Improving Pedestrian and Motorist Safety Along Light Rail Alignments (2009)

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.

12 As discussed in the introduction, LRT is increasing in popularity, and new systems are being introduced in many parts of North America. Although LRT collisions are rare, the safety issues associated with LRT collisions may be severe. LRT agencies are currently subject to the safety methods and standards mandated by the FTA. Chapter 8 of the Manual of Uniform Traffic Control Devices provides guidance for traffic control and signage issues, but the standards and guidance available do not address all the needs of LRT agencies in addressing safety issues. To fill the gaps, agencies have devel- oped their own strategies for improving safety on light rail alignments. The existing body of knowledge concerning LRT safety experience is based on both practical experience and formal research. Local agencies, the police, and other experts who work directly with existing systems have a wide variety of knowledge. Some of this knowledge is based on the available data. Some of it may be anecdotal, based on direct experience of “what works and what doesn’t.” In this project, local exper- tise has been recorded through stakeholder consultations and through site visits and workshops with LRT agencies, police, and related organizations. Where possible, the local expertise is supported and complemented by the available data. In addi- tion to the practical knowledge base rooted in local expertise, various institutions have conducted research into LRT safety related subjects. The research includes reviews and formal, statistical studies. This chapter presents a summary of the current research and state of the practice based on four main sources: 1. The extensive literature review completed in Phase I. The review assembled the documentation available on LRT safety issues and practices. 2. The survey of agencies completed in Phase I. A complete summary of the survey results can be found in Appendix C. 3. Telephone consultation with agencies and SSOs carried out as follow-up to the survey, and repeated in Phase II 4. Observations and additional information from consulta- tions during the site visits. Complete summaries of each site visit can be found in Appendix D. This chapter documents the literature review, and is divided into two sections. The Methodology section summarizes the methodology used to gather the literature and other source material. The State of the Practice section summarizes the find- ings of the literature review, and provides an overview of the current state of the practice. The complete results of the litera- ture review are available in Appendix B. Methodology Literature Review The literature review was designed to collect, review, and summarize published and unpublished information relevant to safety measures, devices, and practices on LRT alignments. The information was gathered from United States and foreign LRT systems. The safety issues considered included at-grade crossings, stations, and safety practices that enhance safety for pedestrians, motor vehicles, and LRT passengers. The full literature review (Appendix B) describes the application of the treatment, the measures taken to improve safety, the quantified safety impacts in terms of changes in the number and/or type of collisions (where available), and any potential caveats that could affect the transferability of the results to any other system. As quantified information was limited, anecdotal evidence of safety improvements and resulting changes to surrogate measures is also described in detail. To obtain the information required for the literature review, the project team searched the following databases: • Institute of Transportation Engineers (ITE) digital library, • Transportation Research Information Services (TRIS), C H A P T E R 2 State of the Practice Methodology and Summary

• International Road Research Database (IRRD), • Organization of Economic Cooperation and Development Library (OECD), • FTA publications, • NTD Safety and Security Reports, • Historical NTD information, • American Public Transportation Association publications (APTA), • Transportation Association of Canada (TAC) library cata- logue, • The European Commission’s Transport website (http://ec. europa.eu/transport/index_en.html), and • Personal and organization libraries of research team members. In addition to searching these sources, the research team attempted to obtain unpublished documents through con- tacts at various North American LRT systems, the FTA, the Transportation Research Board Committee on Light Rail Transit (AP075) and the APTA Rail Transit Standards Oper- ating Practices Committee. Although some contacts provided reports, the reports they provided had already been reviewed during the initial literature review. As a result, no unpub- lished documents were added to the material. The state of the practice summary documents the most sig- nificant information gathered during the literature review. Survey of Agencies The Survey of LRT Agencies was a single online survey that combined the requirements of Task 2 (LRT Collision Data Collection) with Task 4 (LRT Innovative Control Devices and Applications), Task 5 (Assessment of MUTCD Elements along LRT Alignments), and Task 6 (LRT and New Technologies). The survey was implemented online to provide easy access to the target agencies dispersed across the United States and Canada. The survey content and structure were designed to minimize the burden on participants. The survey