copyright 1996, Tracy-Williams Consulting
by John Williams and Kathleen McLaughlin
Actually, in our view the answer is "none." Bikeway plans tend to treat cyclists' needs in a pretty one-dimensional manner. One plan we saw, for example, proposed spending several million dollars on bikeways but nothing to fix up the town's many unsafe drain grates!
On the other hand, the best Comprehensive Bike Plans* we've seen have been done in Australia. For example, the Geelong Bikeplan (the first of the genre), followed by the Adelaide, Sydney, Melbourne and Perth Bikeplans, include serious proposals for improved educational programs, real enforcement work, and good ideas for encouragement projects. The plans were developed through local task forces, which included members from bicycling groups, local police, planning agencies, public works departments, school districts, and safety groups.
*Comprehensive Bike Plans include education, enforcement, and encouragement elements in addition to engineering and planning work.
Go back to list.
Sure. It's been done for years. Demand-actuated systems consist of electrified wire loops buried in the pavement under the traffic lanes. The loops set up an electromagnetic field and are hooked to sending units. When a large mass of metal passes over the field, the sending unit alerts the controller (located in a box on one corner of the intersection), which changes the signal.
To detect bikes, use the right loop and make sure the amplifier is tuned properly. In San Diego, the CalTrans "Type D" loop is used in shared-lane situations and the "Type Q" is used in bike-only lanes.
Go back to list.
The best barrier is routine enforcement. While some communities install bollards, these are easy for bicyclists to miss (visually) and hit (physically). Tom Walsh, Traffic Engineer for Madison, Wisconsin, only uses bollards when there's a proven need. Otherwise, he doesn't use anything more than regulatory signs. If you do use bollards, make sure they're highly visible (reflective tape, overhead lighting, etc). If you need to get through with emergency vehicles, consider a hinged design for the post. Use the Caltrans approach to striping around them.
Diane Bishop reports that Eugene, Oregon, uses a different approach: they split the path entrance into two 5-foot one-way paths. In between the paths, they plant low bushes to discourage motorists; fire trucks and such can drive over these bushes in an emergency. The Ohio Department of Transportation's design guide has a good diagram to give you the idea.
What about keeping motorcycles off paths?
Enforcement is the only way we've seen. If you make a barrier that will keep out a motorcycle, it'll also be a real nuisance for bicyclists and (on multi-use paths) wheelchair-bound pedestrians.
Go back to list.
Hard to say, actually. Bicycle crashes tend to go unreported; only one in 5 car-bike crashes and 1 in 20 non-car-related crashes typically get reported and virtually no one has determined crash rates by comparing their crash statistics with the number of bicyclists using their pathways or the number of miles traveled. As a result, agencies may not know whether their paths are relatively safe or dangerous.
Some people believe that mere separation from auto traffic makes a path safe. But "safety" is the result of a combination of three main factors: the built environment, the bicyclist, and other users.
First, since most serious bicycle crashes result from falls or collisions
with stationary objects, we need to look at whether a path contains any
built-in hazards. A common problem is insufficient clearance to static obstructions.
AASHTO, for example, recommends a minimum of 3 feet between the edge of
the path and any obstructions (1). These may include posts, street furniture,
nearby buildings, supports for overhead structures, overhead structures
themselves, railings, fences, trees, steep drop offs, embankments, and drainage
channels or culverts.
Next, inadequate pathway widths, curve designs, and other features of the layout and construction can increase the likelihood of falls and collisions with pedestrians and other bicyclists. Curves that are too sharp for the appropriate design speed, curves with restricted sight distances, and inappropriate grades (too steep) can all cause crash problems.
Another problem is narrowness; two-way bike-only paths less than 8 feet wide or shared bike-pedestrian paths less than 10 to 12 feet give trail users no margin for error. These minimum widths are not generous, by the way. Progressive agencies tend to go at least 12 feet for shared paths. Another common problem to look for is lack of maintenance. Debris (sand, gravel, rocks, leaves, branches) on the trail-especially on curves, vegetation that encroaches on the edge or hangs over the path should be taken care of on a routine basis.(2)
Finally, since most bicycle-motor vehicle crashes happen at intersections, it is important to look at how the paths relate to the roads. Paths with lots of intersections with the road network or intersections that meet roads at blind curves or odd angles may increase the level of danger for bicyclists.
