In keeping with STB’s generally technocratic ethos, I think tolling I-90 is good policy, not to fund more megaprojects but because it will improve travel times both for people who choose to pay the toll and (more importantly) for transit riders that will no longer have to deal with chronic congestion. Olympia’s tolling rhetoric never seemed to grasp this and viewed it as a purely revenue-generating measure.
However, the Times reported ($) last month that the governor’s transportation package, by funding the new 520 bridge with other revenues, would also “abandon the idea of tolling the I-90 Mercer Island floating bridge.” While I still think this is a bad policy, I surprised myself by not being particularly concerned about it.
That’s not just because of skepticism about the package’s chances. A toll on I-90 a few years ago, now, and for the next few years would indeed alleviate the suffering of transit riders. Processes being what they are, however, any plausible toll would probably begin about when two-way HOV operations start on the bridge in 2017. At that point general-purpose lane congestion becomes theoretically irrelevant to bus riders; six years after that, and East Link will make it entirely irrelevant.
Stepping beyond transit for a moment, it would be great for freight to flow freely and for people to choose between an inexpensive but fast transit ride and a more expensive, even faster drive. But I can also accept a cheap, fast transit ride vs. a cheap slog through traffic. If freight and SOV interests can’t, they certainly have the platform to demand what they need without transit advocates concerning themselves with it.
Have all the decisions now been made with respect to how I-90 will operate when E-Link opens.
1. Roadway configuration – Ramps HOV 2 or 3 Plus and NO tolling for now.
2. Which buses will truncate to a LINK station (S.Bellevue or Mercer) and why. Pushing riders onto Link surely effects the boardings shown in the EIS. Has that changed, in light of the uncertainty of how Mercer Stn will be oriented?
3. Frequency of trains and make up of consists. I’ve heard and read all sorts of things over the years from you can only have one train at a time on the floating section which is why the EIS only shows 9 minute headways and 2 car trains but the latest Ops Plan shows 6 minute headways in both directions in the peak, using 4 car trains at times. That’s a huge difference.
All the above is from memory, but wonder if it’s nailed down in one document. The latest post on density from LA and discussion on E.Link (vis a vis I-90 capacity and tolling issues) is really getting convoluted. Of course, a stop and go I-90 can’t hurt Link ridership.
1. The R8A project will result in 3 GP lanes and 1 HOV lane each direction in the outer roadway (mainline). East Link will have exclusive use of the center roadway (express lanes). The working assumption seems to be HOV 2+. Most existing ramps are being retained/reconfigured with the exception of the 77th Ave SE reversible ramp (closed) and the D2 ramp from Rainier to 5th Ave and the DSTT entrance (light rail only last I heard).
2. Not determined yet and won’t be for years, likely 2021 or 2022 as East Link operation gets closer. The current very early working assumption is for I-90 routes (212, 214, 218, etc.) to terminate at Mercer Island and I-405 (111, 114, 566/7) routes to go to South Bellevue. Among other reasons, this is due to ramp configuration. For I-90 routes, there are HOV direct access ramps at 80th Ave SE on Mercer Island, while getting to South Bellevue requires “swimming” from the inside HOV lane after serving the Eastgate freeway station to the outside Bellevue Way exit. This is doable (ST 556 does this) but a potential safety problem due to the short distance involved. The I-405 ramps to I-90 and Bellevue Way make it much easier for I-405 traffic to access South Bellevue, although it requires “swimming” on I-405 but that happens anyway.
There is no issue with the orientation of Mercer Island station; it is in the center roadway between 80th and 77th Avenues and hasn’t changed at all. Bus access options at that station continue to be discussed and nothing has been decided.
3. Take the following with a huge grain of salt because it is hearsay: ST does not seem to have any issues with multiple fully-loaded 4-train consists on the bridge simultaneously.
I’ve asked the question about about multiple trains on the bridge directly to ST staff and have not gotten a straight answer beyond they’re not worried. If they do have a problem or concern, they’re not saying (publicly), but neither have they confirmed that such a thing is not possible. It bears watching, but even if it is a problem proper sequencing would minimize this. At this time I have no reason to think headways are severely limited.
