Pedestrian Safety Relative to Traffic-Speed Management (2019)

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.

44 3.1 Introduction The preceding chapter discussed research findings relevant to pedestrian safety and traffic speed, including the strong, positive correlation between higher vehicle speeds and pedestrian injury severity and a toolbox of potential countermeasures (Rosén and Sander 2009; Tefft 2013). However, research alone is insufficient for helping jurisdictions understand and address the issue of pedestrian safety and traffic speed—the work has not been underway for long enough in enough places to create standard strategies for doing so, nor to provide robust documentation about which strategies are the most useful for various contexts. The profession also lacks a widespread understanding of how best to measure the success of these efforts, given that pedestrian crashes are relatively rare in any one place and are widely understood to be underreported (Stutts and Hunter 1998, Lopez et al. 1999). In light of these gaps, this NCHRP synthesis project seeks to understand the state of the practice among agencies actively working to improve pedestrian safety via managing traffic speed. This chapter describes the results of the in-depth interviews with transportation profes- sionals (the methodology is described additionally in Chapter 1) about their efforts to address pedestrian safety via traffic-speed management. The interviewees were selected from an initial screening survey conducted to understand the breadth of such efforts and to ensure that non– Vision Zero cities also working on this topic could be included. The survey asked respondents about their policies and plans, efforts, and metrics used to manage speed for pedestrian safety. The 33 people who responded reported a wide variety of efforts, including Vision Zero poli- cies and action plans, engineering improvements, and collaborative work with community partners. Eleven jurisdictions actively working in this area were selected for in-depth interviews to learn more about their efforts with the aim of filling this knowledge gap and providing information to other jurisdictions interested in implementing speed-reduction efforts to increase pedestrian safety. Table 3-1 and Figure 3-1 give an overview of the interviewed jurisdictions and whether each is a Vision Zero city. The interviews included questions about the following (the full interview script is included as Appendix C): • Overview of efforts to improve pedestrian safety through addressing traffic speed, • The role of community and professional partners, • Challenges to implementing efforts, • Resistance to efforts and how that has been addressed, • Relationships with state DOTs, and • Lessons learned. C H A P T E R 3 State of the Practice

State of the Practice 45 Where practical, conclusions drawn from the aggregate interviews are presented; additionally, selected agencies are highlighted as case examples in sidebars for specific ideas and approaches. Appendix D provides summaries of the interviews conducted for this synthesis. The following sections delve into the major themes that emerged through the interviews, drawing broader conclusions when possible. First, events that catalyzed speed-reduction efforts are reviewed, followed by a summary of speed-reduction efforts. The next section dis- cusses findings about partnerships and community support, as well as resistance to attempted speed-reduction efforts. Subsequently, metrics that jurisdictions are using or have used to measure their programs’ effectiveness will be described. Lastly, broader lessons learned will be addressed. 3.2 Catalyzing Events Speed-reduction efforts and campaigns can be initiated through several avenues. For the juris- dictions interviewed, the catalyzing events seemed to fall into three major categories: community- driven advocacy efforts, adopted policies and plans, and specific crash events or patterns. Jurisdiction State Population (2016) Vision Zero City Arlington VA 230,050 No Calgary Alberta, Canada 1,392,609 No Chicago IL 2,705,000 Yes Durham NC 263,016 Yes Fremont CA 233,136 Yes Green Bay WI 105,139 No Los Angeles CA 3,976,000 Yes Nashville TN 659,000 Yes Portland OR 639,863 Yes San Francisco CA 864,816 Yes Seattle WA 704,352 Yes Table 3-1. Jurisdictions interviewed. Figure 3-1. Locations of interviewed jurisdictions.

46 Pedestrian Safety Relative to Traffic-Speed Management 3.2.1 Community-Driven Advocacy Community-driven advocacy efforts take various forms. Nearly all the jurisdictions inter- viewed have some sort of community or resident program where individuals can report traffic concerns to the transportation bureau or public works department. However, jurisdictional staff are also aware that community-driven efforts, while powerful, sometimes focus more on outlying traffic hazards (e.g., a few people going excessively high speeds on a neighborhood street) versus more systemic speed issues (85th percentile speed of 35 mph on a 30-mph street) that staff want and need to focus on. Additionally, residents who feel empowered to report issues to a jurisdiction may not be representative of the entire community. Depending on the number of concerns reported and how much jurisdictions use community feedback to inform project prioritization and direction, safety efforts may focus disproportionately in locations or neighborhoods that staff hear about from their constitu- ency. Staff work to balance these needs and concerns and to take advantage of opportunities to try new things. A particular case of community-driven catalysis occurred in Nashville, where the City created three “Walking Districts” as part of their speed-reduction efforts (see sidebar for more infor- mation). The City and county (locally known as Metro) had already been discussing whether to reduce the de facto speed of 30 mph in Nashville, but had agreed that trying to do it in one citywide push was neither resource efficient nor feasible. As a result, the City decided to choose three distinct areas, defined as Walking Districts, where they would focus their efforts. These areas were selected in large part because they already had well-organized traffic-safety advocacy groups and efforts underway. These were also neighborhoods that had, for years, been feed- ing comments and concerns into resident request programs. The City took advantage of these conditions to roll out their program and, in turn, had little resistance since the neighbors were already energized and educated about such efforts. Nashville: Walking Districts “Walking Districts” were introduced in Nashville to help residents see and experience what an environment focused on lowering vehicle speeds and promoting walking would feel like. While the City sees the benefits in exploring these practices citywide, piloting the concept and practice of the Walking Districts allowed programs to begin in neighborhoods and areas where there was already support for the concept. The Walking Districts have signage lowering the speed limit to 20 mph on residential streets and 25 mph on collectors, down from the previous 30 and 35 mph speed limits, respectively. In addition to lowering speed limits, other traffic-calming features such as improved crosswalks at major intersections, increased enforcement, educational flyers, and educational sessions with neighbors were implemented. The neighbors within the pilot districts were essential to the success and startup of the program—the idea was brought to the city by a Vanderbilt University physician looking to get more of her clients to walk. This advocacy was essential to further- ing momentum for Nashville’s speed-reduction efforts, which is continuing as more neighborhoods are looking to participate in the Walking Districts program.

