Bicyclist narrowly avoids opening door -- Centre St, Jamaica Plain, Mass.

This bicyclist had to swerve as the oblivious motorist flung open her door on Centre Street in the heart of the business district where parking turnover is high and traffic is slow.

This Thursday the City of Boston will present its proposal for bike markings on South Street and Centre St from Forest Hills to Jackson Square in Jamaica Plain. The proposal will be:

* door-zone bike lanes (DZBLs) from The Monument to Lakeview St, except for a two-block section where the width dips below 44 feet, where shared lane markings (SLMs) are proposed;

* DZBLs on the uphill side only, and on the downhill side a narrow lane with an SLM in the middle, from Hyde Square to Jackson Square and from Sedgwick Street to Jamaica Street.

* SLMs elsewhere.

The parking lane be striped at 7 feet from the curb throughout the corridor. Some additional marking to indicate the door zone may be discussed.

Most people agree that riding in the door zone is not a good idea (except when deliberately going very slowly). But some DZBL proponents argue that a) DZBLs actually move people away from parked cars, and b) you won’t get people to ride outside of the door zone.

To use their own words:

a) “First, [the additional distance between bicyclists and parked cars found in the Hampshire St. Study is] 2.4 inches further away, not 2, and that’s merely the median (average) distance   It’s clearly in the author’s conclusion that a narrowing of the spread of distances was the significant finding. Namely, that far fewer people were riding very close to the doors. And when we talk about doors and the door zone, keep in mind that when you’re on the street and a door opens, inches definitely count.”  (Pete Stidman)

b) “The problem that you are trying to solve is how to get bicyclists to ride completely outside the door zone. Without removing parking or somehow changing the dimensions of the roadway, I don’t think any type of facility and/or education will be effective in doing so.” (Charlie Denison)

I have some responses to both points:

a) The Hampshire Street study in Cambridge found that bicyclists moved away from the curb after marking a solid line at 12 feet from the curb — but they moved by an average of less than 3 inches. A San Francisco study found that shared lane markings moved bicyclists away from the door zone by an average of 8 inches. The lead author of the Hampshire Street Study has agreed with my assessment that a wheel track 11 feet from the curb, not 10 feet, is the proper cut-off for being outside the door zone 85% of the time with a 6″ margin to spare. Only 11% of bicyclists in the study were riding this far away. I am currently doing some re-analysis of the study data. One reason that few were out of the door zone is that they did the survey at rush hour, mostly approaching intersections. Traffic on Hampshire Street regularly queues up for several blocks approaching red signals in the peak period. So some of the bicyclists in the door zone may have been slowly passing cars on the right instead of being stuck in traffic for several light cycles.

b) Boston has added Shared Lane Markings in the middle of the lane in several places, and proposes to do so in the Centre-South corridor–where there is only one travel lane in each direction. These markings are supposed to direct bicyclists to ride in the middle of the lane. If it is possible to encourage people to ride in the middle of the lane, surely it is possible to encourage them to ride outside the door zone where they will impede traffic much less (at those times where traffic is moving faster than bicycle speed). A bicyclist who rides with his or her wheel at 11 feet from the curb is safely out of the door zone and in most places in the corridor is leaving 10 or more feet for a 6-foot wide passenger car to pass without changing lanes. (In fact, most drivers will crowd or cross the center line in order to provide the bicyclist with plenty of extra room when passing.)

I propose a solid white line at 10 feet from the curb with a shared lane marking completely to its left. This will provide clear guidance that bicyclists are expected to stay out of the door zone. Another San Francisco study showed that using a 9 foot parking lane instead of a 7-foot one moves the average spacing of parked cars about 4 inches further from the curb. I don’t think this a significant concern. If it is, then a solid white line could be marked at 7 feet from the curb with a series of diagonal lines (gore stripes) between the lines at 10 and 7 feet — to indicate the buffer zone to be avoided. This design would be used consistently throughout the corridor, instead of the confusing alternation between DZBL and SLM that is proposed.

To my knowledge, no city has yet tried a SLM to the left of a line indicating the door zone (as distinct from a BL with the symbol to the right of the line, or an SLM with no line).  There is an opportunity to find out how effective this configuration is in pulling people away from doors. In any case we know that SLMs, even without a line, can be more effect than DZBLs in pulling bicyclists away from the door zone.

Please come to the meeting to support shared lane markings instead of door zone bike lanes. Defend the right of bicyclists to use as much of the road as needed for safety and travel. Don’t lure bicyclists into the danger zone. The details:

Thursday, April 29^th

Agassiz School cafeteria
20 Child Street, Jamaica Plain, Mass.
6 pm to 8 pm

“Bike accommodations” will be the first and major topic of the meeting.

