Is The Tunnel Worse Than Nothing?

According to the state, a deep bore tunnel with high tolls means more street traffic than no tunnel at all.
This post is part of the research project: Seattle's Great Viaduct Debate

A few months back, the transportation consulting firm NelsonNygaard released a fascinating report that looked at what might happen to downtown Seattle traffic patterns if the state builds a tunnel through downtown, while imposing a steep toll on drivers who choose to take it. (We’ve already written about that report once before.)

Below, for your viewing pleasure, is the most important image from that report. Click it to see a larger version — and I’ll explain what it means in a moment.

City and state figures show that a tolled tunnel will snarl downtown traffic, compared with the “Streets, Transit, and I-5″ plan.

There’s a lot of information in the chart, but you can ignore much of it.  The only thing I want you to pay attention to are the orange and yellow bars:

  • The dark orange bars represent the state’s projections for 2015 traffic volumes under the most likely deep bore tunnel scenario, in which drivers would pay $5.00 to use the tunnel during the afternoon rush hour.
  • The yellow bars represent traffic volumes under the “ST5″ or “Streets, Transit, and I-5″ plan, using the same traffic model but with some different inputs.  Under the ST5 plan, the city and state would make substantial investments in transit and “transportation demand management” to reduce traffic volumes, and also make improvements to city streets and I-5 to help improve traffic flows.

The first thing to notice is that the ST5 plan results in lower overall traffic volumes. (Those are the tallest bars to the left; the yellow ST5 bars are lower than the orange tolled tunnel ones.) If you care about greenhouse gas emissions from Seattle’s traffic, this is certainly an important thing to pay attention to.

But the second thing to notice is the length of the orange and yellow bars for center city streets, I-5, and arterials east of I-5.  For all those three corridors, the orange bars are all longer than the yellow bars.  And what that means is simple:  the state’s own traffic models are projecting that a tolled deep-bore tunnel creates worse traffic downtown and on I-5 than the streets-and-transit plan.

It’s simply devastating news for the deep-bore tunnel, because it means that the city and state are predicting that a multi-billion dollar tolled tunnel would actually make downtown gridlock worse.

How is that even possible?

Well, the ST5 plan has several elements that encourage alternatives to cars, by investing in transit and proven transportation demand management strategies.  But the tunnel has no such investments; and worse, it envisions tolls that are so steep—$9 per day for a rush-hour round trip—that the models predict that many drivers will simply avoid the tolls by taking to city streets and I-5.

This point is so damning that I’m sure that someone will claim that the numbers are being cooked. But they’re all official sources, from transportation model runs conducted or authorized by the city and state. The estimated 2015 traffic volumes for the tolled tunnel come from the state’s projections based on the Puget Sound’s transportation demand model; see exhibits 9-11, 9-12, and 9-13 in Chapter 9 of the Alaskan Way Viaduct SDEIS. The numbers for ST5 come from a different run of the exact same transportation demand model, but with slightly different inputs. That model run was prepared for the Alaskan Way Viaduct stakeholder advisory committee, but the consulting firm NelsonNygaard adjusted them upwards (that is, in a direction less favorable to the ST5 plan) to make them more directly comparable to the figures from the SDEIS.

It’s also worth taking a look at Figure 4.7, which shows the projections for 2030.

Projections for 2030 show less traffic on city streets under the ST-5 plan.

Tunnel vision.

Even in 2030, the state is projecting that traffic on “city streets” — the second-darkest bar — will be worse with a tunnel than without one:  115,000 cars per day under ST5, and 131,000 cars per day under the most likely tolling system (Scenario C).  [Note, I slightly edited the paragraph above, based on a helpful suggestion in the comments. I’m talking specifically about the “city streets” section of the bar above.]

This, of course, puts a whole new spin on the ads that are running on blogs around town—suggesting if you’re “sick of gridlock” you should “move forward” by voting to build a tunnel. Perhaps the folks behind the ads should take the time to read the state’s own traffic forecasts.

We are a community-supported resource and we can’t do this work without you!