State of the Practice 47 3.2.2 Adopted Policies and Plans In theory, jurisdictional actions and programs are founded in policies and, in turn, reflected in plans. In reality, this does not always happen; jurisdictions often react to residents or elected officials, and policy guidance can become a lower priority. However, policies remain an impor- tant tool for helping staff organize larger efforts and rationalize and support projects or plans if they are challenged. By far, the most complete and overarching policy supporting reducing speeds to increase pedestrian safety is Vision Zero. Vision Zero is a strategy, first implemented in Sweden, that looks to eliminate all traffic fatalities and severe injuries. As seen in Table 3-1, most of the pro- filed cities have adopted Vision Zero policies or action plans, or both. Although Vision Zero is implemented uniquely by each jurisdiction, speed management is one of its four core actionable strategies. Jurisdictions have taken this strategy and used it as the basis for creating programs, supporting design efforts, and engineering roadways to reduce speeds. Fremont, San Francisco, Seattle, and Portland all specifically cited their Vision Zero action plan as a primary mobilizing force in speed-reduction efforts. Beyond Vision Zero, there are many other plans that may use speed reduction as a tool to improve pedestrian safety. Green Bay is looking to their upcoming Safe Routes for Non-Drivers plan to better understand whether and how vehicle speed is a threat to pedestrians, bicyclists, and transit riders. Arlington specified a desire for lower speeds in their transit station modal plan. Although this plan does not directly link speeds to pedestrian safety, it does make the con- nection between lower speeds and encouraging more residents, employees, and visitors to walk. Chicago and Calgary have speed-reduction policies and actions within their pedestrian plans. 3.2.3 Specific Crashes or Events Pedestrian crashes can mobilize individuals, communities, and elected leaders. Crashes, espe- cially those that result in serious injuries or fatalities, shock and anger residents of the commu- nity and can potentially spur individuals and groups to act, advocate, and demand a change. In San Francisco, Green Bay, Fremont, and Portland, highly publicized crashes and the associated community reaction resulted in additional public support for reducing speeds on roadways. Crashes do not just spur citizens to act. While only a few cities mentioned crashes as a primary catalyst for speed-reduction efforts, many others implied that residents, elected officials, and/or staff saw crash rates as unacceptable and as motivation for further efforts. Staff from San Francisco, Portland, and Green Bay mentioned specific crashes that spurred additional interest in speed reduc- tions; Portland’s City Council went so far as to declare a state of emergency, which allowed a speed reduction without the usual process for lowering a speed limit (see sidebar for more information). Crashes also help inform locations of speed-reduction efforts. Several cities have identified networks of roads with high crash rates and/or severe injury crashes, often called “high-crash corridors,” “high-injury networks,” or “safety-priority streets,” which are targeted for speed- reduction efforts. Traditionally, a network screening method such as the “sliding window” is used to detail the crash rate per length of a roadway, helping jurisdictions identify crash concen- trations. Most jurisdictions that have identified these networks have found that a dramatically disproportionate number of serious and fatal crashes occur on a limited number of streets. For example, in Portland, 80% of overall traffic deaths and serious injuries occurred on the 19% of streets posted at 30 mph or higher (Figure 3-2). In Fremont, all serious and fatal crashes in 2015 occurred on streets with speed limits of 40 mph or over (Figure 3-3). For both of these cities, further analysis of these patterns helped create a high-crash and safety-priority network, respectively.

48 Pedestrian Safety Relative to Traffic-Speed Management Figure 3-2. City of Portland pedestrian high-crash network. (Source: Portland Bureau of Transportation.) Portland: State of Emergency In March 2017, Portland City Council enacted a state of emergency for a portion of SE Division St., a corridor that has historically seen a high number of severe crashes. The state of emergency resulted in a speed-limit reduction from 35 to 30 mph. Under Oregon law, a jurisdiction with road authority can establish an emergency speed, which takes effect immediately. Division St. was already known to be a high-crash corridor, having been the street with the most fatal or serious crashes in the city over the prior 10 years (13 deaths and 117 serious injuries). The emergency ordinance was catalyzed by two pedestrians killed within hours of each other in December 2016. The City also expedited already-planned installation of speed safety cameras along the corridor, which were activated when the 30-mph speed went into effect. Funds received by speeding citation fees help pay for the program and safety improvements on Portland’s high-crash network. The emergency speed provision let the City reduce the speed limit for 120 days. The speed reduction and speed cameras together resulted in change to the 85th percentile speed and resulting approval from the state DOT for a permanent speed change to 30 mph. In addition to reductions in speed, there has also been significantly improved com- pliance at locations with cameras. In April 2017, the City issued nearly 4,300 citations from the Division St. speed cameras. In April 2018, just over 1,800 citations were issued from those same cameras. The state-of-emergency process is rarely used, but is still a tool in the City’s speed management toolbox. In early 2018, the City issued a state of emergency for another corridor, outer SE Stark St., after two pedestrian deaths occurred within 2 weeks.

