Oct 14

Thoughts on Highway 47 – Part Two

Note as of 12/4/2014: This post was originally written during my run for Forest Grove City Council during the 2014 General Election.  It does, however, cover currently active developments with Highway 47.

Thoughts on Highway 47 – Part Two

Part Two of my policy essay series on Highway 47 will focus on my stance with respect to the current plans for improvements, and my take on what should actually be done. To briefly summarize my points from Part One:

  • The real safety issue with the two “flashpoints” of the Highway 47 situation are the intersections with Verboort/Purdin Roads to the north, and Fern Hill Road/Maple Street to the south. The main problems at these intersections is for traffic on the crossroads to determine when it is safe to cross or turn onto Highway 47.
  • The lack of east-west connectivity in Forest Grove poses issues, many of which are interlaced with the systemic problems with Highway 47. Verboort Road and Fern Hill Road form segments of the circuitous east-west routes that many Forest Grove residents use to get to other parts of Washington County, or as alternatives to Highway 8.
  • ODOT’s attempts to fix the Verboort/Purdin intersection in the past decade were “token” efforts that proved largely ineffective, and possibly even detrimental to safety at that location, because they focused on the wrong problem.

Current State and County Plans

Now, as to ODOT’s plans, currently, there are no real plans for the Fern Hill/Maple intersection, though on September 29th, they announced plans to do a “safety audit” in that area. The funds for the audit have not yet been located, and there is no timeline on when this audit will take place.

The Verboort/Purdin intersection on the north end, however, is a different story. In the wake of the 2007 tragedy there, ODOT studied the intersection and came to the conclusion that the best solution was to convert it into a multi-lane roundabout, a project for which they did not have the funding. Friends of the victim’s family had offered to chip in for the full cost of a traffic signal there, a solution that would have solved the main issue of determining when it is safe to cross the intersection from Verboort and Purdin, but ODOT turned them down. Reportedly, they never gave the family a direct explanation for the refusal, though as I noted in Part One, ODOT representatives have made public remarks to the effect that motorists would simply “ignore red lights”, should a signal be installed at that location.

After the April 2014 collision that killed two Pacific University students, ODOT agreed to hold a town hall meeting about the intersection in May, at the behest of Mayor Pete Truax, and in conjunction with Washington County’s Department of Land Use and Transportation (LUT), and Senators Bruce Starr and Betsy Johnson, to discuss the future of the northern segment of the highway. I was at that town hall, and the positive response I received from some city officials, about the comments and questions I posed to the ODOT and LUT representatives, was one of the things that truly convinced me to enter the council race this year.

The plan that the ODOT representatives detailed at the town hall involved expediting the roundabout plans, by getting $750,000 funding in place as soon as possible to cover the engineering costs. They also announced that they planned to work with Washington County LUT to coordinate efforts with the David Hill Road extension, which was also to be accelerated from its original 2018 completion timeline, due to the fact that that road would intersect Highway 47, between Verboort/Purdin and the Sunset Drive/Beal Road intersection. They also announced the latest batch of “token” fixes (the 45mph zone, and the Through-Route Activated Warning Sign system) that I discussed in Part One. In June 2014, ODOT announced that they planned to have Washington County LUT and engineering consultant firm CH2M Hill (the same group responsible for the earlier Verboort roundabouts) handle initial phases of engineering. Construction is currently scheduled for the latter part of 2015.

Are Roundabouts Appropriate for Highway 47?

You may notice the plural—roundabouts—on the above heading. Washington County LUT Director Andrew Singleakis remarked at the town hall meeting in April that the likely treatment for the new David Hill Road intersection with Highway 47 would be a traffic signal. However, since that time, LUT has been quiet about the plans, but all information suggests that they are now considering installing a second roundabout on Highway 47, at this new intersection. Per page 36 of the September 15, 2014 agenda packet for the Forest Grove Planning Commission, which discusses the Silverstone subdivision planned for the immediate vicinity (a subdivision I oppose, by the way), city planning staff note that “[r]oundabouts are proposed for both [David Hill and Verboort/Purdin] intersections”. ODOT’s information page for the Verboort/Purdin intersection also notes that they plan to “[e]valuate the two other nearby OR 47 intersections”, referring not only to the future David Hill connection, but the existing signalized Sunset/Beal intersection. Based on the trajectory of the Verboort/Purdin and David Hill plans, it would not be a stretch to speculate that this short stretch of highway could be the site of as many as three roundabouts, by the time ODOT, LUT and CH2M Hill get done.

