Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
54 Design Guidance This section provides general design guidance related to bicycle lane widths, taking into account a range of road- way and traffic characteristics. The design guidance is based primarily on the results of this research but also takes into consideration the results of previous research. The design guidance primarily pertains to the installation of bicycle lanes on roadways in urban and suburban areas. In the absence of similar data and analyses for rural areas, it is likely that the design guidance is applicable to rural areas; however, appli- cation of the design guidance to rural areas should be done with caution. This section addresses suggested bicycle lane widths as they relate to the following roadway and traffic characteristics: ⢠Parking lane width ⢠Travel lane width ⢠Traffic volume ⢠Vehicle mix ⢠Grade Recall that the 2012 AASHTO Bike Guide indicates that a bicycle lane should range in width between 4 and 8 ft depend- ing on conditions. Under most circumstances, the recom- mended width for bicycle lanes is 5 ft, but wider lanes may be desirable under conditions such as being (1) adjacent to narrow parking lanes with high turnover, (2) in areas with high bicycle use and without on-street parking to allow bicy- clists to ride side by side or to pass each other, (3) on high- speed and high-volume roadways, or (4) on roadways with a substantial volume of heavy vehicles (i.e., trucks). Bicycle lanes as narrow as 4 ft may be used for roadways with no curb and gutter and no on-street parking or on extremely constrained, low-speed roadways with curbs but no gutter where the preferred bicycle lane width cannot be achieved. The AASHTO Bike Guide also states that the recommended width of a marked parking lane is 8 ft, and the minimum width is 7 ft. The 2011 AASHTO Green Book states that the desirable minimum width of a parking lane is 8 ft; however, parking lane widths of 10 to 12 ft may be desirable to provide better clearance from the traveled way and to accommodate use of the parking lane during peak periods as a through-travel lane. Parking lane widths of 10 to 12 ft are also sufficient to accom- modate delivery vehicles and allow a bicyclist to maneuver around an open door of a parked motor vehicle. The Green Book also notes that 7-ft parking lanes have been successfully used on urban collector streets within residential neighbor- hoods, where only passenger vehicles need to be accommo- dated in the parking lane. The suggested design guidance based on the results of this research is presented in the following. It is important to note that throughout this research and analyses, where a buffer space is present, its width is not included in the width of the bicycle lane. Also, no data were included in the analyses that considered the position of two (or more) bicyclists riding side by side or one bicyclist passing another bicyclist in the bike lane. Therefore, the design guidance presented is intended for designing facilities to accommodate a bicyclist riding alone or more than one bicyclist riding in single file behind another. Finally, in developing the design guidance, equal weight is given to designing bicycle lanes to reduce the risk of bicycle crashes involving open doors of parked vehicles and bicycle crashes involving passing vehicles (i.e., moving vehicles in the travel lanes). Bicycle Lane Width This research investigated bicycle lanes ranging in width from 3.5 to 6 ft. In general, there was no practical difference in bicyclist positioning when operating within the bicycle lanes of these varying widths. When adjacent to an on-street parking lane, a majority of the cyclists positioned themselves within the open door zone of parked vehicles, regardless of the width of the bicycle lane. Thus, in the context of design S E C T I O N 5
55 guidance presented in the current Bike Guide, there is no evi- dence to suggest that a 6-ft bicycle lane provides any addi- tional benefits to bicyclists in terms of drawing or moving bicyclists away from the door zone of parked vehicles com- pared to a bicycle lane width of 5 ft, or even as narrow as 3.5 ft or 4 ft. It should be noted, however, that the effect of increasing the bicycle lane width up to 6 ft without making a corresponding reduction in parking lane width (or buffer width) was not fully considered. Similarly, when adjacent to a vertical curb (without a gutter), there was no practical dif- ference in bicyclist positioning when operating within bicycle lanes of 4 ft as compared to 5 ft. The width of the bicycle lane does slightly affect the spread of bicyclist lateral positions, in that narrowing the bicycle lane reduces the variability of bicyclist lateral positions; how- ever, the impact is relatively minor. For example, the average spread of bicyclist lateral positions