State of the Practice 49 3.3 Implementation Efforts The jurisdictions were asked about efforts they have implemented over the past 2 years to improve pedestrian safety by managing speeds. None of those interviewed rely solely on lower- ing the posted speed limit to reduce vehicle speeds. Instead, efforts are usually part of a more holistic suite of policies, programs, design, and operations efforts throughout which vehicle speed reduction is woven as a goal. 3.3.1 Policies and Programs For many jurisdictions, policies offer the foundation for any efforts and help to validate the needs for staff and resources. The adoption of Vision Zero is a key component of many cities’ speed-reduction efforts and is associated with an action plan that defines measurable steps for increasing pedestrian safety through speed reduction. Complete Streets policies were also cited as key components of holistic speed-reduction strategies in several places, including Arlington County and Chicago. Due to their multimodal nature, Complete Streets efforts do not often singularly focus on pedestrian safety, but have the same nexus between lower vehicle speeds and safety for all users. Arlington County has two programs related to their umbrella Complete Streets policy: Neighborhood Complete Streets and Arterial Complete Streets. Policies can also merge into programs. Most of the policy-driven programs around speed reduction included enforcement efforts. Enforcement efforts such as crosswalk enforcement actions and positive ticketing (when coupons or incentives are offered for correct behav- ior) were praised by multiple jurisdictions. However, in many situations these could not be regularly performed due to staff or financial resource restrictions. School zone enforcement programs and ASE efforts were also used with much success, but may be challenging to Figure 3-3. City of Freemont high-crash network. (Source: Fremont DOT.)

50 Pedestrian Safety Relative to Traffic-Speed Management sustain in light of cost and staffing demands. This is especially the case when police officers are needed; many jurisdiction’s police departments have limited traffic enforcement offi- cers and, as such, cannot afford to regularly have them pulled off routine shifts for targeted enforcement activities. 3.3.2 Design and Operations Many jurisdictions explained how the most powerful speed-reduction technique was creating an environment where drivers naturally go the preferred speed, not because of a posted speed limit but because of the roadway design. Table 3-2 shows design efforts and infrastructure that were specified as part of speed-reduction efforts by the interviewed jurisdictions. Jurisdictions not shown to be using a specific design technique to limit vehicle speeds may still be using the referenced design, but for another purpose. Reallocating roadway space and reducing the vehicle-lane size were used overwhelmingly by jurisdictions to signal to drivers that they need to drive more slowly. Even when jurisdictions specified “adding bike lanes,” decreasing vehicle-lane width was a primary reason for the effort (although the increased activity of bicyclists in the bike lane was also cited as slowing drivers down). Ten feet was the preferred lane width, although many jurisdictions used 11-ft lanes for transit or on roadways designated for freight movement. Space created by lane-width reductions has been used for parking, bike lanes, and curb extensions. Arlington County’s design standards are particularly detailed and well developed, although they were not initially developed with speed reduction in mind. As a first-ring suburb of Washington, D.C., Arlington’s success is based on residents, employers, and visi- tors being able to get to, from, and through Arlington easily. Much of their mobility planning Seattle: Vision Zero Action Plan The City of Seattle adopted a Vision Zero action plan in 2015 that included state- ments of support from the mayor, the police chief, and the transportation director on the motives behind and commitment to eliminating deaths and serious injuries on Seattle’s streets. The plan presents an overall vision for safer streets, along with short-term actions and evaluation metrics to understand whether efforts are improving safety as desired. While the plan addressed all the tenants of Vision Zero, there are a few actions directed specifically at speed reduction, including • Creating a 20-mph Zone program to reduce speed limits to 20 mph on residen- tial streets, prioritizing program roll-out on streets with high collision histories, schools, and parks; • Reducing arterial speed limits to 30 mph or lower, in combination with speed- reduction tools such as radar speed signs and road design improvement; and • Improving downtown safety by reducing speed limits downtown to 25 mph. In 2017, the city produced a progress report on completed actions and their effects on speed reductions. More information on both these reports is available at https://www.seattle.gov/visionzero.

State of the Practice 51 is based around access to the rail system (11 METRO stations are in Arlington), and the goal is to get as many people using transit as possible. To do that, starting in the 1960s, a sec- tor plan was developed for each station area that specified sidewalk widths and pedestrian amenities. These sector plans are updated regularly to improve the pedestrian experience and safety, often by changing design standards that influence speeds (e.g., increasing pedestrian crossing density, reducing crossing distances, and reducing vehicle-lane widths). These plans and the associated design standards help Arlington to evaluate consistently how to improve conditions for pedestrians, and to require such conditions at the time of development by virtue of articulat- ing the standards in specific area guides. Most recently, Arlington has attempted to switch from a paradigm of designing roadways for large trucks to designing them for cars. This change decreases lane size and turning radii, forcing large vehicles to be more spatially aware and/or slow down when turning. Figure 3-4 shows an example of pedestrian improvements made due to updated design standards at the intersection of Glebe and Fairfax in Arlington. Some jurisdictions also used operations to force speed reductions. The most common opera- tions efforts were related to signal timing, although techniques varied among those interviewed. San Francisco and Portland regularly use signal timing to encourage a consistently low speed among vehicles, often referred to as the “Green Wave.” Timing lights for a relatively low but consistent speed reduces the likelihood of injury and helps maintain traffic flow, but it also stops motorists with a red light if they are driving too fast. Los Angeles, on the other hand, uses a signal preemption strategy at some locations to slow speeding drivers, by capturing their speed using the built-in radar in a speed feedback sign, and triggering a downstream signal to change to red if the driver is speeding. Los Angeles has also Design Profiled Jurisdiction Ar lin gt on Ca lg ar y Ch ic ag o Du rh am Fr em on t Gr ee n Ba y Lo s A ng el es N as hv ill e Po rt la nd Sa n Fr an ci sc o Se att le Reallocate roadway space—roadway reconfigurations/lane conversions. x x x x x — x x x x x Adjust curb radii. x x x — x x x — x x x Install curb extensions. x x x — x x — — — x x Add or remove on-street parking. x — x — — x — — — — x Reduce lane size (from 12 ft to 10–11 ft). x — x x x x x x x x x Add right side bike lanes. x x x — x x x x — x x Install pedestrian islands. — x x — x x x x — x x Pedestrian-focused street branding/signs. — — — — — — — x — — — Convert peak hour lanes to full-time parking. — — x — — — x — — x — Increase street lighting. — — — — x x — — — — — Speed humps or cushions. x x x — x x x x x x x — = N/A. Table 3-2. Roadway-design strategies used to slow drivers in the profiled jurisdictions.