Jackson Square Bike Lane Concept

A proposal for bike lanes on Centre Street between Lamartine St and Sunnyside St (near Jackson Square) was floated at the March 23 meeting of the Centre South Transportation Action Plan Citizens’ Advisory Committee. The 42-ft curb-to-curb width is just a bit too narrow for the bare minimum bike lanes proposed for Centre Street between Elliot St and Lakeview St. The necessary width would be obtained by eliminating parking on one side of the street only. (See the presentation, pages 56 – 61.) Currently there is about 21 ft (travel lane + parking lane) on each side of the street. The suggested new configuration is an 11 ft travel lane and 4 ft parking lane on one side and an 11 ft travel lane with a 5 ft bike lane, 8 ft parking lane on the other side. The remaining 3 ft would be used to widen the existing 8 ft sidewalk on one side. The proposal would leave on-street parking in front of most businesses that do not have on-street parking. This would be accomplished by switching the side without parking from the south side near Mozart Park to the north side near Bromley-Heath. Nevertheless, at least 37 on-street spaces would have to go.

The proposal would remove the door-zone threat on the side of the street where parking is newly prohibited. And the remaining parking lane + bike lane on the other side would be 13 ft wide instead of the 12 ft proposed for central Jamaica Plain, meaning that 1 to 2 ft of the nominally 5 ft bike lane would be safe for riding, rather than 0 to 1 ft. A slightly wider sidewalk would be nice, although the 3 ft could be used instead as a marked buffer area so the bike lane could be out of the door zone (more on that concept in a later post).

Two alternate proposals would use the space from eliminating a parking lane to widen the sidewalks, and would also narrow the travel lanes to 11 ft (with no bike lanes). In other words, safe cyclists would frequently have to ride in the middle of the lane, because there would be insufficient room for motorists to pass safely, particularly on the side next to on-street parking. These alternate proposals are clearly a dis-improvement from the point of view of bicycling.

But will it be possible to remove any parking? Michael Halle, a member of the CAC, is quoted in the JP Gazette. “Halle noted that the area is right next to a T station and a huge parking lot at Stop & Shop that he described as ‘underused.’  .  .   .  He said if on-street parking can’t be cut here in favor of bikes or pedestrians, ‘it can’t be done anywhere.’”

I know of three places in the Boston area where on-street parking has been removed, in part to improve bicycling conditions:

* Huntington Avenue between Opera Place and Longwood Avenue (adjacent to Northeastern, MassArt, the MFA, and Wentworth).

* Mass Ave, Cambridge, adjacent to MIT.

* North Harvard St, Allston, adjacent to the Harvard Business School (just last year).

(I can’t think of any other examples in Cambridge–anybody else know of any?) What do all of these examples have in common? Parking was removed immediately adjacent to institutions only, not where there were businesses or residents abutting the street. All these institutions have large off-street parking facilities. While the Jackson Square proposal attempts to avoid businesses, it is not really analogous to the three successful removals. We will see if this proposal is implemented.

Meanwhile, one important consideration for bicyclists in this area is that there is a significant downhill grade from Stop and Shop to Lamartine Street. Perhaps parking can be removed just in front of Bromley-Heath, on the uphill side, where bicyclists go slowly. If not, the centerline could be moved two feet so that the downhill side would have a narrow (11 ft) travel lane, which could be outfitted with a Shared Lane Marking in its middle, and the uphill side could have a wide (15′) travel lane (and an optional Shared Lane Marking 12 ft from the curb).  (A door-zone bike lane could be used instead on that side only, although this is not preferable to a shared lane, for reasons I’ve previously described.)

Meanwhile, we’re hoping that the (separate) Jackson Square project will fix the basic problem that the drop-off area needs to be adjacent to the station, not on the other side of the street. Who was the genius who designed that? It’s created delay and danger for probably 23 years now.

Centre Street with and without bike lanes -- the street grew by 10 ft!

The “Centre/South Streetscape and Transportation Action Plan” is proposing bike lanes in the Centre Street business district in Jamaica Plain from Eliot Street to Lakeville Rd. The images from the presentation at the January 2010 meeting of the Citizens Advisory Committee for the project show bicyclists safely out of range of car doors in the existing conditions but clearly within range when bike lanes are added (see images above, doors added to original). Shared lane markings should instead be used on Centre and South Streets to encourage bicyclists to safely use the streets and to discourage motorists from harassing bicyclists. Gore stripes can be used to indicate that the door zone is unsafe. Shared lane markings have already been used at Forest Hills and in Roslindale Square and several other locations in Boston, and are proposed for the majority of the Centre-South Street corridor that is less than 44 feet wide and thus considered too narrow for bike lanes.