within a 6-ft bike lane was 3.0 ft, while the average spread of bicyclist lateral positions within a 3.5- to 4.0-ft bike lane was 2.6 ft. Thus, narrowing the bicycle lane by 2 ft reduced the average spread of bicyclist lateral positions by 0.4 ft. Therefore, in terms of accommodating bicyclists within a bicycle lane, there does not appear to be a distinct advan- tage of providing a wider bicycle lane compared to a nar- rower bicycle lane, at least when considering bicycle lane widths between 3.5 and 6.0 ft. Widening or narrowing the bicycle lane does not necessarily move bicyclists away from the door zone of parked vehicles, nor does it practically effect the spread of bicyclist lateral positions within the bicycle lane. However, evidence suggests that providing a buffer space between the parking lane and the bicycle lane is desirable. When a buffer was provided between a bike lane and a parking lane, bicyclists positioned themselves further away from the door zone of parked vehicles, and as a result, a higher percent- age of bicyclists were within the effective bike lane compared to when no buffer space was provided. The recommended buffer space is at least 1.5- to 2-ft wide and preferably marked with white diagonal cross hatching or chevron markings. The Man- ual on Uniform Traffic Control Devices requires that a buffer space wider than 4 ft be marked with chevrons (or diagonal cross hatching), while the National Association of City Trans- portation Officials requires chevrons or diagonal cross hatch- ing for a buffer space of 3 ft or wider. The study scenarios evaluated in this research did not include any scenarios with a buffer space wider than 2 ft. In each scenario that included a buffer, a buffer was present between the parking lane and the bicycle lane. In one sce- nario (Y-11), a buffer was also present between the bicycle lane and the travel lane. Based on the scenarios evaluated, the placement of the buffer spaces across the scenarios, the distribution of the bicyclists within the roadway cross sec- tion, and the placement of passing vehicles, evidence suggests that it is more desirable to install the buffer space between the parking lane and bike lane where on-street parking is permit- ted. In addition, data suggest that for parking lanes widths of between 7 and 9 ft and a buffer space of only 1 to 2 ft, a sizable portion (40% to 60%) of bicyclists may still position them- selves within the door zone of parked vehicles. Thus, when adjacent to narrow parking lanes, it is desirable to provide a wider buffer space up to a maximum of 4 ft. Caution should be used in marking too wide of a buffer space since this may result in motor vehicles using the buffered bike lane even if it is properly marked. In summary, providing a buffer space between a parking lane and bike lane offered distinct advantages over simply providing a wider bike lane. Parking Lane Width This research investigated parking lanes ranging in width from 7 to 9 ft. From a bicyclistâs perspective, the primary interest was to determine if the parking lane width influ- enced how close drivers parked their vehicles to the curb, which affects the overall displacement of the vehicle from the curb and potential placement of an open car door. Although a general trend in the data suggests that drivers park their vehicles closer to the curb as the parking lane narrows from 9 ft to 7 ft, the results are not statistically different. The data also show that for parking lane widths of 7 ft, approximately 5% to 15% of parked vehicles extend beyond the limits of the parking lane. Therefore, to accommodate a larger percent- age of drivers, a parking lane width of 8 ft is suggested for when a bicycle lane is adjacent to the on-street parking. An 8-ft parking lane allows more of the roadway cross section to be designated for use by bicyclists and motor vehicles in a bicycle lane and the travel lanes compared to a 9-ft parking lane. When the roadway cross section is limited or if there is a desire to install a buffered bike lane, a 7-ft parking lane may be used adjacent to a bicycle lane. For parking lanes 7- to 9-ft wide, the open door zone width of parked vehicles extends approximately 11 ft from the curb, assuming the 95th-percentile parked vehicle displacement and an open door width of 45 in. Thus, where bike lanes are adjacent to parking lanes 7- to 9-ft wide, the design of the bike lane should encourage bicyclists to ride outside of this door zone area (and account for the width of the bicyclist). Travel Lane Width This research investigated travel lanes ranging in width from 10 to 18 ft. The widest travel lane adjacent to a bicycle lane was 14 ft. During the field data collection, few passing vehicles were observed encroaching into the bicycle lanes for most of the study scenarios, even from the narrowest 10-ft
