About the Austin colored bike lane marking report

Please read the Austin, Texas colored bike-lane report, Effects of Colored Lane Markings on
Bicyclist and Motorist Behavior at Conflict Areas in connection with this review of it.

My comments:

In the Executive Summary:

The report says:

Conflicts are common on facilities where a motorist must cross a bicycle lane in order to access a right turn bay and where highway exit ramps cross major arterials that have a bicycle lane.

A conflict is a situation in which one road user must take abrupt action to avoid a collision with another. Colored markings are sometimes used to mitigate the problems with what can accurately be described as conflict zones: zones where unusual yielding rules apply, where sanctioned movements require looking in unusual directions, and/or where sight lines are restricted.

On the other hand, colored markings are sometimes used in what I would describe as merge zones, or crosswalks, or bike lane extensions across ramps — places where the motorist or bicyclist must yield right of way according to normal rules, with no unusual impediments.

In other reports, the researchers have describe route choices as “avoidance maneuvers” and merges into an adjacent lane as “encroachments”. Here, they elevate normal traffic maneuvers to “conflicts.” Conflicts may indeed sometimes occur during normal traffic maneuvers, as a result of mistakes in driving, but that doesn’t elevate the locations of these conflicts into “conflict zones”.

Examples of both normal traffic movements and conflict-generating ones are discussed in this report. The distinction deserves to be made clear.

All in all, the researchers’ repeated use of terms describing hazardous and unlawful maneuvers to describe normal and reasonably safe maneuvers raises serious questions about their understanding of the traffic laws, and traffic operation.

Quoting again:

For this study, safety was defined along the following lines:

1. the bicyclist used the bicycle lane to approach the conflict area,
2. the bicyclist used the bicycle lane to negotiate the conflict area,
3. the motorist yielded to the bicyclist when crossing the colored lane area, and
4. the motorist used a turn signal when crossing the conflict area.

Only the last two of these four criteria relate clearly to safety. Whether the first two do depends on whether the markings are properly located considering traffic conditions at the time and place.

To ensure that the data collected reflects the effectiveness of the treatment alone, no educational or outreach campaign was conducted.

This issue poses a legitimate quandary. Markings must be understood to be effective. The colored paint here is between ordinary bike lane markings, and that makes their intention clearer. Still, it would have made the most sense to compare the effectiveness of the markings without, and then with, an educational campaign.

However, the colored paint in this study was accompanied by installation of “yield to bikes” signs, which produced a similar conflation of results as an education campaign would. There was no phased installation, and so it can not be determined to what extent the change in behavior resulted from the signs or from the paint.

In the section Background:

Figure 1 in the report shows a situation in which bicyclists taking the route shown have to yield to motorists in the interest of safety, due to a restricted sight line, and not to defeat the purpose of the on-ramp in allowing motorists to accelerate into the flow of traffic. Instead, the motorists are required to yield to the bicyclists, as at a crosswalk.

Research in the U.K. (reported to me by John Franklin) has shown a problem with fatal collisions between motorists entering from high-speed ramps — looking back to merge into traffic — and bicyclists who continue in the normal through-travel lane position. I will not condemn the routing shown — but for the reasons I have given, it does require bicyclists to yield.

In the photo, I also note that there is a sidewalk, but there are no crosswalk markings. What expectations does this treatment create, in connection with the bold bike lane markings? That motorists should stop and look only upon reaching the bike lane? That pedestrians should use the bike lane as a crosswalk?

In the section Site Descriptions

Figure 4 — shows that all of the three lanes on Dean Keeton cross ramps at a very low angle, worsening sight conditions for bicyclists and greatly increasing their crossing times. Compare with the crossing on Naito Boulevard in Portland, Oregon — which has other issues but which crosses at a right angle.

Figure 7 — This shows a more conventional vehicular routing of bicyclists, straight through to the left of an off-ramp. I find it odd that the green paint begins partway into the merge zone rather than at its start.

In the section Experimental Design

The same four criteria for safety are repeated, and again, only the last two relate directly to safety. It could be argued, for example, that at least the slower and more timid bicyclists are safer riding down the right side of the ramp on San Jacinto Boulevard, turning left and yielding to motorists. As motorists are generally traveling faster than bicyclists and approaching from behind, it can be argued that the last criterion givwen, motorist used turn signal, is irrelevant, because bicyclists would not see the signal. An actual measurement of safety would require counting conflicts, in the usual meaning of the term — abrupt braking or swerving either by a bicyclist or by a motorist.

In the section Terminology

The definitions here are generally good, including, this time, a correct definition of “conflict”. However:

If the motorist accelerated to cut off the bicyclist while the bicyclist was in the conflict area or the bicyclist yielded to the car during a yielding event, the yielding event was not described as ‘car yielded to bicyclist’.

The incorrect term ‘conflict area” is used inside the definition of “yielding event”, and there are not separate categories for motorists cutting off bicyclists, and bicyclists yielding to motorists. Not every motor vehicle is a car.

Illegal event – An event was recorded as illegal if the bicyclist acted in a way that qualified as an event, but also acted illegally or very unsafely.

What if the motorist acted illegally?

In the section Results

Tables 1 and 2 show very little difference in bicyclists’ behavior.

Tables 3 and 4 show an increase in yielding behavior and turn signal use by motorists, but the sample size is very small. Note that yielding at the facilities provided is different from that on San Jacinto Boulevard, because motorists are required to slow or stop in order to yield, rather than only to time a merge. At the on-ramps, there is only one way to go, and at off ramps, motorists are coming from behind bicyclists and supposed to yield to them, so it is unclear to me just what the use of turn signals achieves for bicyclists, Signals are, however, useful when merging from the ramps into faster traffic.

The remaining tables show mixed results and minor differences in behavior, generally somewhat improved except for Table 9, which shows a steep reduction in motorist yielding behavior but with an extremely small sample size.

In the section Conclusions and Recommendations

The conclusions in favor of the painting appear overstated to me, and particularly in the light of installation of signs at the same time.

References:

None of the four references is directly to a study of a similar treatment — three of the four relate to generalities about what encourages people to ride bicycles and one is about integrating bicyclists into a pedestrian campus. That is a bit surprising because a study of colored lanes, the Portland blue lane study,

(FHWA version); (Version on Portland, Oregon site)

has not only been published but also is referred to in the text of the report. (page 8, second paragraph). Considering the number of European countries referred to in the Austin report, I would not be surprised if European studies also exist.