Driving After Dark, January/February 2003 Public Roads

Researchers at FHWA are striving to improve nighttime visibility, making roads safer for motorists and pedestrians.

It was 10:27 p.m., and John had been driving a winding rural road for hours. Although he had driven the road before and the traffic was light, the dark cloudy night and the pine trees flanking both sides of the pavement made it difficult to judge upcoming curves. As he approached one curve, he failed to notice a faded yellow-and-black sign until it was too late. Fortunately, his vehicle was going uphill, which moderated his speed, and the curve had a broad shoulder, so when his vehicle veered off the road, he was able to bring it to a stop just seconds before colliding with a tree. John was lucky.

The circumstances leading up to John's near-crash occur every night across the country. The challenges are greater than during the day because many of the visual cues necessary for safe driving, such as warning signs and pavement markings, are harder to see. As a matter of fact, approximately half of all fatalities due to motor vehicle crashes occur at night, even though the traffic volume is substantially less. According to data in Traffic Safety Facts 2000, published by the National Highway Traffic Safety Administration (NHTSA), the fatal crash rate for nighttime driving is three to four times that for daytime driving.

A large proportion of nighttime crashes are single-vehicle, run-off-the-road crashes, suggesting that a driver's inability to recognize delineation, guidance, or warning information may be a contributing factor. Lack of visibility may contribute to crashes involving pedestrians as well. In addition, adverse weather further complicates night driving by rendering pavement markings nearly invisible when covered with only a small amount of water, ice, or snow.

Research over the past two decades has led to a greater understanding of visibility requirements and spurred efforts to improve the night visibility of traffic control devices and the roadway environment in general. Despite the knowledge gained through research, encountering traffic signs or pavement markings that are barely visible at night is still not uncommon.

To provide State and local agencies with information on verifiable improvements that can save lives and reduce the cost of crashes to society, FHWA maintains an ongoing research program for improving the performance of retroreflective signs and markings, assessing the potential impact of new vehicle headlight technologies, analyzing the effectiveness of roadway lighting, and studying other aspects of human vision, visibility, and driver performance.

Crash statistics can shed some light on the impact that time of day has on driver risk. An analysis of the Fatality Analysis Reporting System (FARS) data for 2000 reveals that: (1) 49 percent of all fatal crashes occur at night; (2) 81 percent of fatal crashes occur on dry pavement, both day and night; (3) 40 percent of all fatal crashes involve alcohol as a factor, with more than 60 percent of those occurring at night; and (4) problems with driver vision, vehicle hardware, or environmental conditions are cited as "related factors" in 15 percent of all fatal crashes. Since more than half of all fatal crashes involve only a single vehicle, the final statistic may greatly underestimate the impact of vision and visibility on driving safety.

The General Estimates System (GES), compiled by NHTSA, expands the numbers in the FARS to provide a national database of all highway crashes. For example, in 2001 an estimated 6.3 million highway crashes occurred in the United States, of which roughly 4.3 million involved property damage only, 2.1 million resulted in some injury, and 37,795 resulted in one or more fatalities (with a total of 42,116 deaths). The GES data reveal that more than 70 percent of all crashes occur during daylight. Thus, even though the number of fatalities during the day and at night are about the same, the percentage of fatalities at night is higher because there are less crashes and fewer vehicles on the roads. The percentage of crashes resulting in injuries also is slightly higher at night than the overall percentage of nighttime crashes. These statistics indicate that, on average, nighttime crashes are more severe than those that occur during the day.

Based on estimates of vehicle miles traveled, the overall nighttime crash rate is approximately 1.6 times that of the daytime rate, while the fatal crash rate is three to four times greater at night. Factoring out alcohol-related crashes, the nighttime fatal crash rate is still nearly twice that of daytime. Although other human factors issues, such as fatigue, may account for some of the increased risk, it appears reasonable that visibility (or lack thereof) also contributes to nighttime crashes.

A single year's data provide only a snapshot of any given aspect of driving safety. An analysis of crash trends since 1975 shows an improvement in nighttime safety over time. Between 1975 and 2001, the number of fatal crashes during dark conditions declined steadily, especially compared with fatal crashes under other lighting conditions.

New, tougher attitudes toward drinking and driving, improved vehicle dynamics, greater use of seat belts, and the introduction of passive restraints are among the factors that are thought to have enhanced driving safety. However, the fact that the trends in fatal crashes under other lighting conditions do not parallel those under dark conditions suggests that changes in nighttime visibility also played a role in improving safety. Specifically, researchers point to the increased use of retroreflective materials in signs, delineators, and pavement markings, as well as the development of new headlight systems, as important contributing factors.

Since the mid-1990s, the rate of reduction in fatal crashes under dark conditions has slowed. This slowing in the rate indicates that further progress will require continuing efforts to enhance night visibility.

Traffic signs and pavement markings represent the first line of crucial information for drivers, both during the day and at night. For more than 40 years, the Manual on Uniform Traffic Control Devices (MUTCD) has required that signs and markings be reflectorized (containing treatments that reflect light) or illuminated (lit up by internal or external lights) to make them visible at night. The private sector developed and marketed a variety of sign materials that provide greater luminance under a given level of headlight illumination, and many agencies are using these improved materials.

Despite these efforts, many signs in the highway system still fail to meet the needs of drivers at night, so FHWA is working to establish minimum requirements for retroreflectivity—a measure of the amount of light returned to its source—for traffic signs and pavement markings.

A key facet of FHWA's effort to implement minimum retroreflectivity requirements is a close working relationship with the State and local officials who will be responsible for implementing any new requirements. FHWA recently sponsored a series of workshops to solicit feedback on the minimum retroreflectivity levels and the means to implement the requirements.