The I-90 Integrated Transit Service Study, Appendix B.4 (page 40), shows 2011 PM peak ridership (a 3-hour period) at approximately 4300 across the floating bridge (both directions combined). 2035 FEIS ridership is ~13,700, of which 12,100 are on trains and the balance on buses. The bus integration would shift everyone to trains and result in a slight bump in ridership to 13,800 in 2035, split 7,800 eastbound and 6,000 westbound. 10 minute headways with 4 car trains yields 3,600 pphpd (@ 150 per car), or 10,800 total per direction. 3 car trains gives you 2,700 pphpd, or 8,100 total. Either way, sufficient capacity exists at 10 minute headways out to 2035. Based on ST’s ridership models, less than 10 minute frequency doesn’t seem to be warranted, even at peak, although I can see how it may be necessary on the interlined segment north of IDS.
Thank you so much for posting that Jason. That’s the clearest straight answer I’ve seen.
Any thoughts on getting to Eastgate using the center roadway lanes to I-90, and then the median. I know the section through the slough is suspect (sinking in peat bogs), but wonder if there’s a best guess at a plan A.
Looking at your PM pk, EB 7800 riders on Link over the 3 hours is interesting. That same load could be replicated in double tall buses, running on about 2 minute headways, and save a Link/Bus transfer at either MI stn or S.Bellevue stn. Being buses, those same center roadway lanes would be free for another 2,000 HOV 2+ vehicles in each lane, each direction per hour.
That would have allowed buses to divert at I90/405 into 3 different directions to final destinations. The added HOV center roadway traffic would have added to the overall capacity of the structure, and allowed some HOT lane revenue also.
Of course we’ll never know that, so it’s all pure speculation at this point. I suspect starting with HOT center-lanes would have been politically more doable than tolling the entire structure – you know – getting the camels nose under the tent.
Sorry, coming back to this late after being offline last night.
I have no good ideas for crossing the Slough towards Eastgate. The Slough presents major engineering problems on its own; I saw a WSDOT report that basically says they almost had a major failure of one of the I-90 bridges back in the 1980s when it was discovered the entire structure was shifting (I can’t find the report, but it is referred to in some of the East Link EIS documents). If you look underneath today there are cables joining the sections to keep it from shifting too much. However some of the reports commissioned by Bellevue when the alignment was still being debated claim an alignment across the Slough is technically feasible, if expensive.
A further problem is getting past 405. WSDOT has plans for HOV direct access ramps between 405 and 90, and while I don’t think they’ll ever get built, WSDOT is protective of the physical space needed to build those. The only two viable options I can see are to go straight east across the Slough and deal with the peat bogs and then dive into a bored tunnel under the 90/405 interchange to get to Factoria and then continue tunneling up to Eastgate (eyeballed at ~5% grade). That would be stupidly expensive, but there’s no way you could go aerial over the interchange anyway; WSDOT’s dream HOV ramps are up there and the climb from Bellevue Way would be excessively steep. The other option is to branch off from much further north, near SE 8th possibly, and then run down either the east side of 405 or Richards Road. I’d guess this is the current Plan A for cost reasons alone, plus it would more neatly tie into a potential future line from Kirkland.
As you say we’ll never know about I-90 BRT options, but it goes to show how poorly East Link performs when its capacity would be only marginally necessary by 2035.
One other note to add, while nobody knows exactly what routes would go where and won’t for years, the focus now is on ensuring the necessary infrastructure exists to accommodate different options. East Link is in final design now, so this is the opportunity to make sure we at least have options in the future instead of having more Mt. Bakers or TIBS. To their credit, ST seems to have learned from those.
Thanks again. Lots of piles in the slough to support the 15+ lanes going all over the place through there, for sure. There are few good options for providing an extension to Eastgate and beyond, because the split tails would have to merge, then merge again with 6 minute headways from Central Link. 12 minute headways on either tail kinda sucks in the peak.
Looking at the narrow shoulders and narrowed lanes on the floating sections shows that any break downs or collisions will cause major backups with only a couple of lanes getting through. The 6′ shoulder in the HOV lane puts a disabled bus and the door about half into the HOV lane.
As you can tell, I’m totally underwhelmed by this deployment of mass transit for E.King Co.
When they did the weight test on the center roadway a few years back they used the equivalent of fully loaded 4 unit LRVs so it shouldn’t be a problem
As for restricting service to one consist at a time that will most likely be defined during the test phase once they have determined what the affects are upon a train entering the floating segment.
