Design Guide for Low-Speed Multimodal Roadways (2018)

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.

76 4.1 General Considerations for Traveled Way Design for All Users The discussion in this section provides information and guidance to the design process for the traveled way of any roadway and street project serving a mix of users with a vehicle design speed of 45 mph or lower. 4.1.1 Roadway Uses, Users and Activities in Low- and Intermediate-Speed Environments As discussed in previous chapters of this Guide, designing road and street facilities that effec- tively serve all current and planned users in low- and intermediate-speed environments can be a challenging process. Often, minimum and desirable vehicle-focused geometric design criteria and dimensions are required or suggested by local, state and federal manuals and guidelines. In many project locations, these criteria and dimensions are not attainable, attainable only at significant cost, or providing them may negatively impact conditions for other users of the right-of-way. Nonetheless, considerable flexibility is available to the designer in evaluating and making these often-difficult trade-offs in design criteria and dimensions. Pedestrians are the most vulnerable roadway users because they are at the greatest risk of injury or death in a collision with someone traveling by any other mode. Bicyclists generally travel at slower speeds than motorized vehicles and are inherently more vulnerable in the event of a crash with a car, truck or transit vehicle. Improving the LOS, QOS or performance for one user mode may negatively affect those same elements for one or more of the other modes. Certainly, technical analysis driven by quantitative data can provide useful information to the designer about the impacts of these design choices and modal interactions, but no single pro- cess tool can make absolute choices for a “best fit” design solution to meet a project’s purpose and need. Balancing and blending modal accommodation in the geometric design process often involves technical analysis supported by a qualitative, even subjective, process that involves policy choices, engineering judgment and the use of flexible and unique design approaches. This Guide identi- fies the ranges of flexibility that are available in current national design practice and provides guidelines for how to apply that flexibility. Greater awareness of the flexibility and versatility available in national guidance will help designers overcome many challenges related to both new and retrofit project designs. This chapter addresses criteria for a wide range of elements that may be involved in the design of a traveled way that serves motorized vehicles, including trucks, transit and possibly bicycles, in low- and intermediate-speed environments. Chapter 5 addresses criteria for design of roadside C H A P T E R 4 Traveled Way Design Guidelines

Traveled Way Design Guidelines 77 areas adjacent to the traveled way and which may include facilities for pedestrians, bicyclists, transit access and other elements that support the land use context. 4.1.2 Relationship of Traveled Way and Roadside Environments A roadway design project may involve improvements to the traveled way, to the roadsides, or to both the traveled way and the roadsides. Even if a design project involves only one of these two realms, consideration of the other realm is essential to the design process because of the proxim- ity and interaction of the traveled way with the roadside. Land use context creates a third realm that also should be considered in the design process because the context informs the designer how the land use is served by, and relates to, all users of both the traveled way and roadsides. This chapter addresses design elements and criteria for a traveled way that serves: • Motorized vehicle users (moving and parked), • Bicycle users (in some but not all possible settings), and • Pedestrians (e.g., on shoulders where sidewalks do not exist and shoulders may be the only all-weather travel surfaces available to them). Although shoulders are not substitutes for a well-designed separated pedestrian facility in the roadside, the need may occasionally arise to design shoulders as walkways where roadside space or funding is constrained. The FHWA memorandum titled ACTION: Consideration and Implementation of Proven Safety Countermeasures states that “walkable shoulders (minimum of 4 ft. stabilized or paved surface) should be provided along both sides of rural highways routinely used by pedestrians” (FHWA 2008). This Guide presents design guidance for the traveled way, roadside and intersections in three distinct chapters, but all three design environments usually have many interrelation- ships involving multiple modal users. Therefore, the full right-of-way cross section and intersection design development process should be integrated as alternatives are developed and analyzed. 4.1.3 Understanding Context and Multimodal Relationships Understanding context is considered a necessary element of effective multimodal roadway design (see Chapter 2). This is especially true for the traveled way component of the roadway because of the interactions of non-motorized users who share the traveled way, travel adjacent to the roadway, and cross it at intersections and non-intersection locations. This Guide presents design guidance that requires a thorough understanding of the existing and future context of a project area and how that context affects multimodal activity within and adjacent to the roadway project limits. The application of context also requires a designer to know how to apply geometric design controls and criteria of the roadway to support beneficial interactions between the roadway, roadside and existing and planned multimodal activity gener- ated by adjacent land uses and local modal networks. To the extent possible, land use context and multimodal demand should be considered dur- ing the earliest stages of project development so that these needs can be addressed before the preliminary and final design stages. The designer also should recognize that existing and future context are both important design considerations, much as existing and future vehicle design hour volumes are important for determining vehicle needs. Finally, designers should realize that context—which may be defined initially as a specific context zone category for a project—often will vary within each general context zone. Context zones also may vary throughout the limits of a roadway project. The combination of multimodal functions, land use context, network

78 Design Guide for Low-Speed Multimodal Roadways needs, community goals and other factors may cause the design concept, cross section and design elements to vary throughout the roadway’s length. Project alternatives should emerge from a full understanding of the relationships between the roadway, the adjoining properties, the character of the broader area and other unique proj- ect circumstances, values or objectives. Preferably, a roadway project’s modal priorities and emphasis will have been clearly identified through the project planning and concept develop- ment phases. If this has not occurred, then the design process should address these needs to the extent feasible given the project’s scope, budget, available right-of-way and other considerations. The design process for low- and intermediate-speed multimodal roadways requires an expanded understanding of context beyond elements like on-street parking and access management. Context is highly dependent on many aspects of land use, including building and site design, which can support integrated pedestrian, bicycle and transit activities and environments. 4.1.3.1 Land Use Types Land use is a common criterion for characterizing development and estimating vehicle trip generation, particularly in vehicle-dominated areas. The design process should recognize land use as an important contributor to the overall project context and as a major factor in the selection of design criteria related to the levels of current and projected motorized and non-motorized travel. Land use also is a key factor in selecting and assembling components of the cross section and determining how the available roadway right-of-way is allocated to particular uses. In addition to having a fundamental impact on automobile travel demand, variations in adja- cent land use will affect the width and design of the roadside, the part of the roadway between the curb or shoulder and the edge of the right-of-way, including sidewalks. Differing land uses have differing needs for design elements such as clear sidewalk space, landscaping, street furniture, bicycle parking and so forth. Commercial uses tend to generate higher volumes of pedestrian and bicycle travel than do office or industrial uses. Typically, commercial areas also have a higher volume of delivery trucks and buses and usually have a higher turnover of on-street parking than do residential areas. 4.1.3.2 Property Site Design The ways in which buildings, circulation, parking and landscaping are arranged on a site affect how adjacent roadsides and roadways should be designed. Ideally, roadway designers understand and are familiar with these relationships. The specific elements of site design that contribute to defining context in non-rural areas include: • Building orientation and setback. In contexts where walking has lower priority, such as tra- ditional low-density suburban contexts, buildings typically are less related to the street, either by having large setbacks into private property or by being oriented toward a parking lot rather than the street. By contrast, in contexts with traditional urban and urban core characteristics, buildings will be oriented toward and often adjacent to the roadway, and these contexts will involve a higher priority for pedestrian and bicycle access. The directness of the multimodal connection to the building entry from the roadway, and how the building may be integrated into the roadside, also varies with differing land use types. In some locations, buildings may form a continuous built edge or street wall (see Exhibits 4-1 and 4-2 for rural village and urban context examples). • Parking type and orientation. Parking provided in surface lots between buildings and road- ways typically define a vehicle-dominated context with a lower priority for walking and biking. Conversely, on-street parking, structured parking and parking behind buildings accessed by alleys or side streets is an urban or urban core characteristic. Roadways in these types of contexts will typically have a higher priority for walking and biking.

Traveled Way Design Guidelines 79 Source: North Carolina DOT (2012) Exhibit 4-1. Components of a typical rural village main street. Source: North Carolina DOT (2012) Exhibit 4-2. Components of a typical urban main street. • Block length. Development patterns with traditional urban, urban core and many rural town contexts usually have shorter block lengths with a network of closely connected streets and possibly alleys. Vehicle-focused contexts such as suburban and some urban contexts tend to have larger block lengths, fewer street connections and often no alley access. These types of street patterns typically require longer walking distances and will tend to generate lower pedestrian volumes along higher volume roadways. Typically, shorter block lengths can lead to greater accessibility throughout an area for pedestrians. 4.1.3.3 Site Building Design Building design contributes significantly to context and the priority that the context gives to walking. Building access, height, scale, density, architecture, relationship to adjacent buildings

80 Design Guide for Low-Speed Multimodal Roadways and roadways, and the type of ground-floor land use in multistory buildings affect context and multimodal activity. Contexts that give a lower priority to walking generally are developed to be more internally oriented and may even minimize interaction with adjacent roadsides and sidewalks. The lack of walkability in these contexts is not correlated with building type but with features of building and site design. In a traditional urban, urban core and even rural town context, buildings typically are ori- ented toward the street. Ground-floor uses in these buildings are usually oriented to the pedes- trian passing on the adjacent sidewalk. Building design elements that contribute to an urban context include: • Building height. Buildings are the primary feature of urban contexts that create a sense of definition and enclosure of a roadway—an important urban design element that helps create the experience of being in a city and in a place that is comfortable for pedestrians. Highly multi modal roadways do not require tall buildings. Street trees may be used to provide a similar sense of definition and enclosure in contexts with lower height and less density of buildings. • Building width. Like building height, building width contributes to the sense of enclosure of the roadway right-of-way. There are three elements of width: (1) the percentage of a build- ing’s width fronting the street, (2) the distance between buildings (building separation), and (3) building configurations. As with building height, greater building width along a roadway tends to create a sense of definition and enclosure of the corridor. This feeling of enclosure contributes to drivers feeling more comfortable operating at lower travel speeds. • Building scale and variety. The scale and variety of buildings help define the context and character of a roadway and encourages walking by providing visual interest to the roadway. The scale and variety of buildings should help define the scale of the pedestrian environment. Vehicle-oriented building scale maximizes physical and visual accessibility by drivers and passengers of motorized vehicles, contributing to contexts that discourage walking. • Building entries. Building entries are important in making buildings accessible and interest- ing for pedestrians. To maintain or create traditional urban character, buildings should have frequent entries directly from adjacent roadways to improve connectivity and to break down the scale of the building. Frequent entries from parking lots and secondary roadways should be provided as well. More information on how building design relates to context can be found in the ITE publica- tion, Promoting Sustainable Transportation Through Site Design (ITE 2004) and in the SmartCode from the Center for Applied Transect Studies (CATS 2003). 4.1.3.4 Selecting Context Zone(s) in Roadway Traveled Way Design The design process presented in this Guide uses a group of four context zones as a primary initial consideration in selecting the design elements and criteria for multimodal roadways. As emphasized throughout this document, context helps guide the selection of basic design elements and criteria for low- and intermediate-speed roadways with a mix of motorized and non-motorized users. This chapter focuses on guidelines that help the practitioner identify and select context zones as one of the first steps in the design process. Exhibit 4-3 defines context in five context zone categories (as recommended in NCHRP 15-52). All but the rural context zone are addressed in the Guide. Deciding which context zone to use for a particular project, or combination of contexts, can be difficult for some projects. The designer also should consider land use or modal variations within each context zone when developing cross section and design elements.

Traveled Way Design Guidelines 81 Guidelines for selecting the context to be used to inform the geometric design process include the following: • Consider both the existing conditions and the plans for the future, recognizing that roadway improvements usually last longer than development; • Assess area plans and review general, comprehensive and specific plans, zoning codes and community goals and objectives which may provide detailed guidance on the vision for the area; • Compare the area’s predominant land use patterns, building types and land uses to the char- acteristics presented in Exhibit 4-3; • Pay particular attention to residential densities and building type, commercial floor-area ratios and building heights; • Consider dividing the area into two or more context zones if a range of land use characteristics suggests multiple context zone types; and • Identify current levels of pedestrian, bicycle and transit activity, and estimate future levels and circulation needs based on the type, mix and proximity of land uses. 4.1.4 Multimodal Network Considerations Project design usually takes place at a much smaller scale than design at the network level, but in the design of multimodal projects it is important to understand the network role of the facility on which the project is located. A roadway’s functional classification (arterial, collector, local) as defined in the federal, state and regional transportation planning processes is a primary network consideration. These designations are based solely on the mobility of motorized vehicles; they Context Category Density Land Use Character Setback Typical Level of Multimodal Activity Rural Lowest (few houses or other structures) Agricultural natural resource preservation and outdoor recreation uses with some isolated residential and commercial Usually large setbacks Low or Very Low Rural Town Low to medium (single family houses and other single purpose structures) Primarily commercial uses along a main street (some adjacent single family residential) On-street parking and sidewalks with predominately small setbacks Low to Moderate Suburban Low to medium (single and multifamily structures and multistory commercial) Mixed residential neighborhood and commercial clusters (includes town centers, commercial corridors, big box commercial and light industrial) Varied setbacks with some sidewalks and mostly off-street parking Low to Moderate; possibly High in school and recreation settings Urban High (multistory, low rise structures with designated off- street parking) Mixed residential and commercial uses, with some institutional and industrial and prominent destinations Minimum on-street parking and sidewalks with closely mixed setbacks Moderate to High Urban Core Highest (multistory and high-rise structures) Mixed commercial, residential and institutional uses within and among predominately high-rise structures Small setbacks with sidewalks and pedestrian plazas High to Very High Source: Stamatiadis et al. (2017) Exhibit 4-3. Land use context zones.

82 Design Guide for Low-Speed Multimodal Roadways do not address a facility’s role in the mobility of other modes, or how the facility relates to the community and the adjacent land use context. Ideally, network planning is integrated into a comprehensive planning process that concurrently addresses land use, transportation and environmental needs. In practice, especially in regions with multiple jurisdictions, network planning is often conducted in a piecemeal manner by multiple agencies with different geographic jurisdictions, missions and powers. For the practitioner plan- ning or designing a roadway segment, considering network design and function can lead to solu- tions that help to balance demands for vehicle capacity and support for community goals. The geometric design process should recognize the role of a roadway as part of at least one, and potentially several, plans, from large-scale vehicle-focused network plans (such as the U.S.DOT’s NHS Plan) to more multimodal local neighborhood and corridor plans. The project design pro- cess needs to consider the planned state, regional, sub-regional and neighborhood functions of the roadway facility in relation to context and community goals and values. The design of the individual roadway project, therefore, is guided by both its context and the performance of the network. A multimodal network facility may identify some roadways that emphasize vehicles or trucks and others that emphasize pedestrians and transit. One difficult situation that is often encountered in the design of projects on arterial roadways (and some collector roadways) is the tension between local residents’ and communities’ desire to emphasize livability, character, walkability, bikeability and other non-vehicle mobility goals and transportation agencies’ desire to emphasize vehicle capacity or accommodation of projected regional travel demand. This tension is best addressed through consideration of the broader network and corridor in conjunction with the individual roadway. Network goals and considerations may be informed by several levels of network plans, as described in Exhibit 4-4. The role that the design facility serves in each of these modal network levels will influence the geometric design of that facility. For example, a design project on a state and federal-aid highway route in an urban region may have multiple goals for multiple modes including the multiple layers of planning goals established by federal, state, regional and local agencies. Type of Transportation Network Plan Possible Modal Plan Elements Light Vehicle Freight Transit Bicycle Pedestrian Other Community Goals * Federal: NHS Plan X X State Transportation Plans X X X X X Regional Transportation Plans (MPO/TPO) X X X X X X Local Transportation Plans (County/City) X X X X X Transit Agency Service Plans X Community Plans X X X X X Corridor Plans X X X X X X Neighborhood Plans X X X X X *Land use (context), urban design, housing, community facilities, recreation, parks/open space, utilities, economic development Exhibit 4-4. Planning documents that address modal elements in project design.

Traveled Way Design Guidelines 83 4.1.5 Functional Requirements of Multimodal Roadways Multimodal accommodation can exist on any functional classification of roadway (e.g., arte- rial, collector, or local), but this Guide primarily addresses accommodation needs on arterial and collector roadways. It is typically on those facilities where user types and volumes, vehicle speeds and context interactions combine to present the most challenging conditions to a designer. However, designers should keep in mind that each roadway design is unique, and the ultimate design needs to address the context, objectives, priorities and design concept established for all aspects of the facility and corridor. Consequently, the unique combination of roadway design elements and criteria developed for a project may differ from the Guide’s recommendations for individual design elements and criteria. The process of understanding the relationships between design elements and criteria and balancing them against each other in the design process is the essence of flexibility in the geometric design process. Exhibit 4-5 shows how basic functional requirements and characteristics for a roadway design may vary as project context changes. In itself, a context zone is not the complete indicator of the presence and level of various modes; however, in general the multimodal activity increases as context changes along the continuum from rural to urban settings. The presence of non-motorized users, together with the context, should directly influence the selection of project design elements and criteria. Although the characteristics for multimodal roadways of all functional classifications have common elements, the roadway’s functional classification influ- ences the design characteristics and cross section elements to some degree. Higher order regional routes with longer average trip lengths and higher volumes of vehicles, trucks and buses, will create more vehicle-focused roadway designs even though other users must also be provided reasonable levels of mobility and accommodation. These types of facilities often are the most difficult to design given the significant impacts among modes when particular users are favored in the design trade-off process. 4.1.6 Design Controls for Multimodal Roadways This section identifies the differences in design process controls used where vehicle capacity is the priority consideration versus where higher levels and QOS to pedestrians, bicycle, and context interaction is the priority consideration. As discussed in Chapter 2 of this Guide, design controls are physical and operational char- acteristics that guide the selection of criteria in the design of roadways for users. Some design controls are fixed, including topography and certain user performance characteristics, but many other controls can be influenced through design, and these are determined by the designer. The intent of selecting appropriate design controls is to create design outcomes that best meet the purpose, need and goals of the project. The AASHTO Green Book and several of its supplemental publications identify the func- tional classification and type of project location (urban or rural) as a design control and sug- gest different design criteria for rural and urban locations. Vehicle traffic volumes serve to further refine the design criteria for each location. AASHTO also recognizes the influence that types of locations have on driver and other user characteristics and performance. To reinforce the commitment to designing with all users in mind, the Green Book states that “the designer should keep in mind the overall purpose that the street or highway is intended to serve, as well as the context of the project area” and that “designers should recognize the implications of sharing transportation corridors and are encouraged to consider not only vehicular move- ment, but also movement of people, distribution of goods, and provision of essential services” (AASHTO 2011a).

84 Design Guide for Low-Speed Multimodal Roadways Design Characteristic Typical Multimodal Roadway Design Characteristics by Context Zone Urban Core Urban Suburban Rural Town Design Speed 20 mph–30 mph 25 mph–35 mph 30 mph–45 mph 25 mph–45 mph Target Operating Speed * 20 mph–30 mph 25 mph–35 mph 25 mph–40 mph 25 mph–35 mph Vehicle Lane Widths 10–11 ft. 10–11 ft. 11–12 ft. 11–12 ft. Dedicated Turn Lanes • Can have negative impacts on pedestrians/bikes • Often eliminated on lower-volume streets • Can have negative impacts on pedestrians/bikes • Often eliminated on lower-volume streets • Can have negative impacts on pedestrians/bikes • Special designs may be needed to minimize pedestrian/bike conflicts • Can have negative impacts on pedestrians/bikes • Often eliminated on lower- volume streets Medians • Not typically used due to limited right-of- way • Used on boulevard sections • Not typically used due to limited right-of-way • Used on boulevard sections and for some pedestrian crossings • Used often for arterial roadways • Sometimes used for collector boulevards and at some pedestrian crossings • Rarely used unless for pedestrian crossings On-Street Parking • Frequently used • Parallel is typical, angle or reverse- angle in some settings • Selectively used at lower speed ranges • Generally not used for streets with speeds 35 mph and higher • Rarely used due to safety considerations of speed differentials • Frequently used if land use includes building fronts on or near right-of-way line Curb and Gutter • Typical • Typical • Typical in developed areas • Typical if land use includes building fronts on or near right-of-way line Stormwater Drainage • Typical closed drainage system with curb inlets • Typical closed drainage system with curb inlets • Often closed drainage system with curb inlets in developed areas • May include open drainage channels in undeveloped or low- density areas • Often closed drainage system with curb inlets Shoulders • Rarely used • Rarely used • Rarely used in developed areas • Sometimes used in developing areas • Sometimes used in combination with curbs • Rarely used when building fronts on or near right-of- way line • Sometimes used in towns with low-density land use Roadside Width • Typically wide to serve pedestrians, street furniture, transit stops, landscaping, sidewalk cafes, bike parking, etc. • Usually wide enough to serve pedestrians, transit stops, and landscaping • Wider sections may be needed for areas with street furniture, sidewalk cafes, bike parking, etc. • Typically wide enough for sidewalks separated from the curb, and landscaping • May also include space for separate bike path or shared-use path and transit stops • Typically wide enough for sidewalks and possibly street furniture and landscaping Exhibit 4-5. Typical multimodal roadway design characteristics by context zone.

Traveled Way Design Guidelines 85 Design Characteristic Typical Multimodal Roadway Design Characteristics by Context Zone Urban Core Urban Suburban Rural Town Landscaping/ Green Infrastructure • Typical, although may be limited in constrained settings • Typical, although may be limited in constrained settings • Usually provided at some level, although may be limited or low- maintenance in some settings • Sometimes provided but typically at low levels due to right-of-way constraints Pedestrian Facilities ** • Typically provided both sides with width aligned to volumes • May be adjacent to curb at low speeds • Typically provided both sides with width aligned to volumes • May be adjacent to curb at low speeds, generally separated at high speeds • Typically provided both sides with width aligned to volumes • Typically separated from curb or shoulder due to vehicle speeds • Typically provided with width aligned to volumes Bicycle Facilities • Accommodation typically provided on most streets • If not shared lanes, then type/location of facilities determined by bicycle plan • Accommodation provided on many streets according to bicycle plan • If not shared lanes, then type and location of facility determined by bicycle plan • Accommodation provided on selected streets as determined by bicycle plan • Shared lanes or separate facilities typical as determined by bicycle plan • Accommodation sometimes provided as determined by town plan and roadway network plan Pedestrian/ Bicycle Crossings • Typically provided at most intersections • Some controlled and uncontrolled mid- block crossings • Typically provided at most intersections • Some controlled and uncontrolled mid-block crossings • Typically provided at most intersections • Some controlled and uncontrolled mid-block crossings • Typically provided at most intersections • Some controlled and uncontrolled mid-block crossings Transit Facilities • Typical, some dedicated bus lanes, some fixed guideway in street right-of- way • Many transit stops, some bus pullouts • Some routes with dedicated bus lanes or fixed guideway in street right-of-way • Many transit stops and some bus pullouts • Some routes with bus routes, bus stops, bus pullouts • Infrequent bus access except for possible rural and paratransit access Major and Signalized Intersections *** • Frequent use, with ADA-compliant pedestrian signals standard • Possible transit and bicycle priority • Frequent use, with ADA- compliant pedestrian signals standard • Possible transit and bicycle priority • Frequent use, with ADA-compliant pedestrian signals standard where pedestrian facilities exist • Possible transit and bicycle priority • Occasional use • ADA-compliant pedestrian signals should be standard Driveways • Infrequent due to high land use density and structured parking • Pedestrian and bicycle crossings require careful design • Moderate levels of private access typical • Pedestrian and bicycle crossings require careful design • Moderate to high levels of private access typical • Pedestrian and bicycle crossings require careful design • Infrequent where building fronts on or near right-of- way line • Moderate levels of private access outside of town center • Some properties have open access with no designated driveway access *Design and target speeds are typically the same in urban and urban core contexts. **Pedestrian facilities always assumed to meet ADA guidelines for accessibility. ***Roundabouts may be an appropriate intersection alternative in any context zone if pedestrian and bicycle movements are properly designed. Exhibit 4-5. (Continued).

86 Design Guide for Low-Speed Multimodal Roadways 4.1.6.1 AASHTO Design Controls The Green Book presents the pedestrian’s needs as an important factor in roadway design and recommends that attention be paid to the presence of pedestrians in rural and urban areas. Characteristics of pedestrians that serve as design considerations and controls include walking speed, age (young and old), walkway capacity, special needs at intersections and the needs of persons with disabilities. The Green Book also notes that the bicycle is an important element of the design process and provides guidance for reducing crash risk and considering separate facilities where needed. The Guide for the Planning, Design and Operation of Pedestrian Facilities (AASHTO 2004b) and Guide for the Development of Bicycle Facilities (AASHTO 2014b) expand significantly on the Green Book guidance, presenting factors, criteria and design controls for those modes. This Guide emphasizes pedestrians and bicyclists as a design control in all contexts but particularly in multimodal contexts at design speeds of 45 mph and lower. The Green Book identifies functional classification and design speed as the primary factors in determining roadway design criteria. Green Book design criteria are separated by both func- tional classification and context by rural and urban types. The primary differences between the two contexts are the facility operating speeds, the mix and characteristics of the users, and the constraints presented by the surrounding context. The other design controls and criteria that form the basis of AASHTO design policy guidance focus on the following basic controls: • Design vehicle; • Vehicle performance (acceleration and deceleration); • Driver performance (age, reaction time, driving task, guidance and so forth); • Vehicle traffic characteristics (speed, volume, composition); • Vehicular capacity and LOS; • Access control and management; • Pedestrians and bicyclists; • Safety; and • Environment and economics. In this chapter of the Guide, further discussion of multimodal design considerations for selected AASHTO design controls is provided in specific sections. • Functional classification. Functional classification describes a roadway’s theoretical function and role in the network and governs the selection of certain design parameters, although the actual function can vary significantly in urban, suburban and rural town contexts. Functional class may influence some aspects of the roadway, such as its continuity through an area, trip purposes and lengths of trips accommodated, level of land access it serves, type of freight service and types of public transit served (see “Functional Requirements of Multimodal Road- ways” in this chapter). It is important to consider these functions in the design of the roadway, but the physical design of the roadway must also support and integrate with the context of the project area and network functions of all other modes. • Vehicle capacity and LOS. The conventional design process uses traffic projections for a 10-year, 20-year, or even longer design-year period and attempts to maximize vehicular operations within that timeframe. A context-based, all-user approach to design also takes vehicle traffic projec- tions and LOS into account, but the demand and service to all other users are also projected over that same timeframe, consistent with any planned changes in context. Only after understand- ing the future needs of all users and the surrounding land use and community changes can the process of balancing the level, quality and performance of all users be accomplished. Design that considers all users may emphasize one user over another depending on the con- text and circumstances. Capacity and vehicular LOS certainly play a role in selecting design criteria, but they are only two of many factors the design practitioner should consider and prioritize in the design of multimodal roadways. In many communities, roadway capacity on

Traveled Way Design Guidelines 87 selected corridors and roadway segments is considered a lower priority than other factors, such as walking and biking accessibility, economic development or historical preservation. In those locations, vehicle LOS is a lower priority and higher levels of vehicle congestion are considered acceptable. This community-driven approach to design may sometimes conflict with partner agencies that may focus on conventional design outcomes (such as when a state highway also serves as an important community street). • Design speed. Under the conventional design process, many roadway corridors have been planned and designed to serve high speeds and high traffic volumes. As the contexts and character of these roadways in urban, suburban and rural town settings change over time, the vehicle speeds allowed and encouraged by the vehicle mobility design often become a signifi- cant concern to non-motorized users who share the right-of-way, adjacent property owners, surrounding neighborhoods and even entire communities. Posted speed limits established for these roadways using the conventional “85th percentile” method may yield speeds higher than the target operating speed for the facility and are often inappropriate for the multimodal activity, land use context, and community desires for the area. In these cases, traffic engineers and police departments are often tasked with using traffic control devices and increased speed enforcement to attempt to reduce operating speeds. The design speed for a facility should generally be the preferred target operating speed for the facility, consistent with the need to provide safe and convenient accommodation of all current and anticipated users. (See the sec- tion on “Speed Management” for additional discussion of methods for managing operating speeds that are considered too high for a facility.) • Design vehicle. The design vehicle influences the design criteria selection for lane width and curb return radii. A conservative approach is to select the largest design vehicle (often WB 50 to WB 67) that could be expected to possibly use a roadway, regardless of the frequency of that use. Given that lane widths and curb return radii can have significant impacts on the crash risk and mobility of non-motorized users, a multimodal design process should include an evaluation of the trade-offs involved in selecting one design vehicle over another. In urban and some suburban settings, it is not always practical or desirable to choose the largest design vehicle that might occasionally use the facility. Wider lanes and corner radii can easily reduce pedestrian visibility, increase roadway crossing distances, and increase speed of turning vehicles, all of which may be inconsistent with the com- munity vision and the goals and objectives for the roadway. In contrast, selection of a smaller design vehicle in the design of a facility regularly used by large vehicles can invite frequent operational problems. The selected design vehicle should be of a type that will use the facility with considerable frequency. This Guide suggests that two types of design vehicle be considered: – A design vehicle, which must be regularly accommodated without encroachment into the opposing traffic lanes, and – A control vehicle, which must be accommodated infrequently, but for which encroachment into the opposing traffic lanes, multiple-point turns, or minor encroachment into the road- side is acceptable. Regional freight mobility plans may be helpful in determining future truck volumes and types. Ideally, the designer should obtain classification counts to determine the frequency and size of large vehicles and how these numbers may vary over the design period as context changes. In urban core and urban contexts, the selected design vehicle is often a single-unit truck or possibly a transit bus. These vehicle types typically provide higher levels of design flexibility in designing accommodations for multimodal users. 4.1.6.2 Other Design Controls Other elements in the conventional design process also are important as design controls in multimodal design (e.g., horizontal and vertical alignment, stopping and decision sight distance,

88 Design Guide for Low-Speed Multimodal Roadways and access management). These design controls will normally be applied in multimodal design as they are in conventional design practice. • Horizontal and vertical alignment. Speed affects the criteria for horizontal and vertical cur- vature, so design is dependent on the desired operating or target speed. For multimodal road- ways, careful consideration must be given to the design of alignments to balance safe vehicular travel with a reasonable operating speed. The Green Book provides guidance on the design of horizontal and vertical alignments for urban streets. • Sight distance. Sight distance is the distance that a driver can see ahead in order to observe and successfully react to a hazard, obstruction, decision point or maneuver. Adequate sight lines remain a fundamental requirement in the design of multimodal urban roadways. The criteria presented in the Green Book for stopping and signalized stop- and yield-controlled intersection sight distances should be used in multimodal roadway design. • Access management. The management of private and public access to multimodal roadways influences geometric design, establishing criteria for intersection and driveway spacing, raised medians and median breaks, and vehicle movement restrictions through various channeliza- tion methods. Public and private access points also create conflicts for pedestrian, bicycle, transit and vehicle users, and those impacts should be assessed in the design process. The Green Book (AASHTO 2011a), the TRB Access Management Manual (TRB 2016a) and NCHRP Report 659: Guide for the Geometric Design of Driveways (Gattis et al. 2010) provide extensive guidance on access management and design. 4.1.6.3 Design Controls for Pedestrians and Bicyclists On roadways with existing or anticipated high levels of pedestrian and bicycle usage, project design should provide appropriate roadside and bicycle facilities in the traveled way and/or roadside. These facilities must be coordinated with the other design elements in the traveled way and roadside, and they must be sensitive to project context. As a result, in some projects the design requirements for bicyclists and pedestrians may function as design controls that significantly influence the prioritization of design elements for all users of the right-of-way. For example, in some projects requirements for bicycle lanes may be considered a higher priority than a landscaped median, on-street parking or vehicle travel and turn lanes. 4.1.6.4 Design Controls for Freight and Transit Vehicles Both freight and transit vehicles are classes of motorized design vehicles, but they can have significantly different operating characteristics in the right-of-way. Their specific needs and the impacts of design alternatives on their operation should be considered in the design process of multimodal roadways. Freight vehicles typically are considered a part of the motorized traffic demand and generally do not receive special design accommodation beyond ensuring that basic roadway geometry— horizontal curves, lane widths and turning radii—are appropriate for the type and number of freight vehicles existing or anticipated. For a roadway designated as an official freight route, the project design should also ensure that the range of freight vehicles routinely expected can be effectively served. Bus transit is the dominant form of public transportation in most urban and suburban areas. Most bus transit operates in mixed traffic on streets. Generally, designs that make traffic move faster and more safely will improve bus speeds and service reliability. Roadway geometry should be adequate for bus movement, and pedestrian access to transit stops should be convenient. Preferential treatment for transit (dedicated lanes, stations, and priority at traffic signals) may be desirable in some situations. In those cases, the benefits to transit riders should be balanced with

Traveled Way Design Guidelines 89 the effects on roadway traffic. Treatments and priorities for bus transit can vary depending on specific traffic, roadway and environmental conditions. Regardless of the type of treatment, the geometric design and traffic control features of a project should adequately and safely accom- modate all expected vehicles. Design guidelines, standards and practices for transit accommodation have evolved. Much recent guidance has addressed the needs and requirements of specific transit modes, such as buses, rapid transit and light rail transit (LRT). For example, the Guide for the Geometric Design of Transit Facilities on Streets and Highways (AASHTO 2014a) provides design practitioners with a single, comprehensive resource that documents and builds on past and present experience in transit design in streets and roadways. As discussed throughout this Guide, the design of multimodal roadways emphasizes allocat- ing the right-of-way to accommodate all modes in a reasonable manner. This allocation should be driven by consideration of modal priorities based on LOS and QOS, on performance metrics for each and all modes, and as defined by the surrounding context and community objectives. This approach should result in a well thought out and rationalized design trade-off process, the only method by which all of these factors can be appropriately understood, assessed and weighed against each other in a project’s design. 4.1.7 Users with Disabilities in Multimodal Design Pedestrian access routes should contain continuous and clear pedestrian pathways for indi- viduals with mobility, visual, hearing or other disabilities. Many existing urban- and suburban- context roadways were originally designed primarily to serve motorized vehicles, and where pedestrian facilities have been provided or added, these facilities have not always been designed to accommodate pedestrians with a disability. If a roadway project does not provide the proper design of sidewalk widths, ramps, clearances, slopes, surfaces, streetscape furniture, traffic sig- nals, street crossings and transit stops, the facility may be inaccessible to some disabled users. In those situations, the disabled users may have no options but to travel in vehicle travel lanes or directly adjacent to travel lanes with no physical separation, and either of these options is highly undesirable. Additional discussion of the history of and legal implications of accessible design within the public right-of-way is provided in Chapter 2 of this Guide. The PROWAG (U.S. Access Board 2011) and a supplemental notice that addresses shared-use paths (U.S. Access Board 2013) provide guidance on accessible design. The PROWAG calls for a pedestrian access route to provide a 4-ft. minimum continuous clear width, a maximum grade consistent with the road grade, a maximum 2-percent cross slope, and a “firm, stable, and slip- resistant” surface (U.S. Access Board 2011). These accessibility guidelines greatly influence the design strategies for all pedestrian facilities, including sidewalks, shared-use paths, street cross- ings, curb ramps, signals, street furniture, transit stations, on-street parking, loading zones and more. Key guidance from the PROWAG and other reference documents is presented in Achiev- ing Multimodal Networks: Applying Design Flexibility and Reducing Conflicts (FHWA 2016a), and can be summarized as follows: • Sidewalks. Sidewalks should provide a continuous circulation path and connect pedestrians to accessible elements, spaces, and facilities. Where narrower than 5 ft., a 5-by-5-ft. mini- mum passing space is required at 200-ft. maximum intervals (U.S. Access Board 2011, Advi- sory R302.4). To increase maneuverability, additional space should be provided at “turns or changes in direction, transit stops, recesses and alcoves, building entrances, and along curved or angled routes, particularly where the grade exceeds 5 percent” (36 CFR Part 1190, U.S. Access Board Advisory R302.3). The Urban Street Design Guide (NACTO 2013) is a good source for additional information on sidewalk access routes.