included questions related to collision data, and questions designed to determine the availability of data. Topics included: • LRT and roadway characteristics, • Vehicular and pedestrian traffic volume data, • Details of 35 LRT treatments, • Observations of risky behavior or near misses between LRV and motorists and pedestrians (e.g., videotapes from CCTV cameras), • Inventory of treatments with dates of implementation, • Use of traffic control elements identified in Chapter 10 of the MUTCD, and • Use of new technologies. The survey questions were pilot tested for usability and clar- ity by having a remote participant (a staff member from the Utah Transit Authority) complete the survey using a talking out loud protocol while on the phone with the research team. The purpose of the pilot test was to provide sufficient insight to val- idate the usability of the survey and to determine the approxi- mate time required for completion. Following the user testing session, the survey was revised to improve the survey’s ease of use and clarity. The revised survey was submitted to the Panel for review on October 20, 2006. The final survey incorporating the Panel’s comments was launched on November 6, 2006. In addition to developing the survey, the project team iden- tified a survey contact at agencies across North America. At the request of the TCRP Panel, representatives from new LRT sys- tems that were about to become operational were also con- tacted. In total, 37 transit agencies were asked to participate in the survey. The location of the agencies contacted is shown in Figure 2, and the names of the systems are listed in Table 1. A list of stakeholders was developed. The list contained 86 contacts, and included at least two representatives from each of the 37 transit agencies shown in Table 1. It was important for every agency to have more than one contact in order to facilitate the collection of various types of sample data (e.g., collisions, volumes, geometric design, etc.). Before the repre- sentatives from the various transit agencies were invited to participate in the survey, the survey team made introductory phone calls to initiate contact. Every individual contact received an initial e-mail invitation and a supporting phone call at the start of the survey. The sur- vey was scheduled to run from the beginning of November until the end of December 2006, but due to a lack of response from the transit agencies during the early stages, the time period was extended until mid-January 2007. To maximize the sur- vey response rate, the research team made follow-up phone calls to encourage participation from agencies that had not completed or responded to the survey by the end of December. The phone calls to LRT agencies continued through early Jan- uary with the last survey response received on January 6, 2007. In total, survey responses were received from 24 different LRT agencies (Table 2). Telephone Consultations In Phase II of the project, the project team began making consultation calls with representatives from local agencies and SSO offices. A list of 22 SSO representatives and 32 local agency representatives was provided by FTA. The 54 represen- tatives were contacted between February and May 2008, and were asked about the following topics: • Data collection practices; • Relationship between the SSO and the local agencies over- seen by the SSO; 13

• Data transfer activities (between local agency, NTD, and SSO); • Availability of data for research purposes; and • Interest in participating in the upcoming site visits, where appropriate. In total, 14 consultations were completed with representatives from SSOs and local agencies. Consultations with SSOs focused on the SSO role and data transfer, while calls to agencies focused on procuring additional data, the form of their relationship with the SSO, and their willingness to participate in site visits. Site Visits From May to July of 2008, the project team conducted five site visits to LRT agencies across the United States. The site visits usually had two components: 1. A tour of the LRT system, accompanied by a knowledge- able agency representative where possible; and 2. A meeting with LRT agency staff and other stakeholders as available. The meetings ranged from informal one-on-one meetings to more formal stakeholder workshops, depend- ing on the agency’s preference and availability. The site visits gave the project team the opportunity to observe different safety treatments and problematic locations (as identified by the respective agencies) in situ. They also allowed the project team access to the knowledge and experi- ence of the experts who run these systems every day. The agencies chosen for the visits were selected from the 24 agencies that responded to the survey. An initial screening identified locations that had implemented a number of dif- ferent safety treatments. Different alignment types, histories of operation, and geographic locations were also targeted. Seven agencies were contacted directly to request site visits. The five agencies who responded and who were visited were: 1. Utah Transit Authority, Salt Lake City, Utah; 2. Metro Transit, Minneapolis, Minnesota; 3. Hudson-Bergen Light Rail Line, Hudson County, New Jersey; 4. San Francisco Municipal Railroad, San Francisco, Califor- nia; and 5. Santa Clara Valley Transportation Authority, Santa Clara County, California. A complete summary of the findings of the five site visits is included in Appendix D. 14 Figure 2. Location of LRT systems in the United States and Canada included in the survey.

