation is valid, of course, but you would need to start your cost calculation for building a transit system similar to the Morgantown PRT from a somewhat lower starting point than the inflation-adjusted $60 million per mile figure you cited.

      Bottom line, yes, automated systems need to be fully grade-separated, but a fully grade-separated PRT system should cost substantially less than any fully grade-separated alternative.

  14. before the STB concludes the SLU line is doing well, it could consider the opportunity cost of the service subsidy. in fall 2009, service were extracted from the south-first Link integration to subsidize the SLU line operation. (wat is about 17k annual platform hours?) on Seattle routes those hours could have attracted more than 40 rides per platform hour (680,000 or substantially more than the one-half million). each streetcar hour cost more than a bus hour, so it is not a one for one comparison. some of the streetcar hours could have been spent to improve the headway on Route 70 that already extends to the U District and penetrates downtown Seattle better than the streetcar. Routes 26-28 and 70 each provide 12 to 15 minute headway service within three blocks of the streetcar line, so coverage is not the issue. as others have posted, the SLU line is only 1.3 miles long and has 15-minute headway, so it does not provide much advantage to pedestrians.

    1. Headways on the 70 are already at 10 minutes at rush hour, 15 minutes during the day. You can’t run diesel busses at much less than 10 minute headways without platooning and making traffic suck even more; ETBs are even worse due to the their inability to overtake. At some point it makes sense to invest in something with more capacity per vehicle.

  15. Route 70 now has 12-minute peak headway. between 1940 and 1963, etb routes 7 and 8 served Eastlake with a combined peak headway of 2.5 minutes. yes, at some point it make sense to invest in something with more capacity, but that corridor has not reach that point and a slow Nickels-Drago-Vulcan streetcar is not the something.

    1. According to the schedule it’s 10 min:

      http://metro.kingcounty.gov/tops/bus/schedules/s070_0_.html

      A center-lane center-platform SLUT tram running in traffic will carry 50% to 75% more people (sitters + standees) than a 60′ bus due to the increased usable floor area, and, more importantly, off-board payment and four step-free doors at optimal locations in the carriage will make it possible to get these people (+ bikes/baby strollers/wheelchairs) on and off without the delays that plague bus loading. It does all of this without requiring Metro to increase service hours for this corridor, taking any parking, or dieselizing an electric route.

      As for what the trolley schedules from 70 years ago say, I don’t care. Bus service — especially ETB service — platoons like mad at those headways, that’s a matter of empirical fact. I frequently take routes 1/36, 2, and 13 off peak to and from Downtown to Lower Queen Anne when they (with the 3 & 4) have a combined headway of five minutes. They are almost never on time, and usually show up in bunches.

      Finally, what else do you propose? There’s no dedicated ROW or a bus-only lane in the foreseeable future in Eastlake. You have to run in traffic. Tell me what is going to do better than a streetcar.

  16. streetcars have to funded. where is the funding? Vulcan has done their part. extending the streetcar to the U District would have some extraordinary costs: Fairview bridge, Eastlake water main, University Bridge. SDOT could start by letting Route 70 stop in lane.

    2. The Fairview bridge is already scheduled to be replaced with a modern one that will be able to take streetcars. University bridge requires only relatively minor work. None of the costs are “extraordinary” — they’re pretty normal for rail construction in a city.

      Funding is tight right now, no doubt. I’m not suggesting we should necessarily go out and build this streetcar in the next year. I’m looking three to four years ahead.

      I am suggesting that this corridor is (a) already dense from Westlake to Mercer, from Valley St to Boston St, then again from the I-5 bridge to the end of the 70 wire (b) has an excellent chance of becoming denser in the rest of its length and (c) is already operating close to its maximum bus capacity at peak times.

      Constructing a streetcar would (d) expand capacity to accommodate existing ridership with room to grow, (e) encourage the densification in (b), and (f) capture additional choice riders who might not take a bus. All of these are excellent public transit goals.

      This could be funded from a variety of sources including the citywide TBD that’s not maxed out yet. Presumably there will come a point in time when the city is less broke. When that time comes, a streetcar in this corridor would be an excellent investment in the future.

Comments are closed.