56 travel lane. Similarly, few passing vehicles likely encroached into adjacent travel lanes to the left, especially when encroach- ment involved crossing the centerline of the roadway. Thus, based on these field observations, travel lanes between 10 and 12 ft in width were found to be appropriate when adjacent to a bicycle lane. This is consistent with previous research (Potts et al., 2006) that indicates the use of travel lanes narrower than 12 ft on urban and suburban arterials does not necessarily increase expected crash frequencies and that geometric design policies should provide flexibility for use of lane widths nar- rower than 12 ft. With respect to wide curb lanes, this research investi- gated travel lanes of 16 and 18 ft in width on streets without on-street parking. Marking a bicycle lane of 4 or 5 ft in width on such a facility may have some advantages in dis- tinguishing allocation of roadway width and minimizing the potential for operation of two motor vehicles side by side, but there was no practical difference in the bicyclistsâ posi- tioning between the scenario with a 4- or 5-ft marked bicycle lane (narrowing the travel lane to 11 to 14 ft) and a scenario with a wide curb lane and no marked bicycle lane. On streets without on-street parking and travel lanes of 16 and 18 ft in width, whether a marked bicycle lane is provided or not, the effective bike lane is, for practical purposes, the same, and almost all bicyclists will position themselves within the effec- tive bike lane. Traffic Volume This research included study sites with traffic volumes ranging between 14,800 and 29,000 vpd. The data show that as traffic volume increases, bicyclists move away from vehicles in the travel lane and position themselves closer to parked vehi- cles or the curb. In the analyses that were performed, the traffic volumes were categorized as lower ADT (15,000 to 17,000) and higher ADT (29,000). It was found that bicyclists positioned themselves approximately 1.5 to 2.5 ft closer to parked vehicles or the curb at the higher ADT level compared to the lower ADT level. As such, on streets with ADTs above 20,000 vpd, additional displacement of bicyclists due to traffic volume should be considered when determining the allocation of street width between parking lanes, bicycle lanes, and travel lanes. In particular, consideration should be given to desig- nating additional street width to bicyclists and/or providing a buffer to account for the additional displacement of bicyclists at higher traffic volumes. Vehicle Mix This research included study sites with the percentage of trucks in the vehicle mixes ranging from between 2% and 20%. Similar to traffic volume, the data show that as truck percentage in the vehicle mix increases, bicyclists move away from vehicles in the travel lane and position themselves closer to parked vehicles or the curb. In the analyses that were per- formed, the truck percentages were categorized as low (<10%) and high (16% to 20%). It was found that bicyclists positioned themselves approximately 2.5 to 3.0 ft closer to parked vehicles or the curb at the higher truck percentage level compared to the lower truck percentage level. As such, on streets with truck percentages above 10%, additional displacement of bicyclists due to trucks should be considered when determining the allocation of street width between parking lanes, bicycle lanes, and travel lanes. In particular, consideration should be given to designating additional street width to bicyclists and/or pro- viding a buffer to account for the additional displacement of bicyclists at higher truck percentages. Grade This research included a supplemental grade study in which cyclists pedaled up a moderate grade of 3% to 4%. The average observed back-and-forth sway of the cyclists was approximately 6 in., while their deviation from a straight-line trajectory was typically between 3 and 4 in. Given that so few bicyclists position themselves within 6 in. of the outside edge of a marked bicycle lane, there is not sufficient evidence to suggest the need to widen a bicycle lane on moderate to steep upgrades to account for potential back-and-forth sway of cyclists while pedaling up the grade. Allocation of Total Roadway Width Based on the research results, Table 19 provides guidance for suggested lane widths for total roadway widths measuring 44 to 54 ft curb to curb, based primarily on the percentage of bicyclists riding within the effective bike lane and the esti- mated central positioning of bicyclists, while accounting for traffic volume, truck percentages, and the presence/absence of a buffer. The suggested lane widths are not the direct result of a single analysis performed as part of this research but are based on the combined information collected during the research. Table 19 is most applicable to urban and suburban two- lane undivided roadways, with constrained roadway width and on-street parking, and with a posted speed limit 30 mph. The roadway could function either as an arterial or collec- tor roadway. For all locations, engineering judgment needs to be exercised when selecting the final allocation of road- way width, taking into consideration the safety, mobility, and accessibility of all roadway users. The results of this research are most applicable to assist in providing design guidance for allocation of lane widths for total roadway widths mea-
