Another arterial (Manchester Road at East Meadow Lane in Ellisville), shown in the photo above, has adequate, if minimal, lane/shoulder-sharing width despite the center turn lane. The rolled curb here is steep enough that a bicycle wheel would be more likely to skid than to mount it, and the gutter pan (at least in the foreground) could constitute a hazard as the asphalt pavement ages. Pedestrians are able to walk on what appears to be a narrow MoDOT shoulder.

The photo above also was taken on Manchester Road, at Ballwin Government Center. Manchester Road here is elevated on filled land to avoid a steep descent and ascent. The curb profile is similar to that in the previous picture, but here a guardrail intrudes far enough into the roadway to increase shy distance for bicyclists. The gutter pan in the foreground has deteriorated and is strewn with gravel, producing a bumpy ride and the possibility of a diversion-type fall on the joint between the gutter pan and the asphalt. There is a park at the bottom of the hill at the right side of the picture. Pedestrians have worn down the grass to the right of the guardrail, at the top of a tricky slope.

In this location, a shallower curb profile without a gutter pan and a guardrail placed back from the edge of the roadway would have been preferable. Pedestrians could be directed to the bottom of the hill, since the road visible at the right side of the picture connects with Manchester Road at each end. Careful attention to details of design for bicyclists and pedestrians could solve many of the problems shown here at little or no added expense.

Arterials in commercial districts

The photo below, taken on Lindbergh Boulevard in Florissant, shows the Boulevard's cross-section in a commercial strip mall. The two bicyclists on the sidewalk are the girls shown in the previous section on education.

Lindbergh Boulevard here has a cross section of five travel lanes including a wide center turn lane. In a three-lane cross-section with a center turn lane, motorists will merge partway into the center lane to overtake bicyclists. In a five-lane section as shown here, motorists in the outside lane are more reluctant to merge to their left, since they may be merging into the path of overtaking vehicles.

The outside travel lanes shown in the photo are of marginally adequate width for bicycle/motor vehicle lane sharing. The MoDOT shoulders here are disused, with deteriorating pavement, as I observed also in many other locations, and not usable for bicycling The curb between the outside travel lane and MoDOT shoulder is at approximately the location where bicyclists would prefer to ride to allow bicycle/motor vehicle lane sharing even with large trucks and buses.

The sidewalk in this area is discontinuous. Where the sidewalk exists, it is set far back from the roadway. The sidewalk has many blind intersections with driveways, where sidewalk and driveway traffic are hidden from each other by vegetation, signs and piles of merchandise. While I do not recommend bicycling on sidewalks, some bicyclists, particularly novices and children such as those in the photograph, will do it.

Strip development, with the many commercial driveways, produces less favorable conditions for bicycling than do shopping malls with combined driveways. Some of the unusual St. Louis design features such as the removal of usable road space onto the elevated MoDOT shoulder exacerbate this problem.

 Arterial intersections

The photo below, taken at Washington and Lindbergh in Florissant, shows some typical intersection design elements of arterials in St. Louis county. Lanes are relatively narrow to allow for separate turn lanes, and the "pork chop" triangular islands narrow the usable lane width for bicyclists further. Also note the MoDOT shoulder in the foreground, being used for parking, the storm grating which forces bicyclists in the right turn lane to ride within the stream of motor traffic, and the absence of designated crosswalks and pedestrian signals. Since the storm gratings are frequent along the entire roadway, bicyclists must ride to the left of the fog line to be predictable and safe from grates.

One very positive design element at intersections is the use of quadrupole traffic signal actuator loop detectors (below). This type of loop is much better at controlling stray detection than the commonly used dipole loop. Sensitivity of the quadrupole loop may be raised high enough to detect a bicycle without detecting a large vehicle in the adjacent lane. Whether these detector loops were in fact adjusted to detect bicycles, I do not know. I have heard a report from a bicyclist that at least some of the actuator loops are not adjusted to detect bicycles.

During my inspection tour, I visited two pilot projects to improve intersection conditions for non-motorized travel. The intersection shown below has a long right-turn lane, and to the left of it a bike lane for through travel. The bike lane correctly encourages bicyclists to merge to the left of right-turning traffic. For such a lane to be feasible, the traffic islands at the intersection must allow sufficient width.

On Manchester Road in Ballwin (photo, below), Terry Byatt has been instrumental in promoting construction of sidewalks, and in having a short segment of the center turn lane converted into a median, to allow a place of refuge for a pedestrian crossing. This treatment stands in strong contrast to the rarity of pedestrian crossings, and even more so, of median refuges, on the wide St. Louis County arterials.

Note, however, the storm grating in the foreground. The pedestrian improvement has worsened conditions for bicyclists, who must merge left to avoid the grating. But because of the traffic island, motorists in the inner lane can not merge to their left to overtake other motorists who are overtaking bicyclists.

Storm gratings

At some time in the past, the Missouri Department of Transportation specified parallel-bar storm gratings which could swallow a bicycle wheel up to the hub. The standard storm grate design has been modified twice since.

The first modified design is shown in the photo above, taken on Lindbergh Boulevard in Florissant. Cylindrical steel bars approximately 1/2 inch in diameter were welded across the tops of the parallel bars. This grating could not swallow a wheel. The parallel bars guide bicycle wheels to some degree, and the 6 inches between crossbars make the ride very bumpy. All in all, bicyclists still avoid riding over this type of grating if at all possible, though the potential for a disastrous crash is much reduced.