Door Zone Buffer Gore Stripes (Paris)

Centre Street Dimensions
The drawings shown at the beginning of this article are taken from the presentation by the consultant team delivered at the January 28, 2010 meeting of the Citizens Advisory Committee. The first drawing shows the existing conditions on Centre Street between South Huntington and Monument Square. The second drawing shows how a bike lane might be inserted into the existing space without removing any parking or travel lanes or narrowing the sidewalk. Upon close examination, it becomes clear that the travel lanes in the second drawing are 15 to 16 feet wide, rather than the marked dimension of 10.5 feet. Someone merely inserted two 5-foot bike lanes without taking that space away from the travel lanes, as if the curb-to-curb width of the street could grow by 10 feet to accommodate the bike lanes. This mistake falsely suggests that more space will exist once bike lanes are installed.

The drawings have been altered by the addition of red rectangles indicating the approximate position of open doors on the parked cars. Note that in the first drawing, the bicyclists and cars shown have ample room to clear an open door. In the second drawing, the bicyclist centered in the bike lane gets hit if anyone opens a door suddenly.

The Door Zone

Correct Riding Position is Outside of Bike Lane (on Lane Line)

The bike lane would be mostly or entirely within the door zone, as can be seen in the graphic at right showing the same dimensions proposed for Centre Street. Bicycling in the door zone—within a door’s width of parked cars—is not safe. When riding faster than a walking pace, a bicyclist cannot stop in time to avoid a suddenly opened door. Although car occupants are legally responsible for making sure it is safe before opening a door, no prudent bicyclist can rely on 100% compliance with this newly adopted rule, and must ride as if a door could open at any time. “Dooring” crashes can be very serious, due to possible impacts with sharp metal and glass and a fall on to hard pavement. In the worst case, the bicyclist falls and is hit by a passing motor vehicle; documented fatalities have occurred this way while the bicyclist was riding in a presumably safe bike lane. The bicyclist must leave room for the entire bicycle to clear a door. Dana Laird was in the left part of the bike lane in Central Square, Cambridge, and her handlebar was in the door zone by only a few inches. But this was sufficient to throw her and her bicycle to the left, where a passing bus ran her over. (Other bicyclists killed by car doors opening in a bike lane include David Smith, Sixth Avenue in Manhattan, 2007, and Clinton Miceli, LaSalle St, Chicago, 2008.) Cambridge subsequently conducted a study that claimed to show that bicycle lanes keep bicyclists away from the door zone, but did not in fact show this (see Hampshire Street Study below). Like Central Square, the Centre-South business district has an elevated dooring risk because of high parking turnover.

Advantages of Shared Lane Markings

Shared Lane Marking in Roslindale Square

Where there is on-street parking, shared lane markings are specifically designed to position bicyclists away from the door zone (see note below). However, there are also several other advantages of shared lane markings compared to bike lanes for a corridor such as Centre-South Street:

* The same design can be used for the entire length of the corridor, even where it is a few feet narrower than the minimum that is supposed to be safe for bike lanes.

* Motorists will leave more room when overtaking bicyclists. When there is a dividing line, motorists assume it is sufficient to stay to the left of the line, whereas when overtaking in the same lane they leave as much room as possible, including using the left half of the road when necessary. (A greater average passing distance without bike lanes was documented in Evaluation of Shared-Use Facilities for Bicycles and Motor Vehicles, Transportation Research Record No. 1578, pages 111-118; William W. Hunter, John R. Feaganes, Raghavan Srinivasan, Conversions of Wide Curb Lanes: The Effect on Bicycle and Motor Vehicle Interactions. Transportation Research Record. Volume 1939 / 2005 Parkin J, Meyers C. The effect of cycle lanes on the proximity between motor traffic and cycle traffic. Accid Anal Prev. 2010 Jan;42(1):159-65. The last study concludes, “The results suggest that in the presence of a cycle lane, drivers may be driving within the confines of their own marked lane with less recognition being given to the need to provide a comfortable passing distance to cycle traffic in the adjacent cycle lane.”)

* When a bike lane is striped the remaining portion of the pre-existing travel lane becomes the “car lane,” at least in the perception of the vast majority of road users.

