In Seattle we tend to like alliterative self-effacement. Since we already have the Mercer Mess, let’s talk a little about the Denny Dilemma (disaster? debacle? despair?) Pick your favorite ‘D’.
Route 8 provides critically important mobility. A rare non-CBD/crosstown trunk route in central Seattle, it connects Uptown, Cascade/SLU, Capitol Hill, Madison Valley, the Central District, and the Rainier Valley. Especially on its east-west segment, perhaps no route better exemplifies our city’s stated goal of connecting dense urban villages.
The 8 is also one of our least reliable routes, frequently subject to severe delays and bunching, a problem especially severe eastbound in the PM peak. Though Route 8 is a local route its reliability problems are almost entirely due to I-5. I-5’s construction cut off Capitol Hill from Cascade and Uptown, funneling all traffic onto Denny, whereas in the downtown core the grid was better preserved via Madison, Seneca, Pike, Pine, and Olive. If topography and local access problems weren’t enough, the only ramp onto I-5 southbound between Mercer St and Spring St is at Yale Ave, a two-block length section of street that serves exclusively to queue cars seeking to get from Denny to I-5 southbound. Backups routinely stretch over a mile, all the way back to Seattle Center. Eastbound #8s that depart their Lower Queen Anne terminus on time can be up to 30 minutes late reaching Capitol Hill, just over a mile east!
In an era of austerity, fixing Route 8 is no easy task. Denny is very narrow – limited to 2 lanes in each direction except for a turning lane at Fairview Ave – and there is very little room to widen it. HOV/BAT lanes would reduce general-purpose vehicle capacity (slightly more than) 50%, and they would do little for transit anyway as eastbound Route 8 must stay in the I-5 access lane in order to serve its stops at Dexter, Westlake, Fairview, and Stewart. So what can we do with a severe problem and very little money to fix it?
A simple and cheap idea after the jump…
- Move the eastbound #8 stop at Denny/Westlake to the near side of the intersection.
- Remove the eastbound #8 stop at Fairview. Far fewer riders board at Fairview than at either Westlake or Stewart. (see Bruce’s excellent ridership chart)
- Close Yale Ave between Denny and Stewart and convert it to a pedestrian plaza. This section of Yale is less than 100’ long.
- Make Yale local-access only between Stewart and Howell, running one-way westbound.
- Move the southbound stop at Stewart/Yale back 100′ to serve the new pedestrian plaza.
- Restrict right turns from Stewart onto Denny during peak hours and create a right-hand BAT lane on Stewart between I-5 and 9th Ave. This frees up all buses from delays related to Yale and protects tunnel bus access to Convention Place via 9th Ave.
- Redirect I-5 access to Howell St via a choice of 3 alternatives: Minor, Boren, or 7th.
- Move the bus lane on Howell from center-right to center-left, allowing for the two right lanes to store cars waiting to get on I-5. At the lane drop just north of Yale, merge the bus lane to the right to serve the stop at Howell/Yale and provide the same access to the I-5 express lanes.
I’ve included a map to help visualize how this might work:
Drivers that currently queue in a single lane on Denny would instead distribute themselves in 3 directions to access I-5 at Howell/Yale, effectively ramp metering their access while keeping them a bit more out of the way of local Denny traffic. An eastbound Route 8 would serve the near-side stop at Westlake/Denny, immediately move to the left-hand lane, bypass remaining I-5 traffic, and pull back into the right hand lane to serve the pedestrian mini-plaza at Denny/Stewart.
The cost? Remove one bus stop, move two bus stops, change signal timings, add new signage for drivers, and create a (very small) pedestrian plaza.
Drawbacks? In addition to remaining congestion betweeen Queen Anne Ave and Westlake, the absence of a fully-modeled operational analysis limits my ability to predict traffic impacts on Howell. Some diversion to Mercer could be expected, but we must assume that I-5 demand will not diminish significantly. Effective signalization could prevent excessive merging-related problems on Howell and keep transit moving.
I don’t pretend that these changes would fix Denny, but I do think the idea is worthy of full study. I have little doubt that transit would benefit greatly from changes like this.