90 Design Guide for Low-Speed Multimodal Roadways • Curb ramps. Curb ramps facilitate pedestrian access between sidewalks and street crossings, and between sidewalks and accessible on-street parking. Curb ramps may be perpendicular or parallel to the pedestrian access route, or a combination of both, with a maximum running slope of 8.3 percent. The PROWAG allows for different maximum cross slopes depending on the traffic control in place at the crossing (36 CFR Part 1190, U.S. Access Board Advisory R302.6). Ramps should align with pedestrian crossings; the use of apex curb ramps (i.e., diag- onal ramps in the center of a corner radius) should normally be a last resort, as these ramps direct pedestrians into the middle of the intersection and away from the crosswalk. Each curb ramp must include a landing/turning space for wheelchair maneuverability and a detectable warning surface to alert pedestrians with a visual disability that they are entering or exiting the roadway. Detectable warning surfaces must include truncated domes to provide tactile feedback and must exhibit visual contrast with adjacent surfaces (e.g., light on dark or dark on light). Detectable warning surfaces should be placed at the back of the curb, unless otherwise specified by PROWAG (36 CFR Part 1190, U.S. Access Board Advisory R305.2). Detectable warning surfaces are also needed at blended transitions (i.e., crossings with a run- ning slope less than 5 percent), raised crossings, and at pedestrian crossing islands. • Street furniture. Street furniture (benches, trash receptacles, bike racks, newspaper racks, etc.) should not be placed within the continuous pedestrian access route. Obstructions near the access route should be detectable by cane. Protruding objects, such as wall- or pole-mounted items, must be limited because they can be difficult to detect and avoid (36 CFR Part 1190, U.S. Access Board Advisory R402). • Street crossings. Street crossings continue pedestrian access route across travel lanes at inter- sections and mid-block locations. According to Section 3B.18 in the MUTCD (FHWA 2009b), a variety of striping designs may be used to denote the pedestrian crossing, but high-visibility ladder-style crosswalks with longitudinal lines are recommended due to their improved vis- ibility. The roadway design should ensure that adequate roadway sight distance is available in advance of the pedestrian crossing to provide the proper visibility for approaching motorists and bicyclists. Sight distance should be increased as vehicle operating speeds increase. • Intersections. Intersections are special design situations and additional treatments should be considered that minimize multimodal conflicts by reducing motorist turning speeds and improving motorist yielding rates. Curb extensions are able to shorten traveled way crossing distances, prevent illegal stopping/parking in close proximity of the crosswalk, and further increase visibility of pedestrians to motorists, particularly on roadways with on-street park- ing. Raised crossings enhance visibility and provide an additional traffic-calming benefit to encourage motorist yielding behavior. Pedestrian refuge islands break up long mid-block crossings and help pedestrians address directional conflicts one at a time. • Signals. At signalized intersections, accessible pedestrian signals communicate the location of the pedestrian actuator (usually a pushbutton) and the direction and timing of “WALK” and “DON’T WALK” intervals in a non-visual format. Section 4E.09 in the MUTCD defines non- visual as one or more “audible tones, speech messages, and/or vibrating surfaces” (FHWA 2009b), whereas the PROWAG defines non-visual as both “audible tones and vibrotactile surfaces” (36 CFR Part 1190, U.S. Access Board Advisory R209). Section 4E.08 of the MUTCD advises that designers should separate pedestrian actuators by at least 10 ft. and locate each near a level landing or a blended transition to “make it obvious which pushbutton is associ- ated with each crosswalk” (FHWA 2009b). Section 4E.06 of the MUTCD recommends that walking speeds slower than 3.5 ft. per second be considered when determining pedestrian clearance times to accommodate older pedestrians and pedestrians with disabilities (FHWA 2009b). Signal timing should allow pedestrians to cross both sides of the street during a single cycle. Designers should place an ADA-compliant actuator at pedestrian crossing refuge islands for slower moving pedestrians to call the signal if they cannot cross the street in a single cycle.

Traveled Way Design Guidelines 91 • Surface treatments. The PROWAG requires planar and smooth pedestrian access route surfaces. Uneven unit pavers, rough bricks, and hand-tooled concrete control joints cause uncomfortable or even painful vibrations for people using wheeled mobility devices. Efforts should be taken to minimize vertical discontinuities between unit pavers, vault frames, gratings and points where materials intersect. Refer to the PROWAG and U.S. Access Board Advisory R302.7 for specifications for vertical discontinuities and horizon- tal openings. Saw-cut concrete control joints and wire-cut bricks are design methods to help reduce vibrations. 4.1.8 Aging Users in Multimodal Design The proportion of U.S. drivers aged 65 years and over is expected to increase significantly in coming years. This means that a steadily increasing proportion of drivers and pedestrians will experience declining vision; slowed decision-making and reaction times; exaggerated dif- ficulty when dividing attention between traffic demands and other important cognitive tasks; and reductions in strength, flexibility and general fitness. Although the effects of aging on people as drivers and pedestrians are highly individual, design practices that explicitly recognize these overall changes will better serve this growing segment of the nation’s population. The Older Driver Highway Design Handbook, first published in 1998, focuses exclusively on older drivers (FHWA 1998). Subsequent editions in 2001 and 2014 expand the handbook’s focus to older drivers and pedestrians. The current reference, Highway Design Handbook for Older Drivers and Pedestrians (FHWA 2014d), provides designers and practitioners with a prac- tical information source that links aging road user performance to highway design, operational and traffic engineering features. This FHWA handbook supplements existing standards and guidelines in the areas of highway geometry, operations and traffic control devices. Guidance priority in the handbook focuses on the intersection environment, reflecting aging drivers’ most serious and persistent crash problem area, as well as the greatest exposure to risk for aging pedestrians. Planning Complete Streets for an Aging America is another resource available to the design practitioner (Lynott et al. 2009). It provides project planning, design and operations guidance to address the special needs of aging users of the right-of-way in several specific areas including: • Approaches to roadway design and engineering for older users, • Best practices for making streets work better for older travelers, and • Key design elements for older driver and pedestrian safety. Another design guidance resource is the website maintained by ChORUS (Clearinghouse for Older Road User Safety) at https://www.roadsafeseniors.org/. A collaborative project involving the Roadway Safety Foundation and supported by the FHWA and NHTSA, the goal of ChORUS is to help communities improve conditions and safety for older road users by providing informa- tion on proven and cost-effective design features and crash countermeasures addressing older user needs, such as: • Retroreflective signage that helps older drivers navigate at night, • High-visibility crosswalks that allow drivers to more easily see pedestrians, • Specialized pedestrian signal operations, and • Left-turn lanes that improve sight distance at intersections and help prevent right-angle crashes. ChORUS has been developed to provide quick and easy access to design guidelines for the aging population, technical documents, case studies and success stories, and information about innovative financing solutions.

92 Design Guide for Low-Speed Multimodal Roadways 4.2 Traveled Way Design Element Guidelines for All Users This section of the Guide provides principles and guidance for the design of a roadway’s trav- eled way with multiple users in low- and intermediate-speed environments. On roadways with shoulders and curbs, or shoulders in lieu of curbs, the traveled way as defined also includes the shoulders. Design elements of the traveled way include mid-block crosswalks and mid-block bus stops. The guidance in this chapter is intended to be used in conjunction with the guidance for the roadside in Chapter 5. The traveled way is made up of the central portion of the roadway, as seen in Exhibit 4-6. It contains the design elements that allow for the movement of vehicles, transit, freight and often bicycles. The traveled way also is where vehicles interact with the roadside through on-street parking and access driveways. Design of the traveled way and roadside is influenced by the adja- cent land use context. Exhibit 4-6 shows a typical urban context with building frontages at or near the back of right-of-way. Fundamental design principles for the traveled way include development of a cross section that remains relatively uniform along the length of the roadway and its improvement projects. Traveled way design also incorporates the cross section transitions that result from removing or adding lanes, moving vehicles laterally through lane shifts, and changes in design elements or dimensions. This section of Chapter 4 addresses considerations in cross section development, then discusses the following key design elements for the roadway traveled way: • Vehicle travel lane widths, • Curbs and shoulders, • Bicycle facilities, • Transit facilities, • Medians and median landscaping, • Parking configuration and width, • Geometric transition design, • Mid-block pedestrian/bicycle crossings, and • Pedestrian/bicycle refuge islands. Exhibit 4-6. Components of a typical urban/suburban street.

Traveled Way Design Guidelines 93 4.2.1 Traveled Way Cross Section Considerations The goal of traveled way cross section development is to provide an objective and balanced assessment of the impacts, trade-offs and benefits of each alternative on each user mode using the traveled way. These include users that travel along the roadway, users who must cross the roadway (e.g., vehicles, pedestrians and bicyclists) and users that enter or exit the traveled way by access points such as intersections and driveways. This evaluation process requires careful consideration of the project’s overall purpose and need, community goals and preferred perfor- mance outcomes for all users. In this Guide, Chapter 3 discusses the process of determining and balancing user service levels and Chapter 5 provides design guidance for the roadside portion of the traveled way. The design of multimodal traveled ways should consider not just multimodal transportation objectives, but other community and environmental objectives as well. 4.2.1.1 Consider All Cross Section Design Elements Concurrently The designer always must keep in mind that, although the traveled way typically is the portion of the roadway with the highest total user volumes and speeds, it is but one portion of the overall roadway cross section. Design decisions in the traveled way often have significant impacts on users of the roadside as well as users of intersections. For projects that are scoped to include all these areas a trade-off analysis typically is required to balance all cross section design elements across all three design realms. For some projects, preferred project outcomes and context considerations may place more emphasis on roadside user accommodation than the primary users of the traveled way (typically motorized vehicles). For example, the road diet concept removes one or more vehicle travel lanes from the cross section so that the right-of-way may be reallocated to other cross section priorities such as bicycle lanes, center turn lanes, on-street parking, medians, wider sidewalks, landscaping or other priorities. A specific discussion of the road diet concept is presented in another section of this chapter. Evaluating cross section alternatives should include a comprehensive analysis of applicable issues and options using selected criteria (e.g., modal capacity; modal accessibility; geometric alignment; design concept; costs; right-of-way; environmental, social and economic impacts; operations; safety and so forth). Ideally, the selection of a preferred cross section alternative will be a consensus-based process involving all project stakeholders. The design process should include a process for selecting, refining and building consensus on cross section alternatives. A successful selection of a preferred alternative is one that is compatible with the context(s), reflects the prioritized needs of all users and best achieves the performance objectives and vision established for the corridor. 4.2.1.2 Cross Sections: Variations and Transitions Within Project Limits Many roadway design projects pass through more than one context zone or involve variations within a context zone. This situation can occur even when the roadway functional classification of a roadway (arterial, collector or local) does not change within the project limits. These varia- tions in context zones or features will typically suggest application of different design elements, criteria and cross sections that best align with those context situations. Changes in context zone and features can indicate that the presence, needs and priorities of roadway user modes may also change along the project limits. For example, a rural arterial roadway that passes through a small village or town often will experience several transitions: from a fully rural context on the approach to the town/village to a town/village-edge context with increasing amounts of developed property and land use density, and to a town/village cen- ter environment with intersecting city streets and increases in building density with little or no

94 Design Guide for Low-Speed Multimodal Roadways setbacks from the right-of-way line. These context changes usually indicate increasing numbers of non-motorized users traveling along and across the roadway, a need for pedestrian sidewalks and ADA-compliant ramps, turn lane additions to serve an increased number of driveway and intersection turning movements, a possible change from shoulders to curb and gutter, poten- tially on-street parking, and traffic control changes such as traffic signals and lower speed limits. Significant adjustments to the traveled way, roadside and intersection designs may be needed to best serve the changing user and context conditions at transition locations. Traveled way transi- tions may require the provision of a smooth taper of appropriate length where lanes or shoulders change width, lanes diverge or merge, or lanes have been added or dropped. In multimodal road- way design, however, transitions extend beyond the traveled way geometric design requirements to reflect changes in context and multimodal activity. Transition locations also can be designed to provide visual, operational and environmental cues of upcoming changes in: • Functional emphasis (e.g., from vehicle-oriented to pedestrian- and/or bicycle-oriented); • Context (e.g., between a rural highway and a rural town context); • Roadway type, particularly where functional classification and speed changes; and • Width of roadway (e.g., narrowing or widening of lanes, use of raised medians or decreases or increases in the number of lanes). Considerations for designing effective transitions include: • Using the established guidance such as the MUTCD and the Green Book to properly design, mark and sign geometric transitions; and • Designing transitions on a tangent section of roadway, avoiding areas with horizontal and vertical sight distance constraints. If the traveled way transition is intended to correspond to a context change, community district change or speed zone change, the transition design principles may need to include: • Providing a transition speed zone. • Providing visual cues to changes in context or environment. Techniques may include chang- ing traveled way features (e.g., raised medians, curb extensions, on-street parking and/or alternative pavements), roadside features (e.g., landscaping, street furniture, and/or street lighting) and intersection features (e.g., alternative pavements, raised intersections and/or enhanced crosswalks). • Changing the overall traveled way width to better align with the context, roadway type and traffic characteristics. Approaches to adjusting the traveled way width can include reducing vehicle through and/or turn lanes, reducing lane widths and installing curb extensions at intersections and mid-block pedestrian crossings. 4.2.2 Curbs and Shoulders On low- to moderate-speed roads, designers may choose to place shoulders, curbs, or some- times a combination of both, on the outside of the traveled way. In urban and suburban contexts, most roadways are planned to have curbs and gutters at the edge of the traveled way. Curbs and gutters serve to: • Delineate the roadway edge, • Protect the pavement edge, • Facilitate roadway drainage, • Assist in managing driveway access, and • Provide some level of separation between vehicles and sidewalks, bikeways, roadside appur- tenances and adjacent properties.

Traveled Way Design Guidelines 95 Whether paved or gravel, roadway shoulders typically are associated with rural highway and road classifications, and in some low-density suburban areas where open drainage facilities are preferred over closed drainage underground systems. Paved shoulders like those generally used for higher volume and higher-speed facilities are not generally used for low-speed urban and suburban contexts. On higher-speed facilities, shoulders provide • Space for drivers to perform evasive maneuvers or to recover when drifting from travel lanes, • Storage for stopped or broken-down vehicles, • Locations for emergency vehicles to stay out of the traveled way, • Areas for traffic enforcement activities, and • More room for maintenance activities. Curbs serve different purposes from shoulders, and the choice of which to use depends on the designer’s priorities for the road based on consideration of how the road will be used and who will be using it. Paved shoulders significantly reduce maintenance costs and are proven to reduce crashes. With or without curbs, paved shoulders also can provide space for low levels of pedestrian and bicycle travel, which facilitates safer passing behaviors and improves comfort for all users. Shoulders are not suggested by the Green Book for urban areas, and most roadways in those contexts are built using curbed (or closed) cross sections, especially where right-of- way is limited. All users should be considered to develop the most appropriate design given the intended use of the shoulder. Some designers choose to include a curb on the outside edge of a shoulder. Curbs also are frequently used along raised medians and islands on the inside of the traveled way in urban and suburban contexts. Designers have flexibility in determining when to use curbs and gutters on a roadway and when to pave shoulders, as well as designing shoulder widths and placement of rumble strips on higher-speed facilities. 4.2.2.1 Current AASHTO Policy and Guidance The Green Book notes the frequent use of curbs on all types of low-speed urban roadways to control drainage, provide protection for pedestrians and, in general, allow greater flexibility in how the available roadway width can be used. Curbs may be placed at the edge of the traveled way on low-speed streets, but an offset of 1 ft. to 2 ft. is preferred (AASHTO 2011a). Curbs can be designed with vertical or sloping faces. A curb and gutter section may be part of a longitudinal drainage system. Curbs used on the outside of a shoulder can help with drainage, reduce roadside erosion and control access. Vertical curbs, generally 6 in. to 8 in. tall, are used to discourage drivers from leaving the road- way. Vertical curbs should not be used on high-speed roadways because an errant vehicle that strikes the curb may become airborne and overturn. Vertical curves can provide some measure of comfort to pedestrians using sidewalks adjacent to the roadway because vehicles tend to shy away from them. Sloping curbs, which generally are 4 in. tall or lower, are designed to be mount- able by emergency or other vehicles when needed (ITE 2009a). Generally, curbs used to delineate channelizing islands or medians should be offset at least 1 ft. to 2 ft. from the edge of the traveled way. Vertical curbs used intermittently along a roadway should be offset by at least 2 ft. according to the Green Book (AASHTO 2011a). Care should be taken when curbs are used in conjunction with traffic barriers, such as on bridges. In those applications, vertical curbs should not exceed 4 in. in height. Sloping curbs are preferred and should be located flush with or behind the face of the barrier. If curbs are improperly located, vehicles can strike them and become airborne, either striking the barrier and overturning or vaulting over it. Curbs should not be used with concrete median barriers.

96 Design Guide for Low-Speed Multimodal Roadways The Pedestrian Facilities Guide (AASHTO 2004b) emphasizes the role a curb can play in improving comfort and safety for pedestrians. It points out that curbs provide a clear delinea- tion between the space intended for motorized vehicles and the space intended for pedestrian use. When pedestrian facilities are located adjacent to the roadway, or with only a narrow buffer strip separating them from the roadway, vertical curbs are preferred over sloping curbs because they do more to deter motorized vehicles from crossing them. Curbs along pedestrian facilities should discourage drivers from parking on sidewalks because this blocks the pedestrian route and could compromise pedestrian safety. The Green Book states that shoulders are desirable on any roadway and that “paved shoulders’ advantages include providing a space for pedestrian and bicycle use, for bus stops, for occasional encroachment of vehicles, for mail delivery vehicles, and for the detouring of traffic during construction” (AASHTO 2011a). It recommends further that shoulders be 2 ft. to 8 ft. wide on urban arterials and collectors where sufficient right-of-way width exists. A minimum shoulder width of 4 ft. clear of rumble strips is recommended when the shoulder will be used to accom- modate pedestrians and bicyclists. The Green Book does not provide minimum shoulder-width recommendations for local streets The AASHTO Bicycle Guide states that paved shoulders are a good way to accommodate bicyclists, especially on roads with higher traffic volumes. The Bicycle Guide states, “Adding or improving paved shoulders can greatly improve bicyclist accommodation on roadways with higher speeds or traffic volumes, as well as benefit motorists. . . . Creating shoulders or bike lanes on roadways can improve pedestrian conditions as well by providing a buffer between the sidewalk and the roadway” (AASHTO 2014b). Where bicyclists are expected to use a shoulder that has no curb or vertical obstruction, the shoulder should be at least 4 ft. wide and continuous along the length of the roadway and through intersections. Where a curb or other roadside barrier is present, the shoulder should be at least 5 ft. wide to accommodate a “shy” distance. Wider shoulders may be desirable where bicycle volumes are high or vehicle operating speeds are greater than 50 mph. Designers may wish to use the bicycle level of service (BLOS) model in the AASHTO Bicycle Guide, which includes factors for vehicle speeds, traffic volumes and lane widths to determine the appropriate shoulder width. 4.2.2.2 Principles and Considerations Regarding Use of Curbs and Shoulders Shoulders may provide safety and operational benefits for all users by providing space for emergency, maintenance, or broken-down vehicles to stop outside of the traveled way, and by providing a space for pedestrians and bicyclists to travel outside of the space used by motorized vehicles. Considerations for the use of curbs and shoulders on low- to intermediate-speed roads include the following: • Curbs can be used to encourage vehicles to remain within the traveled way and can provide comfort and protection to pedestrians using adjacent sidewalks. They can also be used in conjunction with a gutter pan to assist with roadway drainage. Curbs should only be used on roadways with low to intermediate speeds. • Gutters, shoulder or edge-line rumble strips, and pavement markers can be obstacles to a bicyclist using the shoulder. The shoulder width provided for cyclists should be clear of these features. Therefore, shoulders may need to be wider to accommodate space for cyclists along with other desired traffic control and design features. • Where unpaved drives or roads meet paved shoulders used by bicyclists, it is desirable for the drive or road to be paved for some portion of the approach to the shoulder. Where practical, the paved section should be sloped away from the roadway to help prevent gravel or other loose roadway materials from spilling onto the shoulder and impeding bicycle travel.

Traveled Way Design Guidelines 97 • Most shoulders are not considered pedestrian facilities because they are not intended for use by pedestrians; however, an occasional pedestrian may use the shoulder when necessary (e.g., when a vehicle breaks down). If a paved shoulder is intended to serve as a pedestrian facility, it should be considered a pedestrian access route and designed to meet accessibility requirements. • Walking-along-roadway crashes occur when pedestrians who are walking along the travel lanes or shoulder of a roadway are struck by vehicles traveling on the roadway. Most of these types of crashes occur where sidewalks are not present. Adding paved shoulders to a roadway where pedestrians walk along a grass shoulder might not improve pedestrian safety because the paved shoulder may attract pedestrians closer to the travel lanes; however, widening existing paved shoulders has been shown to reduce walking-along-roadway crashes (FHWA 2001a). 4.2.2.3 Use of Traveled Way Shoulders for Pedestrian and Bicycle Use The U.S.DOT defines a walkway as a continuous way designated for pedestrians and sepa- rated from motorized vehicle traffic by a space or barrier. U.S.DOT also notes that a traveled way shoulder provides a gravel or paved highway area for pedestrians to walk next to the road- way, particularly in rural areas where sidewalks and pathways are not feasible. Except where expressly prohibited, pedestrians may legally walk on roadway shoulders. Most highway shoul- ders are not intended for use by pedestrians, but they can accommodate occasional pedestrian use. If a shoulder is intended for use as a pedestrian access route “it must meet ADA require- ments for pedestrian walkways to the maximum extent possible” (AASHTO 2004b). For more information, refer to the design topic on “Accessibility.” When accommodation of pedestrian travel is needed, separate roadside pedestrian sidewalk or path facilities should be provided. While traveled way shoulders are not substitutes for a well- designed pedestrian facility, there may occasionally be a need to design shoulders as walkways where roadside facilities are not provided and space is constrained. U.S.DOT encourages all state and local agencies to consider providing and maintaining paved shoulders or walkways along both sides of streets and highways in urbanized areas, particularly near school zones and transit locations, and where there is frequent pedestrian activity (FHWA 2008). 4.2.2.4 Recommended Practice The design and use of curbs and shoulders is highly situation-specific. Ideally, the design pro- cess will include the consideration of available right-of-way, likely roadway users and intended uses, travel speeds and aesthetics. However, the following guidelines can be applied to most multimodal roadways in low- and intermediate-speed contexts: • Vertical curbs should be used when pedestrian facilities are adjacent to the roadway or sepa- rated from the roadway only by a narrow planted strip, as long as the roadway design speed is 45 mph or lower. • Where shoulders are used to accommodate bicyclists, at least 4 ft. of shoulder width that is clear of rumble strips, raised pavement markers and gutter pans should be provided. • Even where a gutter pan is not used, the presence of a curb should increase shoulder width by 1 ft. • Pavement resurfacing offers an opportunity to reallocate roadway space. In some cases, designers should consider reducing lane widths to provide more paved shoulder width suit- able for bicyclists. For example, in a retrofit situation, the AASHTO Bicycle Guide suggests that a 10-ft. or 11-ft. travel lane with a 3-ft. or 4-ft. shoulder for bicyclists is preferable to a 12-ft. travel lane with a 2-ft. shoulder (AASHTO 2014b). • Including paved shoulders in the design of new and reconstructed roadways is cost effective and should be considered on rural and suburban arterial roadway projects. This affords the

98 Design Guide for Low-Speed Multimodal Roadways best opportunity to get a 4-ft. or greater paved shoulder in place. This is also the time to con- sider other treatments (e.g., separated bike lanes, shared-use paths and sidewalks) that may be more desirable in urban and suburban contexts with higher bicycle and pedestrian demand. 4.2.2.5 Rumble Strips and Rumble Stripes Center line rumble strips (CLRSs) and shoulder rumble strips (SRSs) are FHWA-proven safety countermeasures for reducing roadway departure crashes, including head-on and run-off-road crashes. A rumble strip becomes a “rumble stripe” when an edge line or center line pavement marking is placed on it. Designers have flexibility regarding the placement and configuration of roadway rumble strips. Therefore, it is important that rumble strips be designed with bicyclist safety in mind. The AASHTO Bicycle Guide recommends providing a 4-ft. clear space from the rumble strip to the outside edge of a paved shoulder, or a 5-ft. clear space to an adjacent curb, guardrail or other obsta- cle (AASHTO 2014b). A reduced-length rumble strip measured perpendicular to the roadway or edge-line rumble strips (sometimes called rumble stripes) can be considered to provide additional shoulder width for bicyclists. The AASHTO Bicycle Guide recommends providing 12-ft. mini- mum gaps in rumble strips spaced every 40–60 ft. to allow bicyclists to enter or exit the shoulder as needed (AASHTO 2014b). Designers should consider longer gaps for contexts where bicyclists are traveling at relatively high speeds. Designers may also consider bicycle-tolerable rumble strips. Even though the strips can be made more tolerable, they are not considered rideable by bicyclists. Additional information on rumble strip and rumble stripe design can be found in the AASHTO Bicycle Guide, the Decision Support Guide for the Installation of Shoulder and Center Line Rumble Strips on Non-Freeways (FHWA 2001b), and the State of the Practice for Shoulder and Center Line Rumble Strip Implementation on Non-Freeway Facilities (FHWA 2017). In constrained locations with a paved shoulder width less than 4 ft., designers should consider placing rumble strips at the far-right edge of the pavement to give bicyclists additional space near the edge of the lane. Results from NCHRP Report 641: Guidance for the Design and Application of Shoulder and Centerline Rumble Strips indicate that there may not be a practical difference in the effectiveness of rumble strips placed on the edge line or 2 ft. or more beyond the edge line on two-lane rural roads (Torbic et al. 2009). 4.2.3 Vehicle Travel Lane Widths The criteria provided in the Green Book (AASHTO 2011a) describes design width values for through travel lanes, auxiliary lanes, ramps and turning roadways. There are also recommended widths for special-purpose lanes such as continuous two-way left-turn lanes. AASHTO also provides guidance for widening lanes through horizontal curves to provide for the off-tracking requirements of large trucks. Lane width in the Green Book does not include shoulders, curbs and on-street parking areas. AASHTO notes that speed is a primary consideration when evaluating potential adverse impacts of lane width on safety on high-speed two-lane highways because drivers may have more difficulty staying within the travel lane. On any high-speed roadway, the primary safety concerns with reductions in lane width are crash types related to lane departure, including run-off-road crashes. Conversely, the Green Book notes that in a reduced-speed urban environment, the effects of reduced lane widths are different and the design objective is often how to best distribute limited cross-sectional width to maximize safety for a wide variety of roadway users (AASHTO 2011a). Lane widths may be adjusted to incorporate other cross-sectional elements, such as medians for access control, bike lanes, on-street parking, wider sidewalks, transit stops, and landscaping. The

Traveled Way Design Guidelines 99 recommended ranges for lane width in the urban, low-speed environment (less than 50 mph) normally provide adequate flexibility to achieve a desirable urban cross section without requir- ing a design exception. Although 12-ft. lanes have been used historically for vehicle travel lanes, the Green Book notes that 10-ft. travel lanes are acceptable in low-speed (45 mph or lower) environments (AASHTO 2011a). The Green Book’s guidelines for different types of vehicle travel lane widths are provided by roadway functional classification. Those guidelines for low- and intermediate-speed roadway contexts are summarized in Exhibit 4-7. 4.2.3.1 Recommended Practice Design decisions for lane widths are influenced by a wide range of factors, including: • Type of travel lane (through, left or right auxiliary, two-way left turn); • Functional classification of the facility; • Travel demand; • Actual and desired operating speed of the facility; • Adjacent facilities in the right-of-way (e.g., medians, bicycle lanes, parking lanes, transit lanes); • Presence and level of non-vehicle users; and • Context of the surrounding area. Parking lanes and lanes incorporating transit operations are addressed in separate sections of this chapter. Based on consideration of guidance in the Green Book and other source documents, the research team developed recommendations for lane widths for low- and intermediate-speed facilities across the range of land use contexts discussed in Chapter 3 of this Guide. The suggested lane widths are provided in Exhibit 4-8. These width recommendations serve as a starting point in the multimodal design process; many other considerations and factors (discussed later in this section) may suggest higher or lower dimensions. Decisions on lane width should be made by the knowledgeable design professional after having evaluated the trade-offs for various cross section alternatives. 4.2.3.2 Lane Width Selection: General Principles and Considerations A wide range of potential considerations beyond those listed in Exhibit 4-8 may influence the selection of appropriate lane widths for a specific design project. This section discusses these Lane Type All Classes Local Urban Street Urban Collector Street Urban Arterial Street Range Minimum Preferred Minimum Preferred Minimum Preferred Through Lane 9–12 ft. 9 ft. 10–11 ft. 10 ft. 11–12 ft. 10 ft. 11–12 ft. Through Lane (Industrial) 11–12 ft. 11 ft. 12 ft. 11 ft. 12 ft. 10 ft. 11–12 ft. Left/Right Turn/ Auxiliary Lane 10–12 ft. 9 ft. 10–12 ft. 10 ft. 11–12 ft. 10 ft. 11–12 ft. Two-Way Left- Turn Lane 10–16 ft. N/A N/A 10 ft. 11–12 ft. 10 ft. 11–12 ft. Source: AASHTO 2011a Exhibit 4-7. Green Book suggested lane widths for urban low- and intermediate-speed facilities.

100 Design Guide for Low-Speed Multimodal Roadways potential considerations and provides design guidance for the various types of travel lanes used in low- and intermediate-speed environments. Where streets are designed in areas with a significant level of existing or planned use by non- motorized users, excessive street width can create barriers for pedestrians and encourage higher vehicular operating speeds. Wide streets can reduce the level of pedestrian interchange that supports economic and community activity. Wide streets also discourage crossings for transit connections, and the overall width of the street can affect the building height-to-width ratio, a vertical spatial definition that is an important visual design component of many urban streets. Although vehicle lane width is only part of the overall width of the street, it is often cited as the design element that most adversely affects the comfort, convenience and safety of pedestrian crossings. In fact, many factors affect pedestrian crossing safety and exposure, including the number of lanes, presence of pedestrian refuges, curb extensions, walking speed and conflicting traffic movements at intersections. In establishing the most appropriate vehicle lane width for a particular low- or intermediate- speed facility, the designer should consider the needs, safety and operational impacts of alter- native widths to all legal users of the roadway facility. Some key factors that will influence lane width selection on a specific facility will include: • Total traveled way width. The width of the traveled way should be adequate to accommodate through and turning traffic lanes, medians, curbs and appropriate clearances from curb or barrier faces. The width of the traveled way affects users’ perceptions of the speed and volume of the street. Wide streets with multiple travel lanes may be perceived as a barrier to cross- ing where frequent crossings are desired and encouraged. Wider lanes contribute to wider traveled ways and larger intersections, which create longer crossing distances for pedestrians and bicyclists, increased exposure time to vehicle traffic, and the need for longer traffic signal clearance intervals. The total number and width of travel lanes selected should be based on a Context Zone Lane Type Suggested Lane Widths for Target Operating Speeds by Context Zone* 20 mph 25 mph 30 mph 35 mph 40 mph 45 mph Urban Core Through 10 ft. 10 ft. 10 ft. 10 ft. N/A N/A L/R Turn 9–10 ft. 9–10 ft. 9–10 ft. 10 ft. N/A N/A TWLTL 10 ft. 10 ft. 10 ft. 10 ft. N/A N/A Urban Through 10 ft. 10 ft. 10 ft. 10 ft. 10–11 ft. 10–11 ft. L/R Turn 9–10 ft. 9–10 ft. 9–10 ft. 10 ft. 10–11 ft. 10–11 ft. TWLTL 10 ft. 10 ft. 10 ft. 10 ft. 10–11 ft. 10–11 ft. Suburban Through N/A 10 ft. 10 ft. 10–11 ft. 11–12 ft. 11–12 ft. L/R Turn N/A 9–10 ft. 9–10 ft. 10–11 ft. 11–12 ft. 11–12 ft. TWLTL N/A 10 ft. 10–11 ft. 10–11 ft. 10–11 ft. 11–12 ft. Rural Town Through 10 ft. 10 ft. 10–11 ft. 10–11 ft. 11–12 ft. 11–12 ft. L/R Turn 9–10 ft. 9–10 ft. 10–11 ft. 10–11 ft. 11–12 ft. 11–12 ft. TWLTL 10 ft. 10 ft. 10 ft. 10–11 ft. 10 ft. 10 ft. * 1. On low- and intermediate-speed facilities with a mix of users, the selected design speed and the desired operating speed are typically the same value, except on higher volume principal arterials where design speed may be 5 mph above the desired operating speed. 2. On roadways primarily serving industrial uses, minimum lane widths should be 11 ft. 3. On roadways with high percentages (>5%) of large trucks and buses, outside lane widths should be a minimum of 11 ft., including any usable gutter width. Exhibit 4-8. Suggested lane widths for various context zones and speed ranges.