Bicycle paths tend to attract less experienced bicyclists. And these riders tend to have more crashes. The skill levels, knowledge, and behavior of typical path users would likely influence crash rates. While the data is far from perfect, one report (5) gives crash rates per million miles for three types of riders: elementary school cyclists (720), college-affiliated riders ( 510), and adult club cyclists (113). Of the three groups, the first two are most likely to find bicycle paths attractive riding environments.
For bicyclists, the ideal might be to have a path to themselves. However,
this is seldom possible. Typically, bicycle paths are shared-use paths,
used by pedestrians, rollerbladers, skateboarders, and joggers, in addition
to bicyclists. In general, the more "shared use" there is, the
more stressful and, potentially, the more dangerous the path is for any
given user type. In Calgary, Alberta, the riverfront bike-pedestrian paths
have gotten so popular that the City Parks Department is building separate
paths for bicyclists near what used to be shared-use paths. The older paths
are now reserved for pedestrians.
With respect to bicyclists and pedestrians, one study (3) concluded that "The incompatibility of the physical design parameters and the operational and behavioral characteristics of each mode preclude the development of shared pathways for pedestrians and bicycles in high density urban areas. The characteristics of bicycle travel in this environment is more compatible with vehicular traffic than with pedestrian traffic."
Go back to list. Go to Kaplan
There are some options here. The one we particularly like these days is the "vane" design, which has curved bars perpendicular to the direction of travel. According to the Neenah Foundry catalog,1 "hydraulic tests prove the new Vane Grate will accept more water than any of the conventional grate styles under virtually all flow conditions." They also say that it "is being used in grate replacement programs in many cities throughout the country, not only because of its storm water capacity but also for its bicycle-safe qualities." That's our top choice.
Other options include curb-face inlets. These eliminate the grate entirely from the surface of the road. But the roadway may have to be sloped to move water towards the inlet. This can cause handling problems for cyclists if the slopes are extreme.
In a pinch, you can weld 1-inch x 1¦4-inch cross-straps over a bad parallel-bar drain grate (the ones that can catch your wheel). Weld them at a 6- to 8-inch spacing. But keep in mind that this is a very temporary measure, especially in the Snow Belt, where snow plows can pop the straps off like crazy.
The Manual on Uniform Traffic Control Devices2 (MUTCD) says you can stripe warning stripes around a bad grate. Don't bother. Paint wears off and there's no guarantee that a bicyclist will notice the striping and avoid the grate. Replace the grate.
Go back to list.
Actually, no. Because they increase bike-car conflicts at intersections and are designed for pedestrian speeds, sidewalk bikeways actually make things worse. In Eugene, Oregon, for example, "The average accident rate for the three sidewalk bicycle route sections is 1.8 accidents per 100,000 bicycle miles per year. This is nearly three times the average for the signed lanes or striped lanes. These facilities are significantly more hazardous for bicycle-motor vehicle accidents."1
Palo Alto's 1974 study showed a 54 percent increase in the car/bike crash rate on streets where sidewalk bikeways had been installed.2 They said "After over a year of operationassessment of field observations and an analysis of accident data indicate that this particular concept presented an undesirable high-risk situation."
Further, for years the available design standards have recommended against them. The 1991 AASHTO Guide for the Development of Bicycle Facilities says "Providing a sidewalk bicycle path is unsatisfactory for a variety of reasons."3 Among those mentioned: low pedestrian-oriented design speeds, conflicts with pedestrians, fixed obstructions (fire plugs, parking meters,), intersection conflicts with turning and crossing motorists, and sight obstructions (bushes and fences at corners and driveways). Their comments are taken from the 1978 CalTrans report Planning & Design Criteria for Bikeways in California4.