15 Table 1. List of LRT systems surveyed. Map No. Locations System 1. Austin, TX ASG ( Capital Metropolitan Transportation Authority All Systems Go!) ( Proposed system with construction scheduled to start in summ er 2006. The system was tested in Spring 2009 and was not yet in service at the time of publication.) 2. Baltimore, MD MTA-MD (Maryland Transit Administration) 3. Boston, MA MBTA (Massachusetts Bay Transportation Authority) 4. Buffalo, NY NFTA (Niagara Frontier Transit Authority) 5. Ca md en, NJ NJT (New Jersey Transit – River LINE) 6. Cleveland, OH GCRTA (Greater Cleveland Regional Transit Authority) 7. Dallas, TX DART (Dallas Area Rapid Transit) 8. Denver, CO RTD (Regional Transit District) 9. Galveston, TX GIT(Galveston Island Transit) 10. Houston, TX Metro (Metropolitan Transit Authority of Harris County) 11. Jersey City, NJ NJT-HBLR (New Jersey Transit – Hudson-Bergen Light Rail) 12. Kenosha, WI KT (Kenosha Transit) 13. Los Angeles, CA LACMTA (Los Angeles County Metropolitan Transportation Authority) 14. Memphis, TN MATA (Memphis Area Transit Authority) 15. Minneapolis, MN MT (Metro Transit) 16. New Orleans, LA NORTA (New Orleans Regional Transit Authority) 17. Newark, NJ NJT-NCS (New Jersey Transit – Newark City Subway) 18. Norfolk, VA HRT (Ham pton Roads Transit) (Currently under construction. Expected completion date is in early 2009) 19. Philadelphia, PA SEPTA (Southeastern Pennsylvania Transportation Authority) 20. Pittsburgh, PA PAAC (Port Authority of Allegheny County) 21. Portland, OR TriMet (Portland TriMet) 22. Sacram ento, CA SRTD (Sacramento Regional Transit District) 23. Saint Louis, MO/IL BSDA (Bi-State Development Agency) 24. Salt Lake City, UT UTA (Utah Transit Authority) 25. 26. San Diego, CA SDTI (San Diego Trolley Inc.) NCTD (North County Transit District) 27. San Francisco, CA SF Muni (San Francisco Municipal Railway) 28. San Jose, CA SCVTA (Santa Clara Valley Transportation Authority) 29. Seattle, WA WFSC (King County Metro) 30. Tacoma, WA ST (Sound Transit, Link) 31. Tampa, FL HART (Hillsborough Area Regional Transit) 32. Calgary, Alberta C-Train 33. Edm onton, Alberta Edm onton Transit System 34. Ottawa, Ontario O-Train (diesel-powered pilot project) 35. Toronto, Ontario TTC (Toronto Transit Commission) Streetcars 36. Vancouver, British Columbia CLCO (Canada Line Rapid Transit Link – RAV Line) 37. Waterloo, Ontario Region of Waterloo (Waterloo LRT) (In the environmental assessm ent stag e ) San Diego, CA State of the Practice Summary TCRP Report 69 indicated that LRT systems in North America are generally safe. A crash at any given crossing is a rare event, but when a collision occurs at an LRT crossing, the outcome is often severe (1). The following sections provide information about current safety practices along the LRT alignment. LRT Exposure to Pedestrians and Motor Vehicles describes the influence of LRT exposure. The next section iden- tifies the top safety issues, and LRT Safety Treatments provides basic information about the types of treatments available to LRT agencies. LRT Exposure to Pedestrians and Motor Vehicles A number of TCRP research projects have discussed safety in the LRT alignment. TCRP Report 17 (TCRP Project A-05)