Traveled Way Design Guidelines 101 balance of community objectives, the street’s role in the overall network and the existence or lack of parallel roadways across which traffic can be balanced. • Functional classification. The Green Book states that “while the accommodation of bicy- clists, pedestrians, and transit users is an important consideration in the planning and design of highways and streets, the functional classification of a highway or street is primarily based on motorized vehicle travel characteristics and the degree of access provided to adjacent prop- erties” (AASHTO 2011a). Higher order classifications serving urban areas such as principal arterial, minor arterial and collector roadways often have multiple and even competing roles in the urban street system. The Green Book goes on to say that “even though many of the geo- metric design values could be determined without reference to the functional classification, the designer should keep in mind the overall purpose that the street or highway is intended to serve, as well as the context of the project area” (AASHTO 2011a). For these reasons, the designer must be able to fully consider and balance design criteria such as travel lane width in consideration of the mobility, safety and convenience of all modes and users in the design process of these functional classifications across a broad range of network contexts and com- munity priorities. • Design and control vehicle. As discussed in this chapter under “Design Controls for Multi- modal Roadways,” lane widths should consider the selected design and control vehicles for a project. The safety and operational impacts of a selected lane width also should be evaluated against the various types and sizes of vehicles expected and the frequency with which they are expected to use the facility. Some practitioners will conservatively select the largest design vehicle that could use a thoroughfare (e.g., WB 50 to WB 67), regardless of the frequency, although that is typically not the most cost-effective design solution in low- and intermediate- speed settings. Selecting too large a design vehicle can lead to wider cross sections and intersec- tions, creating negative impacts on other users, particularly crossing pedestrians and bicyclists. Context-sensitive design emphasizes an analytical approach in the selection of a design vehicle, including evaluation of the trade-offs involved in selecting one design vehicle over another. • Vehicle capacity. Lane widths may affect vehicle capacity on some facilities. The HCM suggests that lanes narrower than 12 ft. reduce vehicle capacity and therefore vehicle LOS on higher- speed facilities (TRB 2016b); however, recent studies have shown these impacts to be minimal or non-existent on low- and intermediate-speed roadways in urban settings when lanes are at least 10-ft. wide (Potts, Harwood and Richard 2007; Rahman et al. 2017; Potts 2006). • Lateral clearance. A wider lane width provides more lateral clearance between vehicles trav- eling in opposite directions on two-lane facilities or traveling in the same direction on four- lane facilities. It also provides for more clearance to on-street parked vehicles, vertical curbs in outside lanes or raised medians, and fixed objects that may exist behind either of those curbed spaces. Lateral clearance also can impact operating speeds, with research showing that vehicle operating speeds are generally decreased by smaller roadside lateral clearance distances (Dixon et al. 2008b). • Design speed. Design speed is a critical input to the design process for many geometric ele- ments, particularly on high-speed (>45 mph) facilities. For most of these elements, however, the relationship between the design speed and the actual operating speed of the roadway is weak or changes with the magnitude of the design speed. The relationships between lane widths and vehicle speed are complicated by many factors, including time of day, the amount of traffic pres- ent, and even the age of the driver. Within the low- to intermediate-speed range, it is generally recommended to use somewhat wider lane widths for higher design speeds (i.e., 40–45 mph) than for the lower design speeds (i.e., 20–35 mph). • Operating Speed. General agreement exists among design and traffic engineers with urban and suburban geometric design and operations experience that operating speeds tend to decrease as lane widths decrease to dimensions that create discomfort for drivers and make side-by-side driving more difficult. Although no definitive research establishes the relationships between

102 Design Guide for Low-Speed Multimodal Roadways these two variables, a study by Fitzpatrick et al. (2003) found that on suburban arterial straight sections away from a traffic signal, higher speeds should be expected with greater lane widths. The study identified several variables other than the posted speed limit that show some sign of influence on the operating speed on tangent sections. These variables include access density, median type, parking along the street and pedestrian activity level. • Vehicle safety. With limited exceptions that may represent random effects, research studies have shown no effect of lane width on vehicle safety on urban and suburban roadways in low- and intermediate-speed settings. As a result, the chapter on urban and suburban arterials in the HSM does not include a CMF for lane width on urban and suburban arterials (AASHTO 2010). On low- and intermediate-speed facilities, the risk of lane-departure crashes is less, and the design objective usually becomes how to best distribute limited cross-sectional width to maximize safety for a wide variety of roadway users. With vehicle mixes that contain substan- tial numbers of large trucks or buses, however, safety considerations would generally suggest a wider curb lane to more safely accommodate those wider vehicles. • Pedestrian safety. Many design professionals believe that, in general, pedestrian safety is improved as vehicle lane widths are reduced because the shorter traveled way crossing dis- tances reduce exposure time to vehicles and because of the reduced vehicle speeds typically induced by the narrower lane widths. • Bicycle safety. Bicyclists experience the same safety benefits as pedestrians when crossing narrower lane widths. An equally important, if not more important, consideration in the case of bicycles is the relationship of the travel lane to bicycle traffic either within the lane or adjacent to it. The AASHTO and NACTO bicycle guides (AASHTO 2014b and NACTO 2014) provide extensive guidance on the design of bicycle accommodation within the traveled way, including recommended widths for both shared vehicle-bicycle lanes and striped bicycle lanes placed between the travel lane and vertical curb. When a parking lane is present, these lane width relationships become even more sensitive, and painted buffer strips sometimes are added between the vehicle, bicycle and parking lanes. • Space for other facilities. Using narrower lanes on urban and suburban arterials can provide space for incorporation of other features that are positive for operations and safety, including medians, turn lanes, separate or shared bicycle lanes, parking lanes, bicycle lane buffers, wider sidewalks, enhanced border landscaping and context amenities. Other potential considerations in the evaluation and selection of lane widths include: • Curb lane widths should be measured to the face of the curb unless the gutter and catch basin inlets do not accommodate bicycles and motorized vehicles. To preserve available width for the best use, however, inlets should be designed to safely accommodate bicycle and motorized vehicle travel. • In most cases, as a part of a thoughtful, integrated design of suburban or urban arterials and collectors, travel lane widths between 10 ft. and 11 ft. do not negatively impact overall motor- ized vehicle safety or operations and have little, if any, measurable effect on vehicular capacity (Harwood 1990). The study found one exception where 10-ft.-wide travel lanes should be used with caution, which is on undivided four-lane arterial roadways (Harwood 1990). • Lanes greater than 11 ft. generally should not be used on roadways with high levels of pedes- trian and bicycle activity, as they may encourage higher speeds. • Roadways designated as major truck or transit routes through urban areas may require the use of wider lane widths for specific lanes, with 11 ft. generally being the minimum width used. • Where adjacent lanes are unequal in width, the outside lane should be the wider lane to accommodate large vehicles and bicyclists (but only where bicycle lanes are not practical), and to facilitate the turning radius of large vehicles. • Using wide curb lanes for bicyclist accommodation is not considered an effective means of accommodating bicyclists in urban areas; normally, bike lanes or shoulders are preferred.

Traveled Way Design Guidelines 103 • Where wider curb lanes are required, the designer should consider balancing the total width of the traveled way by narrowing turn lanes or medians to maintain the same overall pedestrian crossing width. • Additional lane width may be necessary for receiving lanes at turning locations with tight curves. • As an alternative to providing wider lanes for an entire route, wider lanes can be considered along horizontal curves to accommodate vehicle off-tracking based on a selected design vehicle. The Green Book provides guidance on widening for vehicle off-tracking (AASHTO 2011a). • Many fire districts require a minimum 20-ft. clear traveled way. This minimum usually can be achieved on urban roadways with two or more lanes without medians, but it may present challenges on streets with one travel lane in each direction separated by a median. • The HSM’s safety performance functions for low-speed streets are not sensitive to lane width (AASHTO 2010). Findings from the available research are mixed regarding the effects of narrow lanes on crashes on urban streets. In some cases, narrow lanes appear to reduce crash rates, but in other cases, narrow lanes appear to increase crashes and in other cases, a particular width has lower crash rates than wider or narrower widths. Whatever the lane width, the potential for vehicle crash rates should be evaluated with the safety of vulnerable users in mind. 4.2.4 Bicycle Lanes/Accommodations Bicycle travel within the traveled way can be accommodated using several design approaches. Selecting the appropriate design can depend on many factors, including roadway classification, vehicle speeds, user volumes, available right-of-way width, cross section (curb, shoulder or no shoulder), modal network plans, community plans and context. The width of the street and the speed and volume of adjacent traffic are typically the most critical factors in providing safe bicycle accommodation. If adequate facilities cannot be pro- vided, then the safety of bicyclists, motorists and pedestrians may all be compromised. In urban and suburban areas, the designer also may need to coordinate bicycle facilities with on-street parking, increased levels of public street and driveway access, transit facilities and signalized intersections. Bicycle travel should be part of the roadway design considerations for most low- and intermediate- speed urban and suburban roadways, and for many rural roadways as well. Bicyclists vary in their level of skill and confidence, trip purpose and preference for facility types; accordingly, their mobility needs can vary quite significantly. In urban and suburban areas, bicycle facilities should encompass a system of on- and off-street interconnected routes, paths and roadway facilities that provide for safe and efficient bicycle travel. Not all urban and suburban roadways will include dedicated bicycle facilities. Except for high-speed freeways and other roadways where bicycling is specifically prohibited, bicyclists are normally permitted to use any street for travel, even if designated bicycle facilities are not provided. Design accommodations for bicycles typically are determined by a community or regional master bicycle plan to ensure overall connectivity, including selection of the best streets for implementation of varying levels of bicycle facility priorities. The absence of designation in a master bicycle plan does not exclude the roadway designer from considering the bicycle mode and providing bicycle facilities if bicycles are allowed and the need currently exists or will exist with future context and network conditions. Bicycle accommodation in the traveled way typically occurs through the use of dedicated bicycle lanes, cycle tracks and separated bike lanes.

104 Design Guide for Low-Speed Multimodal Roadways 4.2.4.1 Bicycle Lanes • Conventional bike lanes. Bike lanes designate a preferential space for bicyclists using pave- ment markings and signage (Exhibit 4-9). The bike lane is located adjacent to motorized vehicle travel lanes and flows in the same direction as motorized vehicle traffic. Typically, bike lanes are located on the right side of the street between the adjacent travel lane and curb, road edge, or parking lane. Bike lanes facilitate predictable behavior and movements between bicyclists and motorists. • Buffered bike lanes. Buffered bike lanes are conventional bicycle lanes paired with a des- ignated buffer space separating the bicycle lane from either the adjacent motorized vehicle travel lane and/or the parking lane (Exhibit 4-10). MUTCD guidelines allow for buffered bike lanes as buffered preferential lanes (FHWA 2009b). The buffering provides more space (“shy distance”) between the motorized vehicles and the bicyclists and allows bicyclists to pass other bicyclists without encroaching into the adjacent motorized vehicle travel lane. • Contra-flow bike lanes. Contra-flow bike lanes are designed to allow bicyclists to ride in the opposite direction of motorized vehicle traffic (Exhibit 4-11). They convert a one-way traffic street into a two-way street with one direction used by motorized vehicles and bikes, and the other by bikes only. Contra-flow lanes are separated using yellow center-lane striping. This design introduces new design challenges and may introduce additional conflict points, as motorists may not expect oncoming bicyclists. • Left-side bike lanes. Left-side bike lanes are conventional bike lanes placed on the left side of one-way streets or two-way median divided streets (Exhibit 4-12). Left-side bike lanes offer advantages along streets with heavy delivery or transit use, frequent parking turnover on the right side, or other potential conflicts that could be associated with right-side bicycle lanes. The reduced frequency of right-side door openings lowers dooring risk. 4.2.4.2 Separated Bike Lanes A separated bike lane (or path) is an exclusive facility for bicyclists that is located within or directly adjacent to the traveled way and is physically separated from motorized vehicle traffic with a vertical design element. Separated bike lanes are differentiated from on-street bike lanes Source: NACTO (2014) Exhibit 4-9. Typical on-street bike lane.

Source: NACTO (2014) Exhibit 4-10. Typical on-street buffered bike lanes. Source: NACTO (2014) Exhibit 4-11. Typical contra-flow bike lane.

106 Design Guide for Low-Speed Multimodal Roadways by the vertical element. They are differentiated from shared-use paths (and sidepaths) by being closer to the vehicle travel lanes and the fact that they are bicycle-only facilities. According to the Separated Bike Lane Planning and Design Guide, separated bike lanes are also sometimes called “cycle tracks” or “protected bike lanes” (FHWA 2015f). • Cycle tracks. A cycle track is an exclusive bicycle facility that combines the user experience of a separated path with the on-street infrastructure of a conventional bike lane. A cycle track is physically separated from motor traffic and distinct from a pedestrian sidewalk. By separat- ing cyclists from motor traffic, cycle tracks can offer a higher level of security than bike lanes and are attractive to a wider spectrum of the public. Cycle tracks have different forms but all provide space intended for use exclusively or pri- marily by bicycles, and all are separated from motorized vehicle travel lanes, parking lanes and sidewalks. Where on-street parking is allowed, cycle tracks typically are located to the curbside of the parking area (in contrast to bicycle lanes). Cycle tracks may be one-way or two-way, and may be at street level, at sidewalk level or at an intermediate level. • One-way protected cycle tracks. One-way protected cycle tracks are bikeways that are at street level and use a variety of physical methods for protection from passing traffic, as shown in Exhibit 4-13. A one-way protected cycle track may be combined with a parking lane or other Source: NACTO (2014) Exhibit 4-12. Typical left-side bike lane. Source: 3rd&5th Avenues project webpage (City of Phoenix n.d.). Photo © Small Giants, LLC. Used by permission. Exhibit 4-13. One-way protected cycle track.

Traveled Way Design Guidelines 107 barrier between the cycle track and the motorized vehicle travel lane. This design reduces the risk of dooring compared to a bike lane, and it eliminates the risk of a doored bicyclist being run over by a motorized vehicle. • Raised cycle tracks. Raised cycle tracks are bicycle facilities that are vertically separated from motorized vehicle traffic (Exhibit 4-14). Many raised cycle tracks are paired with a furnishings zone between the cycle track and motorized vehicle travel lane and/or pedes- trian area. Raised cycle tracks may be at the level of the adjacent sidewalk or set at an intermediate level between the roadway and sidewalk to segregate the cycle track from the pedestrian area. • Two-way cycle tracks. Two-way cycle tracks are physically separated cycle tracks that allow bicycle movement in both directions on one side of the road (Exhibit 4-15). A two-way cycle track may be configured as a protected cycle track at street level with a parking lane or other barrier between the cycle track and the motorized vehicle travel lane, and/or as a raised cycle track to provide vertical separation from the adjacent motor- ized vehicle lane. Source: NACTO (2014) Exhibit 4-14. Typical raised cycle track. Source: New York City DOT Exhibit 4-15. Typical two-way cycle track.

108 Design Guide for Low-Speed Multimodal Roadways 4.2.4.3 Current AASHTO Policy and Guidance • Green Book. The Green Book states, “. . . the designer should be familiar with bicycle dimen- sions, operating characteristics and needs” (AASHTO 2011a). It also references the existence of different types of bikeway facilities. The Green Book does not note recent design treatments (e.g., separated bike lanes) and provides little guidance on the type, selection or design of bicycle facilities; rather, it directs readers to the AASHTO Bicycle Guide (AASHTO 2014b) for appropriate design guidance for bicycle accommodation. • AASHTO Bicycle Guide. The Bicycle Guide provides guidance on bicycle lanes, shared-use paths, bike trails and other related facilities. It does not specifically address separated bike lanes in the right-of-way because it was published prior to the widespread adoption of this facility type. The guide does provide design guidance regarding on-street bike lanes that may apply to separated bike lane designs, including a preferred bike lane width of 5 ft. to 7 ft. (1.5–2.1 m) and a desired 1 ft. (0.3 m) of additional width (shy distance) from vertical curbs and specifications for bicyclist operating dimensions (AASHTO 2014b). The Bicycle Guide discourages a common configuration of separated bike lane designs where the bike lane is placed between the parking lane and the curb, stating that “such placement reduces visibility at driveways and intersections, increases conflicts with opening car doors, complicates maintenance, and prevents bike lane users from making convenient left turns” (AASHTO 2014b). Subsequent guidance featured in the FHWA Separated Bike Lane Planning and Design Guide (FHWA 2015f) provides design solutions to mitigate all of these concerns. The Bicycle Guide also provides guidance on the design of sidepaths, a treatment with similar operational attributes to two-way separated bike lanes. The preferred width of sidepath facili- ties is 12 ft. (3.6 m), with a preferred minimum of 10 ft. (3.0 m). In constrained conditions, a sidepath may function in as little as 8 ft. (2.4 m). On this topic, the Bicycle Guide expresses cau- tion about potential operational challenges, offering a list of specific conflicts that may apply to some two-way separated bike lane facilities (AASHTO 2014b). The guide concludes that one- way paths on both sides of the street, which may operate similarly to directional separated bike lanes, “can reduce some of the concerns associated with two-way sidepaths at driveways and intersections” (AASHTO 2014b, 5–11). In addressing the design of shared-use paths, the Bicycle Guide recommends a minimum paved width for a two-directional shared-use path of 10 ft. (3.0 m), and in rare circumstances notes that a reduced width of 8 ft. (2.4 m) may be used. It recommends wider pathways, 11 ft. to 14 ft. (3.4 m to 4.2 m), in locations that are anticipated to serve a high percentage of pedes- trians (30 percent or more of the total pathway volume) and higher user volumes (more than 300 total users in the peak hour) (AASHTO 2014b). • PROWAG. To complement the design recommendations in the AASHTO Bicycle Guide, the U.S. Access Board supplemented its rulemaking on public rights-of-way in the PROWAG to also cover shared-use paths (U.S. Access Board 2013). The proposed rights-of-way guidelines, which address access to sidewalks, streets and other pedestrian facilities, provide requirements for pedestrian access routes, including specifications for route width, grade, cross slope, surfaces and other features. The Access Board proposes to apply these and other relevant requirements to shared-use paths and the supplementary rulemaking would add provisions tailored to shared-use paths into the rights-of-way guidelines (U.S. Access Board 2013). 4.2.4.4 Additional Guidance • Manual on Uniform Traffic Control Devices (MUTCD). The MUTCD does not directly refer to separated bike lanes, but does describe a class of facilities called “preferential lanes” (FHWA 2009b). Preferential lanes are exclusive-use lanes for a particular vehicle type, and bicycle lanes are a recognized preferential lane type (MUTCD Section 3D-01). Preferential lanes may be barrier-separated from general-purpose travel lanes (MUTCD Section 3D.02). Application

Traveled Way Design Guidelines 109 of FHWA guidance related to barrier-separated preferential lanes formed the basis for the first separated bike lanes and has been expanded on in later FHWA guidance. • Separated Bike Lane Planning and Design Guide. This FHWA design guide provides a com- prehensive set of guidelines related to the preferred and minimum dimensions of separated bike lanes (FHWA 2015f). The guide stresses that “designing separated bike lanes is still evolv- ing and until various configurations have been implemented and thoroughly evaluated on a consistent basis, design flexibility will remain a priority” (FHWA 2015f). Separated bike lanes can be one-way directional facilities, generally traveling in the same direction as adjacent traffic, or two-way bidirectional facilities, offering two-way travel on one side of a street. The preferred clear travel width of a directional bike lane is 7 ft. (2.1 m), with a minimum width of 5 ft. (1.2 m). Total clear width between the curb and vertical buffer element should be at least the width of the fleet maintenance vehicle. Narrower separated bike lanes do not provide good opportunity for passing and may meet resistance from some bicyclists. The preferred clear travel width of a bidirectional separated bike lane is 12 ft. (3.6 m). No mini- mum width is specified in the FHWA guide. The separation buffer dimensions may vary based on vertical buffer element type and on the presence of on-street parking. Adjacent to parking, the minimum buffer width is 3 ft. (0.9 m). Other vertical buffer types may function within as little as 1.5 ft. (0.45 m). The Separated Bike Lane Guide describes eight forms of separation that can be used as vertical elements in the buffer area of the bike lane. The selected form of vertical separation is chosen based on the presence of on-street parking, overall street and buffer width, cost, durability, aesthetics, traffic speeds, emergency vehicle and service access and maintenance. (FHWA 2015f). Exhibit 4-16 consolidates and summarizes the guidelines provided for widths of separated bike lanes in low- and intermediate-speed contexts. 4.2.5 Separated Bike Lanes: Principles and Considerations for All Users FHWA states that bicycle “facilities should accommodate people of all ages and abilities” and encourages “transportation agencies to go beyond the minimum requirements, and proactively pro- vide convenient, safe, and context-sensitive facilities that foster increased use by bicyclists” (FHWA n.d.a). Toward this goal, attributes of well-designed separated bike lanes include the following. • Adequate width. Because the channelizing nature of separated bike lanes prevents opera- tional flexibility for bicyclists to avoid hazards or pass slower bicyclists, separated bike lanes should be designed to allow two people to ride side-by-side and/or pass other users. Separated Bike Lane Area Absolute Minimum Width Preferred Minimum Width Vertical Buffer Area* 1.5 ft. (0.45 m) 3 ft. (0.9 m) One-way Clear Travel Area 5 ft. (1.5 m) 7 ft. (2.1 m) Two-way Clear Travel Area 10 ft. (3.0 m)** 12 ft. (3.6 m) *Minimum vertical buffer area widths vary in response to type of vertical buffer. **Minimum two-way clear travel area widths here are based on AASHTO recommendations for sidepaths. In constrained conditions, an 8-ft. (2.4 m) minimum width may be appropriate. Sources: AASHTO 2014b, FHWA 2015f Exhibit 4-16. AASHTO/FHWA recommended separated bike lane widths.

110 Design Guide for Low-Speed Multimodal Roadways • Forms of separation. Separated bike lanes provide a physical separation from motorized vehicles by a curb, a raised median or a vertical element. The design of the separation should be based on the presence of on-street parking, overall street and buffer width, cost, durability, aesthetics, traffic speeds, emergency vehicle and service access, and maintenance. Generally, raised medians are preferred because they provide permanent curb separation; however, they are costly and may impact drainage, so raised medians are most commonly installed as part of full roadway reconstruction projects. Delineator posts or other lower-cost vertical elements can be ideal for retrofit projects where existing curb lines remain. Designers should consider the crashworthiness of separation types. Fixed objects in the roadway generally are not recommended and some movable objects, such as planters, may not be appropriate on higher-speed streets. According to the AASHTO Bicycle Guide, on lower-speed streets, separation types “need not be of size and strength to redirect errant motorists toward the roadway” (AASHTO 2014b, 5–11). • Bike lane elevation. Separated bike lanes may be designed at any elevation between the street level and the sidewalk level. Many factors contribute to the selection of bike lane elevation, including drainage, accessibility, usable bike lane width, intersection frequency, curbside con- flicts, maintenance and separation from pedestrians and motorized vehicles. Nonetheless, the final decision often is dictated by the construction technique (e.g., retrofit versus reconstruc- tion). A separated bike lane elevation may transition throughout a corridor in response to changing conditions (e.g., raising to sidewalk level at driveways and lowering to street level at major intersections); however, designers should avoid frequent transitions to preserve a comfortable bicycling environment. • Visibility. Clear sight lines should be provided between the roadway and the separated bike lane in advance of driveways and intersections. • Clear pedestrian interactions. Additional width should be provided in the buffer area to accommodate pedestrian access to vehicles, commercial loading activity or an accessible aisle adjacent to accessible on-street parking spaces. • Maintenance. The ability to maintain the separated bike lane free of debris is crucial to pre- serving the functionality of the facility. When building separated bike lanes, municipalities must consider how they will be swept and, if applicable, plowed during snow events. Designers should consider facility designs that provide a clear width compatible with maintenance equipment. 4.2.5.1 Recommended Practice • Design guidance. Exhibit 4-17 provides recommended widths of separated bike lanes for low- and intermediate-speed streets in the urban contexts. These recommendations are based on FHWA and AASHTO guidelines, and have been adapted for this Guide to account for three levels of non-motorized multimodal accommodation. • Implementation guidance. Separated bike lanes may be configured on the left side of one- way or median divided streets. Consider a left-side running separated bike lane on corridors with high-frequency transit routes, where there are fewer driveways, intersections or conflicts on the left side of the street, or where on-street parking is located on only the right side of the street. If accessible on-street parking is provided, an accessible aisle must be provided to allow users to access the sidewalk. The separated bike lane width may be reduced to accom- modate the accessible aisle. Designers can consult the Separated Bike Lane Guide for common configurations (FHWA 2015f). • Small Town and Rural Multimodal Networks (FHWA 2016e). This FHWA resource is intended for transportation practitioners in small towns and rural communities. It applies existing national design guidelines to a rural setting and highlights small town and rural case studies. The document addresses challenges specific to rural areas, recognizes how many rural

Traveled Way Design Guidelines 111 roadways are operating today, and focuses on opportunities to make incremental improve- ments despite the geographic, fiscal and other challenges that many rural communities face. It provides information on maintaining accessibility and MUTCD compliance, while encourag- ing innovations such as “Yield Roadways” and “Advisory Shoulders” (dashed bicycle lanes). The document notes that, as of 2016, an approved Request to Experiment document is required to implement Advisory Shoulders (FHWA 2016e). 4.2.6 Transit Facilities This section of the Guide identifies the key elements of transit facilities and operations that affect the design of roadways, and provides information on other resources that contain detailed design guidance for integrating roadway and transit design. Many urban and suburban roadways accommodate public transportation, as do some rural roadways. The types of transit services accommodated on roadways range from local and express bus service in less dense urban areas to bus rapid transit (BRT), trolleys, streetcars and LRT. These transit services can be accommodated within a dedicated right-of-way, in the roadway right-of-way, or in mixed-flow lanes. In all cases, the design of the roadway needs to consider the special requirements of transit vehicles, running ways and operations, current and future plans, and pedestrian and bicycle access. 4.2.6.1 Types of Transit Operating in and Adjacent to Roadways The various types of public transportation systems that use urban, suburban and rural road- ways have differing physical and operating characteristics that will establish the design controls and geometric design parameters in roadway design. It is important for the roadway designer to understand the dimensions and capabilities of the type of transit that currently uses (or is Multimodal User Priority Level * Separated Bike Lane Zone Width** LOW Multimodal Priority Vertical Buffer Area 1.5 to 3 ft. (0.45 – 0.9 m) One-Way Clear Travel Area 5 ft. (1.5 m) Two-Way Clear Travel Area 10 ft. (2.4–3.0 m) *** MODERATE Multimodal Priority Vertical Buffer Area 3–4 ft. (0.9–1.2 m) One-Way Clear Travel Area 7 ft. (2.1 m) Two-Way Clear Travel Area 10 ft. (3.0 m) HIGH Multimodal Priority Vertical Buffer Area 3.0–6.5 ft. (0.9–2.0 m) One-Way Clear Travel Area 10 ft. (3.0 m) Two-Way Clear Travel Area 12 ft. (3.6 m) *Design and operating speeds should be commensurate with multimodal priority. Moderate and high multimodal priority designs should typically have design speeds of 35 mph and lower. **Total clear width between the curb face and vertical element should be at least the fleet maintenance vehicle width. Clear widths narrower than 7 ft. (2.1) may require specialized equipment. (FHWA 2015e) ***In constrained conditions, an 8 ft. (2.4 m) minimum width may be appropriate. Source: Adapted from information in AASHTO (2014b), FHWA (n.d.a) and FHWA (2015d) Exhibit 4-17. Recommended separated bike lane widths by multimodal priority level.

112 Design Guide for Low-Speed Multimodal Roadways planned to use) the roadway, and how the transit vehicles, their operation and their stops and stations will affect the design of the roadway. 4.2.6.2 Transit Facilities on Roadways Transit on urban roadways can utilize many operating configurations, including: • Mixed-flow travel lanes; • Transit or high-occupancy vehicle (HOV) lanes in the median or adjacent to mixed-flow lanes that are used for transit either full-time or during peak periods; • Reversible or contra-flow dedicated transit lanes (in the median or in outside travel lanes); • A dedicated and separated transit way in the median, inside travel lanes or outside travel lanes; • A dedicated and separated transit way elevated on structures above the roadway; and • Transit-only streets, busways, or transit malls. Each operating configuration requires that the roadway designer understand the right-of-way requirements of the transit facilities and their interactions with traveled way lanes, intersections, pedestrian facilities and, where present, bicycle facilities. Safe and convenient access between transit vehicle stops and stations and their riders (generated by parking, pedestrian and bicycle facilities) is a critical element of designing for all users in transit environments. Transitways within roadways also affect the required roadway right-of-way width because transit systems can employ multiple lanes of sets of tracks. Transit stops and stations can have multiple design requirements depending on required peak user demand at the stop/station, the transit vehicle loading/unloading operation, the frequency of service and other factors. Transit stops typically relate to bus operations and imply more modest passenger loading/unloading requirements that may range from a single bus stop sign to dedicated waiting areas with benches and shelters. Conversely, for fixed transit services such as trolleys and LRT systems, a transit station generally implies heavier loading/ unloading volumes with more substantial passenger amenities such as much larger loading/ unloading areas, ticketing facilities, restrooms or other services. Stations may accommodate multiple vehicles or have integrated intermodal facilities. The roadway designer needs to coor- dinate with the responsible transit agencies to identify the appropriate running way configu- ration, transitions, and the locations and design of stops and stations in and adjacent to the right-of-way. 4.2.6.3 Consideration for Changing Transit Conditions When designing roadways that are identified as future transit corridors, the designer will need to consider several factors to reserve the appropriate right-of-way and ensure the design is rela- tively easily converted to accommodate transit and transit access. In addition to specific design issues, the practitioner may need to consider planning considerations such as: • Changing transit types over time, • Plans for changes in stop or station locations and spacing to meet changing context and future development, and • Possible changes in transit routing. 4.2.6.4 Transit Design Guidance The predominant transit facility type encountered in roadway design is the fixed-route bus transit stop. This Guide provides general design guidance and considerations for this most com- mon design situation. The Guide does not present design guidance for other types of transit facilities or their integration into roadway facilities, although several excellent design guidance resources are discussed at the end of this section.

Traveled Way Design Guidelines 113 4.2.6.5 Bus Stop Design Considerations The most typical transit facility accommodation in urban, suburban and some rural roadways is the bus stop. This section of the Guide presents general guidance for the planning and design of bus stops based on current national guidance. When designing a roadway project, the local transit agency also should be consulted to ensure the designer understands the transit agency’s specific service needs and that those needs are addressed by the design solution. 4.2.6.6 General Principles and Considerations Bus stop locations must address both traffic operations and passenger accessibility issues. If possible, the bus stop should be located in an area where typical amenities (e.g., benches or shelters) can be placed in the public right-of-way as needed. A bus stop location should always consider user access, potential ridership, traffic and rider safety and bus operations elements that may require site-specific evaluation. Well-lit, open spaces visible from the street create a safer environment for waiting passengers. Designing Walkable Urban Thoroughfares (ITE 2010a) provides a thorough discussion of con- siderations in locating and designing bus stops that should be coordinated with other elements and functions of roadway design and operations. Those considerations include safety elements for drivers and transit users, and efficiency of bus operations. Key placement considerations include the following: • The preferred location for bus stops is the near side or far side of an intersection, where pedestrian accessibility may be available from both sides of the street and the cross streets. Connections to intersecting bus routes also occur at intersections. • Bus stops may be placed at a mid-block location on long blocks or to serve a major transit generator. • At mid-block bus stop locations, crosswalks should be placed behind the bus stop so that pas- sengers do not have to cross in front of the bus, where they are hidden from passing traffic. • Bus stops should be placed to minimize the difficulties associated with lane changes and weav- ing maneuvers of approaching vehicles. Where it is not acceptable to stop the bus in traffic and a bus pullout is justified, a far-side or mid-lock curbside stop is generally preferred. 4.2.6.7 Bus Stop Design Bus stop design in the right-of-way should attempt to include the following minimum elements for passenger accessibility, safety and comfort (ITE 2010a and NCHRP Project 15-48 research team): • In roadsides with a detached sidewalk (planting strip between curb and sidewalk), the design should: – Provide a landing area adjacent to the curb for a minimum distance of 34 ft. in length and a minimum of 8 ft. in depth (from face of curb); – For stops serving smaller buses, smaller landing areas may be acceptable; and – Provide a connecting pathway from pedestrian throughway to landing area. • Provide convenient pedestrian pathways/access ways to and from adjacent buildings. • Locate the bus stop so coach operators have a clear view of passengers and waiting passengers can see oncoming buses. • Minimize driveways in and adjacent to the bus stop area. • Locate street furniture more than 2.5-ft. tall in a way that provides motorists exiting nearby driveways clear visibility of the street. • Passenger boarding areas should have a pad with a smooth, broom-finished surface to accom- modate high heels and wheelchairs, and must have high-strength capacity to bear the weight of a shelter. For aesthetics, textured or decorative paver tiles can be used in combination with a concrete pad. The slope of the pad should match the slope of the adjacent sidewalk and allow drainage of the pad (2 percent maximum per PROWAG requirements).