  1. Kamala Rao says:

    Thanks for crunching these numbers, Clark. I find it very concerning that Gov. Gregoire is implying that the deep bore tunnel will reduce Washington’s carbon footprint, according to a recent article:

    “Gregoire spoke with reporters during a stop on her trade mission to Europe after she visited a tunnel constructed by the Madrid company slated to build one in Seattle:

    ‘It removes any doubt … about why Washington State needs to move in this direction. It was done for them to make their city more sustainable. It was done because they wanted to get rid of some of their carbon footprint.’ “

    • Clark Williams-Derry says:

      The brain in Spain…

      There is simply no evidence for the Governor’s claim. I don’t claim to understand her motivations here. Perhaps it’s wishful thinking, or perhaps it’s a simple misunderstanding, or perhaps it’s just a lawyer’s natural inclination to build a case rather than seek the truth.

      That said, I’m not (yet) convinced that the ST5 plan reduces overall GHGs, compared with a tolled tunnel. The tunnel boring machine will obviously consume a lot of energy, but construction is generally a small part of overall GHG emissions from transportation projects. Obviously, the models predict that there will be less traffic through downtown under ST5, but it’s not clear to me what the impacts are on traffic outside that corridor; perhaps some of the trips that wouldn’t be made through downtown would still be made elsewhere.

      Regardless, as far as I can tell the governor’s GHG claims are entirely speculative. I know of no evidence for the Governor’s claim, and plenty of evidence that suggests she’s simply wrong. If anyone knows what she’s talking about, please share. (But if the Governor is making a factual claim for which there’s no evidence, then shame on the Governor.)

  2. Adam Parast says:

    Just one comment because I see this happen a lot. Higher volumes don’t neccesrily equate to “traffic” if you mean delay/congestion by “traffic”. Delay/congestion and GHG are a function of volume but also a factor of distribution of trips througout the day, trip length, roadway capacity, etc. Lower volume generally means lower congestion/GHG but when you are making such a dramatic change to the transportation network it gets more complex. I can send you a presentation from PSU on this subject.

    • Clark Williams-Derry says:

      I agree!

      Actual congestion effects of the tunnel plan are addressed elsewhere in the NelsonNygaard report. Here’s an example of the effects of a tolled tunnel:

      “Analysis conducted as part of the Partnership Process for the Surface and Transit
      Scenario B (four-lane Alaskan Way) estimated northbound travel time on Fourth Avenue between Edgar Martinez Way and Cedar Street at 12 minutes. This compares with SDEIS travel time estimates for northbound traffic on Fourth Avenue between Royal Brougham Way and Battery Street of 12 minutes for a non tolled bored tunnel and 16 minutes for a tolled tunnel (Toll Scenario C).”

      In short, the models suggest that the tolled tunnel not only creates higher overall traffic volumes on city streets, but longer travel times. To me, the most troubling of the state’s findings was that the tolled tunnel could actually create rush hour backups on SR-99 itself at the south end, with drivers lining up on SR-99 to get off at the last possible exit to avoid eth tolls.

      “Drivers using the bored tunnel for 2015 Bored Tunnel Toll Scenarios A and C are projected to have slightly longer travel times than they would for the 2015 Bored Tunnel due to expected backups on the SR99 mainline. These back-ups would be due [to] heavier off-ramp volumes just before the bored tunnel, which would increase delay at intersections at the ramp termini.”

      I don’t know how much confidence to have in these models. I’m pretty skeptical of them — particularly the projections for future traffic volumes. Still, it seems to be a case where the state cites models to justify building roads (“the models show that traffic is going through the roof”) but then ignores the models when they show that their preferred solution makes the problem worse. Live by the sword, die by the sword.

  3. NotSmartEnoughToRefute says:

    A friend writes:

    “Wow. That’s a straight-up terrible article. Talk about misleading. It kinda had me until Fig. 4.7, a which point he compares the ST2 “plan” (which is where, by the way?) with the tunnel example C, and says there’s less traffic on city streets. Well, that’s true if you consider the tunnel a street. But it’s lower for everything else, and HALF the ST2 for AK Way. I’m now convinced – though not in the way you were hoping, I’m sure.”

    I would appreciate your rebuttal. Thanks.