57 suring 44 to 54 ft curb to curb. The guidance generally reflects that a buffer space provides distinct advantages over simply providing a wider bike lane and that providing a buffer space on both sides of the bike lane may help bicyclists to ride within the effective bike lane on roads with higher traffic volumes or truck percentages. Table 19 does not provide design guidance for total road- way widths greater than 54 ft or less than 44 ft. For total road- way widths greater than 54 ft, designers have more flexibility to provide wider lane widths and need less guidance due to the availability of space (e.g., additional width can be allo- cated to the travel lane or parking lane). On the other hand, for total roadway widths less than 44 ft, conditions are so con- strained that based on the analysis results, it is suggested that bike lanes not be marked but rather a shared lane be provided adjacent to a parking lane and/or the roadway be marked with a shared-lane marking. Table 19 does not provide guid- ance on where it might be more appropriate to install or use a shared-lane marking rather than a bike lane. Also, Table 19 does not consider or address cross sections with a two-way lane, left-turn lane, or multiple lanes in the same direction of travel, although the general suggestions are still appli- cable. Concepts for designing Complete Streets could also be considered when determining the final allocation of road- way width. Table 19 provides several design options for total roadway widths of between 44 and 50 ft. For total roadway widths of 52 ft or more, design decisions concerning allocation of lane widths can be made independent of traffic volumes and truck percentages, and the same is true for total roadways widths of 46 ft or less. However, for total roadway widths of between 48 and 50 ft, several different suggested lane widths are pro- vided depending on the expected traffic volumes and truck percentages for the roadway. The design guidance provides suggested lane widths for four categories of traffic volumes or truck percentagesâthat is, low and high traffic volumes and low and high truck percentages. The range of traffic vol- ume and truck percentage categories used to evaluate the effect of roadway characteristics on the central positioning of bicyclists should be used as a rule of thumb for distinguish- ing values for low and high volumes and truck percentages in Table 19, as follows: ⢠Low volume: 15,000 to 17,000 vpd, ⢠High volume: 29,000 vpd, ⢠Low truck percentage: <10%, and ⢠High truck percentage: 16% to 20%. For example, a threshold value of 20,000 vpd may be reason- able to distinguish between low and high traffic volumes. Simi- larly, a threshold value of 10% trucks in the vehicle mix appears to be a reasonable value to distinguish between low and high truck percentages. The design guidance shown in Table 19 suggests that the combined width of the buffer area(s) and bike lane be a mini- mum of 5 ft and a maximum of 9 ft. Caution should be used in designing a buffer area and bike lane with a combined width greater than 9 ft because it may promote the use of this portion of the roadway by motor vehicles, even when prop- erly marked and designated as a bike lane. Widths (ft)âOne Direction of Travel Curb to Curb (ft) Travel Conditions1 Parking Lane Buffer Bike Lane Buffer Travel Lane Curb to CL 8 3* 4 2 10 27 54 All conditions 7 3* 4 2 10 26 52 All conditions 7 2* 4 2 10 25 50 High volume or high truck percentage 7 3 5 0 10 25 50 Low volume and low truck percentage 7 1.5 4 1.5 10 24 48 High volume or high truck percentage 7 3 4 0 10 24 48 Low volume and low truck percentage 7 2 5 0 10 24 48 Low volume and low truck percentage 7 2 4 0 10 23 46 All conditions 7 0 5 0 10 22 44 All conditions 7 1** 4 0 10 22 44 All conditions * May consider combining buffers to create a 4-ft buffer between parking and bike lanes. ** Caution that striping of double white lines may cause confusion. 1 The suggested threshold for distinguishing between low and high traffic volume is 20,000 vpd, and the suggested threshold for distinguishing between low and high truck percentage is 10% trucks in the vehicle mix. Note: CL = center line. Table 19. Suggested lane widths for urban and suburban two-lane undivided roadways with on-street parking and constrained roadway widths.