focused on light rail operating on-street at low to moderate speeds, while TCRP Report 69 (TCRP Project A-13) investi- gated light rail safety for operations in semi-exclusive rights-of- way at speeds greater than 55 km/h (35 mph). TCRP Project D-09 provided more general information about the design of transit vehicles and facilities, both bus and light rail. Several other studies of specific safety problems and treatments have been published by TCRP and other organizations. LRT alignments are typically categorized into three right-of- way types for planning purposes. These classifications are also used for operations and safety. TCRP Reports 17 and 69 and TCRP Project D-09 used this classification system to distin- guish between different types of light rail alignments. The clas- sification system is useful because the type of alignment and the resulting level of exposure to vehicles and/or pedestrians have significant safety implications. TCRP Report 17 provides this information about the three basic alignment classes (2): 1. Type a. Exclusive alignments use full grade separation of both motor vehicle and pedestrian crossing facilities. Exclusive alignments eliminate grade crossings and oper- ating conflicts, and maximize safety and operating speeds. 2. Type b. Semi-exclusive alignments keep the LRT apart from road vehicles and pedestrians, except where road vehicles and pedestrians intersect at an at-grade crossing. Operating speeds on segments that do not have automatic crossing gates are governed by vehicle speed limits on the streets or highways. On Type b segments where the right-of-way is fenced, operating speeds are maximized (based on geomet- ric limits), but these higher speeds are typically maintained only for short distances, often on segments between grade crossings. 3. Type c. Non-exclusive alignments allow for mixed flow operation with motor vehicles or pedestrians, resulting in higher levels of operating conflicts and lower-speed oper- ations. Non-exclusive alignments are often found in down- town areas where there is a willingness to forgo operating speeds in order to access areas with high population den- sity and many potential riders. Table 3 summarizes the alignment classification set out in TCRP Report 69 (1). That report provides more detailed descriptions of each subcategory. TCRP Report 17 considered the safety issues and implica- tions associated with the different types of light rail alignment, 16 Map No. Locations System 2. Baltimore, MD MTA-MD (Maryland Transit Administration) 5. Camden, NJ NJT (New Jersey Transit – River LINE) 8. Denver, CO RTD (Regional Transit District) 10. Houston, TX Metro (Metropolitan Transit Authority of Harris County) 11. Jersey City, NJ NJT-HBLR (New Jersey Transit – Hudson-Bergen Light Rail) 12. Kenosha, WI KT (Kenosha Transit) 13. Los Angeles, CA LACMTA (Los Angeles County Metropolitan Transportation Authority) 14. Memphis, TN MATA (Memphis Area Transit Authority) 15. Minneapolis, MN MT (Metro Transit) 19. Philadelphia, PA SEPTA (Southeastern Pennsylvania Transportation Authority) 20. Pittsburgh, PA PAAC (Port Authority of Allegheny County) 21. Portland, OR TriMet (Portland TriMet) 22. Sacramento, CA SRTD (Sacramento Regional Transit District) 23. Saint Louis, MO/IL BSDA (Bi-State Development Agency) 24. Salt Lake City, UT UTA (Utah Transit Authority) 25. 26. San Diego, CA San Diego, CA SDTI (San Diego Trolley Inc.) NCTD (North County Transit District) 27. San Francisco, CA SF Muni (San Francisco Municipal Railway) 28. San Jose, CA SCVTA (Santa Clara Valley Transportation Authority) 29. Seattle, WA WFSC (King County Metro) 30. Tacoma, WA ST (Sound Transit, Link) 32. Calgary, Alberta C-Train 33. Edmonton, Alberta Edmonton Transit System 35. Toronto, Ontario TTC (Toronto Transit Commission) Streetcars Table 2. List of LRT agencies responding to the online survey.