    • Clark Williams-Derry says:

      “NotSmart”? Don’t count yourself short!

      My responses —

      First, as to “where is the plan.” It’s on the WSDOT website, since WSDOT staff was deeply involved in developing the plan and modeling the traffic implications.

      The ST5 plan, as modeled by the City and State, was created as part of the the Alaskan Way Stakeholders’ Advisory Committee process. That Committee, which was convened by SDOT and WSDOT, included a mixed group of stakeholders: business leaders, transportation geeks, road users, etc. The WSDOT website has a description of the process. A summary of the “ST5″ plan that was modeled by WSDOT is available on the WSDOT website, as is a nifty map that explains the specific street and transit upgrades that were included in the WSDOT modeling effort. There’s more, including a list of the specific elements that were included in the model.

      After studying the issues for a year, the AWV Stakeholders’ committee narrowed down the options to two alternatives: ST5 (with an option for a tunnel later, if ST5 didn’t work as expected), and a new elevated highway.

      The Governor and Mayor Nickels simply rejected these recommendations, and opted for a deep bore tunnel instead. The WSDOT’s timeline clearly shows this: the stakeholder committee narrowed the options to 2, and the governor and the mayor chose a third option that hadn’t passed the committee’s review.

      Second, as to Chart 4.7: When I discuss “downtown traffic” I’m referring specifically to the part of the blue bar labeled “City Streets.” In that bar, you can see the numbers to which I’m referring: 115,000 vehicles on “City Streets” under the ST5 plan in 2030, and 131,000 vehicles per day on “City Streets” under tolled tunnel Scenario C. I’m simply reading numbers off the NelsonNygaard chart here, not trying to be misleading in an way. Sorry if it appears that way.

      I do agree that it’s a bit arbitrary to separate out “Alaskan Way” from “City Streets.” Yes, given the geography and limited connections between Alaskan Way and the rest of the street grid north of Pioneer Square, they serve slightly different functions: the city streets distribute cars & transit into the business core, and Alaskan Way is mostly a bypass route around the downtown core. Still, isn’t Alaskan Way also a “City Street”? But after looking more closely at the details, I thought that NelsonNygaard was right to keep them separate, since the ST5 plan would make major improvements to Alaskan Way so that it can carry more traffic. (See the map.) The same is true for I-5. The tunnel plan does not include extensive upgrades of AWV or I-5 through downtown, because the tunnel itself sucks up all the money. Perhaps the lack of improvements on Alaskan Way are one reason that the models predict that traffic diverted from the tunnel by tolling goes onto city streets rather than Alaskan Way.

  4. Tim! says:

    When I wrote to WSDOT in April asking why ST5 was no longer an option under EIS review, I got this reply:

    “The I-5/surface/transit hybrid alternative studied as part of the 2008 Stakeholder Advisory Committee process was measured against the screening criteria and did not advance for further review because it didn’t meet the objective of providing capacity for the future. Investments on I-5 were needed to accommodate traffic shifted from the viaduct, leaving little room for future regional and state growth. In addition, travel times for trips through downtown on Alaskan Way would be 10 to 15 minutes longer with this alternative. Some analysis of this alternative is included in our 2010 Supplemental Draft Environmental Impact Statement’s Transportation Discipline Report, in Appendix A starting on page 399.”

    Can you unpack this? I don’t even know which end is up.

    • Clark Williams-Derry says:

      That’s a good question. Here’s what I THINK is happening (though I may be no better judge than you).

      For the SDEIS, they are modeling an untolled tunnel. Under that alternative, there’s not much traffic diversion onto city streets, and the tunnel can “handle” the anticipated traffic.

      But the reality is that there’s no money to build a tunnel unless they raise about $400 million in tolls. To raise that much money, they have to put on $5 tolls in the afternoon rush hour and $4 tolls in morning rush hour. With tolls that high, only 38,000 cars or so take the tunnel each day, and lots and lots of traffic gets diverted to city streets. (That’s what the models say, anyway.)

      So the thing that creates the traffic problem for downtown is the tolling scheme.

