|
Article Excerpt
INTRODUCTION
The modern roundabout intersection (see Figure 1) is a staple of transportation engineering in many parts of the world (Brilon, 1991; Brown, 1995; Jacquemart, 1998), with an estimated 30,000 to 40,000 roundabouts worldwide. The first U.S. roundabout was constructed in 1990, and by 2004 the number of U.S. roundabouts had grown to an estimated 500 to 1000, with many more planned (see Kittelson and Associates, n.d.). Although general guidelines exist (Federal Highway Administration, 2000; Ourston & Doctors, 1995; State of Maryland Department of Transportation, 1995), the design of U.S. roundabouts has not been standardized, with U.S. designers being influenced by differing practices employed in Europe (e.g., Brilon, 1991; Brown, 1995; Guichet, 1997; Sawers, 1996) and elsewhere (e.g., Troutbeck, 1993). At roundabouts, exiting traffic does not stop unless yielding to pedestrians crossing the exit lane or lanes. Entering traffic yields to vehicles in the circulatory roadway and to pedestrians in the entry lane's crosswalk. At a typical roundabout, pedestrian crosswalks are located one to three car lengths from the circulatory roadway. Roundabouts typically have raised or painted splitter islands on each leg to separate the entering and exiting traffic, create deflection in vehicle pathways, and provide a pedestrian refuge midway through a street crossing. The inability of vehicles to travel at high speeds through most roundabouts and the fact that vehicles rarely approach each other at right angles are credited for roundabouts' low rates of severe vehicular crashes (Bared, 1997; Federal Highway Administration, 2000; Flannery. 2001).
[FIGURE 1 OMITTED]
This paper considers roundabouts' potential impact on the pedestrian travel of individuals who are blind. Concern has been raised about the safety of roundabout crosswalks for people with disabilities (Crawford, 2002; U.S. Access Board, 2001/2003) as well as for children and older adults (Alphand, Noelle, & Guichet, 1991). The U.S. Americans With Disabilities Act requires that public rights-of-way, including sidewalks and crosswalks, be accessible to pedestrians with disabilities. Anecdotal evidence suggests that blind pedestrians may avoid some roundabouts but not others (Guth, Long, & Ashmead, 2000; Jacquemart, 1998). Other than the present study, however, we know of no systematic efforts to determine whether roundabouts may be more problematic for blind pedestrians than for sighted pedestrians.
The empirical literature about pedestrian-vehicle crashes at roundabouts is very limited; most have compared roundabouts with traditional intersections, and none has mentioned blind pedestrians. The general conclusions of this literature are that two-lane roundabouts are about as safe for sighted pedestrians as traditional intersections carrying similar traffic loads, whereas single-lane roundabouts may be safer (Alphand et al., 1991; Brude & Larsson, 2000; Lalani, 1975; Maycock & Hall, 1984; Retting, Persaud, Garder, & Lord, 2001; Schoon & van Minnen, 1993, 1994; Seim, 1991; Tumber, 1997). An interpretive problem with the available studies is that pedestrian usage has rarely been taken into account. If roundabouts are to some extent avoided by pedestrians, then comparisons of crash rates between roundabouts and other types of intersection may be confounded by differences in pedestrian exposure. Nevertheless, given the low absolute frequency of reported vehicle-pedestrian crashes and the relatively small numbers of pedestrians with visual and mobility impairments, it is unlikely that unique difficulties that roundabouts may pose for those pedestrians would be revealed by crash data. Other empirical approaches are needed to evaluate the accessibility of roundabout crosswalks to blind pedestrians. One such approach is to compare the performance of blind and sighted participants on street-crossing subtasks, such as determining an appropriate time to begin crossing the street. This was the approach taken in Experiment 1.
One factor that affects the accessibility of a roundabout for all pedestrians is the availability of vehicular gaps at the crosswalks that are long enough to cross safely. This paper provides, to our knowledge, the first report of distributions of traffic gaps from the pedestrian perspective. The occurrence of long or "crossable" gaps does not by itself ensure accessibility. Pedestrians must be able to accurately perceive the duration of a gap, and they must relate this information to their knowledge about the distance of the crossing and their walking speed.
In Experiment 1 we measured gaps in traffic at crosswalks at three roundabouts. We related judgments about these gaps made by blind and sighted participants to the time required to cross to the splitter island. We also compared the delays with which blind and sighted participants detected crossable gaps, and we compared the numbers of crossable gaps that blind and sighted participants failed to detect. Participants did not actually cross the street. Instead, they stood near a crosswalk and used push buttons to indicate when they would have initiated a crossing.
EXPERIMENT 1
Method
Participants. The 12 adult participants included 6 who were totally blind and 6 with normal vision (20/20 with or without correction, and full visual fields). Three of the blind participants normally used dog guides as their primary mobility aids and 3 used long canes. Participants' ages ranged from 23 to 56 years and were similar between groups. All participants routinely traveled independently in urban areas. Four blind participants and all sighted participants reported previous but infrequent experience negotiating roundabouts. Four of the sighted participants were orientation and mobility (O&M) instructors, and 1 was a graduate student in O&M. The frequent street-crossing experience of these participants enabled us to compare persons skilled at using vision to make judgments about traffic with persons skilled at using hearing. We did not have the blind participants use their dogs or canes during the trials because the presence of these aids could have affected the characteristics of gaps experienced by the blind and sighted participants. However, this issue is considered further in Experiment 2.
Sites. Data were collected at three roundabouts in the Baltimore, Maryland, metropolitan area. The largest and busiest of these was at Towson, an urban double-lane roundabout located in a busy shopping/business district along a commuter route. It has five legs with an oval-shaped central island and a circulatory roadway with dimensions of 140 x 260 feet (42.6 x 79.2 m). Daily traffic volume, according to state traffic officials, was approximately 40,000 vehicles. Data were collected at painted crosswalks for the two-lane entry and exit lanes of Dulaney Valley Road on the north side of the roundabout. These crosswalks are about one to two car lengths from the circulatory roadway.
The second site was West Street at Taylor Avenue in Annapolis. The circulatory roadway is 168 x 188 feet (51.2 x 57.3 m), and daily volume was approximately 24,000 vehicles. Data were collected at the two-lane red-brick crosswalks at the entry and exit lanes of West Street on the west side of the roundabout, about two car lengths from the circulatory roadway.
The third site was Wilkins Avenue at Hilltop Road, at the entrance to the University of Maryland Baltimore County (UMBC). This single-lane roundabout is 125 feet (38.1 m) in diameter and carries approximately 12,000 vehicles per day. Data were collected at the Hilltop Road leg of the roundabout, where it leads into the campus. There were sidewalks but, unlike the other two sites, there was not a marked crosswalk. Data were collected about two vehicle lengths from the circulatory roadway.
Task. Participants used an indicator task to identify when they judged that it was possible for them to cross from their position on the curb to the splitter island before the arrival of the next vehicle at the crosswalk. Participants were instructed to assume that drivers would not yield to them if they initiated a crossing and to base their judgments on their normal walking speeds. As noted earlier, participants did not actually cross the street.
Each trial lasted 2 min, marked by the words "go" and...
|