58 Table 19 suggests that suggested parking lane widths on roadways measuring 44 to 54 ft curb to curb should be 7 ft. The Green Book states that 7-ft parking lanes have been used on urban collector streets within residential neigh- borhoods, but in most other situations, the desirable mini- mum width of a parking lane is 8 ft (as was also suggested based on the results of this research). Given that a buffer and in some cases two buffers are suggested in conjunction with the designated bicycle lane under constrained condi- tions, providing a 7-ft parking lane adjacent to a buffered bicycle lane on a wider range of facility types than simply urban collector streets within residential neighborhoods seems appropriate. Finally, for streets where on-street parking is prohibited, the analysis results from this research indicate that the mini- mum bike lane width should be 4 ft, measured from the face of curb or vertical surface to the center of the bike lane line, for roadway widths of 32 ft or greater (measured curb to curb) and may be appropriate for roadway widths as narrow as 28 ft. For roadways with higher volumes or higher truck percentages, a bike lane width of 5 ft is desirable. It is also worth mentioning that although this research did not evalu- ate bike lane widths as narrow as 3 ft, Hunter and Feaganes (2003) concluded that marking a 3-ft bike lane provides ben- efits over a wide curb lane. Along sections of roadway with curb and gutter or guardrail, the usable width of the bike lane should be considered when determining the desired width for the bike lane. Limitations of the Research This design guidance needs to be considered within the context of the research. In particular, it should be noted that the suggested allocations of roadway widths are based on data collected along streets with posted speed limits of 30 mph. The speeds of motor vehicles in the travel lane adjacent to a bike lane likely affect the comfort and positioning of bicyclists within the bike lane. Therefore, the suggested allocations of roadway widths should be used cautiously for the design of roadways with motor vehicle speeds outside of the range of 25 to 35 mph and, in particular, for higher-speed roadways. In addition, data were collected only at five sites in two cities. It would have been desirable to collect data at more sites in additional cities. This would have permitted a wider range of roadway characteristics to be evaluated and analyzed. As such, a limited dataset was used to generalize results and make them applicable to other communities/cities. It should also be recognized that physical and financial constraints typically exist, so agencies must do the best that they can within their means and with available resources. This is to say, if physical and financial constraints did not exist, from a motorist and bicyclistâs perspective, it would be desirable to provide 12-ft travel lanes, 7-ft bike lanes with buf- fers, 10-ft parking lanes, and so forth. Such lane widths and cross sections would provide additional separation between vehicles/bicycles within adjacent lanes; however, such wide cross sections could also result in undesirable consequences such as increased speeds of motor vehicles and increased crossing distances for pedestrians. However, all agencies must deal with the realities of financial limitations, and particu- larly during the construction or reconstruction of urban and suburban streets, right-of-way constraints limit the amount of total roadway width that can be allocated to accommo- date a range of users. Thus, decisions must be made regarding allocation of roadway width to comfortably serve all road- way users. While it may not always be practical to provide an ideal design, cross section, or allocation of total roadway width, the reality is that, in some situations, lane widths may be what would be less than preferred in an effort to balance the needs of all roadway users. For example, the resulting effective bike lane widths may be less than the physical, mini- mum, or preferred operating space of a bicyclist as defined in the current edition (2012) of the AASHTO Bike Guide, or the minimum shy distance separating a vehicle from another vehicle (or bicyclist) as recommended in the Roadside Design Guide (AASHTO, 2011) may not be achievable, especially under constrained conditions. Finally, it should be emphasized that the objective of this research was to develop a set of suggestions for bicycle lane widths for various roadway and traffic characteristics under the overall guiding principle to provide suggestions on how wide the bicycle lane should be in cases where a decision to include a bicycle lane has been made. It should be made very clear, as evident by the data collected and analysis results, that the design guidance presented herein does not eliminate the potential risk to bicyclists striking open car doors of parked vehicles or being struck by passing vehicles, nor does the design guidance eliminate the potential for encroachment of passing vehicles into adjacent (motor vehicle) travel lanes to the left. When a facility is designed, whether it meets or does not meet current guidelines, it is important to fully under- stand the risks associated with all road users. This report sheds light on the risks faced by road users for a certain range of roadway and traffic characteristics where bicycle lanes may be provided. Given the objective, scope, and limitations of this research, it is understood that some of the design guidance suggested from this research could be viewed as controversial. The basic question that has to be posed is, âParticularly for constrained or fixed roadway widths, which facility type is most desir- able from a bicyclist perspective: a shared lane, a marked shared lane, or a bicycle lane?â This research did not answer this basic question, but rather focused on providing design guidance for a bicycle lane given the decision that a bicycle
59 lane will be installed. Proceeding from this research, roadway designers and transportation agencies have several options concerning the use of the suggested design guidelines. They can (1) accept the design guidance suggested and incorporate the design guidance for bicycle lanes within their local design practices, (2) interpret the data and analysis results differ- ently than what has been presented and develop their own design guidance for bicycle lanes, or (3) reject the suggestions (and potentially focus on designs for a shared lane or a road- way with a shared-lane marking, or where on-street parking is permitted, eliminate the parking in favor of a dedicated bicycle lane). However the results of this research and design guidance are viewed, it should be remembered that as stated in the Foreword of the AASHTO Green Book, good highway design involves balancing safety, mobility, and preservation of scenic, aesthetic, historic, cultural, and environmental resources. A design policy is not intended to supersede the need for appli- cation of sound principles by knowledgeable design pro- fessionals but is intended to provide sufficient flexibility to encourage independent designs tailored to particular situa- tions, and engineering judgment is to be exercised to select appropriate design values.