      It strikes me that the state is trying to have it both ways: the SDEIS models an untolled tunnel, and says that traffic will be dandy! Meanwhile, they’re putting steep tolls on the tunnel, which renders the SDEIS projections moot. As far as I can tell, there’s just a ton of institutional pressure to build something, which is screwing up any hope for a reality-based decision.

      I don’t know if there’s a “solution” here. Maybe the state or the city “find” another $400 million to remove the tolls. Or maybe you do what the Stakeholder Committee recommended for the tunnel: build it only after you whether ST5 works. Then you get a chance to see how traffic adjusts — which means you could right-size the tunnel, or eliminate it entirely if it turns out to be too costly to be worth it.

      • Wells says:

        Clark, consider 2 points regarding tolls and Wsdot’s surface/transit hybrid:

        With no toll, the main body of traffic ‘not discouraged’ from using the DBT is that which originates or terminates along the Interbay/Ballard corridor. That traffic (some percentage of its 35,000 daily average, say 20,000) is redirected from the straight, short, commercial Elliott/Western corridor circuitiously through residential Queen Anne on the Mercer corridor. As if the Mercer Mess wasn’t bad enough, adding this traffic makes it worse. Plus, the entire I-5 to Elliott Mercer corridor becomes a major thoroughfare with a new traffic pattern that adds still more traffic.

        IOW, the DBT plus Mercer West adds traffic that ruins Mercer Phase One (which looks decent alone) and ruins residential Queen Anne. Toll or no toll, central city traffic gets worse.

        The 2nd point is about how Wsdot RIGGED their surface boulevard studies to exaggerate thru-put travel times. The total number of stoplights between Valley and Atlantic studied is 28. 13 stoplights between King and Pike. 5 stoplights north of Denny. 5 stoplights south of King. 5 stoplights in Lower Belltown. The arrangement is Alaskan Way/Western Couplet.

        Wsdot directors and department heads know that a minimum of 13 stoplights for the entire corridor is possible. No stoplights north of Denny, 2 stoplights for Lower Belltown, 9 stoplights between Pike and King, and 2 stoplights further south.

        Wsdot leaders also know that the Alaskan Way/Western Couplet is nonsense. It will only create a chaotic mess of traffic as motorists trying to park conflict with thru-traffic on both Alaskan Way AND Western, not forgetting that mass pedestrian crossings add to the traffic congestion.

        WSDOT is led by crooks hellbent on further ruining Seattle’s street network with the worst traffic imaginable, committing criminal negligence and dereliction of duty on behalf of automobile-related business interests. Douglas MacDonald and Paula Hammond will look perfect in orange jump suits.

      • ScottH says:

        It seems like most of the anti-tunnel arguments based on traffic all rest on the tolling plan. But tolling is a lot more flexible than the physical infrastructure. Once the tunnel is there, it’s there for the indefinite future to provide more mobility options than there would be if it didn’t exist. As to the tolling plan, isn’t it plausible that if diversion becomes too big of a problem that the state could reduce the toll and pay back the money over a longer period of time? Or perhaps they might end up tolling a longer segment of the 99 corridor at a lower rate? Or maybe even start tolling I-5–because that road will need funding for repaving etc. anyway. I don’t see the problems with the current tolling plan as a deal-breaker for building a long-term physical asset like the tunnel because I’m confident the tolling system can adjust, and I think it’s likely to do so in a pretty environmentally friendly way in the not too distant future.

  5. peter herford says:

    Every plan has its plusses and minuses. I have watched and listened and read the debate for decades now. Every group has its studies and all studies show varying results. That’s the best you can do with studies about events that have not happened. Probabilities. Everyone who embraces a study quite naturally picks the study most favorable to their position. Everything from reasoned debate to demonizing has characterized the run-up to the tunnel. Every decision that has led to the final decision to build the tunnel has been fought every inch of the way by opponents. They may, in the end, be right. Personally I am not convinced by the alternatives. Like many citizens I am amazed that it has taken this long to accomplish so little. But finally there has been political leadership that includes risk taking and the tunnel moves forward. It is fortunate that an earthquake has not come along, yet, to prove the instability of the Viaduct the hard way. Are those whose arguments did not carry the day going to continue to take every opportunity to delay or block construction? Projections of traffic flows with factors such as tolls or no tolls and the level of tolls thrown in are half science and two-thirds art. Yes, the same can be said of the studies that convinced the Governor and others to support the tunnel, but at some point a decision is made, the debate closes and the tunnel boring machines need to get to work. If the tunnel succeeds in handling traffic and raising enough revenue to help pay the bills, the tunnel supporters can crow – but shoudn’t. Meantime, are there not other more important issues than the endless tunnel debate?