In the photo above, there is a curb inlet adjacent to the storm grating. A curb inlet is a preferred type of storm drain for bicyclists, since it need have no effect on the roadway surface. The reason for the "belt and suspenders" situation, with a curb inlet and storm grating adjacent to one another at this location, is unknown.

The grating modification shown in the photograph above proved unsatisfactory: the ends of the cylindrical bars could bend up and puncture the tires of motor vehicles. As a result, the design was modified to the one shown in the photo below. The direction of travel is from the upper left to the lower right in the photo.

Notice that the parallel bars have been drilled to receive the cylindrical crossbars, which are now about an inch below the level of the pavement. This has solved the problem of flat tires for motorists, but created much worse problems for bicyclists. The parallel bars now can guide a bicycle wheel as the tracks guide a railroad train, possibly causing the bicyclist to lose balance and cause a diverstion-type fall. The likelihood of getting a flat tire or damaging the bicycle's rims by riding over one of these gratings is also high. While I was in St. Louis, a bicyclist told me that he had flatted both tires in a single encounter with such a grating.

Bicyclists will ride farther from the road edge in order to avoid such gratings. In effect, a roadway with such gratings is narrower by the width of the gratings, and bicyclists must often merge into the path of motor vehicles when a lane would otherwise be wide enough to share.

Bicycle-safe and hydraulically efficient gratings are well-known and available from commercial foundries. During a break at the August 14 seminar, I spoke with an engineer from Kansas City, who told me that his city would be specifying bicycle-safe drains. His is an example which, in my opinion, all of Missouri should follow.

General conclusions:

In St. Louis County since 1950, almost all attention to transportation needs has been focused on private automobiles. Arterial roads vary widely in their suitability for bicycling. Alternative routes on local and collector roads appear to exist in many cases, but many trip generators are not connected by convenient and attractive routes. In addition, many parts of St. Louis County are quite hilly, and avoiding excessive climbing is an important consideration for utility bicycling.

The availability of maps indicating preferred through routes can help even in the short term. However, it will be necessary to improve arterials, particularly by increasing outside lane/shoulder width, or to provide connections such as shortcut bicycle paths which establish continuity of routes on local and collector streets.

Fortunately, conditions for bicycle travel are not nearly as bleak in St. Louis County as they are for pedestrian travel. As a vehicle operator, a bicyclist is able to integrate far better with the system of roads than a pedestrian can in an environment lacking sidewalks and pedestrian crossings.

Recommendations:

• Develop a master plan for a grid of through bicycle routes. This should look at all types of facilities and potential facilities including existing arterial routes, local and collector streets, and potential trails as shortcuts and links. In suburban areas, special attention should be paid to the construction of bicycle/pedestrian links between cul-de-sac developments, which can increase the attractiveness of non-motorized transportation by making trips shorter.

• Develop and market maps. Since the through routes preferred by bicyclists are often not the same as those used by motorists, maps are helpful to make those routes known to the bicycling population. In other American cities and states, such maps have typically been developed in the private sector, though sometimes with public assistance or seed money. Bicycle clubs generally can provide a wealth of information about desirable routes.

• Install bicycle-friendly storm gratings. The Missouri standard grating is unsatisfactory, effectively narrowing the roadway. Bicycle-friendly and hydraulically efficient designs exist and should be used both in new construction and to retrofit old installations.

• Eliminate, or narrow, concrete gutter pans. Gutter pans wear and shift position differently from the asphalt pavement, resulting in a hazardous step edge.

• Reclaim the MoDOT shoulder and increase outside lane width. A shoulder can be used more flexibly as bicycle travel space (whether or not designated as a bicycle lane), and/or as parking if it is at the same level as the travel lanes.

• Incorporate consideration of bicycling into decisions about installation of center turn lanes and medians. Center turn lanes are sometimes justified, but sometimes a median is more desirable, as in the Ballwin example illustrated above. Outside lane/shoulder width should be considered when specifying a center turn lane or median. The current design with outside lanes no wider than inside lanes does not account even for shy distance for motorists in the outside lane. I saw many roads with a center turn lane wider than any of the other lanes, even though the center turn lane is hardly ever occupied and could be used most of the time to provide shy distance. A 14-foot outside lane/shoulder is adequate for bicycle/motor vehicle lane sharing. If large trucks and buses are prohibited, 12 or 13 feet may be adequate. Alternatives to a center turn lane include a narrower median, and prohibiting left turns.

• Provide adequate width adjoining traffic islands. Islands should allow room for a bicycle and a motor vehicle to travel side by side.

• Test traffic signal actuators for bicycle sensitivity. Improve design further. The existing quadrupole actuator loops are capable of detecting bicycles but may not have been adjusted to a high enough sensitivity level. Actuator loops need either to be extended into the typical path of bicyclists, or their location needs to be marked, especially when repaving over them renders them invisible. The diagonal (California D-type) loop is even more satisfactory than the existing quadrupole loops; other detection means such as video motion sensors also might be considered, and have the advantage of being able to detect bicycles of largely non-metallic construction..

• Promote driveway consolidation in commercial districts. The shopping center or shopping mall model, with fewer driveways, is preferable to the strip mall because it reduces the number of conflict points.

• Pay special attention to areas where bicycle use is already high. University City, in particular, has a high level of bicycle use, and it is here where improvements would have the greatest impact, at least initially.