A number of city residents with whom I’ve discussed the situation, even those who have expressed that they are fine with the existing county-built Verboort Road roundabouts from 2003, are concerned by the prospect of even a single roundabout at the Highway 47/Verboort/Purdin intersection. Their trepidation stems from a feeling that roundabouts do not belong on a high-volume rural state highway. While some residents and officials are simply happy that the state is finally doing something at that intersection, the general impression that I have gotten is that many Forest Grove (and Banks) residents would prefer a signal over a roundabout at this intersection. With the types of changes already made at the intersection as part of ODOT’s 2005 turn lane project, almost all the preliminary work required for a proper signalized intersection is already in place, and the average signal, per ODOT’s signal brochure, costs between $300,000 and $500,000—at most, only two-thirds of the cost of merely designing the roundabout, to speak nothing of building it.

As far as the full costs through the construction phase, ODOT has thrown out a figure of $3,000,000-$5,000,000 figure, but per the project page, this figure is apparently only for a single-lane roundabout, rather than the two-lane setup the agency is actually leaning toward building. Per ODOT’s page, the two-lane design would “significantly impact the final footprint of the project” and “could increase the cost by 30 to 50%”. In other words, it could potentially cost as much as $7.5 million for that single roundabout—a whopping 15 to 25 times the average cost of signalization. The electricity and maintenance savings that many roundabout proponents cite is in fact minimal. Again referring to ODOT’s signal brochure, those costs typically run in the range of $4,000 per year. Even assuming the most favorable figures for the roundabout cost—the more favorable $500,000 signalization versus a $3 million roundabout—the savings will not be recouped until the year 2640.

However, this is not the only glaring problem with putting roundabouts on Highway 47. While there are a number of existing policy studies out there which show very positive safety records for yield-controlled “modern” roundabouts, including the Insurance Institute for Highway Safety’s continued studies, my own preliminary investigation shows a strikingly different picture.

First, to lay bare the most essential claim made by roundabout proponents, the Insurance Institute for Highway Safety report suggests that intersections “converted from traffic signals or stop signs to roundabouts have found reductions in injury crashes of 72-80 percent and reductions in all crashes of 35-47 percent”. Citing Isebrands and Hallmark (2012), which studied “19 higher-speed rural intersections (speed limits of 40 mph or higher) that originally had stop signs on the minor approaches and were converted to roundabouts”, they “found a 62 percent reduction in all crashes and a 85 percent reduction in injury crashes.” The lower speeds, both in approach and in the circular roadway itself, along with the reduced number of conflict points, are the most oft-cited factors.

In Oregon, the most commonly used metric for representing the severity of traffic accidents is the “KABCO” system, described on p. 22 of the PDF version of the state’s Highway Safety Investigation Manual, which ODOT uses for their statistics. Their ways of notating each severity level in the KABCO system differs depending on the type of report, but the below chart breaks this down:


Code Short Description Long Description
K Fatal injury
A Incapacitating Injury/Major Injury “Prevents person from walking, includes severe lacerations, broken limbs, abdominal injuries.”
B Non-Incapacitating Evident Injury/Moderate Injury “Evident to observers, lump on head, bruises, cuts.”
C Possible Injury/Minor Injury “Limping, momentary unconsciousness.”
O No Injury/Property Damage Only (sometimes also referred to by the acronym “PDO”)

Now, let’s take a look at some statistics. The roundabout for which the most extensive data exists is the one at the base of the Youngs Bay Bridge in Astoria, where US Highway 101 meets Oregon Highway 202. This roundabout opened in October 2002, and features two travel lanes around most of its circumference.

Astoria Roundabout

Roundabout at US Highway 101 and Oregon Highway 202 in Astoria, Oregon

The highest approach speed is a mere 35mph, from the Highway 202 leg, while the Highway 101 legs have a posted speed of 30mph. Being an intersection between two state highways, I was able to obtain crash data back to the 1990s, including a before period of October 1993-September 2002, and an equal after period of October 2002-September 2011. Here’s the KABCO breakdown:


Period K A B C O Total Crashes Total Injury Crashes Total Injuries Truck Involved
10/1/1993-09/30/2002 (Before) 0 1 4 5 15 24 8 10 2
10/1/2002-09/30/2011 (After) 0 1 3 19 16 58 18 23 11

As you can see, rather than the 35-47% reduction in crashes one would expect from reading the studies, for an intersection in an urban setting, the total number of crashes appears to have jumped from 24 before to 58 after, meaning an increase of 142%. Even more problematic, the number of injury crashes has not decreased by 72-80%, but has instead increased by 125% , from 8 to 18. The number of injuries sustained in those crashes also increased, from 10 to 23 (130% increase), and the severity of those injuries did not go down. The majority of accidents pre-roundabout were property damage only, while injury accidents are the majority post-roundabout. We still see a single Class A injury both before and after the roundabout installation. Class B injuries dropped insignificantly (4 to 3), and Class C increased sharply (5 to 19).