People who haven't looked into the problem often have a hard time imagining that sidewalk bikeways could be dangerous. But if they when they realize that most car-bike crashes happen at intersections, that sidewalk bikeways tend to reduce visual contact between bicyclists and motorists, and that bikes generally travel between 8 and 20 mph, the reasons become clear.
Go back to list. Go back to Bike
A wide curb lane is an outside through lane that is wider than normal. "Normal" means 12 feet. By adding extra width to the curb lane, you make it easier for motorists and bicyclists to share the road. How much extra width is needed? According to a study done by the Maryland State Highway Administration, "Highway departments should promote increased lane widths within the normal construction program on closed section highways and bridges on which bicyclists are expected and permitted to 15.0 - 15.5 feet depending on traffic and bicycle volumes."1 This confirms the informal consensus we've found in the field: 15 feet is a good lane-sharing width, especially on urban roads with speeds under 45 mph.2
The Highway Capacity Manual3 makes an interesting point on the wide curb lane situation. It gives "passenger-car equivalents" for bicycles in traffic lanes of different widths. In a lane less than 11 feet wide, a bicycle is equivalent to either 1.0 or 1.2 cars, depending on whether the bike's movement is unopposed or opposed. [Unopposed means not crossing another direction of travel; opposed means crossing another direction of travel; a left turning bike, for example, is "opposed" by oncoming traffic.] In a lane between 11 and 14 feet wide, a bike is worth 0.2 or 0.5 (unopposed vs. opposed). But in a lane wider than 14 feet, a bike is equivalent to 0.0 passenger cars, regardless of opposition. In other words, in a narrow lane, bikes add to the congestion while in a wide curb lane, they do not.
Go back to list.
If you're going to do this, it should be plenty wide. The AASHTO Guide1 says you need 13 feet (with an 8-foot parking lane) or 15 feet (with a 10-foot parking lane). Baltimore County did an experiment some years back in which they photographed a cyclist passing a parked car that had its door open.2 A vertical flag was placed at different distances from the curb; researchers noted that 13 feet was the minimum width in which a bicyclist could pass the car door without hitting the flag.
One interesting idea we've seen is to stripe a 15-foot curb lane with a rush hour parking prohibition. During rush hour, it acts as a wide curb lane. In off-peak hours, there's plenty of room for bikes to get by parked cars. The ones we've seen don't include special pavement markings (except, possibly, parking "T"s) but the space is certainly welcomed by the bicyclists.
Go back to list.
Manhole covers, if they're not flush with the pavement, can cause no end of trouble for bicyclists. Whether a cover is sunken into the roadway or raised above its surface, the end result can be the same: cyclists losing control, falling, and hurting themselves and damaging their bikes.
Sometimes, a previously level cover may end up low if the road around it is repaved but the manhole isn't adjusted. Fortunately, there are specs for such projects. One good one we found comes from the Montana Public Works Standard Specifications. This standard gives a smooth ride for the bicyclist.
Go back to list.
Adopt tougher standards and enforce them. When utility companies and governmental agencies dig up the streets, they sometimes neglect to patch correctly. As a result, the patch may be uneven or may fail in the future. And badly done patches are hazardous for bicyclists.
The City of Palo Alto, California, has adopted stringent and detailed bicycle-friendly patching standards. If a utility company digs up a street and patches it, the patch must be smooth and must not fail for at least a year. The City has assembled a packet of information2 on their policies and they make it available for free to interested parties.
By the way, here are the standards for pavement quality on bike lanes or bike routes from CalTrans' Planning & Design Criteria for Bikeways in California (p. 37):1
Grooves* Steps** Parallel to travel < 1/2 inch wide < 3/8 inch high Perpendicular to travel - < 3/4 inch high
*Groove: A narrow slot in the surface that could catch
a bicycle wheel, such as a gap between two concrete slabs.
**Step: A ridge in the pavement, such as that which might exist between the pavement and a concrete gutter or manhole cover; or that might exist between two pavement blankets when the top level does not extend to the edge of the roadway.
Go back to list.