52 Pedestrian Safety Relative to Traffic-Speed Management implemented “rest on red” on some of their roadways, where lights remain red until triggered by a vehicle once they reach the intersection in question. 3.3.3 Effectiveness of Efforts As explained above, jurisdictions often use a suite of treatments in their efforts to reduce speeds and improve pedestrian safety. This method makes it difficult to identify which effort is the most effective and results in the largest gains in pedestrian safety. However, jurisdictional staff indicated a belief that ASE and signal manipulation were two key strategies. These are described further below. Automated Speed Enforcement in Speed Zones (Seattle) Beginning in 2012, Seattle installed fixed automated speed cameras to enforce the 20-mph school zone speed limit at four elementary schools. The program expanded in September 2014 to an additional five schools and in September 2015 to five more schools, for a total of 14 schools with speed photo enforcement (see Figure 3-6). These cameras capture speeding vehicles only when flashing beacons are operating during the hours when students are arriving and leaving school grounds. This work was motivated by a prior ASE pilot that Seattle conducted in 2010, which showed dramatic changes in mean speeds, 85th percentile speeds, and percentage of vehicles triggering the cameras over the pilot period (see Figure 3-5). Since the program began, the average number of traffic violations per camera per day has decreased by 64%, average speeds have decreased by 4%, and 90% of people who receive and pay for a citation do not reoffend. The effectiveness of automated speed cameras has also been found in other jurisdictions; Portland’s Vision Zero Annual Report 2017 stated a 59% reduction in speeding over a year and a half of having speed cameras on four corridors (PBOT 2018a). April 2002 (Before) April 2016 (After) Figure 3-4. Intersection improvements made to the intersection of Glebe and Fairfax due to updated design standards. (left): April 2002 (before); (right): April 2016 (after). (Source: Arlington County.)

State of the Practice 53 Arlington: Revised Design Standards Arlington relies heavily on detailed design standards to create the sorts of places they want to see and tailor development to specific neighborhood characteristics. In addition, Arlington’s continued success as an inner-ring suburb to Washington, D.C., is highly dependent on having pedestrian access to transit. As such, Arlington staff strive to improve their design standards to encourage walking to transit as well as to create an environment where pedestrians are and feel safe. Their design standards incorporate speed-reduction methods such as narrowed turn radii, reduced lane widths, and increased crossing density. See the table below for a summary of design standard changes made to reduce vehicle speeds. Standard or Practice Previous Standard Current Standard Curb return radii 25 ft standard 15 ft standard with adjustment is needed to accommodate design vehicle (city bus or 30-ft SU truck) Lane widths 11 ft standard, 12 ft preferred 10 ft standard, 11 ft if lane is used by transit bus Residential street widths with on- street parking (both sides) 36 ft standard 28 ft standard if ADT is 1,500 or lower Right-turn lanes and slip lanes Frequently provided; slip lanes Only considered where at least 250 vehicles turn right per hour; removing slip lanes wherever feasible Speed limits and design speeds on primary arterials Posted at 30 or 35 mph. Design speed 5 mph higher than posted Posted at 25 or 30 mph; design speed equal to posted speed Crosswalk markings Typical pair of 6-in.-wide white lines, 10-ft width High-visibility markings (alternating 24-ft-wide bars) at crossings of arterial streets, near schools, and midblock; 10-ft to 20-ft width On-street bicycle accommodations Signed routes; began to mark 5-ft-wide bike lanes around year 2000 Where feasible, build protected or buffered bike lanes on arterial streets