According to Gene Hawkins, division head of the Operations and Design Division of the Texas Transportation Institute, workshop participants indicated that agencies need the flexibility to choose an implementation method that best fits their resources and capabilities. "As the workshops progressed, we found that agencies wanted to have options other than measuring sign retroreflectivity as a means of providing reasonable nighttime visibility," Hawkins says. "Because of this, the focus shifted from maintaining minimum retroreflectivity values to providing reasonable nighttime visibility of signs. The implementation options we discussed in the workshop include visual inspections, expected sign life, measured sign retroreflectivity, inspection panels, and control signs."

FHWA also plans to develop guidelines for sign inspection, establish corollary requirements for sign color, and update guidelines for sign design and application, such as the Standard Highway Signs manual. Through these activities, FHWA intends to assist agencies in initiating practices that will assure the continued in-service performance of traffic signs.

Similarly, FHWA is developing minimum requirements for the retroreflectivity of pavement markings. FHWA proposed minimum levels and solicited feedback from participating agencies. One ongoing research project looks at the interactions among driver performance, pavement marking visibility, and the use of retroreflective raised pavement markers (RRPMs), which are seen by drivers as points of light reflected back from the road with their headlights. The purpose of the study is to determine if adding RRPMs will substantiate using lower levels of retroreflectivity in pavement markings, yet still provide the same level of safety. Another study is investigating the visibility of new pavement markings under wet conditions to develop data that can be used in benefit-cost analyses of the new materials.

FHWA and the Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, VA, recently completed a long-term research project on night driving and highway lighting requirements for older drivers. Using Virginia Tech's Smart Road—a stretch of road built for testing highway technologies—researchers established several lighting designs to investigate the effects of varying illuminance, luminance, and uniformity of lighting on the ability of older drivers to detect objects in the roadway. Researchers intend to establish a clear relationship between the various lighting design parameters and driver performance, which will lead to the installation of cost-effective lighting systems that provide the greatest overall roadway safety. The results of the research will be used to revise the recommendations in the American National Standards Institute publication, American National Standard for Roadway Lighting (ANSI/IESNA RP-8-00), as necessary.

A separate study to be conducted on the Smart Road in the coming year is an evaluation of crosswalk lighting designs. "The current crosswalk lighting recommendations are more than 20 years old," says Dr. Ronald B. Gibbons, a lighting research scientist at Virginia Tech. "So we're looking at new technologies to establish updated recommendations for vertical and horizontal crosswalk illumination."

Based on a scan tour of roadway lighting practices in Europe, the study will explore how simple changes in the design of crosswalk lighting, using existing roadway luminaires, can yield significant improvements in the visibility of pedestrians to drivers. According to Gibbons, one of the most important lessons researchers learned from the European tour was the value of lighting vertically, rather than putting more light into the area itself.

"What you'd like in a crosswalk is very little horizontal illumination, which is light shining straight down on you," Gibbons says. "You'd want to maximize vertical illumination, which will make the pedestrian appear bright against a dark background and provide the maximum contrast. What we're looking for is the compromise between horizontal and vertical because we're providing illumination from one light source."

A key difference between this study and previous research is that it is intended to develop crosswalk lighting guidelines that have minimal impact, or none, on implementation costs and maintenance operations.

A vehicle's headlights provide the primary illumination of the roadway scene for drivers. Although specifying the required performance of headlights is outside the purview of FHWA, evaluating the impact of vehicle headlights on driver performance is of great interest. FHWA and the Virginia Polytechnic Institute and State University are evaluating several new headlight technologies as part of ongoing research on enhanced night visibility.

The research team is developing a computer program to model the appearance of the roadway scene under different headlighting and pavement marking schemes. The modeling enables the researchers to do a rapid analysis of the impact of headlight technologies on the visibility of the roadway environment, including traffic signs, pavement markings, and pedestrians. The computer program also will evaluate the impact of glare from oncoming vehicles. An accurate, validated modeling tool will enable FHWA to provide meaningful input to NHTSA as it develops headlight requirements.

Improvements in roadway designs, barriers, vehicle dynamics, and the visibility of pavement markings and signage are reducing the risk of serious crashes on the Nation's roadways. Through continued research into technologies, materials, and behaviors that can enhance nighttime visibility, FHWA aims to build on past successes in reducing nighttime crashes and achieve verifiable improvements in roadway safety—saving money, and ultimately, saving lives.

Kenneth S. Opiela is a highway research engineer and member of the roadway team in FHWA's Office of Safety Research and Development at the Turner-Fairbank Highway Research Center (TFHRC) in McLean, VA. He holds a doctorate in civil engineering from the Wayne State University and is a registered professional engineer in Michigan.

Carl K. Andersen is the manager of the Photometric and Visibility Laboratory in the Office of Safety Research and Development at TFHRC. He holds a master's degree in physics from the Naval Postgraduate School, Monterey.

Greg Schertz is a safety engineer in FHWA's Central Federal Lands Highway Division in Denver. He has a bachelor's degree in civil engineering from Texas A&M University and is a registered professional engineer in Texas.

For more information about FHWA's activities to enhance nighttime visibility, contact one of the authors: Kenneth S. Opiela, 202-493-3371; Carl K. Andersen, 202-493-3366; or Greg Schertz, 303-716-2094.

January/February 2003

Driving After Dark by Kenneth S. Opiela, Carl K. Andersen and Greg Schertz

Driving After Dark
by Kenneth S. Opiela, Carl K. Andersen and Greg Schertz