and suggested the following sequence for route alignment choices in order of desirability (2): • Exclusive alignment (Type a), • Separate right-of-way (Type b.1), • Median alignment protected by barrier curbs and/or fences (Types b.2 and b.3), • Median alignment protected by mountable curbs and striping (Type b.4), • Operation in reserved transit malls or pedestrian areas (Types b.5, c.2, and c.3), and • Operation in mixed traffic (Type c.1). After considering safety, some additional issues can also be addressed. For example, Type a alignments, where the LRT is completely separated from the road and pedestrian network, allow LRVs to reach high speeds, but may be difficult for rid- ers to access from surrounding areas. These types of alignment are most often served by park-and-ride lots or other transit modes. Type b and Type c alignments create more exposure to safety issues, but they offer the advantage of providing more direct access to a variety of land uses (3). This report is concerned with the interactions of pedestri- ans and motor vehicles with LRT alignments, and addresses both Type b and Type c alignments. It does not address Type a alignments as Type a alignments are designed to eliminate pedestrians and motor vehicle interactions, except in unusual or extraordinary circumstances (e.g., trespassing). In all the various systems visited, the LRT staff noted con- siderably different operating behaviors between the down- town street-running sections and the more suburban restricted right-of-way (ROW) sections. In at least one case, LRT staff commented that transit operators felt that they noticeably relaxed when entering a restricted ROW section after navigat- ing a much more complex mixed traffic median-running environment. Many agencies reported that left-hand turn collisions are a significant issue. These collisions are avoided in areas where left hand turns are physically restricted. Phys- ical separation of general traffic from the LRT remains the best way to prevent collisions, but it is not always the most efficient or desirable from a cost or ridership perspective. This is because the needs of other modes and the need for access must be considered when designing a light rail system. Phys- ical separation, where feasible, can include a combination of separate alignments, grade separation at intersections, and fencing/barrier systems. Top LRT Safety Issues The objective of the literature review and the consultations with LRT operators was to identify the most significant safety issues along LRT alignments. Identification of safety issues is a useful step in the selection of safety treatments, as it permits the selection of specific treatment(s) for a problem rather than general or default measures. The first three lists of safety issues presented in this section summarize the main LRT safety issues according to TCRP Report 17, TCRP Report 69, and the site visits conducted for this project. The lists are extensive because of the need to recognize a wide variety of different alignments and local considerations, and because much of the information is anecdotal rather than quantitative in origin. The lists have no statistical significance, but provide a broad view of the types of issues that LRT agencies and SSOs are facing and trying to mitigate. The fourth list condenses the three lists into the top five safety areas facing LRT agencies. TCRP Report 17 (2) investigated 10 transit agencies with operating speeds of less than 35 mph (55 km/h). The authors identified some common safety-related problems faced by 17 Class Category Description of Access Control Exclusive Type a Fully grade separated or at-grade without crossings Type b.1 Separate right-of-way Type b.2 Shared right-of-way, protected by barrier curbs and fences (or other substantial barriers) Type b.3 Shared right-of-way, protected by barrier curbs Type b.4 Shared right-of-way, protected by mountable curbs, striping and/or lane designation Semi-exclusive Type b.5 LRT/pedestrian mall adjacent to parallel roadway Type c.1 Mixed traffic operation Type c.2 Transit-only mall Non-exclusive Type c.3 LRT/pedestrian mall Source: TCRP Report 69 (1) Table 3. LRT alignment classification.

LRT agencies through agency interviews, collision analysis, and field surveys. These were: • Pedestrian safety: – Trespassing on tracks – Jaywalking – Station and/or cross-street access • Side-running alignment • Vehicles operating parallel to LRT’s ROW turning left across tracks: – Illegal left turns – LRV pre-emption violating motorists’ expectation of protected left-turn signal phases • Traffic control: – Passive turn restriction sign violations – Active turn restriction sign violations – Confusing traffic signal displays – Poor delineation of dynamic envelope • Motor vehicles on tracks • Crossing safety (right-angle crashes) • Poor intersection geometry TCRP Report 69 (1) investigated 11 LRT lines with operat- ing speeds greater than 35 mph (55 km/h). The common safety-related problems identified were: • System division: – Vehicles drive around closed automatic gates – LRV operator cannot visually confirm whether gates are working – Slow trains share tracks/crossings with faster LRVs and near-side LRT station stops – Motorists disregard regulatory signs at LRT crossings and grade crossing warning devices – Motor vehicles queue back across LRT tracks from a nearby intersection controlled by STOP signs (MUTCD