54 Pedestrian Safety Relative to Traffic-Speed Management However, many jurisdictions are not allowed to use automated speed cameras or are restricted because of state regulations and/or privacy concerns; several interviewees cited this restriction as a limitation for their speed-reduction efforts. Signal Manipulation Signal timing was used by multiple jurisdictions to regulate speeds, as well as to require high- speed traffic to stop. San Francisco sets progressions depending on street design. If there are bike lanes on a street, the signal progression is set at 12 to 14 mph; for arterials, two-thirds are set at 20 mph. In Los Angeles, signal timing adjustments to curb speeding have been especially effective when signals are close to one another (e.g., one to two blocks apart). In contrast to San Francisco and some other jurisdictions, the Los Angeles program is not intended to encourage a certain speed; it is intended to discourage speeding where roadway conditions do not otherwise send that signal and redesign is not a viable option. Additionally, Los Angeles uses “rest in red” signals to reduce speed by requiring vehicles to stop or slow as they approach an intersection. In many jurisdictions, lights on major roads rest or remain in green until a vehicle on a side street triggers the light to change. “Rest in red,” conversely, has the signal show red indications in all directions until it is triggered by a loop to turn green. Another program involves corridors that have signals that flash yellow overnight for the mainline traffic (side streets are red and then can trigger the arterial signal to red, so they receive a green) to provide a visual cue for drivers to slow down. Initial data from this proactive treatment show speed reduction in Los Angeles, and other jurisdictions have linked this treat- ment to a reduction in injury crashes (Jordan 2014). In addition to this practice, Los Angeles is piloting technology from speed feedback signs connected to signals. In these situations, if the radar catches an approaching vehicle traveling at too high of a speed, it will trigger the upcoming signal to turn red. The sign can also send an alert to the Los Angeles Police Department (LAPD) that speeding vehicles are in the area without including any identifying information; this can prompt LAPD to patrol the area and discourage further speeding. Although this technology is currently being tested, initial results are encourag- ing and will likely lead to a more widespread rollout in the future. Before Enforcement End of Enforcement Figure 3-5. Changes in speed before and after installation of ASE cameras in Seattle. (Source: Washington Traffic Safety Commission 2011.)

State of the Practice 55 Figure 3-6. Locations of school speed zone safety cameras in Seattle. (Source: Seattle DOT.)

56 Pedestrian Safety Relative to Traffic-Speed Management 3.4 Partnerships Creating and fostering partnerships is essential to large-scale behavior change such as speed reduction. Partners, especially those that are local, often have critical relationships with stakeholders who may be affected by speed-reduction efforts and may also have access to resources and data that help facilitate a greater understanding of the problems and potential solutions. For example, some business improvement districts see a benefit from speed-reduction efforts, as slower speeds allow for a more pedestrian-friendly environment and an increase in foot-traffic shopping. The advantage of having a partner with interests seemingly unrelated to transportation safety (e.g., economic development, small business promotion) can help broaden the perspectives of those who may be initially opposed to speed-reduction efforts. Neighborhood/civic associations and Safe Routes to School programs can serve similar advocacy roles. Partnerships can also serve a more functional role. Police departments are necessary partners for any speed-reduction efforts that include enforcement, and traffic engineering and/or public works departments are needed to ensure streets are designed, built, and operated to achieve desired outcomes. In Portland, for example, the police are pivotal partners in focusing enforce- ment efforts on the most dangerous behaviors (like speeding) on the most dangerous streets and ensuring geographic distribution of enforcement throughout the city. During traffic enforce- ment, police use traffic stops as an opportunity to educate people on traffic safety, often offering safety education classes in lieu of a citation. In San Francisco, the transportation agency and the public health department have col- laborated to promote and communicate the Vision Zero and speed-reduction messages and programs, as well as share some of the technical aspects of the effort. The public health depart- ment evaluates the safety data and created the San Francisco High-Injury Network because of its unique access to health data that the transportation agency cannot obtain. Table 3-3 displays the various types of local and regional relationships the interviewed jurisdictions have formed to help carry out their goals of speed reduction and increased pedestrian safety. The sidebars on the following pages describe unique partnerships that Fremont, San Francisco, and Portland have used to further their speed-reduction efforts. 3.5 Garnering Support Obtaining public and institutional support is an essential piece in furthering any type of jurisdictionwide effort. This is especially the case for efforts like improving pedestrian safety that, despite broad support in theory, may hinder one’s daily routine by increasing vehicular travel times, causing diversion, increasing chances of citations, etc. The interviewed jurisdictions emphasized the need for trust and clear lines of communi- cation between the jurisdiction and the public for speed-reduction projects to be successful. Successful efforts began with the clear identification of a problem on which the commu- nity and elected officials agreed. This consensus about the problem was integral to a proj- ect’s or program’s success—if the community or those elected were not convinced of the problem’s existence or seriousness, the project could be derailed, removed, or serve to disempower the community and potentially create active resistance (see Chicago’s story in the sidebar).

State of the Practice 57 Once the problem was identified and validated by community concerns and data, many juris- dictions tried to get official written support by some sort of community organization. Efforts in Arlington, Durham, and Nashville moved forward largely because of support by formal civic and neighborhood associations. Arlington and Chicago sought out and received support from busi- ness improvement districts, who believed that speed-reduction projects would improve safety and were important to economic vitality. San Francisco: Public Health as a Partner When San Francisco staff began their Vision Zero efforts, they realized there was a data gap—almost a quarter of San Francisco General Hospital (SFGH) patients who were admitted for an injury received while walking or biking were not represented in the state database. Thus began the partnership between the department of public health and the transportation agency. San Francisco is unique in that it has only one level 1 trauma center, so all serious and fatal injuries from traffic injuries were receiving treat- ment in one place. This creates an unparalleled source of data that can be used to inform Vision Zero strategies, programs, and projects. The jurisdiction’s Vision Zero program has an epidemiologist who leads the Trans- portation-related Injury Surveillance System, funded by the San Francisco Municipal Transportation Agency and SFGH. This system combines and vets data by the City and County of San Francisco, police reports, hospitalizations and trauma activations, emergency medical service, and the medical examiner’s office into one location to create a comprehensive and reliable data source for evaluating safety in the City. Design Profiled Jurisdiction Ar lin gt on Ca lg ar y Ch ic ag o Du rh am Fr em on t Gr ee n Ba y Lo s A ng el es N as hv ill e Po rt la nd Sa n Fr an ci sc o Se att le Police department x x x x x x x x x x x Public works department x x — — x — x x — — — Universities/institutions x — x — — x — — — — — Local TDM marketing x — x x x — x — x x x Jurisdiction traffic engineering — x x — x — — x — — — Civic and neighborhood associations x — x — — — — x — — — Business improvement districts — — x — — x — — — — — Department of public health — — x — — — x — — x — County agencies — — x — x x — — — — — Adjacent jurisdictions/regional efforts — — x — x — — x — — Vision Zero or project-specific technical advisory committees — — x x — — — — x — — Bike/pedestrian advocacy groups x x x x — x x x x x x Schools/Safe Routes to School programs x x x — x x x — x — x — = N/A. Table 3-3. Local and regional partnerships specified by interviewed jurisdictions.