    • Morgan says:

      Rebuilding a highway through downtown is one of the most important environmental issues of the decade, and I find it disturbing that we can’t find studies that demonstrate that the tunnel would benefit our transportation system, our budget or our environment. As I observe the debate, I find it disturbingly lacking in substance.

      All of the studies that I know about demonstrate that the tunnel option is a poor traffic solution for which we will pay a premium of more than a billion dollars. The only transportation arguments in favor of the tunnel that I’ve seen are limited to individual impacts, like “my commute will be better” or “my business will be helped”. For the vast majority of us, the tunnel will be very, very expensive yet not provide much benefit.

  6. Paddy StClair says:

    Any project that doesn’t include a retrofit of the viaduct is about something else.
    Property values, perhaps even the environment, are the real goals, so any
    Talk about traffic impact is beside the point. Always has been, always will be

    • Wells says:

      There really is no reinforcing the derelict AWV. However, a more elegant replacement viaduct is possible and should be on the table. Mayor Mcginn’s surface boulevard option does not rule out a replacement viaduct nor the more sensible Cut/cover tunnel eventually if necessary.

      The mayor is also correct to begin the seawall replacement as soon as possible. However, Seattle’s environmentalists are easily misled to believe that a Southern California-style beach and artsy-fartsy pier overhangings are a good idea, nevermind how global warming leads to sea rise and more severe storms. James Corner Fields proposals are as ludicrous as Sculpture Park.

  7. Linnea Hirst says:

    Where would the money come from? Is there assurance that the state would contribute to a surface/transit system? Metro and Sound Transit don’t have the money they would need to seriously ramp up transit, and I haven’t heard that the state would kick in current tunnel money to increase transit rather than build a tunnel. (But perhaps I’ve just missed all that conversation.)

  8. Tim McGuire says:

    It seems things must get much worse before they will get better. The problem with that is we’ll all have to live through the “worse” part.

  9. Eric Doherty says:

    All of the ‘alternatives’ on the table seem to assume (and drive) increased driving or vehicle miles traveled (VMT). That means increased oil consumption (largely from Canada’s tar sands or other dirty unconventional sources) and increased greenhouse gas emissions at a time when they need to be radically reduced.

    Is seems to me that the ‘choice’ here is between greater and lesser amounts of public money going to make the greatest problems humanity faces worse.

    Maybe a really significant oil price shock is needed to focus some attention on reducing oil dependence and carbon emissions. Maybe the result would be something like

  10. CML says:

    I’m curious earlier blog posts suggested that Sightline had serious concerns with the State’s model projects for future traffic and congestion, but now this post suggests that State model output of 2030 traffic and congestion are reliable sources – thoughts on this change of perspective?

    • Clark Williams-Derry says:

      Great question.

      I still have serious concerns about the state’s models, though most of the concerns are about projections of future traffic volumes. That is, the models predict that traffic volumes on all urban roads will always go up, and that major roads will be swamped with traffic in another 20 years. But the reality over the past few decades has been much, much different; traffic on many major urban roads has remained flat, while VMT per capita in the urban NW has fallen.

      That said, I don’t know enough about the models’ predictions of specific destination, route or mode choices to make any informed comments. (The models start out by predicting total trip generation — the step I’m most skeptical about — and then parcel out those trips by destination, mode, and route. The first step (trip generation) has proven unreliable, but I simply don’t know how the other steps fare.)

      But the real thing to remember here: I’m NOT endorsing the state’s predictions, or the model results. I’m pointing out a contradiction in one of the most potent pro-tunnel arguments.