So you’re arguing that transit only matters between South Bellevue and Seattle? It’ss still a problematic policy, and it douses water on the PSRC’s regional plans for systemwide tolling. Arguably, this is bad for transit everywhere for all time. Or at least until state political leadership decides to stop pandering.
Let’s just toll everything and be done with it… Toll gates at the north and south side of downtown Seattle and on both bridges. Interstate truck traffic can use 405. Toll all cars equally and use that to find transit, that way if you don’t want to pay tolls you ride Sounder, Link, or a bus. Run Sounder all day every 15 minutes in each direction, push the freight over to the UP tracks, run an extra track from Black River to King Street. Run Link every 2 minutes. Run a bus every 20 minutes. Problems solved.
Do this but at the same time make sure the legislation includes some sort of decrease in gas tax or sales tax. Start the move away from using the wrong tax mechanisms for transportation. Both of those are already decreasing anyways and it will make people more receptive to new taxes if it’s not just additions.
“Run Sounder all day every 15 minutes in each direction, push the freight over to the UP tracks”
At $50m a time slot being paid to BNSF, you’ll need about $3bil just for trackage rights, with the UP smacking their lips for an equal amount to ‘rent’ their right of way, provided ST adds the 2nd and 3rd mainlines. Kent got screwed and will never recover. I hope the I-90 folks fare better than we did.
“Let’s just toll everything and be done with it…”
New York City couldn’t pass a “cordon toll” (and they really need it). So good luck passing it in Seattle.
London (UK) did successfully pass a “cordon toll”.
And even London wasn’t able to keep the western extension of the congestion charge.
Will the changes to 2-way HOV and light rail bring about a net decrease in vehicles getting on and off the bridge? If not it would be great to have tolling that would reduce the peak traffic north of Rainier & MLK (some of it no doubt attempting to avoid I-5 backups that also begin right at the I-90 exit and then push south), which would then make it more acceptable to reduce the number of lanes there and make crossings from the Mt. Baker light rail station to the transit center easier, benefit future Mt. Baker town center projects and so on. Or would a decrease in cars crossing the bridge quickly induce an equal amount of replacement traffic going to and from destinations to the north and south of I-90 in the same area?
Once upon a time, there was a legistature-approved study of HOT lanes on I-90 from Issaquah to Bellevue. I’ve not been able to find whether the study was completed or what the recommendation/likely next steps were. Anybody know what ever happened to that?
HOT lanes off the bridge can manage some of the access to the bridge, and surely in a more politically palatable way. Managing at least some of the access to the bridge buys us a lot of the benefits of tolling the bridge itself (at least westbound).
I do not understand why people keep talking about this. In the absence of an increase in capacity or replacement of a functionally obsolete structure, the bridge simply cannot be tolled. It was built using Federal Interstate Highway Act funding and they are protected from tolls except for the specified reasons.
Someone suggested “Oh, just end I-90 and I-405.” I seriously doubt that the Federal Highway Administration would put up with that for a minute. Washington designed I-90 to connect with I-5 from the very beginning of planning for the system. Washington (D.C.) would see it for the fraud it would be.
Oh, and the inclusion of a reverse direction HOV lane does not equal “increased capacity”. There are two lanes across the East Channel and three across the main body of the lake which are being removed for the LRT system.
It’s increased capacity if you’re typically travelling in the opposite direction of the express lanes.
Not true at all.
The old 1956 (?) policy prohibiting tolling on interstates has been repeatedly watered down in recent years (including just recently), and the R’s in the other Washington are increasingly oriented towards allowing generalized tolling in urban areas since doing so is becoming increasingly disconnected from the concept of rural tolling.
That said, the addition of full time bi-directional HOV lanes and LR to the bridge surely counts as increased capacity and would allow tolling even under existing policy.
We will see what the R’s come up with in Oly when they actually have to find revenue to pay for their transportation plan. It is easy to say you will do it with efficiencies and other motherhood type statements, but actually making it work is another matter. And exempting highway projects from sales tax effectively just takes money from the general fund and social services and will never get past the gov’s desk.
That said, the main reason for tolling I-90 should be “congestion management” and not “revenue”. Since the toll went on SR-520 the number of toll evaders driving N/S on I-5 and I-405 through Bellevue and Seattle has significantly increased congestion on those routes and on adjacent streets. Put a toll on I-90 and you would return balance to the cross lake commute and reduce congestion on I-90, I-405, and I-5.