114 Design Guide for Low-Speed Multimodal Roadways • Use landscaping near the passenger boarding area to maximize passenger comfort, but place any landscaping far enough back from the curb face to prevent interference with bus or pas- senger visibility. All landscaping should be located so as not to obstruct the shelter canopy or obscure sight lines at the bus stop. Shade trees are desirable, and the preferred location is at the back of the sidewalk. • Maintain at least 5 ft. of clearance between bus stop components and fire hydrants. • Locate bus stops where there is a standard curb in good condition. Bus stops are designed with the assumption that the bus is the first step from the landing. Access to the bus is more difficult for elderly and mobility-impaired passengers if the curb is absent or damaged. • Surround all street furniture by at least 4 ft. of horizontal clearance wherever possible for access and maintenance between components. • Provide at least 10 ft. of clearance between the front edge of a pedestrian crosswalk and the front of a bus at a nearside bus stop, and 5 ft. between the back edge of a crosswalk and the rear of the bus at a far-side bus stop. • Avoid placing a bus stop so that the bus wheels will cross over a catch basin as it pulls to the curb. Crossing a catch basin can cause the bus to lurch and possibly throw off passenger bal- ance. Over time, bus operations at the stop also can contribute to excessive settlement of the catch basin’s structure. To avoid splashing waiting passengers as the bus pulls to the curb in wet weather, consider draining away from the curb. • Design for and place clear notifications of parking restrictions (either curb markings or NO PARKING signs) at bus zones. A lack of parking restrictions affects bus operations, traffic movement, safe sight distance and passenger access. For example, inadequate or poorly marked parking restrictions may lead to situations when buses are unable to use the curb and sidewalk to deploy its lift in order to board or alight wheelchair passengers. 4.2.6.8 Design Bus Vehicle The bus is one of the design vehicles used in roadway design for urban roadways with transit routes. Some transit agencies use smaller, urban-scaled transit vehicles (e.g., 30-ft. to 32-ft. coaches), and use of vehicles with the smallest possible turning radii should be encouraged. Most fleets use standard coaches with the design specifications described in this section of the Guide. Critical dimensions of anticipated bus types should be identified, including their turning radii requirements. For a 40-ft. coach or a 60-ft. articulated bus, the minimum inside radius is 21 ft. to 26 ft. and the minimum outer radius is 44 ft. to 48 ft. Turning templates should be used in the design of facilities to identify the curb return radius and required pavement width to avoid vehicle encroachment into opposing travel lanes. Additional allowance should be made for any bicycle racks added to the front of the bus. Typically, the addition of a bicycle rack adds about 3 ft. to the length of the bus. 4.2.6.9 Bus Pullout Design in the Traveled Way Bus pullouts (turnouts) are desirable only under selected conditions because of the delay cre- ated when the bus must reenter traffic. Bus turnouts are typically used only on roadways with higher operating speeds of 40 mph or higher. Typical advantages and disadvantages of a bus pullout include the following (Kimley-Horn and Assoc. 2004): Advantages • Allows traffic to proceed around the bus, reducing delay for general traffic; • Maximizes vehicular capacity of roads [particularly important on roadways that prioritize high-volume vehicle mobility]; • Clearly defines the bus stop; • Passenger loading and unloading can be conducted in a more relaxed manner; and • [Reduces] potential for rear-end crashes.

Traveled Way Design Guidelines 115 Disadvantages • More difficult to reenter traffic, increasing bus delay and increasing average travel time for buses; • Uses additional space and may require right-of-way acquisition. Buses also may have difficulty pulling parallel to the curb, reducing accessibility, and they may face a greater crash risk pulling back into traffic from the pullout than buses stopped in the traffic lane. 4.2.6.10 Bus Pullout Placement and Design Guidelines prepared for the transportation authority of Orange County, California, also provide the following information on bus pullout placement and design (Kimley-Horn and Assoc. 2004): The far side of an intersection is the preferred location for warranted turnouts. Nearside turnouts typi- cally should be avoided because of conflicts with right-turning vehicles, delays to transit service as buses attempt to reenter the travel lane, and obstruction of pedestrian activity as well as traffic control devices. The exception would be where buses would use a right-turn lane as a queue jump lane associated with a bus signal priority treatment at an intersection (where a far-side pullout is not possible). Turnouts in mid-block locations are not desirable unless associated with key pedestrian access to a major transit- oriented activity center and subject to the general warrants above. The Orange County guidelines further recommend that pullouts (turnouts) should be placed at signalized intersections where the signal can create gaps in traffic allowing the bus to reenter the street. Twelve (12) ft. width is desirable to reduce sideswipe accidents; 10 ft. width is consid- ered minimum. Where bike lanes are present, and where bus layovers occur, bus pullouts should be wide enough so that the buses do not impede the bike lanes (Kimley-Horn and Assoc. 2004). Typical urban bus pullouts are usually made up of an entrance taper (40 ft. to 60 ft.), stopping area (40 ft. to 60 ft. per each standard and articulated bus, respectively) and exit taper (40 ft. to 60 ft.). 4.2.6.11 Bus Stop Passenger Boarding Area Passenger boarding areas should comply with PROWAG guidance: • Door clearance: Minimum of 5 ft. wide (along the curb) by 8 ft. deep (from face of curb to back of boarding area); • Distance between front and rear boarding area: 18 ft.; • Surface material: Stable, firm and slip resistant; • Slope: Does not exceed 1 ft. vertical over 20 ft. horizontal (5 percent); • Cross slope: Does not exceed 1 ft. vertical over 50 ft. horizontal (2 percent); • Clear throughway width: 48 in. maintained in boarding area; and • Vertical clearance: 84 in. maintained in boarding area. Additional local agency requirements for boarding areas may also need to be met. 4.2.6.12 Additional Transit Design Guidance for the Traveled Way The following transit facility design guidance documents are available to assist the roadway designer in understanding more specific design and operations guidance for all types of transit accommodation in the traveled way, including operating characteristics, service efficiency, and many related design elements and criteria. Those documents include: • Guide for Geometric Design of Transit Facilities on Highways and Streets (AASHTO 2014a). This comprehensive reference to current practice in the geometric design of transit facilities on streets and highways covers the following facilities: – Local buses, express buses and BRT operating in mixed traffic, bus lanes, and HOV lanes, and bus-only roads within street and freeway environments; and – Streetcars and LRT running in mixed traffic and transit lanes, and within medians along arterial roadways.

116 Design Guide for Low-Speed Multimodal Roadways The guidelines are based on a review of relevant AASHTO, TRB, and ITE documents, and of design reports provided by various transit agencies. • This Guide. Chapter 3 discusses the implications of basic vehicle characteristics on road- way design, summarizes basic roadway design requirements, and contains general capacity guidelines. The controls and guidelines apply to bus facilities operating on freeways, streets, and in separate rights-of-way. They also cover streetcar and light rail operations within street rights-of-way. Chapter 6 addresses pedestrian and bicycle access to transit. Riders must be able to reach bus stops or train stations comfortably, safely, and by the most direct routes. Access to stops and stations can be gained by walking, riding a bicycle, or taking a motorized vehicle (including a bus). Walking from adjacent land uses requires stops and stations to have safe, direct, and accessible pedestrian connections to the adjacent community. Cycling from the surrounding community requires stops and stations to be connected to appropriate bicycle facilities, and to have ample and secure bicycle parking. Chapter 6 outlines planning and design guidelines that achieve these objectives. It also contains some general guidelines for passenger amenities at stations and stops. • Transit Street Design Guide (NACTO 2016). This guide from NACTO provides design guid- ance for the development of transit facilities on urban city streets, and for the design and engineering of city streets to prioritize transit, improve transit service quality and support other goals related to transit. Developed through a peer network of NACTO members and transit agency partners, it incorporates information from other design guidance, city case studies, best practices in urban environments, research and evaluation of existing designs, and professional consensus based on North American street design practice. Building on the Urban Street Design Guide (NACTO 2013) and Urban Bikeway Design Guide (NACTO 2014), the transit design guide details how reliable public transportation depends on a commitment to transit at every level of design. • TCRP Report 183: A Guidebook on Transit-Supportive Roadway Strategies (Ryus et al. 2016). This report addresses ways to improve bus speed and reliability on surface streets while addressing the needs of roadway users that include motorists, bicyclists and pedestrians. The report “(1) identifies consistent and uniform strategies to improve transportation network efficiency to reduce delay and improve reliability for transit operations on roadways; (2) devel- ops decision-making guidance for operational planning and functional design of transit/traffic operations on roads that provides information on warrants, costs, and impacts of strategies; (3) identifies the components of model institutional structures and intergovernmental agree- ments for successful implementation; and (4) identifies potential changes to the MUTCD and related documents to facilitate implementation of selected strategies” (Ryus et al. 2016). • TCRP Report 175: Guidebook on Pedestrian Crossings of Public Transit Rail Services (Fitzpatrick et al. 2015b). Presenting an array of engineering treatments to improve the safety of pedestrians using light rail, commuter rail and streetcar services, this guidebook “presents pedestrian crossing issues associated with the National Environmental Policy Act of 1969 and the Americans with Disabilities Act; summarizes readily available decision flowcharts used to make decisions regarding pedestrian treatments at rail crossings; presents information for 34 pedestrian treatments used at rail crossings, grouped into eight appropriate categories; and includes four case studies that examine specific decisions with respect to pedestrian-rail cross- ings” (Fitzpatrick et al. 2015b). It is supplemented by the contractor’s final research report, TCRP Web-Only Document 63: Treatments Used at Pedestrian Crossings of Public Transit Rail Services (Fitzpatrick et al. 2015c). • TCRP Report 117: Design, Operation, and Safety of At-Grade Crossings of Exclusive Busways (Eccles and Levinson 2007). Exclusive busways in separate rights-of-way may have at-grade cross- ings with roadways or pedestrian and bicycle facilities. TCRP Report 117 provides guidelines for the safe design and operation of at-grade crossings of exclusive busways. The guidelines can assist transit, traffic engineering and highway design agencies in planning, designing, and operating

Traveled Way Design Guidelines 117 various kinds of busways through roadway intersections to enhance safety while maintain- ing efficient transit and highway operations and minimizing pedestrian delay. Guidance is included for at-grade intersections along busways within arterial street medians; physically separated, side-aligned busways; busways on separate rights-of-way; and bus-only ramps. Highway intersections, mid-block pedestrian crossings and bicycle crossings are discussed. • TCRP Report 112/NCHRP Report 562: Improving Pedestrian Safety at Unsignalized Cross- ings (Fitzpatrick et al. 2006). The research team that developed this joint transit/highway report provides useful guidelines for selecting pedestrian crossing treatments for unsignalized intersections and mid-block locations. Based on key input variables (e.g., pedestrian volume, street crossing width and traffic volume), quantitative procedures presented in the report can be used to generate a recommended treatment from among four crossing treatment categories. The report also suggests potential revisions to the MUTCD pedestrian warrant for traffic control signals. Findings of the study include that the crossing treatment type affects motorist compliance and that factors influencing the effectiveness of a crossing treatment include the number of lanes being crossed and posted speed limit. Appendices to the report (some available online) provide useful information and tools for improving pedestrian safety at unsignalized crossings. • TCRP Report 19: Guidelines for the Location and Design of Bus Stops (TTI et al. 1996). The primary objective of the research for this report was to develop guidelines for locating and designing bus stops in various operating environments. The guidelines presented will assist transit agencies, local governments, and other public bodies in locating and designing bus stops that consider bus patrons’ convenience, safety and access to bus stop sites along with safe transit operations and traffic flow, and include checklists of factors that should be considered. 4.2.7 Medians The median—the area of the roadway that separates opposing lanes of vehicle traffic—can vary significantly in width and purpose. A median can be open (with pavement markings only), depressed (e.g., with grass or landscaping) or channelized (e.g., raised medians or islands). Operational and safety benefits of medians include limiting conflict points, reducing certain crash types (e.g., head-on collisions), providing pedestrian and bicycle crossing refuge, provid- ing space for left-turning and crossing vehicles, storing plowed snow and collecting stormwater runoff. On low- and intermediate-speed roadways in urban and suburban contexts, medians typically are used to provide these same benefits, with added emphasis placed on landscaping, pedestrian and bicycle refuge, lighting and utilities. In urban and urban core contexts, fixed transit may operate in a median, and parking may also exist within or adjacent to medians. In addition to the operational and safety functions of medians, well-designed and landscaped medians can help create tree canopies over travel lanes and serve as a focal point of the street or an identifiable gateway into a community, neighborhood or district. Flexibility in the design of median width revolves around the median’s function, appurte- nances and landscaping to be accommodated in the median and available right-of-way. Designers need to consider trade-offs between the provision of a median and other design elements, particularly in constrained rights-of-way. 4.2.7.1 Multimodal Safety Considerations of Medians Raised medians provide many benefits to all users of the roadway. These benefits include (FHWA 2013e): • Reducing crashes of motorized vehicles by 15 percent; • Decreasing delays (> 30 percent) for motorists;

118 Design Guide for Low-Speed Multimodal Roadways • Increasing capacity (>30 percent) of roadways; • Reducing vehicle speeds on the roadway; • Providing space for landscaping within the right-of-way; • Providing space to install additional roadway lighting, further improving the safety of the roadway; • Providing space to provide supplemental signage on multilane roadways; and • Potentially reducing cost, as raised medians can be less expensive to build and maintain than pavement. FHWA also encourages the addition of medians and refuge islands because they can increase pedestrian, bicycle and motorized vehicle safety, helping to solve multiple challenges faced by transportation agencies. Medians allow pedestrians and bicyclists to cross one direction of traffic at a time, often allowing them to focus on just two to three lanes rather than having to anticipate traffic for the entire width of the road. Medians also provide a space to install improved lighting at pedestrian and bicycle crossing locations. Improved lighting has been shown to reduce night- time pedestrian fatalities at crossings by 78 percent (FHWA 2013e). Sufficiently wide raised medians and refuge islands can reduce the delay incurred by pedes- trians waiting for a gap in traffic to cross. Shorter delays translate into fewer pedestrians taking risks by crossing through perceived openings in the traffic stream. On a four-lane roadway with average daily traffic (ADT) of 5,000 vehicles, medians can reduce pedestrians’ delay waiting for a gap from 41 seconds to 9 seconds, or 79 percent (Dowling et al. 2008). Crossing the street can be a complex task for pedestrians, especially under nighttime conditions. Pedestrians must estimate vehicle speeds, adjust their walking speeds, determine the adequacy of gaps, predict vehicle paths, and time their crossings appropriately. Drivers also face challenges: they must see pedestrians, estimate vehicle and pedestrian speeds, determine the need for action, and react in a timely fashion. Providing raised medians or pedestrian refuge areas at pedestrian crossings at marked crosswalks has demonstrated a 46 percent reduction in pedestrian crashes. At unmarked crosswalk locations, pedestrian crashes have been reduced by 39 percent (FHWA 2008). Installing raised pedestrian refuge islands on the approaches to unsignalized intersections has had the most impact reducing pedestrian crashes. Medians can be especially beneficial in relation to transit stops, as many transit stops are located along higher-volume arterials at uncontrolled crossing locations. Providing medians can make these crossings safer and more appealing to existing and potential transit users. 4.2.7.2 Getting Pedestrians Safely Across the Street FHWA strongly encourages the use of raised medians or refuge areas in curbed sections of multilane roadways in urban and suburban areas, particularly in areas that mix a significant number of pedestrians with high volumes of traffic (more than 12,000 vehicles per day) and intermediate or high travel speeds (FHWA 2008). FHWA guidance further states that medians or refuge islands should be at least 4-ft. wide (pref- erably 8-ft. wide for accommodation of pedestrian comfort and safety) and of adequate length to allow the anticipated number of pedestrians to stand and wait for gaps in traffic before crossing the second half of the street (FHWA 2008). On refuges 6-ft. wide or wider that serve designated pedestrian crossings, detectable warning strips complying with the requirements of the ADA must be installed (U.S. Access Board 2011). 4.2.7.3 Current AASHTO Policy and Guidance Chapter 4 of the Green Book addresses the motorized vehicle safety and operational aspects of providing medians and their design considerations. The Green Book’s chapter on local roads

Traveled Way Design Guidelines 119 and streets notes that local urban streets often do not have medians, but where provided they are primarily used to enhance the environment and to act as buffer strips. The chapter on collector roads and streets notes that medians generally are not provided on rural collector roadways, but offers significant discussion of median use for vehicle traffic operations and safety purposes on urban collectors. The chapter on rural and urban arterials discusses the use and design of medi- ans on rural arterials, but not in the context of their relationship to multimodal accommodation (AASHTO 2011a). The discussion of urban arterials in Chapter 7 of the Green Book provides considerable infor- mation on the use and design of medians and their vehicle traffic operations and safety consid- erations. The discussion notes that medians are a desirable feature of arterial streets and should be provided where space permits. Relevant guidance regarding median relationship to all right- of-way users includes the following (AASHTO 2011a): • Like refuge islands, medians can benefit pedestrians and bicyclists by allowing them to cross one direction of traffic at a time, provided they are at least “1.8 m [6 ft.] wide when they will be used by bicyclists”; • “Where intersections are relatively infrequent (e.g., 1.0 km [0.6 mi] or more apart) and there is no developed frontage to generate pedestrian crossing needs, the median width may be varied by using a narrow width between intersections for economy and then gradually widening the median on the intersection approaches to accommodate left-turn lanes”; and • “A raised curbed median may be used on low-speed urban arterial streets. This median type is used where it is desirable to manage access along an arterial street by preventing mid-block left turns. Raised curbed medians provide a refuge for pedestrians and a good location for signs and other appurtenances.” 4.2.7.4 General Principles and Considerations for Medians General principles and design considerations for medians include the following: • Landscaping on medians should be designed in a manner that does not obstruct sight distance triangles for any user mode; • If medians are provided at intersections as refuge, they should be wide enough to accommo- date groups of pedestrians, wheelchair users, bicyclists and people pushing strollers; • On roadways where median dimensions need to remain continuous and left-turn lanes are provided, medians should be 16–18 ft. wide to allow for a turn lane plus pedestrian refuge; • When designing median width, use an appropriate design vehicle for left turns and U-turns; • At intersections in urban areas, set the width of medians only as wide as necessary to provide the desired function (accommodation of longitudinal left turns, pedestrian refuge and so forth) so that the intersection does not lose operational efficiency and to prevent vehicles crossing the median from using the width inappropriately (e.g., side-by-side queuing, angled stopping and so forth); • On multilane roadways, a median of 6 ft. to 8 ft. in width can be helpful to a crossing pedes- trian and more desirable than the same width added to another element of the roadway; • In low-speed urban contexts, raised medians should be constructed with vertical curbs to provide refuge for pedestrians, access management, and a place to install signs, utilities and landscaping; • In snow conditions, raised medians can improve delineation of the median, and if emergency access is a concern, mountable curbs should be considered in special locations; • Narrow medians (4-ft.-wide or less) should only be used to restrict turning movements, to separate opposing directions of traffic and to provide space for traffic control devices; • Where flush medians are desirable to maintain access to fronting property (e.g., suburban commercial corridors), consider using textured or colored paving or stamped concrete for

120 Design Guide for Low-Speed Multimodal Roadways the median lane, interspersed with raised landscaped islands to channelize turning traffic, divide opposing lanes of traffic and provide pedestrian refuge where appropriate (such as at mid-block and intersection crossings); • At lower urban operating speeds (25 mph to 30 mph), generally no need exists to provide an offset between the median curb face and the travel lane; • Unless a gutter pan is required for drainage, pave the inside travel lane up to the face of the median curb; • If gutter pans are needed, use 6-in. to 1-ft. gutter pans unless typical flow requires more and avoid placing catch basins in median gutters; • Design the median “nose” using state, local or AASHTO guidelines, ensuring proper end treatments to guide vehicles away from the median and pedestrian refuges; • Design median turn lanes, tapers and transitions using state, local or AASHTO guidelines for intersection design; and • At intersections with significant pedestrian crossings, and where the median is wide enough, extend the median nose beyond the crosswalk to provide an enclosed pedestrian refuge. 4.2.7.5 Trees and Landscaping in Medians An important aspect of roadway landscape design is the treatment of trees and other fixed objects in roadsides and medians. Integrating trees into the design of a facility has many advan- tages. Trees provide a visual edge to the roadway that helps guide motorists. Trees also add to the aesthetic quality of a highway. In urban and suburban areas, trees soften the edges of arterial and collector streets, and they can be an important aspect of community identity. In general, roadway designers must balance safety with other community values when considering facility design and tree preservation. If sight distance is a concern, taller trees with lower branches that are trimmed or low-growing plants can offer landscaping options along both the roadway edge and in medians. Generally, a tree with a trunk diameter greater than 4 in. (measured 4 in. above the ground line) is considered a fixed object along the roadway. Because most trees grow larger than this, their placement along the roadway and in medians needs to be carefully considered. Factors that affect tree placement include the roadway context, design and anticipated vehicle operat- ing speeds, volumes of all users, roadway cross section, and placement of barriers. Chapter 10 in the AASHTO Roadside Design Guide (AASHTO 2011b) provides considerable discussion of roadside safety in urban and restricted areas and specifically addresses minimum offsets to both frangible and rigid objects beside the traveled way. Balancing the safety needs of all users can be challenging in designing urban and suburban roadsides and medians, but solutions can be found using the guidance in this document. Additional information and mitigative strategies for trees in the public right-of-way can be found in: • NCHRP Report 500: Guidance for Implementation of the AASHTO Strategic Highway Safety Plan, Volume 3: A Guide for Addressing Collisions with Trees in Hazardous Locations (Neuman et al. 2004), and • Landscaping of Highway Medians at Intersections (CUTR 2013). 4.2.8 Mid-block Pedestrian and Bicycle Crossings When a pedestrian or bicyclist crosses the roadway, this is considered a “crossing.” Most crossings occur at intersections, but pedestrians and cyclists sometimes find it more convenient to cross at a mid-block crossing location, particularly on roadways where adjacent signalized

Traveled Way Design Guidelines 121 intersections are widely spaced or where the nearest intersection crossing location creates sub- stantial out-of-direction travel. Mid-block crossing opportunities also may be preferred because these locations involve no turning vehicles; pedestrians and bicyclists can focus their attention on through-moving vehicles while attempting to cross. It is important for designers to consider the existing, anticipated and desired use of potential crossing locations both at mid-block and at intersections. Major considerations include the land uses on either side of the street and walking distances with and without a crosswalk. The AASHTO Pedestrian Facilities Guide emphasizes the importance of mid-block crossings in areas where intersections are spaced relatively far apart and pedestrian generators exist on both sides of the street: “Mid-block crossings are preferred because pedestrians should not be expected to make excessive or inconvenient diversions in their travel path to cross at an intersection” (AASHTO 2004a). In all locations, enhanced pedestrian crossing treatments should be considered based on the number of vehicle travel lanes and the speed and volume of vehicular traffic. Where a pattern of mid-block crossings is present or anticipated, consideration should be given to providing a marked crosswalk. Properly designed and visible mid-block crosswalks encourage mid-block crossings to occur at the same locations (rather than being scattered along the block) and provide better visibility of pedestrians and bicyclists to drivers. According to the AASHTO Pedestrian Facilities Guide, some of the situations where marked mid-block crossings are most appropriate include: • “Superblocks” where spacing between adjacent signalized intersections exceeds 600 ft.; • Traffic generators located across the street from each other at a mid-block location; • Schools with entrances near a mid-block location; and • Mid-block transit stops (AASHTO 2004b). Designers may use a variety of treatments to create convenient and comfortable crossings for pedestrians. These treatments include median crossing islands, crossing signs and mark- ings, advance yield/stop lines, rectangular rapid flash beacons (RRFBs), pedestrian hybrid beacons and traffic signals. Existing guidance encourages the use of engineering judgment to develop a justification for the installation of a marked crosswalk, pedestrian hybrid beacon, traffic signal or other crossing treatment. The MUTCD includes flexibility for the designer to consider factors besides traffic volume during an engineering study to justify the installation of a beacon or traffic signal. The MUTCD also suggests that even if a traffic signal warrant is met, other treatments (at the designer’s discretion) may be more appropriate to create a safe crossing (FHWA 2009b). 4.2.8.1 Current AASHTO Policy and Guidance The Green Book provides little guidance on the provision and design of mid-block pedestrian/ bicycle crossings, and generally refers the reader to the AASHTO Pedestrian Facilities Guide (AASHTO 2004b) and the AASHTO Bicycle Guide (AASHTO 2014b) for guidance on pedes- trian and bicycle facilities, respectively. The Pedestrian Facilities Guide states that pedestrians should be encouraged to cross road- ways at intersections because drivers have a greater expectation of encountering pedestrians at intersections than at mid-block crossings. This guide also recognizes that mid-block crossings have fewer conflict points between vehicles and pedestrians, which is a safety advantage over crossings at intersections. The Pedestrian Facilities Guide encourages that mid-block crossing locations be designed to increase drivers’ awareness of the location, increase drivers’ expecta- tion of encountering pedestrians and encourage pedestrians to cross at the designated location (AASHTO 2004b).

122 Design Guide for Low-Speed Multimodal Roadways The Pedestrian Facilities Guide also identifies three important distinctions between mid- block crossings and intersection crossings: • Many more potential crossing locations occur mid-block than at intersections; • Motorists are less likely to expect pedestrians crossing at mid-block; and • Pedestrians with visual impairments have fewer audible clues for determining the best time to cross mid-block (AASHTO 2004b). Based on these three distinctions, the AASHTO guide provides the following design consid- erations for designated mid-block crossing locations (AASHTO 2004b): • Make the crossing location convenient for pedestrians. Mid-block crossing opportunities should be provided at locations where intersection crossing locations are not available or are inconvenient for pedestrians to use. Mid-block crossing locations should be conveniently placed to encourage pedestrians to use them rather than other unmarked mid-block locations that may seem more convenient. • Alert drivers to potential crossings as they approach the crossing location. Drivers should be warned of pedestrian crossing locations in advance, and the mid-block crossing locations should be highly visible to approaching drivers. Mid-block crossing locations should be lit at night to improve driver awareness and pedestrian visibility. Drivers should have clear lines of sight to the crossing location so pedestrians who are crossing or waiting to cross are visible. The approach to the crossing location should encourage drivers to reduce their speed prior to the crosswalk. Drivers should be given plenty of time to recognize the presence of a pedestrian and stop in advance of the crosswalk. • Make pedestrians aware of the opportunity to cross. Provide aids for pedestrians with visual impairments to recognize the presence of a mid-block crossing location and the best opportu- nities for crossing. Auditory and tactile information should be provided for pedestrians with visual impairments since cues at an intersection crossing location (such as the sound of traffic stopping and starting) are not always available mid-block. • Alert drivers and pedestrians to their responsibilities and obligations at a crossing location and provide opportunities to meet these responsibilities/obligations. Use MUTCD guidance to estab- lish a legal crossing location. Vehicle approach, pedestrian approach, and traffic control design should provide pedestrians with clear messages about when to cross and drivers about where to yield. Where necessary, a refuge area should be provided for pedestrians to complete the cross- ing in stages. Traffic control devices can be used to create gaps in traffic for pedestrians to cross. The AASHTO Bicycle Guide states that the task of designing a mid-block crossing location between a pathway and a roadway involves consideration of several variables, including: • Anticipated mix and volume of path users; • Speed and volume of motorized vehicle traffic on the roadway being crossed; • Configuration of the road; and • Amount of sight distance that can be achieved at the crossing location (AASHTO 2014b). The Bicycle Guide also presents the following geometric design guidance related to mid-block bicycle crossing locations: • The mid-block crossing should be conspicuous to both road users and path users; • Sight lines should be maintained to meet the needs of the traffic control provided; • All approaches to the mid-block crossing location should be on relatively flat grades; • Mid-block crossing locations should intersect the roadway at an angle as close to perpen- dicular as practical to minimize the exposure of crossing path users and maximize sight lines; • The least amount of traffic control that is effective should be selected; and • Designated mid-block crossing locations should be located a sufficient distance outside the functional area of adjacent intersections (AASHTO 2014b).

Traveled Way Design Guidelines 123 4.2.8.2 Principles and Considerations of Mid-block Pedestrian/Bicycle Crossings General principles and considerations regarding the provision and location of mid-block cross- walks discussed in ITE’s Designing Walkable Urban Thoroughfares include the following (ITE 2010a): • Consider providing a marked mid-block crossing location where protected intersection cross- ings are spaced greater than 400 ft., or so that crosswalks are located no greater than 200 to 300 ft. apart in high pedestrian volume locations; • Consider mid-block crossing opportunities when significant pedestrian demand exists to cross a street between intersections (e.g., to connect to major generators or transit stops); and • Locate mid-block crosswalks at least 100 ft. from the nearest side street or driveway so that drivers turning onto the major street have a chance to notice pedestrians and properly yield to pedestrians crossing the street. A 2005 FHWA report titled Safety Effects of Marked versus Unmarked Crosswalks at Uncontrolled Locations: Final Report and Recommended Guidelines evaluated the safety considerations of pro- viding marked versus unmarked crosswalks at uncontrolled locations. This report provides the following guidance (FHWA 2005): Marked pedestrian crosswalks may be used to delineate preferred pedestrian paths across roadways under the following conditions: • At locations with stop signs or traffic signals to direct pedestrians to those crossing locations and to prevent vehicular traffic from blocking the pedestrian path when stopping for a stop sign or red light. • At non-signalized street crossing locations in designated school zones. Use of adult crossing guards, school signs and markings, and/or traffic signals with pedestrian signals (when warranted) should be considered in conjunction with the marked crosswalk, as needed. • At non-signalized locations where engineering judgment dictates that the number of motorized vehicle lanes, pedestrian exposure, average daily traffic (ADT), posted speed limit, and geometry of the location would make the use of specially designated crosswalks desirable for traffic/pedestrian safety and mobility. Marked crosswalks alone (i.e., without traffic-calming treatments, traffic signals and pedestrian signals when warranted, or other substantial crossing improvement) are insufficient and should not be used under the following conditions: • Where the speed limit exceeds 40 mph. • On a roadway with four or more lanes without a raised median or crossing island that has (or will soon have) an ADT of 12,000 or greater. • On a roadway with four or more lanes with a raised median or crossing island that has (or soon will have) an ADT of 15,000 or greater. The FHWA report also provides a summary table (reproduced here as Exhibit 4-18) contain- ing recommendations for installing marked crosswalks and other needed pedestrian improve- ments at uncontrolled crossing locations. Pedestrians and bicyclists crossing at mid-block need to see and be seen. According to NACTO, the following principles should be considered to improve the visibility near mid-block crossing locations (NACTO 2013): • Use vertical elements, such as trees, landscaping, and overhead signing to help identify crosswalks and islands to drivers; • Restrict parking in advance of a crosswalk; • Install a curb extension; • Stop lines in advance of mid-block crosswalks should be set back so that a person crossing the street is visible to the second driver when the first driver is stopped at the stop line; and • Medians or safety islands create a two-stage crossing for pedestrians. Mid-block crossing locations should not be provided where the horizontal or vertical align- ment of the roadway limits drivers’ sight distance, the view of the pedestrian approach to the crossing or the view of the crossing location itself.