      The argument goes: “without a tunnel, the traffic models show that traffic through downtown will be a nightmare.” But that’s true only for models that didn’t include tolling. Once the models factor in tolling diversion, with tolls at the rates necessary to raise the money the state says it needs to build the tunnel, those same models show that a tolled tunnel will make downtown traffic worse than it would be under the solution they rejected.

      So they’re having things both ways: touting the model results that support their point of view, but ignoring the (arguably more relevant) results that contradict it.

      To me, they either need to drop reference to model results entirely (and put forth other reasons for building the tunnel) or admit that the models actually show that a tolled tunnel will make traffic downtown worse than the alternatives (and, again, argue for the tunnel for other reasons). Either way makes for fact-based argumentation, and they might win those arguments on the merits. But the way they’re choosing to make the argument does damage to public discourse, since the argument that they’re making is flatly contradicted by the evidence that they’re relying on.

  11. David Sucher says:

    I don’t need much convincing that the Tunnel would be a disaster and good work on the State’s own data.

    But it’s unfortunate that you don’t mention the Retrofit which is the only practical and politically-possible solution.

    And to those who keep insisting that the Retrofit wouldn’t work, please tell WSDOT and SDOT that the Spokane Street Viaduct couldn’t be repaired and expanded.

  12. Loren Drebin says:

    Unlike Mr. Sucher, I still need convincing that the tunnel would be a disaster. I’m trying to make an informed decision on this, but finding alot of hyperbole, snarkiness, and even vitriol. I’m not rich, not a corrupt politician, not an evil greedy developer, but I find the tunnel plan to have some merit, and problems as well. The issue is not made clearer by claims that the tunnel would be worse than doing nothing. While your headline asks the question (but cmon, like the Stranger, you are implying that it is so), but your article is a comparison to the ST5 plan (which also has merit, but I would like to see an analysis of its impact, particularly during construction).

    • Clark Williams-Derry says:

      I think it depends on the definition of “disaster.”

      I’m not necessarily convinced that building a deep bore tunnel would be a “disaster.” But the evidence I’ve reviewed suggests that…

      ** The state’s own models predict that it would make traffic on city streets, particularly in Downtown Seattle, somewhat worse than ST5.

      ** The same is likely true for I-5. ST5 has money for upgrading I-5 through Seattle to handle additional traffic, but a tolled deep bore tunnel would push lots of traffic onto I-5 without any upgrades to handle the load.

      ** The deep bore creates a scary traffic choke point next to the city stadiums. Because it has no downtown exits, all traffic from 99 S into the city has to go through a few intersections by the stadiums, and then through the Pioneer square area. During one model run, the state found that, during rush hour, traffic from those exits might back up onto 99 itself. I don’t see how this isn’t going to be another Mercer Mess.

      ** The deep bore is expensive. It costs about a billion dollars more than ST5. (Remember, one billion in present dollars is the financial equivalent of having everyone in Seattle write a check today for $1,600 or so. For that much money, you better get a really good road!) There are a heck of a lot of other things that would be better uses of that money.

      ** The deep bore carries high risk. Cost overruns with this sort of project are common, and will likely fall entirely on the city. But if costs begin to escalate, the state may be forced to trim back the project to cut costs. (I base this on conversation I had with the consultants who developed the tunnel cost estimate procedure.) The first things to go would likely be things like bicycle and pedestrian amenities — which would mean that the reasons that some folks give for supporting the tunnel are at risk.

      ** The main beneficiaries of the tunnel are the drivers who use it. So far, tolling studies show that the state would be able to raise about $400 million from them — just a small fraction of the project costs.

      Anyway, I think there are reasons to support the tunnel. Some people will certainly be willing to pay $9 for a rush hour round trip, or less for trips at other times. The state could decide to cut tolls, and pull money from some other projects to service the bonds. The traffic forecasts could be wrong.

      But as it’s currently projected and structured — considering the specific traffic impacts, cost, and risk — it’s hard for me NOT to think of the tunnel as just about the worst available option.

Leave a Comment

Please keep it civil and constructive. Our editors reserve the right to monitor inappropriate comments and personal attacks.


You may add a link with HTML: <a href="URL">text to display</a>