* Bicyclists would not be discouraged from riding in any portion of the right half of the road as necessary to pass slow or stopped vehicles, to prepare a left turn, and to avoid hazards in the road.

* Shared lane symbols will send a visible message of bicyclist legitimacy compared to the existing conditions, without sending the message that bicyclists must be in one little part of the road only. Therefore motorists would be much less likely to honk at, yell at, or assault bicyclists who dare to leave the bike lane.

* By keeping further away from the curb, bicyclists can be better seen by traffic emerging from side streets.

* Motorists will be more likely to merge as far as practicable to the right of the road in advance of making a right turn (as required by traffic law).

* Bicyclists will be less likely to overtake slow or stopped motorists on the right, in part because right-turning motorists will be further right. Because traffic moves slowly though this area, the temptation to pass on the right is high, and would be more so with a bike lane present. Overtaking on the right—except where traffic is stopped and cannot move forward—leaves bicyclists open to a collision with a car turning right, or a car on the other side of the road turning left through a gap in traffic.

Hampshire Street Study

The City of Cambridge conducted a study which claimed to show that bike lanes help to reduce dooring by moving bicyclists away from parked cars. The study showed that bicyclists did on average track 2.4 inches further left in response to a stripe being added between the car area and the bike area. However, 85% to 90% continued to ride inside the door zone. This is not surprising, since in this design, riding outside the door zone requires the cyclist to ride on the bike lane line. The study did not test lane markings showing the door-zone as a dangerous area (e.g., with gore stripes), nor did it test a shared lane symbol sufficiently away from parked cars with a line or gore markings to its right. However, the bike symbol was reduced and placed close to the left side of the lane, in an attempt to encourage bicyclists to ride their instead of the center.

Design Standards and the Door Zone

Shared lane marking considers the door zone

The shared lane symbol was added to the Manual on Uniform Traffic Control Devices in December 2009 and was specifically designed to take into account the width of opening doors (see drawing). The manual requires that shared lane stencils be centered a minimum of 11 feet from the curb where there is on-street parking. (Assuming a minimum shy distance of 6 inches to account for bicycle wobble, the 11 foot minimum is valid if bicycle handlebars are no wider than 24” and the widest extent of an open door is 30”. In fact cruiser bicycle handlebars can be 28” or wider, and some doors open as wide as 45”. Although most passenger cars are around 6’ wide, they can be legally parked 1’ from the curb (or illegally parked much further from the curb), and trucks can be as wide as 8.5’.)

However, bike lane standards make no explicit calculation of the minimum width needed for door clearance, except to add 1’ to the minimum 4’ bike lane width. The AASHTO and Massachusetts standards permit a parking lane next to a bike lane to be reduced from the standard 8’ to a minimum 7’, completely negating the additional foot required in the bike lane. The portion of Centre Street where a bike lane is proposed is generally 45’ but can be as narrow as 43’—so in some areas the bike lane will have to be less than 5’ wide and would be completely within the door zone.

Safety and Perceived Safety

Studies of car-bicycle crashes show that most urban car-bike collisions in daylight are the result of turning and crossing movements, not overtaking of bicyclists by motorists. However, the bicycle lane addresses only the overtaking threat, not the more common dangers. The proposed bicycle lane on Centre Street does not provide any additional width, but merely reallocates the width available. (By contrast, other projects have provided more width for bicyclists by removing travel or parking lanes.) Thus the motorist’s task in overtaking is unchanged. To the extent that bicyclists move further right, bike lanes might make things slightly easier for motorists—at the expense of more collisions with car doors. However, traffic rarely moves much faster than bicycle speed in this corridor, and most people would agree that facilitating faster traffic movement here would not improve safety. Moreover, motorists have had no problem for years squeezing to the center line, or moving partly over it, to safely pass bicyclists. It’s easier than passing double-parked cars or buses sticking out from bus stops.

Many people are deterred from bicycling because of the perception of danger from overtaking traffic. A bike lane mostly in the door zone caters to this fear while increasing real dangers. A shared lane marking, by contrast, provides the legitimacy that bicyclists crave without promoting dangerous behaviors. Some people are so afraid of traffic that no amount of paint on the roadway will induce them to bicycle in the city. It is possible for these people to gradually overcome these fears with the guidance from more experienced bicyclists who show them what they most need to be concerned about to avoid injury. Learning to avoid the practices that “everyone knows” are safer, such as hugging the edge of the road and riding on the sidewalk or facing on-coming traffic is what reduces bicyclist injuries. While road design cannot teach bicyclists how to ride safely, it should not encourage them to ride dangerously.