124 Design Guide for Low-Speed Multimodal Roadways 4.2.8.3 Recommended Practice The design of a mid-block crossing location and its associated traffic control depends on numerous factors, including: • The design of the roadway and roadside, • Characteristics of the road users, • Vehicle, pedestrian and bicycle volumes, • Traffic speed, and • Trip purposes and other factors. These factors will influence how the guiding principles are implemented; however, some guid- ance applies to all mid-block crossing locations. To establish a crosswalk at a non-intersection location, the law requires that white pavement markings consistent with MUTCD guidance must be used. In addition, curb ramps are required at mid-block crosswalks, unless the cross- walk is raised to the level of the sidewalk. The cross slope of a curb ramp should not exceed 2 percent, but can equal the grade of the street where this is not feasible. According to the U.S. C = Candidate sites for marked crosswalks. Marked crosswalks must be installed carefully and selectively. Before installing new marked crosswalks, an engineering study is needed to determine whether the location is suitable for a marked crosswalk. For an engineering study, a site review may be sufficient at some locations, while a more in-depth study of pedestrian volume, vehicle speed, sight distance, vehicle mix, and other factors may be needed at other sites. It is recommended that a minimum utilization of 20 pedestrian crossings per peak hour (or 15 or more elderly and/or child pedestrians) be confirmed at a location before placing a high priority on the installation of a marked crosswalk alone. P = Possible increase in pedestrian crash risk may occur if crosswalks are added without other pedestrian facility enhancements. These locations should be closely monitored and enhanced with other pedestrian crossing improvements, if necessary, before adding a marked crosswalk. N = Marked crosswalks alone are insufficient, since pedestrian crash risk may be increased by providing marked crosswalks alone. Consider using other treatments, such as traffic-calming treatments, traffic signals with pedestrian signals where warranted, or other substantial crossing improvement to improve crossing safety for pedestrians. Source: FHWA (2005) Roadway Type (Number of Travel Lanes and Median Type) Vehicle ADT <9,000 Vehicle ADT >9,000 to 12,000 Vehicle ADT >12,000–15,000 Vehicle ADT >15,000 Speed Limit ** <48.3 km/h (30 mph) 56.4 km/h (35 mph) 64.4 km/h (40 mph) <48.3 km/h (30 mph) 56.4 km/h (35 mph) 64.4 km/h (40 mph) <48.3 km/h (30 mph) 56.4 km/h (35 mph) 64.4 km/h (40 mph) <48.3 km/h (30 mph) 56.4 km/h (35 mph) 64.4 km/h (40 mph) Two lanes C C P C C P C C N C P N Three lanes C C P C P P P P N P N N Multilane (four or more lanes) with raised median*** C C P C P N P P N N N N Multilane (four or more lanes) without raised median C P N P P N N N N N N N * These guidelines include intersection and mid-block locations with no traffic signals or stop signs on the approach to the crossing. They do not apply to school crossings. A two-way center turn lane is not considered a median. Crosswalks should not be installed at locations that could present an increased safety risk to pedestrians, such as where there is poor sight distance, complex or confusing designs, a substantial volume of heavy trucks, or other dangers, without first providing adequate design features and/or traffic control devices. Adding crosswalks alone will not make crossings safer, nor will they necessarily result in more vehicles stopping for pedestrians. Whether or not marked crosswalks are installed, it is important to consider other pedestrian facility enhancements (e.g., raised median, traffic signal, roadway narrowing, enhanced overhead lighting, traffic-calming measures, curb extensions), as needed, to improve the safety of the crossing. These are general recommendations; good engineering judgment should be used in individual cases for deciding where to install crosswalks. ** Where the speed limit exceeds 64.4 km/h (40 mi/h), marked crosswalks alone should not be used at unsignalized locations. *** The raised median or crossing island must be at least 1.2 m (4 ft.) wide and 1.8 m (6 ft.) long to serve adequately as a refuge area for pedestrians, in accordance with MUTCD and AASHTO guidelines. Exhibit 4-18. Recommendations for installing marked crosswalks and other needed pedestrian improvements at uncontrolled locations.*

Traveled Way Design Guidelines 125 Access Board, the same 2 percent maximum for cross slope also applies to the crosswalk (U.S. Access Board 2011). Specific traffic control and supplemental crosswalk features should be considered on higher- speed roadways where pedestrian or bicycle volumes are high, where sight distance is limited, or where there is a history of pedestrian or bicycle crashes. These features may include: • Yield lines or stop bars in advance of the crosswalk. Using these pavement markings 30 ft. to 50 ft. in advance of the crosswalk can reduce the potential for “multiple threat” crashes, wherein one vehicle stops for a pedestrian or bicyclist but blocks the view of the pedestrian or bicyclist from the vehicle approaching in the adjacent lane (Van Houten 1988). As noted in the MUTCD, yield lines or stop bars indicate to drivers the appropriate location to yield or stop so that they do not “place pedestrians at risk by blocking other drivers’ views of pedes- trians and by blocking pedestrians’ views of vehicles approaching in the other lanes” (FHWA 2009b). Normally, in order to increase visibility, parking should be prohibited in between the yield or stop line and the crosswalk. Exhibit 4-19 shows yield lines in advance of an unsignal- ized mid-block crosswalk. • “Yield to Pedestrians” or “Stop for Pedestrians” signs. Tall, narrow signs can be installed at a lane line, the centerline, or in a median at an unsignalized crosswalk to call attention to the crossing location and remind drivers of their legal requirement to allow pedestrians to Source: Adapted from MUTCD (FHWA 2009b) Note: If Stop Here for Pedestrians signs are used instead of Yield Here to Pedestrians signs, stop lines shall be used instead of yield lines. Exhibit 4-19. Examples of yield lines at unsignalized mid-block crosswalks.

126 Design Guide for Low-Speed Multimodal Roadways cross (FHWA 2005). Studies have shown that these signs increase driver yielding compli- ance (Byszeski 2003). The signs frequently are struck, however, requiring maintenance and replacement, and they also may need to be removed for snow removal activities. Similar overhead signs oriented horizontally can be used above the mid-block crossing location (FHWA 2009b). • Raised crosswalk. Because raised crosswalks effectively serve as speed humps for vehicles, they should only be used on low-speed non-emergency routes. They should be clearly marked, and advanced warning should be used to alert drivers to their presence. Raised crosswalks elimi- nate the need for curb ramps, as they allow pedestrians and bicyclists to cross at the level of the sidewalk. Detectable warnings are needed at the transition from the sidewalk to the crosswalk to assist pedestrians with visual impairments who generally rely on curb ramps for guidance. • Lighting. Lighting should be considered where nighttime pedestrian or bicycle crashes are a concern and where little or no roadway lighting exists in the area. According to the FHWA, either direct or back lighting is effective (FHWA 2006b). • Flashers and beacons (constant or actuated). Mounted on the post of advance warning signs, on the post of warning signs at the crosswalk, or on an overhead structure above the cross- walk, beacons and flashers call drivers’ attention to the crossing location. Some beacons flash continuously whereas others are actuated by pushbuttons or sensors. They may be circular or rectangular and have a steady or varying flash pattern. At uncontrolled crossings where a signal or pedestrian hybrid beacon is not warranted, cost prohibitive or deemed unneces- sary, designers may consider supplementing pedestrian, bicycle/pedestrian or school-crossing warning signs with RRFBs. RRFBs have been shown to increase driver yielding rates at the crosswalk (Fitzpatrick 2001). Generally, this treatment should be used with caution at cross- ings with more than two lanes without a refuge. FHWA’s Effects of Yellow Rectangular Rapid- Flashing Beacons on Yielding at Multilane Uncontrolled Crosswalks found an 88 percent average compliance rate for motorists yielding to pedestrians at crossings with RRFBs; this rate was sustained after 2 years (FHWA 2010a). • Pedestrian signals. Pedestrian signals used at mid-block locations must meet MUTCD require- ments and must be designed as accessible pedestrian signals. Pedestrians with visual impair- ments do not have the sound of cross-street traffic to assist in determining when they should cross. To justify the installation of a pedestrian hybrid beacon or traffic signal, the MUTCD has warrants based primarily on pedestrian volumes and vehicle volumes (FHWA 2009b). These warrants are used to help allocate limited financial resources. In some cases, pedestrians may not be crossing the street in sufficient numbers to satisfy the warrant because there are not adequate gaps in traffic or because the pedestrians do not feel comfortable doing so. Where medians are present, actuators should be located in the median to allow pedestrians to reacti- vate the signal if they were not able to complete the crossing. These median signals also should be designed for accessibility (U.S. Access Board 2011). Where mid-block pedestrian signals are closely spaced with traffic signals at intersections, they should be coordinated with nearby intersection signals to increase efficiency and reduce rear-end crashes (Zegeer 2002); otherwise, signal activation should immediately follow actuation. • Speed-reduction treatments. As vehicle speeds increase, the severity of pedestrian and bicycle crashes also increases. For example, a pedestrian is eight times more likely to die when struck by a vehicle traveling at 40 mph than if struck by one traveling at 20 mph (Massengale 2015). Speed-reduction treatments such a curb bulb-outs, narrower lanes, roadside landscaping, and bike lanes may be used in conjunction with lower posted speed limits near mid-block crossing locations. • Grade-separated crossing opportunity (pedestrian bridge or tunnel). When pedestrian or bicycle volumes are very high and interfere with vehicular traffic flow, a grade-separated crossing location should be considered. This design solution also works well between two major mid- block traffic generators, across facilities where pedestrians or bicyclists are not permitted (such as

Traveled Way Design Guidelines 127 freeways) and at locations where an at-grade crosswalk is not ideal due to various site character- istics but where pedestrians or bicyclists tend to cross anyway. When used, grade-separated cross- ing locations must be accessible and must not substantially increase crossing time or pedestrians/ bicyclists will choose to cross at grade rather than use the grade-separated crossing structure. • Medians or crossing islands. The Pedestrian Facilities Guide states that a crossing island should be considered “where the crossing exceeds 60 ft.” (AASHTO 2004a). Raised medians or pedestrian crossing islands are a “proven safety countermeasure” and have demonstrated a 46 percent reduction in pedestrian crashes. Pedestrian refuge areas or islands allow crossings to be completed in two stages and significantly reduce the distance a pedestrian must cross at one time. These features also can provide traffic-calming benefits and prohibit turning move- ments at nearby driveways, even on two-lane roads. FHWA’s Safety Effects of Marked Versus Unmarked Crosswalks at Uncontrolled Locations found that providing raised medians on multi- lane roads “can significantly reduce the pedestrian crash rate and also facilitate street crossing” (FHWA 2005). On roadways with a raised median and volumes exceeding 15,000 ADT, how- ever, a marked crosswalk is considered appropriate only with additional crossing treatments. Crossing islands should be a minimum of 6 ft. wide and situated at locations where bicycles may be crossing (e.g., where a shared-use path crosses a roadway), “10 ft. is preferred to accom- modate a bicycle with a trailer” (AASHTO 2014b). Additional information on pedestrian and bicycle crossing refuge islands is provided in a separate section of this chapter. Additional implementation guidance on many of these treatments can be found in the MUTCD (FHWA 2009b) and the Traffic-Control Devices Handbook (ITE 2009a). Where special traffic control is used, advanced warning signs should make drivers aware of its presence before they encounter the crossing. The AASHTO Bicycle Guide (AASHTO 2014b) provides additional examples of treatments specifically for locations where shared-use paths cross roadways at mid- block. Exhibit 4-20 shows an example of a shared-use path that is stop-controlled for bicyclists at a mid-block crosswalk. 4.2.9 Pedestrian and Bicycle Crossing Refuge Islands Refuge islands provide a protected area for pedestrians and bicyclists within intersection and mid-block crossings. Refuge islands can be either raised (as shown in Exhibit 4-21), or painted flush. In multimodal urban areas, it is desirable that roadways have shorter crossings, so on wide roadways or where less mobile pedestrians need to cross, refuge islands provide a location for pedestrians or bicyclists to wait partially through their crossing. Refuge islands also break up crosswalks at complex multilane and multileg intersections, making it possible for pedestrians to cross in shorter and easier portions. Several types of medians and pedestrian crossing islands exist. Appropriately designed and applied medians and crossing islands improve safety, providing benefits to pedestrians and vehicles in the following ways: • They may reduce pedestrian crashes by 46 percent and motorized vehicle crashes by up to 39 percent (FHWA 2008); • They may decrease delays by greater than 30 percent for motorists (FHWA 2008); • They allow pedestrians a safe place to stop at the midpoint of the roadway before crossing the remaining distance; • They enhance the visibility of pedestrian crossings, particularly at unsignalized crossing points; • They can reduce the speed of vehicles approaching pedestrian crossings; • They can be used for access management for vehicles (allowing only right-in/right-out turning movements); and • They provide space for supplemental signage on multilane roadways.

128 Design Guide for Low-Speed Multimodal Roadways Source: AASHTO (2014b) Exhibit 4-20. Example of mid-block crossing of shared-use path (path stop-controlled for bicyclists). Exhibit 4-21. Raised pedestrian and bicycle refuge island in Kansas City, KS.

Traveled Way Design Guidelines 129 Mid-block locations account for more than 70 percent of pedestrian fatalities (FHWA 2008). Vehicle travel speeds are higher at mid-block, which contributes to the larger injury and fatality rate seen at these locations. More than 80 percent of pedestrians die when hit by vehicles trav- eling at 40 mph or faster, whereas less than 10 percent die when hit at 20 mph or less (FHWA 2008). Installing such raised channelization on approaches to multilane intersections has been shown to be especially effective. Medians are a particularly important pedestrian safety counter- measure in areas where pedestrians access a transit stop or other clear origins/destinations across from each other. Providing raised medians or pedestrian refuge areas at marked crosswalks has demonstrated a 46 percent reduction in pedestrian crashes. At unmarked crosswalk locations, medians have demonstrated a 39 percent reduction in pedestrian crashes. In summary, refuge islands provide the following benefits to pedestrians and bicyclists cross- ing a roadway: • Reduced exposure time. Refuge islands reduce exposure time during the crossing by allow- ing the crossing to be completed in two stages. At multilane crossing locations, pedestrians and bicyclists need to determine a safe gap for several lanes of traffic at a time, which can be a daunting task. Refuge islands allow pedestrians and bicyclists to cross one direction of traffic at a time. • Refuge while waiting for acceptable gap. At unsignalized intersections, refuge islands provide a storage area for pedestrians while they wait for acceptable gaps in traffic before completing the second half of the crossing maneuver. • Accommodation of shorter crossing phase. At signalized intersections, refuge islands pro- vide a storage area for pedestrians to wait for the next available cycle if they are unable to cross the street entirely during a provided crossing phase. This also helps with the efficiency of the intersection LOS by permitting split signal phasing for major turning movements. • Traffic calming. Refuge islands may provide traffic-calming benefits by physically narrowing the roadway and potentially restricting motorized vehicle left-turn movements. 4.2.9.1 Current AASHTO Policy and Guidance The Green Book provides the following limited guidance on the provision of refuge islands: “Where intersections are channelized or a median is provided, consideration should be given to the use of curbing for those areas likely to be used by pedestrians for refuge when crossing the roadway” (AASHTO 2011a). For more specific information on pedestrian and bicycle facilities, the Green Book generally refers the reader to the AASHTO Pedestrian Facilities Guide (AASHTO 2004b) and the AASHTO Bicycle Guide (AASHTO 2014b). According to the Pedestrian Facilities Guide, some situations in which refuge islands are most appropriate include: • Two-way arterial streets with high traffic volumes, high travel speeds, and large pedestrian volumes; • Wide two-way intersections with high traffic volumes and significant numbers of crossing pedestrians; • Two-way collector and local access streets where they function as traffic-calming devices and street crossing aids; and • Complex or irregularly shaped intersections where islands could provide a pedestrian with the opportunity to rest and become oriented to the flow of oncoming traffic (AASHTO 2004b). In addition, depending on the signal timing, refuge islands should be considered at signalized intersections where the crossing distance exceeds 60 ft., but can be used at intersections with shorter crossing distances where a need has been recognized. Median refuge islands should not be used to justify a signal timing that does not allow pedestrians to complete their crossing in one cycle (AASHTO 2004b).

130 Design Guide for Low-Speed Multimodal Roadways The AASHTO Bicycle Guide refers to refuge islands as “crossing islands” and states that raised crossing islands are associated with significantly lower pedestrian crash rates at multilane cross- ings. Although crossing islands can be helpful on most road types, they are of particular benefit at path-roadway intersections in which one or more of the following apply: • High volumes of roadway traffic and/or speeds create difficult crossing conditions for path users; • Roadway width is excessive given the available crossing time; and • The roadway cross section is three or more lanes in width (AASHTO 2014b). 4.2.9.2 Principles and Considerations of Pedestrian and Bicycle Refuge Islands Refuge islands often are desirable in locations where providing the opportunity for a staged crossing (crossing one direction of traffic at a time) would be particularly beneficial to pedestri- ans. Such locations may include unsignalized mid-block crossing locations across a high-volume or high-speed facility with four or more lanes (or that exceed 60 ft.), signalized mid-block crossing locations where users may walk slower than 3.5 ft. per second (e.g., children, older adults, and persons with limited mobility), and in special circumstances, such as near a school (ITE 2010a). Providing raised medians or refuge islands at marked crosswalks has demonstrated a 46 percent reduction in pedestrian crashes (FHWA 2008). The width of the median refuge island is determined by expected bicycle and pedestrian vol- umes, the constraints of the street to be crossed and the surrounding environment. At loca- tions where pedestrian volumes are low, a relatively narrow median may be sufficient; where pedestrian volumes are higher or where crossing can be difficult due to high traffic volumes and speeds, a wider crossing island may be desired to provide space for larger groups of waiting pedestrians (AASHTO 2004b). In general, refuge islands should be large enough to accom- modate as many potential users as possible, including groups of pedestrians or cyclists, tandem bicycles, pedestrians pushing strollers, wheelchairs and even equestrians if they are permitted to use the crossing path or roadway (AASHTO 2014b). Access to the refuge island should be functional and safe for all users, and should be designed to meet the requirements of the proposed PROWAG (U.S. Access Board 2011). A cut-through design or a ramped design large enough to enable a wheelchair to wait on top of the island should be provided. The cut-through width should be the same as the complete width of the crosswalk, or it should at least maintain a minimum clear width. Cut-through ramps should be graded to drain quickly and should include detectable truncated dome warning surfaces so that pedestrians with vision impairments can identify the edge of the street (AASHTO 2004b). At signalized crossing locations, pedestrian actuators should be provided in the median to ensure that pedestrians who start the crossing late in the cycle or who travel slowly are not trapped in the median without being able to activate another crossing phase. Lighting can be used to increase the visibility of the island to motorists and the crossing to pedestrians and cyclists. Landscaping in the median can be used as a traffic-calming strategy to reduce speeds, but care should be taken to ensure that landscaping features do not obscure the visibility of crossing users to motorists (AASHTO 2004b, NACTO 2014). Motorists should be given substantial advanced notice of the median through an approach nose, offset from the edge of the traffic lane (AASHTO 2011a), and through marking, signing, reflectorization, or lighting (Florida DOT 1999). Proper visibility of the median also is critical for snowplow crews. Crossing islands should be well maintained; snow should be cleared in a way that does not block pedestrian or bicycle access, and any debris that accumulates in the median should be cleared frequently (NACTO 2014).

Traveled Way Design Guidelines 131 4.2.9.3 Recommended Practice Where used on two-way streets, median refuge islands should be placed along the centerline of the roadway between the opposing travel lanes (NACTO 2014). Medians expected to be used as pedestrian refuges should have vertical curbs to delineate the pedestrian refuge from the sur- rounding roadway (ITE 2010a). In areas where snow accumulation can occur, retroreflective white or yellow material should be supplemented with reflective delineators visible above the surface of the snow to snow plow crews (AASHTO 2014b). The MUTCD provides guidance for the pavement markings that should be used on the approach to the refuge island. For all medians at intersections, the nose should extend past the crosswalk. The nose protects people waiting on the median and slows turning drivers (NACTO 2013). The refuge area also may be designed with a storage area aligned perpendicularly across the island or via a diagonal or offset storage bay, as shown in Exhibit 4-22. A diagonal storage area has the added benefit of helping to position pedestrians and bicyclists to face oncoming traffic; therefore, it should be angled toward the direction from which traffic is approaching. It should be recognized however, that a refuge island with diagonal or offset storage might complicate the crossing for pedestrians who have visual impairments. If pedestrians will use the storage area (or cut-through), the 45-degree angle of the curb should transition back to being perpendicular to the street to provide proper directional cues for the blind (NACTO 2014). The cut-through or ramp width should equal the width of the crosswalk. Where this cannot be achieved, the crosswalk should not be striped narrower than the cut-through area just to match the width of the median cut-through or ramp (NACTO 2013). Crossing through pedestrian ref- uges must be accessible with channels at street grade, detectable warnings, and audio and visual output at signalized crossings (ITE 2010a). The minimum dimensions for a refuge island serving typical crossings is 6 ft. wide and 20 ft. long, equivalent to at least 120 square ft. A width of 8 ft. or 10 ft. is preferred, as it provides addi- tional separation distance from traffic and additional storage space, and the preferred length of Source: AASHTO (2014b), Figure 5-22 Exhibit 4-22. Crossing island.

132 Design Guide for Low-Speed Multimodal Roadways such a median is 40 ft. (NACTO 2013, NACTO 2014). Refuge islands used for a multi-use trail are recommended to be 10 ft. wide, with a minimum of 8 ft. wide (ITE 2010a). 4.2.10 Parking Configuration and Width On-street parking serves several needs in urban and suburban environments. It supports eco- nomic activity by providing easy access to local merchants in commercial and mixed-use areas and to residents and visitors in residential areas. On-street parking also serves as an indication to motorists that they are entering a low- to intermediate-speed area where increased pedestrian activity should be expected, and it increases pedestrian comfort by separating pedestrians from moving traffic. Usually, on-street parking alone is insufficient to meet all of the parking demand created by adjacent land uses; typically, on-street parking will be used to supplement the off-street parking supply. On-street parking may provide the following benefits when properly designed and located along streets and roadways (ITE 2010a): • Supports local economic activity of merchants by providing proximate access to local uses, as well as visitor needs in residential areas; • Increases pedestrian comfort by providing a buffer between pedestrians and moving traffic helping reduce vehicle splash, noise and fumes; • Slows traffic, making pedestrian crossing safer; • Enables drivers and their passengers to become pedestrians conveniently and safely; • Provides an indication to the motorist that desired operating speeds are reduced and that they are entering a low or moderate travel speed area; • Provides the shortest accessible route to a street-fronting building entrance for pedestrians who have disabilities; • Increases pedestrian activity on the street since people will walk between their parking space and destination, providing more exposure to ground-floor retail and increasing opportunities for social interactions; • Supports local economic activity by increasing the visibility of storefronts and signs to motor- ists parking on the street; • Reduces development costs for small business by decreasing on-site parking needs, particu- larly in urban infill development on small lots; • Requires less land per space than off-street parking and is thereby an efficient and cost- effective way to provide parking; and • Provides space for on-street loading and unloading of trucks, increasing the economic activity of the street and supporting commercial retail uses. Although it is frequently used on urban and some suburban streets and roadways, on-street parking may have negative impacts on traffic operations and safety. These negative impacts can include (ITE 2010a): • A reduction in through traffic capacity and impedance to traffic flow (3 percent to 30 percent decrease in the capacity of the adjacent travel lane, depending on the number of lanes and frequency of parking maneuvers); • Conflicts with bicyclists, especially bicyclists located in adjacent bicycle lanes; • Occupation of street width that could be used for other functions (e.g., bike lanes, wider roadsides); • Visual obstructions near intersections and driveways; and • An increase in crashes.

Traveled Way Design Guidelines 133 Designers should carefully consider the site-specific conditions to determine whether on- street parking is appropriate for a given block or roadway segment. The designer needs to bal- ance traffic capacity and local access needs when deciding where and when to permit on-street parking. Methods are available for minimizing the impact of parking maneuvers on traffic flow. For an example, see the MUTCD (Section 3B.19, Figure 3B–21) for a parallel parking configura- tion that allows vehicles to drive forward into the parking space (FHWA 2009b). 4.2.10.1 Current AASHTO Policy and Guidance The Green Book notes that within urban areas and rural communities, on-street parallel park- ing should be considered to accommodate existing and developing land uses. The Green Book states that the “designer should consider on-street parking so that the proposed street or high- way improvement will be compatible with the land use . . . the type of on-street parking should depend on the specific function and width of the street, the adjacent land use, traffic volume, as well as existing and anticipated traffic operations” (AASHTO 2011a) On urban arterial and collector streets, the Green Book states the desired minimum width of a parking lane is 8 ft., noting that most vehicles parallel park within 6 to 12 in. of the curb face and, on average, occupy 7 ft. of actual street space. It also notes that 10- to 12-ft. parking lanes may be desirable to provide better clearance from the traveled way, accommodate use of the parking lane during peak periods as a through travel lane, and/or accommodate transit opera- tions. Parking lanes that are 7 ft. wide have been successfully used on urban collector streets within residential neighborhoods that strictly accommodate passenger vehicles. On local streets, on-street parking is generally permitted, but specific parking lanes are not usually designated (AASHTO 2011a). Angle parking is acceptable under certain circumstances, but the Green Book notes that angle parking presents special problems because longer vehicles may interfere with the traveled way and introduce sight distance issues associated with certain types of vehicles (AASHTO 2011a). The type of on-street parking for a street or corridor should be selected based on the function and width of the street, the adjacent land use, traffic volume, and existing and anticipated traffic operations. The Pedestrian Facilities Guide primarily addresses issues associated with pedestrians and parking lanes near intersections. Parked vehicles in a parking lane may cause sight obstructions and result in pedestrians stepping into the parking lane to see around parked vehicles before crossing an intersection. Curb extensions can improve sight lines at intersections so that pedes- trians can see and be seen. However, curb extensions may not be practical at all locations where parking is permitted. The Pedestrian Facilities Guide provides guidance on the design of street corners to allow turning vehicles to clear adjacent parking lanes and align with the departing travel lane, as shown in Exhibit 4-23 (AASHTO 2004b). Guidance in the AASHTO Bicycle Guide is consistent with that of the Green Book, indicating that 8 ft. is the desired width of a parking lane adjacent to a bicycle lane, and 7 ft. is the mini- mum width. The Bicycle Guide provides additional guidance related to the design of parallel and diagonal parking adjacent to bicycle lanes. Where parallel parking is permitted but the parking lane or stalls are not specifically designated, the recommended width of the shared bicycle and parking lane is 13 ft., whereas a minimum width of 12 ft. may be satisfactory where parking usage is low and turnover is infrequent (AASHTO 2014b). In areas with high parking demand and sufficient street width, diagonal parking is sometimes used to increase parking capacity and reduce travel speeds, but bicycles lanes should not be placed adjacent to conventional front-in diagonal parking because drivers in the parking spaces have poor visibility of bicyclists in the bike lane. Where diagonal parking is preferred, back-in

134 Design Guide for Low-Speed Multimodal Roadways diagonal parking is recommended to mitigate conflicts associated with bike lanes adjacent to angle parking. Additional benefits of back-in diagonal parking for all roadway users include: • Sight distance is improved between exiting motorists and other traffic; • Passengers, including children, are channeled toward the curb when alighting; and • Vehicles’ trunks can be more easily loaded and unloaded because they are located at the curb and not in the street. 4.2.10.2 On-Street Parking Principles and Considerations Designing Walkable Urban Thoroughfares (ITE 2010a) provides the following general prin- ciples and considerations regarding on-street parking: • The characteristics and functionality of the street, needs of adjacent landowners, applicable local policies, and plans for parking management should determine the need for on-street parking; • Where street parking is needed on higher volume urban streets, parallel parking should be used; • Angle parking may be used on low-speed and low-volume collector streets with ground-floor commercial uses, primarily those serving as main streets; • On-street parking should generally be prohibited on streets with speeds greater than 35 mph; and • Widths of parking lanes are dependent on the context of the area, the functionality of the street, the expected vehicle use, and the anticipated frequency of parking turnover. The ITE design guide also reaffirms the necessity of conforming to accessibility require- ments. Where vehicle capacity (i.e., mobility) needs to be balanced with on-street parking (i.e., accessibility), consider using the curb lane for parking during off-peak periods and for traffic during peak periods. This strategy requires daily enforcement and immediate towing of violators, but it will have an impact on the walkability of the roadside during peak hours. This strategy should be considered where traffic congestion causes significant delays to adjacent residential neighborhoods or in areas with poorly connected networks and limited alternative routes (ITE 2010a). Source: AASHTO (2004b) Exhibit 4-23. Effective turning radius.

Traveled Way Design Guidelines 135 4.2.10.3 Recommended Practice for On-Street Parking Lanes Consistent with current AASHTO policy, the desirable width of a parallel on-street parking lane is 8 ft. in most locations. Within residential areas and/or constrained rights-of-way, 7-ft. parking lanes can be used. When it is desirable to narrow the parking lane to allocate additional space for bicyclists within a bicycle lane, the bicycle lane and/or buffer should be widened con- sistent with the change in parking lane width. Otherwise, if the parking lane is narrowed and the bicycle lane width is left unchanged, the bicycle lane shifts toward the curb, and no additional operating space is afforded to bicyclists (Furth et al. 2010). 4.2.10.4 Additional Guidance A summary of additional guidance for on-street parking lanes includes: • By narrowing the parking lanes (e.g., to 8 ft. or 7 ft. wide), the proportion of vehicles parked closer to the curb is expected to increase (Furth et al. 2010); • For parking lanes 7 ft. to 9 ft. wide, the dooring zone for parked vehicles extends approximately 11 ft. from the curb (see Section 4.2.D on the design of bicycle lanes) (Torbic et al. 2014); • Angle parking is permissible where operating speeds are 25 mph or lower and the delay pro- duced by parking maneuvers is acceptable; • Where practical or on bicycle routes, back-in angle parking is preferable to front-in angle parking; • Trade-offs associated with different angles of parking include that lower angle parking results in fewer parking spaces, higher angle parking requires a wider adjacent travel lane to keep parked vehicles from backing into the opposing travel lane when exiting the park- ing space, and back-in angle parking requires a wider edge zone in the roadside due to the longer overhang at the rear of most vehicles, a narrower travel lane adjacent to parking for maneuvering, and less depth for the parking stall due to the longer overhang of the curb (ITE 2010a); • Reverse (back-in) angled parking requires a wider edge zone in the roadside due to the longer overhang at the rear of most vehicles, but this extra width can be compensated for by the nar- rower travel lane that is needed adjacent to parking for maneuvering and by less depth needed for the parking stall, as the longer overhang is over the curb; • For parallel parking, provide a minimum 1.5-ft. wide offset between the face of the curb and edge of potential obstructions such as trees and poles to allow for the opening of car doors (ITE 2010a); • Unless curb extensions are provided, prohibit parking within at least 20 ft. from the nearside of mid-block crosswalks and the curb return of stop-controlled intersections and within at least 30 ft. from approaches to signalized intersections (ITE 2010a); and • At bus stops, intersections, and mid-block crossings, extend curbs by 6 ft. into the parking lane to improve pedestrian visibility (ITE 2010a). 4.2.11 Traveled Way Transition Design Transitions may refer to a change in the width, cross section, or speed of a roadway, or to the need to laterally shift vehicles in travel lanes. In terms of geometric design, transitions refer to the provision of an adequate taper where lanes shift or narrow, shoulders widen, or lanes diverge or merge, and where deceleration lanes are provided. Geometric transitions usually are required when a change occurs in the roadway type with an associated change in width, particularly where functional classification and speed changes occur and where a narrowing or widening of lanes or a decrease or increase in the num- ber of lanes is introduced.

136 Design Guide for Low-Speed Multimodal Roadways In terms of vehicle operating speeds, where high-speed facilities (50 mph and higher) meet low- or intermediate-speed (45 mph and lower) facilities, there is a transition zone where drivers are expected to slow to a speed suitable for the environment and context zone they are enter- ing. The speed transition zone is not a specific point along the roadway where an abrupt speed change occurs, but rather an extended length of roadway over which travel speed transitions from higher to lower to recognize changes in context, community goals and increased levels of multimodal activity and access points. Designs that encourage gradual speed reductions over the length of the speed transition zone are preferred to designs that result in sudden reductions in speed at the end of the speed transi- tion zone. Well-designed speed transition zones incorporate traffic control devices along with roadway and roadside design features that convey the need to reduce speed and encourage grad- ual reductions in speed (Torbic et al. 2012). Physical changes in roadway alignment or width are the treatments most likely to affect driver behavior and reduce speeds. According to Forbes (2011), driver speeds will decrease as road- way deflection increases, so designers should consider changes in the roadway alignment to physically slow motorists. Gateway treatments, such as roundabouts (a FHWA-proven safety countermeasure), chicanes, raised medians, reduced lane widths, shoulder removal, providing a curb-line and/or including tall vegetation (e.g., hedges, trees) have been shown to be effective at reducing travel speeds approaching a main street (Forbes 2011). Where present, bicycle facilities should be carried through gateway treatments. Exhibit 4-24 illustrates a typical speed transition zone for a high-speed rural highway enter- ing a land use context that is suburban, urban or a small community or town. As discussed by Torbic et al. (2012), the characteristics of the speed transition zone and adjacent segments can be described as follows: • The rural zone. The rural zone consists of a high-speed (i.e., posted speed limit ≥45 mph) rural road with little roadside development, few access points and relatively few features or potential conflicts that require driver attention. Source: Adapted from Torbic et al. (2012) Speed Transition Zone Community ZoneRural Zone T ra n si tio n T h re sh o ld C o m m u ni ty T h re sh ol d Not Drawn To Scale Perception-Reaction Area Deceleration Area Begin Substantive Speed Reduction Design Speed Traffic Volume Access Density Ped/Bike Activity Land Use On-Street Parking Lower Low Low Rural/Low Density No Lower Low Low Rural/Low Density No Varies But Decreasing Increasing Medium Medium Increasing Density & Intensity Unlikely Higher High High Higher Density & Intensity Possibly 45 mph 45 mph 35 mph Exhibit 4-24. High- to low-speed transition zone area.