58 Pedestrian Safety Relative to Traffic-Speed Management Portland: Investing in Community Relationships The City of Portland worked closely with the Asian Pacific American Network of Oregon (APANO) to create and distribute culturally relevant and language- specific street safety outreach and education after multiple pedestrian fatalities in predominantly Asian neighborhoods. The City acknowledges that City staff are not representative—culturally, geographically, or racially—of many of the communities most impacted by unsafe speeds. As such, they have contracted with APANO to train community members in different languages and to share and receive feedback on specific transportation safety projects. The City’s established relationships with advocacy groups also helps community members advocate for action that is meaningful and realistic. After four pedes- trian deaths in less than 2 years in a corridor, two advocacy groups—Oregon Walks and The Rosewood Initiative—submitted a letter asking for the City to declare a state of emergency and a related speed-limit reduction, as well as funding of identified safety projects that were in the pipeline. Having informed and educated community advocates helps to energize speed-reduction efforts and support ongoing staff planning and projects. Fremont: Optimizing Enforcement The City of Fremont’s Public Works Transportation Division worked diligently with their police department to optimize the abilities of their police force, while acknowledging that there were limited resources. The City’s Vision Zero crash analysis showed that most of the speed-related fatal or serious injury crashes occurred between 6:00 and 10:00 p.m. However, the existing police lead for speed enforcement, the motorcycle unit, was slated for daytime patrols. To address this apparent disconnect between need and availability, the City cross-trained their patrol units, which were on patrol during the times of day with historically more speed-related crashes and offenses. Since the change, there has been nearly triple the amount of speed enforcement, going from 1,200 stops a year to 4,300 a year. The City believes that the greatest benefit of this change is the increase in visibil- ity of the police force and education. When possible, the traffic stops are focused on education about the dangers of speed and its relationships to crashes. The City credits this to a cultural change led by the police chief and a mutual under- standing that speed reductions make the City safer for all roadway users.

State of the Practice 59 Bicycle and pedestrian advocacy groups are also instrumental in offering support for speed- reduction efforts and creating support or energy for the jurisdiction. This proved to be critical for Calgary, Seattle, Nashville, and Portland, whose efforts were pushed forward by advocates who lobbied the elected body for change and created energy and demand that jurisdictional staff could not. Despite such community support for their efforts, few of the jurisdictions interviewed were able to complete speed-reduction efforts without any resistance to the changes and the unknowns that those changes could bring. This resistance was magnified if the community or residents did not perceive that there was a problem. The interviewees reported several types of resistance to the speed-reduction projects for pedestrian safety, including • Residents feel that the infrastructure does not fit. In Green Bay, there are many roads that have high speeds and pedestrian traffic, but lack sidewalks. Although sidewalks seem like one of the first steps in improving pedestrian safety—by offering separation and by slow- ing drivers down because of less roadway space—sidewalks have been resisted by some in the community for two main reasons: (1) sidewalk installation and maintenance costs are the responsibility of the property owner and (2) some believe that the sidewalk encroaches onto their property, bringing pedestrians closer to their homes. Many like the more rural feel of the sidewalkless roads and see parking lanes as multipurpose—for parking, walking, and biking. This hesitation about new infrastructure was also reported related to speed-reduction projects that proposed bike lanes as a way to increase street uses, decrease vehicle-lane size or Chicago: Defining the Problem Chicago’s speed-reduction efforts have encountered resistance from some neighborhoods. In one instance along MLK Drive, the City promoted a project that included a road diet. Neighbors pushed back against the project, saying that they liked being able to drive faster to access downtown. Additional pushback came from a separate neighborhood where a similar speed- reduction effort through design was proposed. This time, the neighbors resisted the change believing that it would increase congestion and change the character of the neighborhood. In both situations, City staff realized that the neighborhoods did not agree with the problems as identified through data analysis. In addition, staff found that safety was not necessarily a primary issue for constituents if they had not directly experienced a crash. Staff also observed that neighborhoods were particularly resistant to change if they felt that the city had made decisions for them without their consent. Since these experiences, City staff have increased their efforts to work with community stakeholders to better understand concerns and collaboratively find solutions.