My data, taken from ODOT’s online Crash Data System, can be acquired here:







The traffic volume during the entire period remained steady, with the segments of Highway 101 around the intersection remaining somewhere in the vicinity of 20,000 Annual Average Daily Trips (AADT), sometimes spiking a little bit higher. The two sample points ODOT used near the intersection include a point at Milepost (MP) 4.97, on the bridge south of the intersection, while the north sample point was located at MP 4.09 pre-2005, and at MP 4.27 from 2005 onward. Based on ODOT data, including reference of their Digital Video Log, my sample points for the pre-roundabout Y-intersection were from MP 4.2 to 4.36, while for the roundabout period, I adjusted down to MP 4.3 to 4.35, due to the different footprint of the intersection.  Because of the relatively low fluctuation in traffic volume, it is clear from this data that the “per million vehicles” rate that many engineers use as a standard has gone up post-roundabout.

It is also worth noting that the number of incidents involving trucks at the Highway 101/202 intersection in Astoria increased from by a factor of 5.5 after the roundabout was built, increasing from just 2 in the previous 10 years, to 11 in the aftermath. At the town hall meeting, ODOT Region 2 Manager Sonny Chickering noted that truck drivers were not happy with the Astoria roundabout, and it’s easy to see why. Highway 47 is in fact a lifeline for our freight access to the outside world, and there has been repeated issues with truck rollovers in the existing Verboort roundabouts. Chickering did also comment that the truckers had said they would not oppose the proposed Highway 47/Verboort/Purdin roundabout, due to the recent fatalities at that site.

That, however, was before Washington County LUT had sneaked in a plan to install a second roundabout at the new David Hill intersection, and, along with ODOT, indicated that they were “studying” the Sunset/Beal intersection. In the question I posed to Chickering at the town hall, I also noted the fact that for a few years following the installation of the Verboort roundabouts, traffic volumes increased sharply on Highway 47, indicating a pattern of motorists avoiding the roundabouts. If the truckers decide to avoid the northern portion of Highway 47 to reach Forest Grove, to bypass the roundabouts, where are they going to go?

In studying the accident records of many other roundabouts across the state, while some do in fact appear to have solid safety records as advertised, one trend I also noticed was that a number of them have had Class A injury accidents within the past five years. Among these is the roundabout that was completed at Scholls Ferry Road and River Road by Washington County in mid-June 2012, which saw such an incident just a little over two weeks after opening. With the exception of the Class A injury in the Astoria roundabout, the vast majority of these major injuries have involved motorcycles. Even with the roundabout at Meinecke Parkway and Dewey Drive in Sherwood—both roads with 25mph speeds—we see one of these severe motorcycle accidents.

There are places where roundabouts are a suitable and effective intersection treatment, but given the history of the Astoria roundabout, I have serious reservations about installing one—let alone two or more—on a high-volume, high-speed state highway, especially given the cost. What I would personally advocate for would be to signalize the Verboort/Purdin and David Hill intersections, as an interim measure to immediately improve safety. Washington County LUT has actually installed three new rural traffic signals in the southern part of the county this summer, at the intersections between Scholls Ferry and Tile Flat Roads, Roy Rogers and Beef Bend Roads, and Roy Rogers and Scholls-Sherwood Roads. The Highway 47 intersections in question are very comparable to these intersections.

In the long term, however, we need to figure out just what our road network in Forest Grove and vicinity needs to look like. While the term “urban highway” has been thrown about quite a bit, I don’t believe Forest Grove needs Highway 47 to turn into another Highway 8. We need a highway that can safely sustain higher speeds, allowing for mobility in and out of Forest Grove. Business and industry that would otherwise consider our city see us as being difficult to reach, and out of the way.  It’s because our road network is so poorly connected with the rest of the region, and does not reflect current travel patterns. Especially given the fact that ODOT and Washington County seem to be willing to spend a considerable chunk of change on the problematic multi-roundabout solution, it is possible they may be up for reallocating this funding toward a real solution, such as a modest interchange, which, as the city grows, is going to be a better long-term investment than expensive roundabouts that may need to be torn out. Where that improvement would be situated is something that would need to be examined, and coordinated with efforts to improve our severely lacking east-west connectivity, such as the proposed Evergreen Road extension, which I believe should be a top priority for western Washington County.

In summary, here is my stance with respect to Highway 47:

  • Roundabouts—especially two or more of them—are not viable solutions. The Astoria roundabout statistics seem to show a significant downside of putting multi-lane roundabouts on high-volume freight routes like Highway 47.
  • The Verboort/Purdin and future David Hill intersections should be signalized in the interim, as should the Fern Hill/Maple intersection on the south end, while a more comprehensive solution is determined.
  • That comprehensive solution for the highway needs to prioritize both mobility and safety, and needs to tie in with plans for improved east-west connectivity, making Forest Grove easier to reach from the rest of the region, and increasing our shot at becoming something more than a moribund bedroom community.