Traveled Way Design Guidelines 137 • The transition zone. Located between the rural (i.e., high-speed) and community (i.e., intermediate- or low-speed) context zones, the transition zone consists of two areas: (1) a perception-reaction area, where some physical and operational characteristics of the context area begin to change, conveying to the drivers an impending need to change speed and driving behavior, and (2) the deceleration area, where changes in the roadway, roadside characteristics, land use and access are sufficient that drivers are expected to decelerate to lower operating speeds before entering the developed area. • The community zone: The community zone is the portion of roadway serving the more developed area and has different design characteristics from the other zones, including ele- ments such as a lower speed limit, increased traffic control, on-street parking, bicycle lanes, sidewalks, curb and gutter, higher land use intensity with reduced building setbacks, frequent access points, landscaping, pedestrian and bicycle activity, raised medians, curb extensions, narrow lanes and turn lanes. The transition threshold, separating the rural zone from the transition zone, is the approximate location where drivers first observe downstream signs or features that alert them to upcoming roadway context and speed changes. From a design standpoint, at the transition threshold it is best if the entire speed transition zone is visible to drivers. The community threshold, separating the speed transition zone from the community zone, is where the 85th percentile operating speed should be consistent with the posted speed limit for entering the community. The community threshold should be near the edge of development for the community, and be defined by land use density, the number of access points, and changes in the roadway and roadside design. It may be appropriate to set back the community threshold a few hundred feet from the community zone, but if the com- munity threshold is set back too far from dense development, drivers may not maintain the desired speed through the community. On the other hand, it may be necessary to set the community threshold far enough away from current development to allow for growth (Torbic et al. 2012). 4.2.11.1 Current AASHTO Policy and Guidance The Green Book provides guidance for designing roads in high-, intermediate-, and low- speed environments; however, the Green Book provides little guidance on the design of speed transition zones. Where the Green Book provides guidance on transition design, it focuses on specific geometric elements (e.g., roadway width, lane width or number of lanes) rather than on the context and purpose of the roadway. 4.2.11.2 Speed Transition Zone Design Principles and Considerations Each speed transition zone and community has its own unique characteristics and context. Thus, each speed transition zone must be assessed on a case-by-case basis to select appropri- ate treatments. Several guiding principles should be considered during the design of a high- to low-speed (or high- to intermediate-speed) transition zone. Some general principles can be assembled from existing national and international sources, including: • NCHRP Report 737: Design Guidance for High-Speed to Low-Speed Transition Zones for Rural Highways (Torbic et al. 2012); • NCHRP Synthesis Report 412: Speed Reduction Techniques for Rural High-to-Low Speed Transitions (Forbes 2011); • Speed Management (ECMT 2006); • Guidelines on Traffic Calming for Towns and Villages on National Routes (REV B) (NRA 2005); • Guidelines for Urban-Rural Speed Thresholds—RTS15 (LTSA 2002); and • Reducing Traffic Injuries Resulting from Excess and Inappropriate Speed (ETSC 1995).

138 Design Guide for Low-Speed Multimodal Roadways These general principles include: • More extensive and aggressive treatments produce greater reductions in speed and crash occurrence than less extensive and passive treatments, and combinations of treatments are more effective at reducing speeds and improving safety than a single treatment; • There should be a distinct relationship between the speed limit and the roadway and roadside characteristics; • As the speed limit is reduced, the change in roadway and roadside characteristics should be apparent enough to reinforce the need for drivers to slow their speeds; • Physical changes to the roadway and roadside are preferred to enforcement and education programs because they have more substantial and lasting effects; • In the perception-reaction area of a speed transition zone, warning and/or psychological treatments (i.e., design and/or contextual elements that communicate suitable operating speed cues to drivers) are more appropriate, and in the deceleration area of a speed transition zone, physical treatments are preferred; and • To maintain a reduction in speed downstream of the speed transition zone, additional treat- ments are necessary within the intermediate- or low-speed environment. The example speed transition zone illustrated in Exhibit 4-24 is of a high-speed rural roadway transitioning to a low-speed roadway through a small rural town. Similar speed transition zones may occur in suburban and urban settings. The design principles described in this section of the Guide can apply to a variety of situations, but different types of treatments may be used to reduce speed in the differing environments. 4.2.11.3 Recommended Practice Speed transition zones are intended to convey the need to change speeds due to a change in environment or context. To effectively encourage and reduce speeds, the following practices are recommended for the design of speed transition zones (Torbic et al. 2012, ITE 2010a): • For changes in traveled way width, designing a geometric transition such as a lateral shift, lane addition or drop, lane or shoulder narrowing and so forth, use the established guidance in the MUTCD, in which the length of the transition taper is computed by the following equa- tion: L = WS2/60 (for speeds less than 45 mph), where L equals the length of the transition taper (ft.), W equals the width of the lateral shift or offset (ft.) and S equals the 85th percentile operating speed in mph or posted speed in mph (whichever is higher) or the target speed in new construction projects. • When the traveled way is being widened or lanes added, a transition taper of 10:1 is normally sufficient. Speed-change lanes at intersections (transitions to left-turn or right-turn lanes) usually require a shorter taper and deceleration distance. AASHTO recommends 100 ft. for single-turn lanes and 150 ft. for dual-turn lanes. • Traffic control devices should be used consistent with the current edition of the MUTCD (FHWA 2009b). The MUTCD should be referenced for guidance on the use of “REDUCED SPEED AHEAD” signs and stepped-down speed limits. • Speed transition zones should desirably be designed on tangent sections of roadway to avoid horizontal and vertical sight distance constraints so the entire length of the speed transition zone is visible to drivers. • In the context of a high-speed rural road transitioning into a built-up rural town or suburban/ urban community, use of landscaping elements such as grass, shrubs, and trees that change in composition and degree of formality along the length of the transition zone is recommended to reinforce the changing characteristics of the environments. • In the context of speed transition zones in more suburban and urban environments, roadside design features (e.g., landscaping, curbs, on-street parking, bike lanes, street light standards

Traveled Way Design Guidelines 139 with banners, entry signs, and street furniture) can serve as visual cues to influence driver speeds. Land uses, building styles and setbacks also can provide visual cues. Progressively introducing taller and closer roadside design elements to the traveled way may also reinforce the principle that altering the physical relationship between the width of the road and the height of nearby vertical elements influences a driver’s perception of the appropriate speed. • At the downstream end of the speed transition zone (i.e., community threshold), a gateway can be included. Gateways can be achieved using combinations of design features or unique intersections such as roundabouts. If sidewalks and/or bicycle lanes are not present upstream of the gateway, they should be introduced on the downstream side of the gateway, signaling the potential for increased pedestrian and bicycle activity. • Consider reducing the overall curb-to-curb width of the street within the speed transition zone to convey the change in context. This can be accomplished by reducing the number of through lanes, reducing lane widths, dropping through lanes as turning lanes at intersections, converting through lanes to on-street parking or bicycle lanes, applying curb extensions at intersections and mid-block crossings and providing a raised median. • Prohibit passing within the speed transition zone. Several resources provide information on treatments that may be implemented within speed transition zones to reduce speeds and improve safety, including: • Transition Zone Design (Stamatiadis et al. 2014); • NCHRP Report 737: Design Guidance for High-Speed to Low-Speed Transition Zones for Rural Highways (Torbic et al. 2012); • NCHRP Synthesis Report 412: Speed Reduction Techniques for Rural High-to-Low Speed Transitions (Forbes 2011); • Determining Effective Roadway Design Treatments for Transitioning from Rural Areas to Urban Areas on State Highways (Dixon et al. 2008a); • Evaluation of Gateway and Low-Cost Traffic-Calming Treatments for Major Routes in Small Rural Communities (Hallmark et al. 2007); • Main Street . . . When a Highway Runs Through It: A Handbook for Oregon Communities (Oregon DOT 1999); • Engineering Countermeasures for Reducing Speeds: A Desktop Reference of Potential Effectiveness (FHWA 2014b); • Speed Management: A Manual for Local Rural Road Owners (FHWA 2012b); and • Speed Concepts: Informational Guide (FHWA 2009c). 4.2.12 Intersections Most conflicts between users of the traveled way occur at at-grade intersections, where trav- elers of all modes cross paths with each other. Vehicle conflicts with pedestrians or bicyclists are of greater concern given the greater vulnerability, lesser size, and variable visibility of non- motorists. Therefore, good intersection design should indicate to all modes approaching the intersection what they must do and who has to yield. Roundabout intersections can reduce conflict potential and increasingly are being considered as alternatives to traditional at-grade intersections. Roundabouts often work well for multimodal roadways in low-speed environments. They are addressed in more detail later in this section. The following principles apply to designing intersections for all users: • Free-flowing movements should be avoided; • Intersections should be designed to be as compact as possible while accommodating buses and emergency response vehicles;

140 Design Guide for Low-Speed Multimodal Roadways • Unexpected conflicts should be avoided; • Simple right-angle intersections are best for all users because many intersection problems are worsened at skewed and multilegged intersections; • Access management practices should be used to remove additional vehicular conflict points near the intersection; • Signal timing should consider the safety and convenience of all users; • Intersection designs should integrate geographic constraints; • Special consideration should be given in areas where users include large populations of dis- abled people, elderly people or children; • Designs should encourage proper travel behavior by all users; and • Intersections should be designed with high legibility and clarity. 4.2.12.1 Skewed Intersections Generally, skewed intersections are undesirable because they introduce the following com- plications for all users: • The greater travel distance across the intersection increases exposure to conflicts and length- ens signal phases for people on foot and in vehicles; • Skewed intersections require users to crane their necks to see other approaching users, making it less likely that all users will be seen; and • Obtuse angles encourage speeding around corners. To alleviate problems associated with skewed intersections, design options may include: • Designing or redesigning the intersection closer to a right angle, if possible; • Providing pedestrian refuges if the crossing distance exceeds approximately 40 ft.; • Marking general-use travel lanes and bike lanes with dashes to guide people on bicycles and motorists through a long, undefined area; and • Channelizing paths of travel to maximize predictability and create space for pedestrian islands. Redesigning an intersection to bring it closer to a right angle may require the purchase of additional right-of-way; however, the added cost may be offset by selling land that is no longer needed for the intersection back to adjoining property owners or by repurposing the unnecessary land (e.g., for a pocket park, rain garden, or greenery). 4.2.12.2 Multileg Intersections Multileg intersections (intersections with more than four legs) generally are undesirable and introduce the following complications for all users: • Multiple conflict points are added as users arrive from several directions; • Users may have difficulty assessing all approaches to identify all possible conflicts; • At least one leg will be skewed; and • Users must cross more lanes of traffic and the total travel distance across the intersection is increased. To alleviate problems associated with multileg intersections, design options may include: • Creating a minor intersection farther upstream or downstream (which enables the removal of one or more legs from the primary intersection); • Using roundabouts (though not necessarily mini-circles), as they are a proven safety countermeasure; • Close one or more approach roads to motorized vehicle traffic, while still allowing access for people on foot or on bicycles;

Traveled Way Design Guidelines 141 • Creating pedestrian refuges if the crossing distance exceeds approximately 40 ft.; and • Marking general-use travel lanes and bike lanes with dashes to guide people on bicycles and motorists through a long undefined area. 4.2.12.3 Corner Radii An intersection’s corner radius has a significant impact on the comfort and safety of non- motorized users. Small corner radii provide the following benefits: • Smaller, more pedestrian-scale intersections, resulting in shorter crossing distances; • Slower vehicular turning speeds; • Better geometry for installing perpendicular ramps for both crosswalks at each corner; and • Simpler, more appropriate crosswalk placement in line with the approaching sidewalks. Normally, the smallest practical corner radii should be selected for intersection designs in areas with significant pedestrian and bicycle activity. It is helpful to keep in mind that the actual curb radius (the radius of the constructed curb) differs from the effective curb radius, which is the corner radius required for a vehicle to make a turn. Where on-street parking or bike lanes exist and curb extensions are not present, the actual curb radius may be minimized. When designing corner radii for complete streets, a default design vehicle might be a refuse truck (representing a larger vehicle that is a regular user of the street). The design vehicle can be modeled through use of a single-unit truck. (Larger design vehicles should be used only where they are known to regularly make turns at the intersection, and corner radii should be designed based on the larger design vehicle traveling at “crawl” speed.) Designers also should consider the effects that bike lanes and on-street parking have on the effective radius, increasing the ease with which large vehicles can turn. Under some conditions, multicentered compound curves and circular curves with tangent offsets may be useful alterna- tives to simple circular curve designs for corner radii. Encroachment by large vehicles is acceptable onto multiple receiving lanes. For example, when using a design vehicle larger than a refuse truck, the vehicle should be allowed to turn into all available receiving lanes. Larger vehicles that are infrequent users of the street can be allowed to encroach on multiple departure lanes and partway into opposing traffic lanes. Based on their review of several sources, the project research team identified several general design guidelines for corner radii, including: • The effective curb radius should be 28 ft. to accommodate a single-unit truck (SU-30) design vehicle. • Where parallel on-street parking or bike lanes exist, their width should be factored into the effective corner radius, and the actual curb radius should be 5 ft. to 15 ft. • Where bus routes turn right, the design vehicle should be the typical city bus and the effective curb radius set to accommodate the turn movement. • In industrial areas, or locations with frequent truck movement, the design vehicle should be increased to an intermediate semi-trailer (WB-50) or appropriate, and the effective curb radius should be increased to accommodate the turn movement. The Green Book recommends that the corner radius be designed for the intersecting street type and design vehicle. Guidelines for the right-turning radii into minor side streets usually suggest that they be between 10 ft. and 15 ft. (AASHTO 2011a). Where a substantial number of pedestrians are present, the lower end of the range may be appropriate. A corner radius as low as 15 ft. also may be appropriate on arterial streets carrying heavy traffic volumes. Where buses or large trucks are prevalent, a larger curb radius should be considered.

142 Design Guide for Low-Speed Multimodal Roadways 4.2.12.4 Curb Extensions The project team’s review of multiple references and best practices suggests that, where on-street parking is allowed, curb extensions should be considered to replace the parking lane at crosswalks. Curb extensions should be the same width as the parking lane. Bulb-outs and curb extensions should be designed with two return curves with a radius of more than 10 ft. to allow street sweepers to clean the corners. Because of reduced road width, the corner radius on a curb extension may need to be larger than if curb extensions were not installed. Curb extensions reduce pedestrian crossing distance, resulting in less exposure to vehicles and shorter pedestrian clearance intervals at signals. 4.2.12.5 Crosswalk and Ramp Placement Crosswalks and ramps at intersections should be placed so they provide convenience and safety for people on foot (including dismounted bicyclists). The following suggested best practices identify design and operations techniques to help achieve these goals: • To maximize pedestrian access, allow crossings on all legs of the intersection unless no pedestrian-accessible destinations exist on one or more corners; • To reduce the amount of time that pedestrians are exposed to motorized vehicles, minimize crossing distances whenever possible; • Add room for street furniture, landscaping and curb ramps; • Balance the speed of turning vehicles with pedestrian safety needs; • To allow pedestrians to walk out toward the edge of the parking lane without entering the traveled way, add curb extensions; • Provide marked crosswalks on all legs of signalized intersections, along intersections for official school routes, on all legs of intersections in transit-oriented development districts and at other stop-controlled or uncontrolled locations where significant numbers of pedestrians cross; • Place crosswalks as close as possible to the desired path of pedestrians (generally in line with the approaching sidewalks); • Ensure that adequate sight lines exist between pedestrians and motorists; • Ensure that crosswalks are not placed too far away from the intersection; • When a raised median is present, consider extending the nose of the median past the cross- walk with a cut-through for pedestrians; • Provide one ramp per crosswalk (two per corner for standard intersections with no closed crosswalks); • Ensure that all ramps are entirely contained within the crosswalk, either by relocating difficult ramps or, if necessary, by flaring the crosswalk to capture a ramp that is difficult to relocate; • Align the ramp run with the crosswalk when possible, as ramps that are angled away from the crosswalk may lead some users into the intersection; and • Ensure consistency with the most recent adopted version of the federal ADA guidance and standards. 4.2.12.6 Intersection Grade The Green Book notes that the gradients of intersecting roads should be as flat as practical on those sections that are to be used for storage of stopped vehicles (AASHTO 2011a). For passenger cars, the calculated stopping and acceleration distances on grades of 3 percent or less differ little from the corresponding distances on a level roadway. Accordingly, grades in excess of 3 percent should be avoided; however, where extenuating conditions exist, grades should not exceed about 6 percent. 4.2.12.7 Intersection Sight Distance Intersection sight distance should be calculated in accordance with the Green Book using the design speed appropriate for the streets being evaluated. When executing a crossing or turning

Traveled Way Design Guidelines 143 maneuver onto a street after stopping at a stop sign, stop bar or crosswalk, drivers will move slowly forward to obtain sight distance (without intruding into the crossing travel lane), stop- ping a second time as necessary. Therefore, when curb extensions are used or on-street parking is in place, the vehicle can be assumed to move forward on the second stop movement, stopping just shy of the travel lane and increasing the driver’s potential to see further than when stopped at the stop bar. The increased sight distance provided by the two-stop movement allows parking to be located closer to the intersection. On-street parking should be positioned far enough away from intersections to allow for good visibility of pedestrians preparing to cross the street. The distance should be at least 20 ft. from crosswalks. One way to achieve this positioning is using curb extensions. Sight- lines should be maintained to minimize conflicts between street users. Some guidelines for sightlines include: • Maintain a minimum of 75 ft. of sightline to signal posts at signalized intersections; • Maintain a minimum of 40 ft. of sightline at controlled mid-block crossings; • Maintain a minimum of 40 ft. of sightline at uncontrolled crossings; and • Maintain at least 10 ft. to 20 ft. of sightline to sidewalks for driveway users, and at least 30 ft. of sightline into the street. 4.2.12.8 Turn Lanes The need for turn lanes for motorized vehicle mobility should be balanced with the need to manage vehicle speeds and the potential impact on multimodal facilities such as bike lanes and sidewalks. Turn lanes allow turning vehicles to move over so that through vehicles can maintain their speed. Left-turn lanes are considered to be acceptable in an urban environment since there are nega- tive impacts to traveled way capacity when left-turning vehicles block the through movement of vehicles. Sometimes a left-turn pocket is sufficient, without providing for vehicular deceleration, just long enough for one or two cars to wait out of through-traveling traffic. Left-turn lanes should be considered where appropriate in an urban environment or when being used with a road diet. The number of turn lanes should be minimized and double turn lanes should be avoided where possible. NCHRP Report 745: Left-Turn Accommodations at Unsignalized Intersections (Fitzpatrick et al. 2016) and NCHRP Report 457: Evaluating Intersection Improvements: An Engineering Study (Bonneson and Fontaine 2001) include guidance on where to consider a left-turn lane at unsignalized intersections. The Green Book states that dedicated left-turn lanes should be provided where exclusive left-turn signal phasing is provided. Dedicated left-turn lanes also should be considered if the peak-hour turn volumes are greater than 100 vehicles per hour (vph). Double left-turn lanes should be considered only where the peak-hour turn volumes are greater than 300 vph (AASHTO 2011a) Right-turn lanes can increase the crossing distance for people on foot, increase speed of through vehicles, and increase potential conflicts with people on bicycles; therefore, exclusive right-turn lanes should be used only when necessary. Where heavy volumes of right turns exist, or high volumes of conflicting pedestrian cross- ing movements, a right-turn lane may be the best solution to provide additional vehicle capacity without adding additional lanes elsewhere in the intersection. For turns onto roads with only one through lane and where truck turning movements are rare, providing a small

144 Design Guide for Low-Speed Multimodal Roadways corner radius at the right-turn lane often provides the best solution for pedestrians’ safety and comfort. Right-turn channelization islands between the through lanes and the right-turn lane can enhance pedestrian safety and access at intersections of multilane traveled ways where trucks make frequent right turns. Right-turn channelization islands also offer a good alternative to an overly large corner radius. The following design practices for right-turn lane channelization islands should be used to provide safety and convenience for pedestrians, bicyclists, and motorists: • Provide a yield sign for the slip lane; • Provide a 55-degree to 60-degree angle between vehicle flows, which reduces turning speeds and improves the yielding driver’s visibility of pedestrians and vehicles; • Place the crosswalk across the right-turn lane about one car-length back from where drivers yield to traffic on the other street, allowing the yielding driver to first respond to any potential pedestrian conflict (independently of any vehicle conflict), and then move forward with no more pedestrian conflict; • Consider installing a raised pedestrian crossing to slow vehicle turning movements and improve safety for pedestrians; • Consider installing bollards where a raised pedestrian crossing is constructed within a chan- nelized right-turn lane; and • When installing bollards, place them on either side of the crosswalk, within the raised island and sidewalk and with adequate separation to maintain ADA clearance (U.S. Access Board 2011) without allowing sufficient space for a car to travel through. A channelized island should be designed roughly twice as long as it is wide. The corner radius will typically have a long radius (150 ft. to 300 ft.) followed by a short radius (20 ft. to 50 ft.). When creating this design, it is necessary to allow large trucks to turn into multiple receiving lanes. Often, this design is not practical for right-turn lanes onto roads with only one through lane. This right-turn channelization design is different from designs that provide free-flow movements (through a slip lane) where right-turning motorists turn into an exclusive receiving lane at high speed. Right turns could be signal-controlled in this situation to provide for a sig- nalized pedestrian walk phase. A raised pedestrian crosswalk could also be used to slow vehicles down and make pedestrians more visible. NCHRP Report 457: Evaluating Intersection Improvements: An Engineering Study Guide (Bonneson and Fontaine 2001) includes guidance on where to consider a right-turn lane at unsignalized intersections. Guidance from Designing Walkable Urban Thoroughfares: A Con- text Sensitive Approach (ITE 2010a) states that a right-turning volume of 200–300 vph is an acceptable range for the provision of right-turn lanes at signalized intersections. The Green Book (AASHTO 2011a) provides guidance on the design of components associ- ated with an auxiliary, median or turn lane. When designing a turn lane, it is necessary to con- sider the necessary storage, deceleration and taper lengths. The deceleration length is a function of the brake-reaction distance and the distance required for the approaching vehicle to come to a stop. It is common practice to accept a moderate amount of deceleration within the through lanes and to consider the taper length as part of the deceleration within the through lanes. On urban or low-speed roadways, deceleration lengths are not always needed, as the deceleration of the turning vehicles helps control the speed of through-traveling traffic. The length of the turn lane should be sufficient to store the number of vehicles (or queue) likely to accumulate during a crucial period. On urban roads, where space is limited, a minimum 50-ft. storage distance should be provided to accommodate two vehicles. Where a high percent- age of buses or trucks are expected, this length should be increased.

Traveled Way Design Guidelines 145 The applicable taper design is a function of the design speed and roadway geometrics. For urban areas, short tapers are preferred because they provided more vehicle queue space and are not as crucial for slow vehicle speeds during peak periods. 4.2.12.9 Intersection Controls Two-way stops and yield intersections are the most common forms of controlled intersections. In low-volume areas, however, stop signs can create unnecessary delays by forcing drivers on the minor approach to stop even when no conflicting vehicle exists. Drivers may, as a result, begin to disregard the stop sign, creating potential for crashes or collisions when approaching vehicles are not seen. Yield signs are an alternative to stop signs and may be appropriate when some level of right-of-way assignment is desired but sight distance is adequate for yield conditions. Mini-roundabouts may also may be considered as an alternative intersection control measure; mini-roundabouts keep speeds low along both approaches and provide additional opportunity for landscaping. All-way stops are used where equal volumes of traffic exist along intersecting approaches. All-way stops often are used as an interim measure before an intersection is signalized. Where possible, mini-roundabouts or traffic circles should be considered as an alternative intersection control to allow conflicting side street vehicle movements to occur concur- rently with reduced delay. Signalized intersections provide unique challenges and opportunities for livable communities and multimodal design. On one hand, signals provide control of people on foot and motorized vehicles with numerous benefits. Where signalized intersections are closely spaced, signals can control vehicle speeds by providing appropriate signal progression along a corridor. Traffic sig- nals can reduce conflicts with motorized vehicle traffic when pedestrians or people on bicycles cross major streets, but their operation also can create challenges for non-motorized users. Sig- nalized intersections often have significant vehicular turning volumes, which can conflict with concurrent movements of pedestrians and people on bicycles if not properly addressed through signal phasing and timing. In many situations, roundabouts may offer a safer, more convenient intersection treatment for multimodal accommodation than traffic signals. To improve livability and pedestrian/bicycle safety, signalized intersections should: • Provide signal progression at speeds that support the target speed of a corridor whenever feasible; • Provide short signal cycle lengths, which allow frequent opportunities to cross major traveled ways, improving the usability and livability of the surrounding area for all modes; • Ensure that signals detect bicycles; • Place pedestrian signal heads in visible locations; • Time the pedestrian phase to be on automatic recall at locations and/or times that experience high foot traffic (e.g., central business districts, transit areas, and near schools at certain times of day); • Set signal timing to favor uninterrupted travel for people on bicycles or those using bus tran- sit; and • Where few pedestrians are expected and automatic recall of walk signals is not desirable, place pedestrian pushbuttons in convenient locations per the MUTCD (FHWA 2009b). 4.2.12.10 Roundabouts Roundabouts should generally be considered the first traffic control option at otherwise con- trolled intersections. Roundabouts reduce vehicle-to-vehicle and vehicle-to-pedestrian conflicts and, because they reduce vehicle speed, also reduce all forms of crashes and crash severity. In

146 Design Guide for Low-Speed Multimodal Roadways particular, roundabouts eliminate left-turn and right-angle crashes, which are common crashes at signalized intersections. Other benefits of roundabouts include: • Reduced delay, travel time, and vehicle queue lengths; • Facilitated U-turns; • Improved accessibility to intersections for people on bicycles through reduced conflicts and vehicle speeds; • Options for people on bicycles of differing abilities to navigate the roundabout; • Reduced maintenance and operational costs (primarily maintenance of landscaping and litter control); • A smaller carbon footprint, as no electricity is required for operation and vehicles use less fuel because they spend less time idling and do not have to accelerate as often from a dead stop; • The opportunity to reduce the number of vehicle lanes between intersections (e.g., to reduce a five-lane road to a two-lane road, due to increased vehicle capacity at intersections); • Little to no delay for pedestrians who have to cross only one direction of traffic at a time and do not need to wait for a specific pedestrian crossing phase; • Lowered noise levels; • The central island can vary in shape from a circle to a “square-a-bout” (e.g., in historic areas), ellipses at odd-shaped intersections, dumbbell, or even peanut shapes; and, • The ability to create a gateway or transition between distinct areas. Single-lane roundabouts can vary in size, with central island diameters from 12 ft. to 90 ft. to fit a wide range of intersections and accommodate through movements and different turn move- ments by various design vehicles. As such, they can be used at a large number of intersections to achieve various objectives. Some single-lane roundabouts are constructed to accommodate through movements by large articulated trucks, but do not permit the trucks to make turn movements. However, they do normally accommodate turn movements by single-unit trucks such as ladder trucks and garbage trucks. Pedestrian crossings are improved through the inclusion of splitter islands and slow speed approaches by vehicles. People on bicycles are able to navigate the roundabout by taking the lane or accessing the sidewalk through a ramp. Multilane roundabouts can be considered when single-lane roundabouts are inadequate for the traffic volume. Consideration should be given to using roundabouts that have two through lanes on the major street and a single lane on the minor street (with or without additional turn lanes) before automatically designing a full multilane roundabout. Because these multilane roundabouts are larger than single-lane roundabouts, they often accommodate all turn move- ments by most large vehicles. It is still necessary to confirm the size and movements by the design vehicle(s), however, because these roundabouts often have to accommodate larger trucks or special vehicles. Multilane roundabouts provide bicyclists the option to take the lane or divert onto the side- walk through a ramp. To improve pedestrian safety when crossing the multiple approach lanes, additional treatments can be provided (e.g., pedestrian-actuated RRFBs). Mini-roundabouts have traversable islands and yield control on all approaches, which allows them to function similarly to other roundabouts (FHWA 2013e). They should not be confused with neighborhood traffic circles. Mini-roundabouts are used in low-speed urban environments, where operating speeds are 30 mph or lower and right-of-way constraints preclude the use of a standard roundabout. The design is based on passenger vehicles passing through the roundabout

Traveled Way Design Guidelines 147 without traveling over the central island. To accommodate large vehicles, mini-roundabouts can be designed to include a traversable central island and traversable splitter islands. Neighborhood traffic circles are very small circles that are retrofitted into local street inter- sections to control vehicle speeds, calming traffic on low-volume neighborhood streets. Mini- circles also may be installed in neighborhoods for aesthetic purposes (FHWA 2013e). Typically, landscaping and/or a tree will be located within the central island to provide increased visibility of the traffic circle and enhance the intersection. Neighborhood traffic circles should generally have similar features as roundabouts, including yield-on-entry signage and painted or mountable splitter islands. Larger vehicles can turn left in front of the central island. The design of neighborhood traffic circles is primarily confined to selecting a central island size to achieve the appropriate design speed of approximately 15 to 18 mph (Rodegerdts et al. 2010). Refer to NCHRP Report 672: Roundabouts: An Informational Guide, 2d Ed. (Rodegerdts et al. 2010) for roundabout design guidance. 4.3 Other Design Considerations for All Users in Traveled Way Design Several considerations in roadway design are not cross section elements but may be important to the design process, depending on community goals, context relationships, operational and safety considerations, geography, climate and other factors. Those considerations include: • Speed management; • Access management; • Snow removal and storage; • Stormwater management; • Special roadway designs (e.g., road diets, main streets or shared streets); • One-way streets; • Bridges; • Railroad-highway grade crossings; • Fire and emergency medical services; and • Traffic control devices and operations. 4.3.1 Speed Management Research has shown that higher operating speeds result in higher crash severity, including higher percentages of injury and fatality crashes and more serious property damage. Pedestrians and bicyclists are particularly vulnerable in the event of a crash. Speed is of fundamental importance: the severity of a pedestrian injury in the event of a crash is directly related to the speed of the vehicle at the point of impact. For example, a pedestrian who is hit by a motor- ized vehicle traveling at 20 mph has a 95 percent chance of survival, whereas a pedestrian hit by a motorized vehicle traveling at 40 mph has a 15 percent chance of survival (Limpert 1994). In addition, vehicles traveling at lower speeds have more reaction time, which helps prevent crashes. Designing for reduced vehicle speeds is especially important in urban, urban core and rural town contexts with higher levels of pedestrians and bicyclists. Speed management is an approach to controlling speeds using enforcement, design and tech- nology applications. Although this Guide primarily addresses guidance for the design of streets and roadways, the design of a roadway facility should be closely aligned with its subsequent goals for operation and maintenance. To reinforce the relationship between the geometric design of

148 Design Guide for Low-Speed Multimodal Roadways a street or road and its resultant operating speed, in October 2015 FHWA issued a memorandum titled Relationship Between Design Speed and Posted Speed, which provides this guidance: “In urban areas, the design of the street should generally be such that it limits the maximum speed at which drivers can operate comfortably, as needed to balance the needs of all users” (FHWA 2015c). Traffic calming is a type of speed management that is usually associated with local residential streets, but speed management methods can be used on all types of roadways. Speed manage- ment methods can use technologies that provide feedback to the motorist about their speed, or designs in which the motorist perceives the need for a lower speed. These techniques include signage, signalization, enforcement, street designs and built environments that encourage slower speeds. Other methods include physical devices that force drivers to slow down, such as round- abouts, raised intersections or narrowed sections created by curb extensions and raised medians. Physical devices generally are more effective at changing driver behavior, but they may be more costly to implement and may not be appropriate on all roadways. The design decision to employ a speed management program often requires a multidisci- plinary process, as this approach affects several groups of users and adjacent property own- ers. Developing the program requires input from engineering, emergency services, bicycle and pedestrian advocacy groups, street maintenance providers, law enforcement and transit service providers. Public involvement helps the design team understand how the community uses road- ways and how it perceives various speed management methods. Designing a facility to achieve desired speed results requires knowledge of the existing traffic patterns for all users, including both quantitative and qualitative information. Ideally, speed management should be a consid- eration in all projects serving multiple modes, regardless of the functional classification of the roadway. It is important for a corridor to have appropriate operating speeds within the project limits and through different jurisdictions if the character and context also remain constant. The following lists present “active” (generally physical design features) and “passive” (gen- erally operational and psychological techniques) speed management measures commonly used in the United States on low- and intermediate-speed roadways functionally designated as arterials or collectors. Not all techniques are considered appropriate across all contexts and speed ranges. 4.3.1.1 Active Measures Commonly used active speed management measures include: • Roundabouts, particularly when used within a “roundabout corridor”; • Road diets (reducing the number of lanes by adding medians, converting travel lanes to parking, or adding bike lanes); • Narrowed travel lanes; • Center raised island; • Mid-block neck-downs (narrowing the traveled way using curb extensions, possibly with a center island); • Chicanes (lateral shift design techniques that require vehicles to move out of a straight path); • Smaller curb return radii to slow turning vehicles and the elimination of free-flow channelized right-turn lanes; • Provision of on-street parking where adjacent land uses and activities will generate demand; • Speed tables or humps (not typically used on roadways with speeds above 30 mph; may impact emergency service response routes and times); • Speed cushions or speed platforms (less impact on emergency vehicles than humps and tables); • Raised crosswalks combined with curb extensions to narrow street; and • Speed-actuated traffic signals where a vehicle traveling at excessive speeds will actuate the signal to change to red.