60 Pedestrian Safety Relative to Traffic-Speed Management number, and improve active transportation options. Because the roads where these projects are located are usually high speed and autocentric, there is rarely existing bicycle traffic, and as a result, community members may not think that bike lanes make sense. Proposing bike lanes that the community deemed unnecessary were project deterrents for speed-reduction efforts in Chicago and Durham, among others. • The community was not involved in the decision making. Many jurisdictions experienced unanticipated resistance to speed-reduction projects that staff thought were obvious choices for both location and infrastructure. Staff in Durham realized that they encountered resis- tance because they lacked sufficient data to convince the community of the problem. Staff in Chicago realized that historic injustices between the City and its residents, especially in com- munities of color, created a deep sense of mistrust. This, in turn, resulted in the community pushing back against improvements when they were not involved in identifying the problem or proposing a solution. • The project gets sidetracked. Because speed-reduction projects are often multifaceted, it can be difficult to keep the community focused on solving the problem at hand: pedestrian safety. Instead, communities often get stuck on a piece of the project and how it may negatively affect them. For San Francisco, one issue has been parking removal. In Fremont, large-vehicle drivers informed officials that narrowing lanes made them feel “less comfortable” driving, and slower speeds were perceived as contributing to traffic congestion. These issues created an opportunity for public education and to clarify the Vision Zero approach of safety over speed. For these situations and others, jurisdictional staff needed to address and acknowledge concerns and seek to minimize negative impacts, while also providing information about how these pieces of the project(s) helped increase pedestrian safety. While community resistance was present in each jurisdiction interviewed, it has not stopped the forward movement of speed reduction and pedestrian safety efforts. However, lessons learned from these experiences compelled staff to involve communities more, collaboratively identify the problem and (sometimes) solutions with the community, and know that there might be setbacks along the way. 3.6 Working with State DOTs Another important aspect of addressing pedestrian safety via traffic-speed management relates to relationships with state DOTs. Often, state DOT roads carry high amounts of traffic and have among the highest numbers of pedestrian crashes within a community; yet local jurisdictions lack authority to make substantive changes to those roadways. Many interviewees identified a need for a defined process to collaborate with state DOTs on policy changes such as the ability to use a car, rather than a truck, as the default design vehicle and to more generally design roadways that can meet both local and state needs for safety and mobility. The interviewees also discussed how state DOTs influence local jurisdictions’ speed-reduction efforts through policies and program support. Many of the interviewees stated that their state DOTs have policies supportive of pedestrian safety, but not necessarily specific to speed reduc- tion. This was seen as another area where the two agencies could work together more closely in the future, to ensure that statewide policies and programs—that often need to apply to rural areas—also work for more urban roadways, where pedestrian traffic is much more common. A desire to work with the state DOTs to achieve the ability to use automated enforcement also emerged from the interviews. ASE is a proven countermeasure, yet it is still illegal in most states, as described in Chapter 2. Interviewees also expressed a desire to use automated enforcement for violations like red-light running, which is still against the law in many states. Having the support of the DOT was perceived as a way to change the debate over automated enforcement.

State of the Practice 61 The process of setting speed limits—and in particular the 85th percentile method—featured prominently in the interviews. While multiple interviewees expressed a desire for more flex- ibility to lower speeds, particularly on certain roadways, there appeared to be a lack of clarity about who is ultimately responsible for the constraints around setting speed limits, and thus how to change those constraints. For example, some interviewees believed state DOTs have the responsibility, whereas others indicated that speed limits were set by state law or mandated by the MUTCD. In some cases, as in California, the process involves multiple players and is not straightforward, pointing to a need for greater clarity about how speed limits are set—and how such procedures can be changed—within each state. Relatedly, interviewees expressed a desire for a collaborative process that would allow the local agency to spearhead speed-reduction efforts when state-owned roadways are identified as the sites of high numbers and/or rates of fatal and severe crashes. This process could provide needed clarity for jurisdictions working on or planning for broader speed-reductions efforts, and could in turn support state safety efforts such as Toward Zero Deaths or other programs like the Highway Safety Improvement Program. 3.7 Measuring Impacts of Efforts All of the interviewed jurisdictions use performance measures to evaluate whether there is a pedestrian safety issue that needs to be addressed or if an implemented treatment is having an effect. Measuring speeds (before and after) a project is the most common metric Setting Speeds in California: A Case Example According to NTSB (2017), speed limits are typically set statutorily, and adjust- ments to those statutes are generally made based on the 85th percentile speed of traffic. Increasingly, cities are saying that this method does not allow them to work toward safety as quickly or comprehensively as they would like. Indeed, the recent NTSB report on speeding-related crashes suggests that use of the 85th percentile method can result in unintended consequences such as increasingly higher speeds over time (2017). Yet changing the speed limit–setting process can be complex and time-consuming. The state of California is grappling with this complexity. The DOT’s Traffic Control Devices Committee (CTCDC) has reviewed the topic repeatedly since 2003, several times slightly amending the text in the CA MUTCD, which guides speed-limit setting in the State. However, the process is not as straightforward as it may seem, as the CA MUTCD is linked procedurally to the California Vehicle Code, which is state law and is outside the purview of the CTCDC. Thus, changes to the CA MUTCD, which take time to achieve, then depend on changes to state law, thereby lengthening the process and involving many more parties along the way. At the same time, the CTCDC recognizes the increasing importance of creating additional flexibility within the process, particularly for cities that have adopted Vision Zero within the State (nine cities as of January 2018; CTCDC 2017). Currently, the CTCDC members are working with their appointing authorities to encourage a legislative solution that would allow the CA MUTCD to remain in compliance with state law, while also being responsive to needs articulated by cities throughout the State.