R1-1) – Sight distances are limited at LRT crossings – Motor vehicles queue across LRT tracks from down- stream obstruction – Automatic gate and traffic signal interconnect mal- functions • System operations: – Freight line is shared with LRT – Freight line was converted to LRT line – Collisions occur when second LRV approaches pedestrian crossing – Motorists disregard grade crossing warning devices • Traffic signal placement and operation: – Motorists confused about apparently conflicting flashing light signal and traffic signal indications – Track clearance phasing – Excessive queuing near LRT crossings – Turning vehicles hesitate during track clearance interval – Vehicles queue back from closed gates into intersection – LRT crosses two approaches to a signalized intersection (diagonal crossing) – Motorist confused about gates starting to go up and then lowering for a second time, when a second LRV arrives from the opposite direction – LRT versus emergency vehicle pre-emption – Turning motorists violate red protected left-turn indi- cation due to excessive delay – With leading left-turn phasing, motorists violate red pro- tected left-turn arrow moving on the green phase when trains also turn • Automatic gate placement: – At angled crossings or for turning traffic, gates descend on top of or behind motor vehicles • Pedestrian control: – Limited sight distance at pedestrian crossing – Pedestrians dart across LRT tracks without looking The five agencies consulted during the site visits for this project noted a variety of safety issues. The issues cited by these agencies as being the most important safety concerns are: • Jaywalking between marked crossing locations (i.e., mid- block, at stations, etc.); • Trespassing at stadium stations after events; • Pedestrians crossing against signals and/or against warning devices; • Pedestrian collisions due to a “second train”; • Pedestrian inattention and/or distraction; • Increased severity of pedestrian collisions; • Pinch points on platforms; • Risky behavior by cyclists; • Vehicles trapped inside gates; • Vehicles crossing tracks despite gates, signals, and/or warnings; • Vehicles stopped on tracks due to queuing in peak traffic periods; • Collisions in left turn lanes shared with LRT tracks; • Left turn collisions, especially where the LRT operates in the center alignment; • Right turn collisions, including collisions that occur on unusual alignments or where right turn on red is prohibited; • Sideswipes on Type b.4 and c.1 alignments; • Motorist confusion such as driving on restricted ROW; • The public’s level of respect for LRV is less than that for heavy rail; • System inconsistencies that impact motorist and pedestrian expectations; and • LRV operator error. 18

Every LRT system operates in its own unique context. In gen- eral, however, different categories of rail transit alignment have different types of safety issues. For example, LRT alignments with lower operating speeds (less than 35 mph (55 km/h)) gen- erally have a higher level of interaction between LRVs and pedestrians, cyclists, and motorists. Warning systems and traf- fic control devices for LRT crossings vary between lower and higher operating speeds, and on different sections of the same system. These differences are likely to be reflected in the safety issues experienced. The project team summarized the above issues into five top areas of safety concern that must be addressed along LRT alignments: 1. Motorist, cyclist, and pedestrian inattention; 2. Motorist, cyclist, and pedestrian confusion; 3. Lack of appropriate physical separation between motorists, cyclists, pedestrians, and the LRV; 4. Risky behavior by motorists and pedestrians; and 5. Operator error or lack of information. The five top areas of safety concern were common themes noted in almost all communications with LRT agency staff, and should serve as a basic checklist for addressing safety problems. As sufficient data are not available to determine the relative importance of the safety concerns, it is not possible to indicate which are the most important. It is also worth noting that the concerns are not necessarily independent. For example, during the site visits, pedestrians were observed to use marked and controlled crossings far more often where there was some phys- ical barrier to direct them. It is clear that motorist- and pedestrian-related behaviors are challenges in LRT safety. LRT agencies can deal directly with operator error or lack of information, but they do not have direct influence over motorists and pedestrians. At the macro level, it may be impossible or impractical to provide complete physical separation between LRT and other modes for many LRT systems. LRT agencies and the broader transportation planning authorities must balance the risk of collisions with other needs and considerations for the greatest overall value. At the micro level, there must be a distinction between pre- venting risky behavior due to inattention and confusion, and preventing risky behavior due to purposeful noncompliance. The agencies visited were clear that purposeful noncompliance can be difficult, if not impossible, to prevent. LRT Safety Treatments LRT safety treatments can be divided into two major cate- gories: physical improvements (including traffic controls) to the immediate environment surrounding the LRT, and educa- tion and enforcement programs for LRT staff and the commu- nity that