Traveled Way Design Guidelines 149 4.3.1.2 Passive Measures Commonly used passive speed management measures include: • Synchronized signals to create progression at an appropriate speed; • Radar trailers/speed feedback signs flashing “SLOW DOWN” message when speed exceeds a preset limit (most effective when coupled with enforcement); • Visually narrowing road using pavement markings; • Visually enclosing street with buildings, landscaping and street trees; • Variable speed limits (using changeable message signs based on conditions); • Speed enforcement corridors combined with public education; • Flashing beacons on intersection approaches to slow traffic through the intersection; • Speed limit markings on pavement; • Mountable cobblestone medians or flush concrete bands delineating travel lanes for visual narrowing; • Shared routes using signs and pavement markings (such as bicycle boulevards); and • Automated speed enforcement (including red light enforcement). These speed management measures may be implemented in the design of new facilities, reconstruction, resurfacing and restriping projects. Every design project offers opportunities to implement new design features that may help manage operating speeds to desired values. Additional detailed guidance for some of the more common design elements used for speed management include: • Mid-block neck-downs. Roadway geometry can be altered at mid-block locations to reduce motorized vehicle speeds by diverting the driver’s path of travel. Neck-downs are curb exten- sions on opposite sides of the road, which create a “pinch-point.” They are particularly use- ful on streets with longer block lengths where motorists tend to pick up speed. They can be combined with mid-block pedestrian crossings to enhance pedestrian safety further by reduc- ing crossing distances and increasing visibility. Recommendations for the use of neck-downs include: – Mid-block neck-downs can be used on two-way streets with one lane in each direction, and one-way roads with no more than two lanes. They are sometimes combined with intermit- tent medians to reduce speeds along the length of a roadway. – Vegetation used in the neck-down should generally be low growing and low maintenance. – In locations with mid-block pedestrian crossings, sight distances should be maintained. Design considerations for mid-block neck-downs include: – Where neck-downs provide pedestrian crossings, ADA-compliant curb ramps, tactile warning strips, and cross slopes should be provided; consider other traffic-calming ele- ments such as raised crossings. – Mid-block neck-downs can serve as alternatives to speed tables. – Care should be taken to avoid suddenly squeezing bicyclists into the traffic flow on streets with higher volumes of traffic, particularly in locations with steep uphill grades where bicyclists may be traveling considerably slower than motorized vehicle traffic. – On low-volume, low-speed residential streets, neck-downs can reduce the street to one lane, requiring oncoming drivers to alternate passage through the neck-down, while maintaining enough clear space for fire trucks and other large vehicles. – Designs should consider snow removal operations in snow regions. Mid-block neck-downs offer space to store snow in winter; however, visual cues should alert snowplow operators of the change in the roadway. • Chicanes. A chicane is a design feature that creates “S” curves in the roadway travel paths that drivers must weave through, with the effect of slowing speeds. Chicanes can be created by alter- nating parking from one side of the roadway to the other, as well as through curb extensions.

150 Design Guide for Low-Speed Multimodal Roadways Chicanes provide opportunities to increase sidewalk space and introduce landscaping and green street elements in the right-of-way. Recommendations for their use include: – Chicanes can be used on two-way streets with one lane in each direction, and one-way roads with no more than two lanes. – The amount of horizontal deflection should be based on the proposed design speed of the roadway. – Vegetation used in chicanes should generally be low growing and low maintenance. In loca- tions with mid-block pedestrian crossings, sight distances must be maintained. Design considerations for chicanes include: – Care should be taken to maintain space for bicyclists and to avoid suddenly squeezing bicyclists into the traffic flow on streets with higher volumes of traffic, particularly in loca- tions with steep uphill grades where bicyclists may be traveling considerably slower than motorized vehicle traffic. – Designs should consider snow removal operations. Chicanes offer space to store snow in winter; however, visual cues should alert snowplow operators of the change in the roadway. – Chicanes can serve as alternatives to speed tables. • Center raised islands. A center island can be used to narrow the roadway, reduce motorized vehicle speeds, and improve pedestrian crossings. Center islands also provide opportunities to introduce green elements in the right-of-way, and can be used to absorb stormwater and reduce the heat island effect. Recommendations for their use include: – Center islands with crosswalks and pedestrian refuges improve pedestrian safety and access by reducing crossing distances and enabling pedestrians to cross roadways in two stages. Islands with crossings should be designed with a stagger, or a “z” pattern, forcing pedestri- ans to face oncoming traffic before progressing through the second phase of the crossing. Center islands with crosswalks should meet all accessibility requirements. – Center islands can reduce the risk of head-on collisions and limit left-turn opportunities to desirable locations (e.g., signalized intersections). – Center islands should be carefully designed to ensure proper drainage and maximize the potential for on-site stormwater retention and infiltration. Design considerations for center islands include: – Sidewalks should not be reduced in width and bicycle lanes should not be eliminated to provide space or additional width for islands. – Center islands can be combined with mid-block pedestrian crossings to reduce crossing distances. – Permeable surfaces, street trees, and low-growing (less than 3 ft. at mature height including the height of the curb and earthwork), drought-resistant plant materials should be used wherever safe and feasible. – Plants should be located as far from the curb as possible to prevent exposure to salt and sand in snow regions. – Center islands should be at least 6 ft. wide when used for low plantings, 10 ft. wide for columnar trees and 18 ft. wide for larger shade trees. – Designs should consider snow removal operations. Center islands offer space to store snow in winter; however, visual cues should alert snowplow operators of the change in the roadway. • Speed tables. Speed tables are raised pavement areas that are placed at mid-block locations to reduce vehicle speeds. Speed tables are elongated and have been shown to effectively reduce 85th percentile speeds. Well-designed speed tables enable vehicles to proceed comfortably over the device at the intended speed, but cause discomfort when traversed at inappropriately high speeds. Recommendations for the use of speed tables include: – Speed tables are typically 3 in. higher than the roadway surface and 3 in. below the top of the curb, but can be fully raised 6 in. to the height of the curb.

Traveled Way Design Guidelines 151 – Generally, speed table design provides 22 ft. of length, with 6-ft. ramps and a 10-ft. flat section along the top. They normally extend the full width of the roadway, although some- times they are tapered at the edges to accommodate drainage patterns. – Speed tables should be designed with a parabolic profile or a flat top, with consideration for a smooth transition for bicyclists. – Per the MUTCD, speed tables should be clearly marked with reflective pavement markings to alert motorists and bicyclists of their presence and they can adjust their speed accord- ingly. Design considerations for speed tables include: – Speed tables should not be confused with speed bumps. Speed bumps are typically only 1–3 ft. wide and used in parking lots and are NOT recommended for public streets. – Speed tables 22 ft. in length have a design speed of 25 mph to 30 mph and are easier for large vehicles to negotiate. – Avoid placing speed tables at the bottom of steep inclines where bicyclists travel at higher speeds and may be surprised by their presence. – Speed tables should be utilized in series or supplemented with other traffic-calming mea- sures to effectively reduce travel speeds throughout a corridor or neighborhood. When used alone, speed tables may otherwise result in speed spiking, or when motorists travel at higher speeds between tables. – Designs should consider snow removal operations where appropriate. Visual cues should alert snowplow operators of the change in the roadway. • Pavement treatments. The choice of roadway materials can have significant impacts on traffic safety and speeds, user comfort, vehicle maintenance costs, stormwater management, roadway noise and the heat island effect. Paving treatments include colored pavements and stamped concrete or asphalt. Paving treatments can help reduce speeds and are more com- monly used on streets with high volumes of pedestrians and lower volumes of motorized vehicle traffic, such as shopping districts and main streets. Modern textured pavements are smoother than cobblestones, which helps accommodate bicyclists. Regardless of the material used on the roadway, an accessible and smooth travel path must be provided at crosswalks to accommodate persons with disabilities. Guidance for the use of pavement treatments includes: – Concrete is discouraged where frequent utility cuts are likely and must have joints to allow for expansion; – Pavers should generally not be used in roadway construction except for specially designated low-speed streets; – Care should also be taken to ensure that materials do not settle to different heights; – The use of paving treatments in parking lanes can visually reduce the width of the roadway; – Pedestrian crossings must meet accessibility requirements by providing a smooth, stable and slip-resistant accessible path, and should include the necessary reflective markings as required in the MUTCD; – Pavers should not be used in crosswalks; and – The use of colored pavements for traffic control purposes (i.e., to communicate a regula- tory, warning, guidance message) is narrowly defined by the MUTCD, and may be required to follow FHWA’s experimentation process. Design considerations for pavement treatments include: – Key considerations for the selection of pavement materials include constructability, ease- of-maintenance, smoothness, durability, porosity and color; – Consideration also should be given to the street type, the volumes and types of users (i.e., pedestrians, heavy vehicles, bicyclists), adjacent land uses and stormwater management goals; – Textured pavements are usually an expensive treatment and include long-term mainte- nance responsibilities;

152 Design Guide for Low-Speed Multimodal Roadways – Consider the reflective characteristics of the pavement; high albedo pavements absorb less heat; – Surfaces such as smooth granite, tile or brick should not be used because they create slip- pery conditions for bicyclists and pedestrians in wet weather; – Pavements that resist heaving and rutting should be used for locations where heavy vehicles stand or park, or at locations that are particularly susceptible to wear (e.g., high-volume intersections or steep grades); and – Concrete bus pads should be considered on high-frequency bus routes. As mentioned in this Guide under “Traveled Way Transition Design,” several documents provide detailed guidance on the selection and use of various design and operations techniques for managing speeds on streets and roads to improve safety and benefit vulnerable users such as pedestrians and bicyclists. Important information from these sources includes the following: • Engineering Countermeasures for Reducing Speeds: A Desktop Reference of Potential Effec- tiveness (FHWA 2014b). This publication provides the results of a wide range of design ele- ments and operational features proven to reduce speeds in certain settings. Supported by 54 reference studies, the countermeasure areas include geometric features, surface treatments and markings, signs, narrowing, access controls, and combination measures. • Speed Management: A Manual for Local Rural Road Owners (FHWA 2012b). This manual provides information on how to develop a speed management program that is tailored to meet the needs of local rural road practitioners. This document describes the various elements of a speed management program, including the principles of setting speed limits appropriate for roads within the jurisdiction and various countermeasures that are effective in mitigating speeding as it relates to roadway safety in rural areas. The manual addresses engineering, edu- cation and enforcement strategies. It also provides more specific guidance on traffic control elements, street and road design elements, traffic-calming techniques and gateway design treatments. • Speed Concepts: Informational Guide (FHWA 2009c). The informational guide discusses how the speed at which drivers operate their vehicles directly affects two performance mea- sures of the highway system: mobility and safety. Noting that higher speeds provide for lower travel times and therefore good mobility, this FHWA guide observes that the relationship of speed to safety is not as clear cut. The document also recognizes that the risk of injuries and fatalities increases with speed. Designers of streets and roadways use a designated design speed to establish design features and operators set speed limits deemed safe for the particular type of road; but drivers select their speed based on their individual perception of safety. This guide discusses the concepts of designated design speed, operating speed, and the speed limit and introduces a new concept, inferred design speed. It explains how these elements of speed are determined and how they relate to each other. The publication is intended to help engineers, planners and elected officials better understand design speed and its implications in achieving desired operating speeds and setting rational speed limits. 4.3.2 Access Management Access management is the practice of properly locating, designing and operating access to adjoining properties to reduce conflicts and improve safety while maintaining reasonable prop- erty access and traffic flow on the public street system. This section of the Guide addresses only private driveway access to roads and streets. Driveways, especially busy commercial driveways, can have a significant impact on the adja- cent roadway. Good driveway design should facilitate smooth vehicle egress and ingress to and from the roadway and should provide for safe accommodation of pedestrians and bicyclists.

Traveled Way Design Guidelines 153 Driveway design should consider the roadway functional classification and driveway usage to better accommodate varying roadway contexts, community needs, and existing conditions. In conjunction with the PROWAG (U.S. Access Board 2011), the ADAAG (U.S. Access Board 2002) and the Revised Draft Guidelines for Accessible Public Rights-of-Way (U.S. Access Board 2005) provide specific suggested guidelines for elements such as minimum width, cross slope, grade and edge conditions at the intersection of sidewalks and driveways to be ADA compliant. These guidelines are based on pedestrian needs and do not compre- hensively address safe and efficient vehicle movements at driveways. Recommendations are needed to accommodate accessibility concerns as well as safe and efficient vehicle use of the driveway. NCHRP Report 659: Guide for the Geometric Design of Driveways was developed to provide recommendations for the geometric design of driveways that consider standard engineering practice and accessibility needs and provide for safe and efficient travel by motorists, pedestri- ans, and bicyclists on the affected roadway (Gattis et al. 2010). This guide contains significant detailed guidance on the design of driveways that consider all users, especially those in urban and suburban context areas. 4.3.2.1 Attributes of Bicyclists, Drivers, and Pedestrians (NCHRP 659) The abilities and limitations of the people using the driveway as bicyclists, drivers and pedestrians will affect design choices. Appreciation of the concept of driver workload leads to the objective of trying to limit the number of (1) decisions a driver must make and (2) potential conflicts with different streams of traffic. Acknowledging that rain, fog and nighttime conditions can make physical objects more difficult to detect, a designer tries to create well-defined edges and increase the contrast between various surfaces (e.g., between the driveway opening and the border area). The AASHTO guides for the design of highways and streets, bicycle facilities, and pedestrian facilities provide a discussion of user characteristics (AASHTO 2011a, AASHTO 2014b, AASHTO 2004b). Characteristics of Emerging Road and Trail Users and Their Safety provides data for a wide range of users, including bicyclists and pedestrians (FHWA 2004). Pedestrians on sidewalks routinely cross driveways, particularly in urban, suburban and rural town contexts where pedestrians and bicyclists are most prevalent. TCRP Report 112/ NCHRP Report 562: Improving Pedestrian Safety at Unsignalized Crossings (Fitzpatrick et al. 2006) provides a distribution of the walking speeds of pedestrians under age 60 and over age 60. In both age groups, most pedestrians walk at speeds between 3 ft. and 6 ft. per second. When estimating the time required for a pedestrian to cross the driveway, the designer should make an allowance for the pedestrian who is not starting from the exact edge of the driveway. A pedestrian may be standing 2 ft. or more back from the driveway edge when the pedestrian begins to walk across the driveway, and if groups of pedestrians exist, the starting distance for some people in the group may be even farther back from the curb. Bicyclists also cross the paths of vehicles entering and leaving driveways. On shared-use paths, a design speed of at least 20 mph is typically assumed because speeds above that would not be desirable in a shared-use setting. Grade and wind also can affect the speeds of bicy- clists, with downgrades creating speeds closer to 30 mph. The Urban Street Geometric Design Handbook notes that most bicyclists in mixed-use settings tend to travel within a range of 7 mph to 15 mph, with an average of 10 mph to 11 mph (ITE 2009b). Characteristics of Emerg- ing Road and Trail Users and Their Safety has examined characteristics of a wide range of users, finding that the 85th percentile speed for bicycles is 14 mph, and that for recumbent bicycles is 18 mph.

154 Design Guide for Low-Speed Multimodal Roadways General design considerations for driveways include: • The number of driveways should be minimized (i.e., consolidated whenever possible) to reduce the number of conflict points for pedestrians and bicyclists and also benefit motorists’ safety; • Driveways should normally be designed to look like private driveways, not public street intersections; • Driveways should be located away from intersections; • Driveways should be kept as narrow as possible to minimize exposure to vehicles; • Well-defined driveways clearly mark the area where motorists will be crossing the pedestrians and bicyclists path; • Non-defined vehicle access points with continuous access to parking create a long potential conflict area between pedestrians and motorists, and this added area of ambiguity complicates the motorist’s task of watching for these vulnerable users; • Sidewalks and bicycle facilities behind the curb should clearly continue through the driveway approach; and • The level of the sidewalk should be maintained, and the driveway should be sloped so that the motorist goes up and over the sidewalk. It bears noting that maintaining the level of the sidewalk and sloping the driveway helps with several goals: meeting ADA accessibility requirements will be easier; the fact that the pedestrian has the right-of-way will be clear; and motorists will need to slow down slightly to enter the driveway, which will help improve crossing safety. The following access management approaches can help improve pedestrian and bicyclist safety as well as mobility at access points in the vicinity of signalized and unsignalized urban and suburban intersections: • Provide raised medians on the major roadway to prohibit vehicles from turning left into driveways, thus reducing the number of pedestrian-vehicle conflicts at the driveways; • Construct a channelized island between the inbound and outbound movements at right- turn-only driveways to provide a pedestrian refuge across the driveway; • Minimize the width of the driveway as much as possible to reduce pedestrian crossing dis- tances (i.e., reduce exposure); • Place sidewalks and pedestrian driveway crossings so that pedestrians are visible to the drivers, and drivers are visible to the pedestrians; • Do not block pedestrian-driver sightlines with landscaping or signage; • Include bike lanes and signage, as appropriate, to alert bicyclists that motorists may be entering or exiting a driveway and to alert motorists that bicyclists may be crossing the driveway; • Use colored pavement across driveways in combination with crosswalk markings and audio/ visual treatments (e.g., a signal and/or flashing sign that is activated to alert pedestrians a vehicle is about to cross the sidewalk from an adjacent parking area) for exiting vehicles with limited sight distance; • Restrict inbound vehicle speeds by designing the driveway access with appropriately designed radii; and • Take care to balance vehicle and pedestrian safety: Smaller driveway radii of 25 ft. to 35 ft. are more sensitive to pedestrian movements because motorists must slow down to complete the turn; however, on-street parking and bike lanes can increase the effective driveway radius. 4.3.3 Snow Removal and Storage During and after a snowstorm, most snowplows operate in emergency or “hurry-up” mode, focusing on opening up lanes for vehicles. Often, snow is pushed from the vehicular

Traveled Way Design Guidelines 155 lanes, into bicycle lanes, parking lanes or along the sidewalks, thus making it difficult for bicyclists and pedestrians to use the facilities that have been provided for them. Snow and ice blockages can force pedestrians onto the street at a time when walking in the roadway is particularly treacherous. Adding to the problem, piled snow can create sight distance restrictions. Many localities that experience regular snowfalls have enacted legislation requiring home- owners and businesses to clear the sidewalks fronting their property within a reasonable time after a snowfall occurs. In addition, many public works agencies have adopted snow removal programs that ensure that the most heavily used pedestrian routes (including bus stops and curb ramps at street crossings) are cleared so that snow plows do not create impassable ridges of snow. In some states, snowplow operations clear the entire roadway from curb to curb. After the roadway is cleared, a smaller “snow blow” crew (using equipment such as snow blowers, brushes, pickups and plows) clears pedestrian facilities. In areas that receive regular snow, there will be trade-offs between the recommendations of this report and the efficiency of snowplowing. Some recommended design elements, such as curb extensions and on-street parking, will affect snowplowing operations. These trade-offs need to be clearly communicated in the design process. Moreover, early collaboration with offi- cials in charge of snow removal is imperative for a successful design. 4.3.3.1 Recommended Practices for Snow Removal The following practices are recommended regarding provisions for snow removal in the design of low- and intermediate-speed roadways that serve all users: • Design street-sides to accommodate a normal level of plowed snow behind the curb without blocking pedestrian or bicycle throughways (e.g., by using a wide planting strip or furnishings zone that can accommodate plowed snow); • Design the furnishings zone to minimize the use of objects that interfere with the ability to plow snow onto the street-side (e.g., large raised planters, continuous hedges and large utility and traffic control cabinets); • Keep in mind that plowed snow can wrap around objects including trees, signs and light poles; and • Use hardscape or setbacks that place plantings and trees beyond the anticipated plow line to reduce landscape damage from deicing chemicals. 4.3.4 Stormwater Management Proper management of stormwater on multimodal urban and suburban roadways improves the walking and bicycling environment, aesthetics and community quality. Green stormwater management practices add value and multiple functionality and should be considered in road- way improvement projects. Traditionally, stormwater runoff from roadways and roadsides normally must be collected and transported within the right-of-way; however, not all communities treat stormwater the same way. In some communities, the convention is to collect and carry stormwater run- off via networks of storm sewer pipes that bring it to a treatment plant, which then sends it as an outfall into a water body or possibly to another facility for beneficial reuse. Other communities control stormwater at the source or through roadside treatment control “best management practices” (BMPs). In this Guide, stormwater treatments that use the traveled way are discussed in this section; the use of BMPs is addressed in Chapter 5, Roadside Design Guidelines.

156 Design Guide for Low-Speed Multimodal Roadways 4.3.4.1 Stormwater Drainage Inlet Design On many roads, stormwater drainage design typically involves a curb and gutter and drainage grates that occur on the right edge of the road, creating potential conflicts with bicyclists. The most effective way to avoid drainage grate problems for bicycles is to eliminate them entirely by using inlets in the curb face. Using alternatives to curb and gutter design can provide the same function as standard gutters and grates while not posing an impediment to bicyclists. Where grates are used, however, the following practices will reduce their impact on bicycling safety: • Design considerations. The function of drainage grates is to drain stormwater quickly from the roadway and to provide access to maintenance workers to clean out the inlet. The gutter and grate (or inlet) must be hydraulically effective. Gutters are sloped to direct water flow into the inlet. This keeps water from ponding at the longitudinal joint and undermining the pave- ment. Gutters also protect the curb from being damaged during maintenance and resurfacing. Grates may be rendered useless, however, if they become clogged with debris (e.g., in areas with many deciduous trees). Generally accepted best practices suggest that a clogging factor of at least 50 percent should be assumed for city streets in the absence of local data. • Grateless roadway designs. Curb-face opening inlet designs can be an effective substitute for grates, particularly on roadways with grades of less than 3 percent. For hydraulic efficiency, a depression of approximately 1 in. is needed in the vicinity of the curb-face inlet; this depres- sion should occur gradually so that it does not pose an obstacle to bicyclists. Maintenance access can be placed in back of the curb (on the sidewalk side). Where curb-face inlets cannot be built, slotted linear drain inlets can be used in the shoulder area in lieu of the grate inlets. • Design and placement of drainage grates. Attempts have been made to retrofit bicycle-unsafe grates by welding crossbars onto the parallel bars, but this solution usually is unsatisfactory, and it costs more than initially installing correctly designed grates. Optimally, the roadway should be designed so that the bicyclist does not have to traverse the grate. On roadways with curb and gutter, the grate should not be wider than the gutter pan. If the gutter pan needs to be widened to accommodate a large drainage grate, the taper should be on the outside edge into the planter strip. On roads with bike lanes, the roadway should be designed such that the minimum bicycle lane width of 4 ft. is ideally maintained between the bike lane stripe and the edge of the gutter, or to the face of curb in constrained settings. If 4 ft. cannot be maintained, then a curb-face inlet design for the drainage grate should be considered. On roadways with shoulders, the grate should be placed outside the travel path of the bicy- clist (i.e., 4 ft. of clear pavement should be maintained between the shoulder stripe and the left edge of the drainage grate). If 4 ft. cannot be provided within the existing shoulder width, the shoulder can be widened to accommodate the grate, with the taper on the outside edge, or a narrower grate should be selected. Optimally a 12-in. maximum gutter pan should be used on new construction projects. 4.3.5 Utility Facilities Utility facilities normally are placed in the roadside whenever possible. In many urban and suburban communities, however, the roadside has limited room for utility placement and access, so utilities are placed under the traveled way with numerous utility access points. Utility covers and construction plates can be slippery, and they create changes in surface elevation with the surrounding pavement that present obstacles to bicyclists in both vehicle and bicycle lanes. Covers and plates can be replaced with less-slippery designs or materials, but to minimize their adverse impacts on bicyclists, it is best to design the roadway to locate covers and plates outside the typical path of bicyclists riding on the roadway. Wherever possible, street and road designs should not place manhole and other utility plates and covers where bicyclists

Traveled Way Design Guidelines 157 typically ride (i.e., within the 6 ft. adjacent to the curb, or between 7 and 12.5 ft. from the curb if parking is permitted). Plain steel plates are slippery and should not be used for permanent installation on the road- way. Temporary installations of construction plates on the roadway should endeavor to avoid using plain steel plates if possible. Manufacturers may imprint waffle-shaped patterns or right- angle undulations on the surfaces of steel or concrete covers and plates to achieve an acceptable level of skid resistance. The maximum vertical deviation within the pattern should be 0.25 in., and, if possible, the placement of the construction plates should provide a clear zone for cyclists to avoid the plates. 4.3.6 Special Roadway Design Concepts This section discusses the design of three specialized types of roadway designs: road diets, main streets and shared streets. 4.3.6.1 Road Diets Road diets involve the reconfiguration of one or more travel lanes to calm traffic and provide space for bicycle lanes, turn lanes, streetscapes, wider sidewalks, on-street parking and other uses. Four-lane to three-lane conversions are the most common road diet, but numerous types exist (e.g., three lanes to two lanes or five lanes to three lanes). FHWA identifies road diets as a proven safety countermeasure (FHWA 2008) and an “Every Day Counts” initiative (FHWA n.d.c). Road diets are one approach to rebalancing a street to better meet the needs of all users. A con- ventional approach to evaluate the feasibility of a road diet is to evaluate the impact on vehicles, not people. Guidance at the national level provides the flexibility to apply engineering judgment to assess the project holistically, incorporating performance measures for all modes, community goals and compatibility with area context. Case studies demonstrate that road diets reduce conflicts at intersections, reduce accidents and have minimal effects on traffic capacity and diversion on roadways traveled by fewer than 20,000 vehicles per day. Three-lane roadways can improve emergency response by allowing emergency vehicles to bypass congestion by using the two-way left-turn lane that is typically pro- vided. They also create opportunities for pedestrian refuges at mid-block and intersection cross- ings, and they eliminate the common “multiple threat” hazards pedestrians experience crossing four-lane roads. Other benefits of three-lane roadways include easier egress from driveways (improved sight distance), smaller curb return radius by increasing the effective radius of the road, improvements for transit (by allowing curbside stops outside of travel lane) and buffering of street tree branches from closely passing trucks. Road diets can improve the flow of traffic and reduce travel speeds, particularly when used in conjunction with roundabouts; however, converting four-lane roads to three lanes with a raised median and on-street parking may result in the traveled way’s inability to meet local fire districts’ minimum clear traveled way requirements. According to the FHWA’s Road Diet Informational Guide, the common four- to three-lane road diet has proven safety benefits with “a 19 to 47 percent reduction in overall crashes” (FHWA 2014f). Adding two-way left-turn lanes reduces the number of potential conflict points, while slower operating speeds typical of this type of road diet reduce the severity of crashes that do occur. In addition to the reduction of speed, pedestrian safety benefits include potentially reduced crossing distances, space for refuge islands, and elimination of multiple threat crashes. Often, road diets also result in a dedicated space for standard or separated bike lanes.

158 Design Guide for Low-Speed Multimodal Roadways For more detailed information, design guidance and case studies regarding road diets, designers can refer to these sources: • Road Diet Desk Reference (FHWA 2015e); • Road Diet Informational Guide (FHWA 2014f); • Evaluation of Lane Reduction “Road Diet” Measures on Crashes (FHWA 2010c); and • Road Diet Handbook: Setting Trends for Livable Streets (Rosales 2006). 4.3.6.2 Main Streets Until the 1960s and 1970s, main streets were the principal roadways in many cities. They were corridors where people could park and walk to find all types of goods and services and often served as the center of local commercial, social and civic activities. Over time, many main street districts were replaced with larger scale, automobile-oriented shopping centers that were located closer to suburban areas and away from downtowns. Today, many of those same communities are attempting to revitalize their main street corridors to create places where residents and visitors can again park, walk, shop, eat and interact with other people. Bicycling also is becoming an important part of many revitalized main street corridors. Although main streets vary from community to community, some typical characteristics can be observed. Main streets may be located in any context zone, but they are most commonly found in small towns and the suburban, urban and urban core areas of larger metropolitan areas. They are usually short, multimodal segments of longer arterial or collector streets and will often exist within a grid or interconnected system of local streets that serve the commercial center of town. These areas usually have short blocks, on-street parking, few driveways and buildings served by alleys. Ideally, land uses on main streets will often consist of compact, mixed-use development, usu- ally with a strong retail and entertainment emphasis on the ground floors of buildings and an equal mix of residential and/or commercial office or services on the upper floors. The buildings typically are low-scale and are oriented to the street with no setback. Parking lots or garages often are located behind or to the side of buildings. Public parking consists of on-street parking and may include strategically located parking lots or garages that support a visitor’s ability to park in one place and walk to multiple destinations. The design of main streets will usually include roadsides that support active uses (e.g., street cafes, social interactions, strolling and window-shopping). By tradition and design, main streets are pedestrian friendly and may feature historic or contemporary examples of urban design, public spaces or public art. Many main streets will have no more than two travel lanes, will pro- vide on-street parking and may contain bicycle lanes. Wider main streets may have additional vehicle lanes and center turn lanes, but the on-street parking and roadside features typically remain the same. From a roadway design perspective, a successful main street will include design features that support a main street environment that focuses on local access, convenient parking, comfortable pedestrian and bicycle accessibility, low-speed traffic and good access to transit (where it exists). Tight corner radii will usually be used at intersections, and curb extensions may be used both at intersection corners and in mid-block crossing locations to shorten roadway crossing distances and increase pedestrian visibility. The physical and visual roadway and urban design elements that draw together both sides of the street should encourage and support frequent, safe crossings of the street. Achieving this typical set of goals for a main street environment will often require trade-offs in the design process, particularly when the main street also serves as a regional route and may be a state or federal-aid highway.

Traveled Way Design Guidelines 159 4.3.6.3 Traveled Way Design Considerations for Main Streets In designing the traveled way of a main street project, there are three typical elements to consider: speed, width and parking. The pedestrian-oriented nature of main streets means the target speed should be kept low (25–30 mph) even when the roadways functional classification and role in the network might suggest higher speeds. Lower speeds create a safer environment for pedestrians, bicyclists and on-street parking maneuvers. Restricted sight distances also may exist in main street corridors, making lower design and operating speeds more appropriate. Given that the width of the traveled way affects users’ perceptions of the speed and volume of street traffic, wider streets often act as a barrier to comfortable pedestrian crossings. Wider streets may be required to implement angle on-street parking, which is sometimes used in main street settings to increase the number of overall parking spaces, but the number of travel lanes should normally not be increased beyond two or three lanes. The use of the traveled way and the number of lanes to be provided should be based on a design development process that considers current and future user needs, context, community objectives, the main street’s role in the larger transportation network and the existence of alternative routes. On-street parking is considered an important design element on main streets. It provides a source of short-term parking for adjacent retail and service uses, buffers pedestrians from traffic, creates friction that slows traffic and produces a higher level of street activity. This Guide pro- vides design guidance for on-street parking in the section labeled “On-Street Parking Principles and Considerations.” Roadside design features in main street projects need an appropriate width to accommodate anticipated levels and types of all current and future activity. The clear pedestrian through- way should typically be wide enough, at a minimum, to allow two people to walk side-by-side. The frontage zone should allow for window-shopping, seating, displays and pedestrian activity at building entrances. The furnishings zone should be able to accommodate many functions, including street trees, planting strips, street furniture, utilities, bicycle racks, transit stops (if needed) and possibly even public art. Community goals may include having street cafes for restaurants located on the main street. If local regulations allow restaurants to have street cafes, then the roadside furnishings zone should allow for that. The edge zone of the roadside will need to accommodate frequent car door openings from on-street parking as well as parking meters and signage. Roadside lighting typically provides both safety illumination of the traveled way and intersections and pedestrian-scaled lighting for the pedestrian walkways. When a planned main street also serves as a state highway, especially in smaller towns where rural highways or principal arterials pass through the community’s historical commercial dis- trict, achieving all the community’s main street design objectives may be difficult. These road- ways may be subject to state department of transportation (state DOT) policies and design standards. In these situations, it is best to work closely with the state DOT at the earliest stages of project planning and design. Although the state agency may recognize the value the community places on their main street and be willing to allow for some flexibility in the design process, it may be difficult to achieve consensus on all the community’s desired main street design elements and criteria. The key to successful planning and implementation of main streets that effectively serve all users and meet community objectives is working collaboratively with the state DOT and all other stakeholders to collectively define a vision, goals and objectives from which a guiding project purpose and need are developed. The objective of this collaboration is for the parties to develop an early con- sensus on the design concept plan and any design features that may require special approvals.

160 Design Guide for Low-Speed Multimodal Roadways 4.3.6.4 Shared and Slow Streets Shared streets (sometimes called flush streets or woonerfs) prioritize pedestrian and bicycle movement by slowing vehicular speeds and communicating clearly through design features that motorists must yield to all other users. Shared streets use various design elements to blur the boundary between pedestrian and motorized vehicle space. The right-of-way design should create conditions in which pedestrians and bicyclists can walk or ride on the street and cross at any location, as opposed to at designated locations. This encourages cautious behavior on the part of all users, which in turn reinforces slower speeds and comfortable walking and bicycling conditions. By slowing the travel speed of all modes, shared streets encourage social interaction and lingering. They support a variety of adjacent land uses including commercial and retail, enter- tainment venues, restaurants, offices and residences, while still accommodating commercial loading and transit operations. Shared streets have also been shown to increase economic vitality and vibrancy. Little design guidance for shared streets exists in the United States, but European countries have successfully applied the concept for decades. Additional research is needed to identify design elements and criteria for this unique type of shared-space design for motorized and non- motorized users. Until that research occurs, a designer exploring the use of a shared street should consider the following: • The design, operations, and maintenance of shared streets should always encourage lower vehicle speeds, reducing the likelihood and severity of crashes. • Shared streets are considered self-enforcing roads, designed and operated primarily for pedes- trian traffic. Designs for shared streets should lead to slow vehicular speeds. The maximum design speed should not exceed 20 mph, and the preferred design speed is between 10 mph and 15 mph. • Design details should communicate clearly that the shared street is a multimodal environment where pedestrians are given priority and motorists are guests on the street who must proceed slowly and cautiously. The lack of predictability of all users heightens awareness, thereby cre- ating lower vehicle speeds and reducing conflicts. • The shared street should support adjacent land uses and support economic and livability goals. • Typically, shared streets do not use vertical curbs; that is, the entire street surface is flush, with minimal separation between sidewalks and the traveled way. Although sloping curbs discourage motorized vehicle encroachment, they have limited ability to prevent a vehicle from driving onto the sidewalk. Designers can choose from among several techniques to con- trol drainage and help delineate the roadway edge. The Urban Street Design Guide (NACTO 2013) includes information on such techniques, with sections that cover both residential and commercial shared streets. • Shared streets should carry no more than 100 vehicles during the peak hour for pedestrians to feel comfortable sharing the road with motorists (NCUTLO 2000). If volumes exceed this threshold, designers can consider restricting access for specific vehicle types to reduce vol- umes. If vehicular volumes are too high, pedestrians will avoid the middle part of the street. Depending on the role of the shared street in the transportation network, personal vehicles may be directed to alternative routes while taxis and freight and transit vehicles are allowed. Emergency access should be maintained on shared streets. • At intersections, designers should consider traditional marked crosswalks and detectable warning surfaces in order to alert pedestrians of potential vehicular conflicts. Gateway fea- tures such as signs, raised crossings, raised intersections or curb extensions can be used to alert drivers entering the shared street of the intended use of the space and the appropriate speed.