62 Pedestrian Safety Relative to Traffic-Speed Management for jurisdictions. Many also gather before and after traffic volumes to understand how the project affected traffic flow. Table 3-4 summarizes metrics that jurisdictions specified in the interviews. Although volumes and travel time were gathered consistently by many of the interviewed jurisdictions, speed and crash metrics varied. • Speeds. Most cities relied on the 85th percentile as their speed metric, both because of its use in state approvals for speed-limit setting and its standard use in engineering treatments. However, some jurisdictions used different speed measurements to help the public better understand the status quo of speed and how it changed post project. For example, Nashville reports on average speed, believing that residents better understand average than the 85th percentile. Similarly, Chicago reports on the changes in vehicles going above a certain speed. According to staff, this focuses the conversations on the reduction of high-speed drivers, which is what community members cite in concerns. There are also a variety of ways that jurisdictions gather speeds. Although most use traditional speed study radar techniques, Chicago has been especially creative in gather- ing speeds from other sources. Specifically, the City has used speed data from insurance companies’ smartphone apps, the City’s fleet, and public transit to understand the gen- eral speeds in an area without having to organize and implement an independent speed survey. As mentioned above, Los Angeles has used real-time data from speed radar signs to identify areas of the City with speeders and send out additional patrols. They are also experimenting with feeding that data into signal timing operations in order to trip signals to red and force a stop downstream. • Crashes. Although many jurisdictions used crash locations and rates to help choose, pri- oritize, and evaluate projects, there was variation in the sorts of crash data used, how it was obtained, and how it informed the process. Chicago uses a 5-year average of crashes in order to better focus efforts on severe crashes. Green Bay evaluates the changes in the patterns of types of crashes (e.g., determining the cause behind new crash patterns). Many jurisdictions wished to have better and more complete databases of crashes, but there is often a lag time and data may be incomplete, making shorter term comparisons more difficult. San Francisco has a notably complete system for recording and analyzing pedestrian crashes due to its partnership with the San Francisco Department of Public Health (SFDPH). Because San Francisco has only one Level One Trauma Center, all serious traffic injuries and fatalities are taken there, where more information about the crash is obtained. Metric Profiled Jurisdiction Ar lin gt on Ca lg ar y Ch ic ag o Du rh am Fr em on t Gr ee n Ba y Lo s A ng el es N as hv ill e Po rt la nd Sa n Fr an ci sc o Se att le Before and after speeds x x x x x x x x x x x Before and after volumes x x x x — x x x x — x Change in volume on adjacent streets x — — x — — x — x — — Crash data — x x x x x x x x x x Corridor travel time — x x — — x x — x — x — = N/A. Table 3-4. Metrics for evaluating problems and project success.

State of the Practice 63 3.7.1 Metrics from the Literature The literature review in Chapter 2 found that common metrics for evaluating the effectiveness of an individual treatment in a specific study included average speed, 85th percentile speed, and the percentage of top-end speeders. A few of the most common metrics for evaluating the effect of a countermeasure or policy on safety include the following (FHWA 2016): • Number of fatal or serious pedestrian injuries over 5 years, • Crashes per volume of pedestrians over 5 years, • Number of speeding-related citations (includes those generated from automatic and manual enforcement efforts), and • Number of crashes with a speed-related contributing circumstance (e.g., speeding, reckless driving, or too fast for conditions). The 5-year recommendation derives from the reality that most areas do not experience a high enough number of fatal and serious injuries over a short period to conduct meaningful analysis. Moreover, the inherent randomness in crashes means that numbers can fluctuate annually and even biennially, and looking only at short periods can therefore lead to erroneous conclusions. Looking at 3 and preferably 5 years helps both to control for this randomness and increase sample size. One can also look at the change in number of fatal and serious injuries per population. Note, however, that a change in pedestrian volume is likely to be a more accurate assessment of the injury and/or fatality rate, as populations change often, especially in urban areas, and may there- fore influence pedestrian safety rates in ways that have no actual connection to pedestrian safety as experienced on the roadway. FHWA (2016) suggests that measurements of perceived safety, derived from resident surveys, can be helpful for evaluating effectiveness. Cities like New York City and Seattle are measuring the impact of some treatments via the number of repeat offenders. As discussed above, both of these cities use automated enforcement cameras, making it easy to monitor this metric. In addition to tracking repeat offenders for speed violations, the City of Seattle also tracks the change in the percentage of speeding-related fatal crashes and fatal and serious injury crashes on an annual basis (SDOT 2017b). The City of Portland (2016) lists the following speed-related performance measures for annual evaluation in its Vision Zero Action Plan 2017: • Number of speeding-related citations issued near speed safety cameras compared to baseline, • Percentage decrease of autos traveling over posted speed after speed safety cameras were installed, • Obtainment of local authority for speed reduction, • Number of street segments where posted speeds have been reduced, • Number of high-crash corridors receiving speed improvements each year, and • Percentage decrease of autos traveling over posted speed after improvements. The aforementioned performance measures represent a variety of metrics that jurisdictions can use to measure the success of treatments, program, and policies to improve pedestrian safety. As with all performance metrics, ensuring that accurate and useful data are collected needs to be a key component of a jurisdiction’s countermeasure evaluation plan (FHWA 2016). 3.7.2 Usefulness of Metrics All jurisdictions were concerned about having dependable and consistent metrics. Often, evaluations do not occur or the speed at which a project is implemented does not allow enough time on the front end to gather existing conditions data.

64 Pedestrian Safety Relative to Traffic-Speed Management In addition, even when metrics are gathered, it is difficult to show causality between a certain type of improvement and pedestrian safety. This is mostly because it is rare for only one treatment to happen at a time—often bike lanes go in when lane size is reduced and parking is adjusted, additional crossings may be accompanied by additional street lighting and amenities, or speed-limit reductions occur with complementary design changes and enforcement efforts. That said, metrics are still important to help a jurisdiction understand if a project is work- ing and to what extent. Metrics also serve as a powerful communication tool to help the public and decision makers better understand, advocate for, and continue efforts to impact speed and pedestrian safety.