live or work in the LRT area. LRT safety treatments can be applied system-wide or to specific locations. In many cases, individual treatments are not applied in isolation, but are applied as part of an integrated treatment package. A package of treatments can be effective, as some safety issues cannot be addressed by a single treatment alone, but when a package of treatments is applied, it may be difficult to discern which ele- ment of a package has the most effect on safety. This section provides an overview of physical improvements and programs within the community. Additional information about safety treatments is presented in Chapter 4. Detailed information about specific treatments is presented in Chap- ter 5 and Appendix A, the Catalog of Safety Treatments. Physical Improvements Physical improvements, including traffic control improve- ments, can be divided into two overall types: active and passive measures. Passive measures do not change with the approach of the LRV, whereas active measures react when an LRV approaches. A passive physical treatment might be a warning for pedestrians or motorists about the presence of an LRV. An active physical treatment might be a device that physically pre- vents pedestrians and vehicles from entering the ROW. While passive measures have the advantage of simplicity (for exam- ple, they cannot fail electrically or mechanically), the change that occurs in an active device has the effect of generating atten- tion from the intended audience of motorists, pedestrians, and cyclists. This may add considerably to the safety benefit of the basic message. During the site visits, the project team observed that well-designed active measures appeared more effective than passive measures, and this was also noted in the feedback from agency staff. Active treatments that were not well tuned to their envi- ronment lost impact. For example, in one location the site visit team observed a pedestrian crossing with flashing lights and bells that rang for many seconds longer than necessary and consequently seemed to be ignored by virtually all pedes- trians in the vicinity who crossed the tracks regardless of the warning. The message was certainly clear but the information was treated as incorrect by the pedestrians. However if the bell would start ringing much closer to the arrival time of the LRV, people might take it more seriously. In another location observed, an active second train warning sign had poor con- trast and was essentially unreadable in daylight conditions, so the message was not effectively delivered. Education and Enforcement Programs The second major category of treatments comprises pro- grams within the agency and community. These programs include education and enforcement activities. It is common for 19

agencies to provide safety training programs for LRV operators. Other programs focus on educating motorists, pedestrians, and cyclists about safety near LRT alignments through signage on platforms, educational brochures, and web pages on safety. Community programs have been implemented across the country, many in cooperation with Operation Lifesaver’s Light Rail branch (Operation Lifesaver, www.oli.org). The programs target children and/or adults through different styles and modes of message delivery and different key safety messages. Some jurisdictions have promoted the inclusion of light rail safety information in their state driver’s handbooks. Local enforcement agencies may be involved by means of issuing citations for infringement of the LRT’s ROW and other unsafe behavior. Some agencies are also using speakers at their sta- tions and in the LRV to broadcast public safety messages. Summary The existing body of knowledge regarding LRT safety expe- rience includes both formal research and practical experience. Following an extensive literature review, a survey of local tran- sit agencies, follow-up consultations with local transit agencies, and information gathered during site visits, the project team was able to identify the five most critical areas of safety concern that must be addressed along LRT alignments. These safety concerns are inattention of those approaching the LRT align- ment; confusion of those approaching the LRT alignment; lack of appropriate separation between motorists, cyclists, pedestri- ans, and the LRV; risky behavior by those approaching the LRT alignment; and LRV operator error or lack of information. The statistical information available is insufficient to determine the relative importance of these safety concerns, or the nature of any interrelationships between them. LRT safety treatments can be divided into two major cate- gories: physical improvements to the immediate environment surrounding the LRT, and education and enforcement pro- grams. Appendix A provides a detailed catalog of LRT safety treatments. Physical improvements can be further categorized into pas- sive measures and active measures. In general, well-designed active measures (including traffic controls) that suited their environment were more effective in increasing LRT safety than were passive measures. A wide variety of education and enforcement programs are available, but the effectiveness of these programs is difficult to quantify. While it is possible to implement measures to prevent risky behavior resulting from inattention or confusion, it is virtually impossible to mitigate the impact of deliberate noncompliance. 20