Traveled Way Design Guidelines 161 • Shared street design should use creative means to delineate space for pedestrians with dis- abilities. This can be done by providing a frontage zone along buildings where a traditional sidewalk is located. The frontage zone can be delineated using a distinct paving treatment, drainage infrastructure, trees, street furniture, art, or parking. Paving textures in the frontage zone should be smooth and vibration free, with a minimum of 5 ft. of clear space. Some shared street concepts offer a great deal of flexibility in how the space is designed and used. Without vertical curbs, the street can be closed to offer space for events, or to more com- fortably provide outdoor seating space for cafes and restaurants. Designers can choose from among several options for drainage design and the delineation of space. Through the thoughtful use of urban design principles, these streets can enhance the sense of place and emphasize the pedestrian and bicycle priority of the street. A multipurpose shared street allows for different uses of the space on different days of the week, times of day or seasons, extending the public space at times of celebration, special events or festivals. Sidewalks, parking and vehicle travel lanes can be made available at various times. Movable planters, metal barricades or signs can regulate the use of the space on a temporary or regularly scheduled basis. Slow streets are designed to enhance safety and improve pedestrian and bicycle comfort by achieving low motorist speeds and minimizing speed differentials between motorists and bicy- clists to prioritize bicycle travel. The lower motorist speeds also promote increased yielding to pedestrians crossing the street. Slow streets also may be called bicycle boulevards, quiet-ways or neighborhood greenways. Typically, they are designed for a maximum speed of 20 mph to 25 mph, with the majority of motorists going slower. Slow streets may require the use of traffic- calming measures (e.g., curb extensions, speed tables, gateway treatments, neighborhood traffic circles, textured pavement and chicanes). Design speeds for slow streets typically are established at or below 20 mph. This design speed reduces the speed differential between roadway users, thus providing a higher level of comfort and safety. Good candidates for slow streets include neighborhood residential streets, school walking routes, bicycle routes and shopping streets with high levels of pedestrian activity. Slow streets also are appropriate on streets running adjacent to (or through) parks and public plazas. Lower-speed streets with comfortable pedestrian crossings enhance adjacent public spaces, whereas higher-speed streets and streets with higher vehicle volumes and difficult cross- ings can detract from them. Slow streets often have a narrowed traveled way (less than 18 ft. in total width) that, in some cases, requires oncoming motorized vehicle traffic to yield before proceeding. Alleys demonstrate this strategy for slow street design. Some slow streets will include bollards, plant- ers and other vertical elements in close proximity to the traveled way, encouraging caution as drivers move along the street. Used in conjunction with other features that reduce speed, the removal of traffic controls at intersections is another strategy to produce cautious behavior in motorists (and therefore slower speeds). Various other traffic-calming measures can be used to slow the speeds of motorized vehicles, provide comfortable places for vulnerable road users and encourage motorist yielding. Slow streets are inherently beneficial to pedestrians of all abilities because they produce slower and more cautious behavior on the part of motorists. Design elements of slow streets must meet current accessibility standards. For example, all surfaces within pedestrian areas must be designed and maintained to be stable, firm and slip resistant. Bicycle boulevards (or bicycle priority streets) are a form of slow street. They have lower motor- ized vehicle speeds that are designed to allow bicyclists to travel comfortably in a low-stress envi- ronment. Bicycle boulevards typically give priority to bicycle use and discourage through traffic by motorized vehicles. They are designed to minimize the number of stops that a bicyclist must

162 Design Guide for Low-Speed Multimodal Roadways make along the route. Designers have a great deal of flexibility when designing bicycle boule- vards. They are easier to implement in areas with a grid street network because drivers have the option to choose an alternate route. Bicycle boulevards typically are designated using special signs or pavement markings. More information on bicycle boulevard design can be found in the AASHTO Bicycle Guide (AASHTO 2014b) and the NACTO Urban Bikeway Design Guide (NACTO 2014). 4.3.7 One-Way Streets One-way streets simplify crossings for pedestrians, who must look for traffic in only one direction. Although studies have shown that conversion of two-way streets to one-way streets generally reduces pedestrian crashes, one-way streets tend to have higher speeds, which creates greater chances for fatalities and increased injury severity among the crashes that still occur. If a street is designed as a one-way facility, it should be evaluated to see if additional design elements should be added to reduce pedestrian and bicycle crash potential, especially if the traveled way or lanes are overly wide. Other considerations for one-way streets include the following: • As a system, one-way streets can increase travel distances for motorists and create some confu- sion, particularly for non-local residents. • One-way streets operate best in “street pairs” that are separated by distances ranging from one block to no more than one-quarter mile. Conversion costs can be quite high to build cross- overs where the one-way streets convert back to two-way streets, and to rebuild traffic signals and revise striping, signing and parking meters. • One-way streets work best in downtown areas or other very heavily congested areas. One-way streets can offer improved signal timing and accommodate odd-spaced signals, but signal timing for arterials that cross a one-way street pair is difficult. 4.3.7.1 One-Way/Two-Way Street Conversions Converting a one-way street to a two-way street is an increasingly popular way to manage traffic patterns, simplify access and change the character of a corridor or neighborhood from a “pass-through” facility to a “destination” for motorists. Converting a one-way street to a two- way street also can help reduce motorized vehicle speeds and VMT (because there is less need to circumnavigate multiple streets to reach destinations in dense mixture of land uses) and can provide improved conditions and access for bicyclists and pedestrians. In terms of pedestrian safety, one-way and two-way streets both have benefits, so the decision to convert a two-way street to one-way (or vice versa) is context sensitive. Studies have shown that converting two-way streets to one-way streets generally results in fewer crashes involving pedestrians because there are fewer turning movements. On the other hand, one-way streets tend to encourage higher motorized vehicle speeds, and intersections involving one-way streets may be more confusing for some roadway users, especially non-local residents and child pedes- trians. In addition, left-turning drivers of motorized vehicles may be less cautious when turning from one-way streets and therefore less likely to see crossing pedestrians due to poorer sight lines. In addition, many one-way streets are multilane, which creates a multiple threat condition for pedestrians crossing the road. Converting a multilane one-way street to a two-lane two-way eliminates this safety issue. Two-way streets may reduce vehicle speeds due to increased turning movements and to increased perceived friction along the roadway. If a two-way street will be converted to one-way, the street should be evaluated to see if additional changes should be made. Potential changes can include lane diets, road diets, curb

Traveled Way Design Guidelines 163 extensions, turning radius reductions and use of signal timing that discourages high vehicle speeds. In addition, traffic circulation in the surrounding area must be carefully considered before converting streets to one-way. 4.3.8 Bridges Bridge crossings are significant investments and therefore often occur infrequently. Thus, it is critical that they accommodate pedestrians and bicyclists (FHWA 2016a). A bridge without walking and bicycling access can result in a lengthy detour that makes the entire trip impractical for pedestrians or bicyclists. Safe pedestrian access often can be included at the same time as bicycle accommodations and should be provided on bridges whenever possible, regardless of funding source. Bridges also should accommodate the bicyclists and pedestrians who may travel under them so that they do not create a barrier. Providing pedestrian and bicycle accommodation during initial construc- tion generally costs less than retrofitting. Federal policy often requires safe accommodation of pedestrians and bicyclists, and design guidance provides adequate flexibility on how to accommodate these users. Federal law states: “In any case where a highway bridge deck being replaced or rehabilitated with [f]ederal financial participation is located on a highway on which bicycles are permitted to operate at each end of such bridge, and the Secretary determines that the safe accommodation of bicycles can be provided at reasonable cost as part of such replacement or rehabilitation, then such bridge shall be so replaced or rehabilitated as to provide such safe accommodations” (23 USC §217(e)). Design guidelines for the accommodation of pedestrians and bicyclists on bridges are as follows. • Bridge designs should provide adequate width for current and anticipated pedestrian and bicycle use. Sufficient clear width and usable width should be provided. Clear width is a trav- eled way clear of obstructions such as railings, light poles, or signs (TRB 2016b). The usable width recognizes that pedestrians and bicyclists will not travel at the very edge of a traveled way or immediately against a railing, but need at least 1.5 ft. of shy distance from vertical objects (TRB 2016b). • Both sides of the bridge should accommodate travel for pedestrians and bicyclists. Where bidirectional facilities can be provided, they may reduce conflicts if they limit the number of roadway crossings. Similarly, facilities should be considered for current and anticipated pedes- trians and bicyclists to access the bridge and travel under the bridge. Designers should consider whether to combine pedestrians and bicyclists on a shared-use path or to separate them. • The desirable clear width for a sidewalk on a bridge is 8 ft. (AASHTO 2004a). • The minimum width for one-way bicycle travel is 4 ft. (AASHTO 2014b). • Well-designed bridge railings can contribute to a safer and more positive experience on bridges for people who walk or bicycle. Railing designs should consider a 1.5 ft. shy distance when determining usable width, and establish a height that keeps pedestrians and bicyclists safe. Given that bicyclists have a higher center of gravity, railings should be a minimum of 42 in. high. Where bicyclists’ handlebars or pedals may come into contact with the railing, smooth and wide rub-rails should be installed; furthermore, on bridges that accommodate both vehicular and pedestrian/bicycle travel, only crash-tested railings should be installed (AASHTO 2014b). • Connections from bicycle and pedestrian facilities on a bridge to related roadway features such as shared-use paths, sidewalks, or other infrastructure are a key component of connected

164 Design Guide for Low-Speed Multimodal Roadways networks. Any connection for use by pedestrians must be accessible to people with disabilities. The design should consider the desired route of pedestrians and bicyclists. A common practice is to install switchbacks, which may be the only option in a confined space; however, designs without switchbacks often create a more direct route for the majority of users. Grades must meet accessibility standards and ramps may be required. Where switchbacks are required, the ramp turns should provide generous width to better accommodate turns by bicyclists. Where bicyclists are permitted to use the connection, the ideal design should not require bicyclists to dismount (AASHTO 2014b). • Stairs may provide a more direct connection for pedestrians and bicyclists, but the accessible route provided for persons with disabilities may not be significantly longer. The stairs can be constructed to accommodate bicycles by including a bike channel (a flat ramp parallel to the stairs on which to roll the bicycle). Handrail designs must meet current accessibility standards. Specifically, handrails provided on stairs with a bike channel need to project out from the wall with at least the minimum clearance required by ADA accessibility guidelines, and the handrail must be aligned above the stair nosing where people are walking. Pedestri- ans must be able to easily reach the railing, and the bike channel must not present a tripping hazard for pedestrians with visual disabilities. • Pedestrians and bicyclists may find it difficult to locate bridge access points from the connect- ing street grid. In some cases, access points for people on foot, in wheelchairs or on bicycles are different and more difficult to locate than vehicle access points. Wayfinding signs and markings should direct users of all travel modes to bridge access points. • Including facilities for pedestrians and bicyclists on bridges increases access, but the bridge design itself may reduce future connectivity. Waterways, railroads, and highways may be desir- able corridors for shared-use paths. Whether or not a current plan exists to build a path along one of these corridors, bridge design should consider future accommodations for pedestrians and bicyclists under the bridge. 4.3.9 Railroad-Highway Grade Crossings Many types of railways exist in urban and suburban areas. Although the heavy rail that has traditionally served freight and passenger needs for more than 100 years is most prevalent, new types of passenger rail systems are being increasingly employed in urban areas. These new sys- tems include LRT, modern streetcars and trolleys. In urban areas where pedestrian and bicycle volumes are higher, many—but not all—crossings of higher-speed and higher-volume train tracks with major roadways and streets are grade separated. Newer types of passenger rail systems often operate within rights-of-way that are adjacent to the roadway. The traditional crossings of heavy rail track and roadways present one type of challenge to the designer, whereas the newer, integrated light rail systems offer quite a different challenge. In urban and suburban areas, pedestrians, bicycles and transit vehicles often travel in great numbers and proximity to both heavy and light rail lines, and they must be considered in the design of railroad-highway (railroad-roadway) crossings just as motorized vehicles are. Railroad-roadway grade crossings are designed and controlled to accommodate the vehicles and other users that travel across them. The vast majority of the vehicles consist of automo- biles, buses and all types of trucks. Generally speaking, the improvements made to a crossing with these users in mind will be adequate for special users such as trucks carrying hazardous materials, long-length trucks, school buses and motorcycles. Designing these crossings for safe and convenient use by bicycles and pedestrians presents a different challenge. These users have unique characteristics and special needs that should be carefully considered in the design of railroad-roadway crossings.

Traveled Way Design Guidelines 165 The type, size and design of pedestrian and bicycle facilities along a street or roadway should be carried across the rail tracks such that there is no reduction of service accommodation to those users. In addition, the warning and protection systems used in passive and active cross- ing systems for vehicles should be designed and operated to serve the non-vehicle users as well. 4.3.9.1 Buses Because buses carry many passengers and have performance characteristics similar to large trucks, these vehicles need special consideration. Many of the measures suggested for trucks with hazardous material apply to buses. Railroad-roadway grade crossings should be taken into consideration when planning school bus and transit routes, not least because school bus routes may contribute to added pedestrian or bicycle travel by children, and transit routes may attract similar pedestrian or bicycle travel by commuters of all ages and abilities. Potentially hazardous crossings (e.g., crossings with limited sight distance or horizontal or vertical alignment issues) should be avoided if possible. Crossings along school bus and transit routes should be evaluated by the appropriate roadway and railroad personnel to identify potential dangers and the need for improvements. 4.3.9.2 Motorcycles and Bicycles Although motorcycles and bicycles typically travel at different speeds, these two-wheeled vehicles can experience similar problems at railroad-roadway crossings. Depending on the angle and type of crossing, a bicyclist may lose control if the wheel of the bicycle becomes trapped in a flange way. Particular attention should be given where streetcar tracks bend or turn, where light rail tracks cross a street, or where bicycle lanes or bicycle turning movements cross tracks. Bicycling adjacent to tracks also poses dangers, which become particularly pronounced when a bicyclist must be prepared to swerve to avoid unforeseen obstacles such as opening vehicle doors. Bicycle-friendly track crossings are applicable wherever • Streetcar or light rail tracks turn across a bikeway (including any bike lane, bike boulevard or cycle track), • A bikeway turns across tracks, and • An intersection accommodates bicycle turns (especially where two bike lanes intersect). Bicycle-friendly trackway design is applicable to all mixed-traffic streetcar/trolley and LRT running ways. The Transit Street Design Guide (NACTO 2016) provides several design recommendations for keeping bicyclists safe at rail crossings, including: • Where bicycle paths of travel cross a street-surface rail track, bicyclists must be directed to cross tracks at a high angle. While 90-degree crossings are preferred, 60 degrees is the minimum design angle for bikeways to cross in-street rails. Bicyclists must be able to cross tracks fully upright and not leaning, with perpen- dicular or high-angle approaches established in advance of tracks to allow riders to right themselves. • A bike sneak [a design concept developed to address rail turns across bikeways] is a short section of bicycle lane, protected bicycle lane, or raised cycle track that is bent out (bent toward the sidewalk) to direct bicyclists at a safe angle across turning tracks. Provide bicycle lane markings to direct bicyclists to the right, establishing sufficient space for a safe crossing of rails. Provide intersection markings, at or near a 90-degree angle to the curving track that return bicyclists to the bicycle lane on the opposite side of the intersection without entering the motor vehicle lane. The bike sneak can be marked, raised, channelized, or otherwise protected using a variety of means of separation, depending on the volume of bicyclists and the role of the street in the bicycle network. • Crossing tracks at an angle less than 45 degrees should be discouraged, both on streets with and without a bicycle facility. • Warning signage or markings should be used ahead of an intersection or other rail crossing where the natural travel path of a bicyclist, generally parallel to the lane line or curb-line, would cross the rail at a low angle.

166 Design Guide for Low-Speed Multimodal Roadways • Two-stage turn queue boxes direct bicyclists to cross rails at a safe angle when turning left across tracks, or turning right across tracks from a left-side bikeway. (Refer to the Urban Bikeway Design Guide [NACTO 2014] for guidance on two-stage turn queue box design). • Bicycle lanes should include a buffer at least 3 ft. wide to account for these instances, and prohibitions of dangerous misuse of the bike lane, such as double-parking, must be strictly enforced. Where possible, physically separating bicycle lanes from streetcar lanes is preferred. In addition to cycle tracks, placing rails on raised beds or transitway design treatments, such as rails in raised beds, or vertical separation, prevent bicycles from entering tracks. Vertical separation may be especially desirable in tight spaces. If curbs are greater than 2 in., roll curbs or mountable curbs should be considered. 4.3.9.3 Pedestrians The safety of pedestrians crossing railroads is the most difficult to control because of the rela- tive ease with which pedestrians can go under or around lowered gates. Pedestrians typically seek the shortest path, and they may not always cross the tracks at a designated roadway or pedestrian crossing. Given the variety of factors that may contribute to pedestrian hazards, detailed studies often are necessary to determine the most effective measures to provide for pedestrian access and safety at specific locations. A variety of preventive design measures can be employed as discussed in this section. ADA guidelines for accessible design also provide many geometric features per- taining to pedestrian facilities that address minimum widths and clearances, accessible routes and pedestrian pathways, curb ramps and protruding objects (U.S. Access Board 2002, U.S. Access Board 2005, U.S. Access Board 2011). Although collisions between light rail vehicles (LRV) and pedestrians occur less often than collisions between LRVs and motorized vehicles, they are more severe. Furthermore, pedestrians often are not completely alert to their surroundings at all times, and LRVs operating in a street environment are nearly silent. For these reasons, appropriate pedestrian crossing-control sys- tems are critical for LRT safety. The following pedestrian crossing treatments may be warranted at rail-pedestrian crossings: • Flashing light signal. At non-gated pedestrian-only crossings of semi-exclusive LRT rights-of-way, a flashing light signal assembly with an audible component serves as the primary warning device where LRT operates two ways on one track or on a double track. When the red lenses of the light signal are flashing alternately and the audible device of the signal assembly is active, the pedestrian is required to remain clear of the tracks (Uniform Vehicle Code, Section 11-513, NCUTLO 2000). At gated motorized-vehicle LRT crossings without pedestrian gates, TCRP Report 17: Integra- tion of Light Rail Transit into City Streets recommends that the flashing light signal assembly be used in the two quadrants without vehicle automatic gates (Korve et al. 1996). According to this recommendation, these signal devices should be installed adjacent to the pedestrian crossing fac- ing out from the tracks. The signal assembly includes a standard crossbuck sign and, where there is more than one track, an auxiliary inverted T-shaped sign indicating the number of tracks. • “Second Train Coming” sign. An LRV-activated, internally illuminated matrix sign display- ing the pedestrian crossing configuration with one or more LRVs passing may be used to alert pedestrians to the direction from which one or multiple LRVs are approaching the cross- ing, especially at locations where pedestrian traffic is heavy (such as LRT stations). Alter- natively, an LRV-activated, internally illuminated flashing sign with the legend, “SECOND TRAIN—LOOK LEFT/RIGHT,” may be used to alert the pedestrian that a second LRV is approaching the crossing from a direction that the pedestrian might not be expecting. This sign warns pedestrians that, although one LRV has passed through the crossing, a second LRV is approaching, and that other warning devices (such as a flashing light signal assembly and bell) will remain active until the second LRV has cleared the crossing. TCRP Report 17 recommends that the “Second Train Coming” sign be placed on the far side of the crossing (and on the near side as well if necessary for pedestrian visibility), especially

Traveled Way Design Guidelines 167 when the crossing is located near an LRT station, track junction, and/or multiple track align- ment involving more than two tracks (Korve et al. 1996). When this sign is activated, only one direction—left or right—is illuminated at any time. Furthermore, only one arrow—to the left of “LOOK” or the right of “RIGHT”—is illuminated at any time (the one that points in the direction of the second approaching LRV). Therefore, if two LRVs are very closely spaced so that they will pass through the pedestrian crossing almost simultaneously, TCRP Report 17 rec- ommends that this sign not be activated. In this situation, pedestrians will have no opportu- nity to cross between the successive LRVs, and pedestrians will need to look in both directions. • Dynamic envelope markings. TCRP Report 17 recommends that the LRV’s dynamic enve- lope be delineated at pedestrian crossings in semi-exclusive rights-of-way and along entire semi-exclusive and nonexclusive corridors. According to this recommendation, contrasting pavement texture should be used to identify an LRV’s dynamic envelope through a pedestrian crossing. A solid 4-in.-wide line may be used as an alternative. Tactile warning strips approved by the ADA can be considered a contrasting pavement texture, and their requirement may supersede the use of painted striping or other contrasting pavement texture. TCRP Report 17 recommends that in an LRT/pedestrian mall, the dynamic envelope be delineated in its entirety (Korve et al. 1996). In addition to pedestrian signals (including flashing light signals), warning signs, and dynamic envelope markings, several pedestrian barrier systems have proven effective in reducing colli- sions between LRVs and pedestrians. These barriers, and the transit systems or railroads where they have been successfully installed, include: • Curbside pedestrian barriers. Between intersections in shared rights-of-way, TCRP Report 17 recommends that curbside barriers (landscaping, bedstead barriers, fences and/or bollards and chains) be provided alongside-aligned LRT operations where LRVs operate two ways on a one- way street (contra-flow operations). They may also be provided for one-way side-aligned LRT operations for normal flow alignments (Korve et al. 1996). • Pedestrian automatic gates. Pedestrian automatic gates are the same as standard automatic crossing gates except that the gate arms are shorter. When they are activated by an approach- ing LRV, the automatic gates are used to physically prevent pedestrians from crossing the LRT tracks. TCRP Report 17 recommends that this type of gate be used in areas where pedestrian risk of a collision with an LRV is medium to high (for example, whenever LRV stopping sight distance is inadequate). The preferred method is to provide pedestrian automatic gates in all four quadrants. TCRP Report 17 provides additional detail on the design and operation of this method (Korve et al. 1996). • Swing gates. Sometimes used in conjunction with flashing lights and bells, swing gates alert pedestrians to the LRT tracks that are to be crossed and force them to pause, thus deterring them from running freely across the tracks without unduly restricting their exit from the LRT right-of-way. The swing gate requires pedestrians to pull the gate to enter the crossing and push the gate to exit the protected track area; therefore, a pedestrian cannot physically cross the track area without pulling and opening the gate. TCRP Report 17 recommends that the gates be designed to return to the closed position after the pedestrian has passed (Korve et al. 1996). Swing gates may be used at pedestrian-only crossings, on sidewalks and near stations (espe- cially if the station is a transfer point with moderate pedestrian volumes) where pedestrian risk of a collision with an LRV is medium to high (e.g., where conditions include moderate stopping sight distance, moderate pedestrian volume, and so forth). These gates may be used at pedestrian crossings of either single-track (one- or two-way LRT operations) or double- track alignments. TCRP Report 17 recommends that swing gates be supplemented with proper signing mounted on or near the gates. Such signing can include the “LIGHT RAIL TRANSIT CROSSING/ LOOK BOTH WAYS” sign (where LRVs operate two ways), LRV-activated, internally illuminated

168 Design Guide for Low-Speed Multimodal Roadways warning signs, and/or flashing light signal assemblies. Where LRVs operate using a single- track, two-way alignment, TCRP Report 17 recommends that an LRV-activated, internally illuminated matrix sign or active, internally illuminated sign with the legend, “TRAIN— LOOK LEFT/RIGHT” be installed to supplement swing gates (Korve et al. 1996). • Bedstead barriers. The bedstead concept may be used in tight urban spaces that lack a fenced- in right-of-way, such as a pedestrian grade crossing at a street intersection. The barricades are placed in an offset (maze-like) manner that requires pedestrians moving across the LRT tracks to navigate the passageway through the barriers. TCRP Report 17 recommends that bedstead barriers be designed and installed to turn pedestrians toward the approaching LRV before they cross each track, forcing them to look in the direction of any oncoming LRV. According to this recommendation, the barriers should also be used to delineate the pedestrian queuing area on both sides of the track area (Korve et al. 1996). Bollards and chains accomplish the same effect as bedstead barriers. Bedstead barriers may be used for crossings where pedestri- ans are likely to run unimpeded across the tracks, such as stations or transfer points, particu- larly where pedestrian risk of a collision with an LRV is low to medium (e.g., where stopping sight distance is excellent to moderate, where double tracking is used and where pedestrian volume is low). TCRP Report 17 recommends that the barriers be used in conjunction with flashing lights, pedestrian signals and appropriate signing. Bedstead barriers also may be used in conjunction with automatic gates in high-risk areas (Korve et al. 1996). TCRP Report 17 recommends that bedstead barriers not be used when LRVs operate in both directions on a single track because pedestrians may look the wrong way in some instances (Korve et al. 1996). Pedestrians also may look in the wrong direction during LRV reverse-running situations; however, reverse running is performed at lower speeds, so this should not be a deterrent to this channeling approach. • Z-crossing channelization. The Z-crossing controls movements of pedestrians approaching LRT tracks. Its design and installation turn pedestrians toward the approaching LRV before they cross each track, forcing them to look in the direction of oncoming LRVs. Z-crossing channelization may be used at crossings where pedestrians are likely to run unimpeded across the tracks, such as isolated, mid-block, pedestrian-only crossings, particularly where pedes- trian risk of a collision with an LRV is low to medium (e.g., locations with excellent stopping sight distance, double tracking, low pedestrian volume and so forth.). Z-crossings used with pedestrian signals create a safer environment for pedestrians than Z-crossings used alone. This channelization approach also may be used in conjunction with automatic gates in high-risk areas. TCRP Report 17 recommends that the Z-crossings not be used when LRVs operate in both directions on a single track because some pedestrians may look the wrong way (Korve et al. 1996). Pedestrians also may look the wrong way during LRV reverse-running situations; however, reverse running is performed at lower speeds, so the risk to pedestrians is lower and should not be a deterrent to a Z-crossing channeling approach. • Combined pedestrian treatments. The pedestrian crossing/barrier systems described in this section may be used in combination, depending on the risk of a pedestrian collision with an LRV at the crossing. Pedestrian safety and queuing areas should always be provided and clearly marked. The key design references for design of pedestrian, bicycle, transit and vehicle crossings of rail facilities are: • Guide for Geometric Design of Transit Facilities on Highways and Streets (AASHTO 2014a); • TCRP Report 17: Integration of Light Rail Transit into City Streets (Korve et al. 1996); • Transit Street Design Guide (NACTO 2016); • TCRP Report 183: A Guidebook on Transit-Supportive Roadway Strategies (Ryus et al. 2016); • TCRP Report 175: Guidebook on Pedestrian Crossings of Public Transit Rail Services (Fitzpatrick et al. 2015b);

Traveled Way Design Guidelines 169 • TCRP Report 117: Design, Operation, and Safety of At-Grade Crossings of Exclusive Busways (Eccles and Levinson 2007); and • TCRP Report 112/NCHRP Report 562: Improving Pedestrian Safety at Unsignalized Crossings (Fitzpatrick et al. 2006). 4.3.10 Fire and EMS Considerations Roadway designs must consider the needs of emergency responders driving fire trucks and emergency medical service (EMS) vehicles. Emergency responders seek to minimize response times to save lives; in some situations, seconds can make the difference between life and death. Many design treatments that can make roads and streets safer for pedestrians and bicycles can reduce the traveled way that emergency responders rely on, and some may reduce the operating speeds of motorized vehicles, including emergency response vehicles. However, pedestrian and bicycle deaths and injuries significantly decrease as motorized vehicle speeds decrease. Where vehicle speeds are a concern, applying speed management designs will improve pedestrian and bicycle safety and access, reduce the frequency and severity of vehicle crashes, and add parking lanes. Speed management designs also may provide opportunities to introduce green infrastruc- ture elements to reduce stormwater runoff. Urban roadways are the primary routes for emergency response vehicles, including police cars, fire trucks and ambulances. Common roadway designs that encourage speed and capac- ity can lead to fatality- and injury-producing crashes involving emergency vehicles and other motorists, bicyclists or pedestrians. Although the emergency responder bears the primary responsibility for both response time and reasonable access to incidents within the commu- nity, a balance among the competing interests of access, speed, capacity and safety must be established for the appropriate design of context-sensitive roadways. Stakeholders can work together to find emergency response strategies that create safe and comfortable places for the non-motorist. Emergency vehicle access and operations always should be considered in roadway design. Local operational conditions will vary from community to community, and emergency response strategies are specific to the locale. Consequently, a designer should collaborate with emergency responders to learn their specific needs, response strategies and tactics used on similar streets. Although all emergency responders are concerned with the speed with which they can respond to a call, firefighting equipment generally involves the largest vehicles to be considered. Obtain- ing answers to the following questions will help designers understand and clarify issues when working with fire departments in the design process: • What types of fire apparatus are used in responding to various emergencies that might occur on or adjacent to the roadway? • Does the vehicle type change depending on the location of the emergency (e.g., a home on a suburban residential street versus a high-rise building in an urban core)? • In urban areas with tall buildings, how does the department deploy its ladders? How much width is needed between the vehicle and building? How much clear space is needed adjacent to the building? Are gaps in sidewalk furnishings required to access buildings? Do the emergency responders need to fully extend their vehicle’s stabilizers? • What characteristics of the apparatus affect roadway design (e.g., what specifications are needed to accommodate the wheel turning path, the overhang turning path and the appara- tus width)? • In a block of attached multistory buildings, does the number of stories cause a difference in firefighting tactics that would affect the design of the adjacent street?

170 Design Guide for Low-Speed Multimodal Roadways Fire codes may provide additional guidance on emergency access requirements such as mini- mum traveled way clear widths and minimum space to deploy certain types of equipment, such as ladders, to reach high buildings. The following approaches should be considered in designing traveled ways to accommodate emergency vehicles: • In urban areas with tall buildings, consider placing no-parking zones or staging areas at mid- block to accommodate large ladder trucks. The length and frequency of these zones should be determined with the emergency responder but should be no longer than 50 ft. in order to minimize the loss of on-street parking. • For the design of curb return radii, use emergency vehicles as a design vehicle only if the vehicle would use the roadway frequently (e.g., as a primary travel route from the fire station to locations in its service area). Otherwise, emergency vehicles are generally able to encroach into opposing travel lanes. Consider using demonstration projects in the field to determine or confirm the optimal geometry for firefighting vehicles. • On streets with medians or other access management features, emergency response time may be reduced by implementing mountable median curbs to allow emergency vehicles to cross. • On roadways that have one lane in each direction and medians, consider using bicycle lanes that are at least 6 ft. wide. With bicycle lanes in place, motorists have the opportunity to safely pull into the bike lane if necessary to allow emergency vehicles to pass. • In high-rise building environments, roadway design may be constrained by the required dis- tance between the building face and the centerline of ladder trucks. In many cases, the required distance is 35 ft.; this dimension varies, however, and it should be examined with fire officials. Firefighters are trained in many techniques that address context-sensitive streets, mainly because narrow, low-speed, pedestrian-oriented streets exist in many towns and cities. Many fire departments have experience with historic networks of narrow streets. Their experience can provide a helpful basis for problem solving when new or updated road design projects take place within existing networks of relatively narrow streets. The designer should be particularly sensi- tive to the local fire official’s experience and the operational needs on urban roadways. 4.3.11 Roadway Pavement Markings and Markers In the design of facilities serving bicycle traffic, care should be taken to specify pavement strip- ing that is durable yet skid resistant. Raised pavement markings and markers can deflect a bicycle wheel, causing a bicyclist to lose control. All longitudinal and transverse pavement markings accessible to bicycles within the trav- eled way should be capable of maintaining an appropriate skid resistance under rainy or wet conditions to maximize safety for bicyclists. Normally, thermoplastic markings can meet these requirements. Skid resistance is optimized when the composition of the pavement marking has been modified with crushed glass to increase the coefficient of friction and the maximum thick- ness of the marking is no larger than 100 mils (2.5 mm). Raised reflective markers and non-reflective ceramic pavement markers both can present a vertical obstruction to bicyclists. When reflective markers are necessary as a fog line or placed adjacent to the edge line, they should have a beveled front edge and be placed to the left of the line, outside the shoulder area. Where raised pavement markers cross a bike lane or extensions thereof through intersections, a 4-ft. gap should be provided as a clear zone for bicyclists. At gore areas and other locations with channelizing lines, if raised reflective markers are used to supplement the striping, extra lane width should be provided in the areas where bicycles travel to provide bicyclists with more latitude to avoid the markers. If retroreflective pavement markers are needed for motorists, they should be installed on the motorists’ side of any bicycle lane stripe.