Roadway Design Decisions and Animal-Vehicle Crashes

Keith K. Knapp, P.E., Ph.D.

Assistant Professor

University of Wisconsin - Madison

Engineering Professional Development

432 North Lake Street #713

Madison, WI 53706

Phone: 608-263-6314

Fax: 608-263-3160

Knapp1

Submitted on October 10, 2004

Word Count: 4,389 + 0 figures/tables = 4,389

ABSTRACT

Animal-vehicle crashes (AVCs) are a significant roadway safety problem throughout the world. In the United States (US), for example, it is estimated that more than a million deer-vehicle crashes occur each year, and that the cost of these crashes is over a billion US dollars. The magnitude of this safety problem can be positively and negatively influenced by a wide range of roadway planning, design, and maintenance decisions and agency policies. This connection, however, is rarely discussed or considered by roadway planners and geometric designers (unless an endangered species is involved). The purpose of this paper is to introduce and start a discussion about some of the planning, design, and maintenance decisions and/or policies that can impact the number of AVCs along a roadway. Some design decisions (and the policies that guide them) related to AVCs include posted speed limits, roadway curvature and cross section (e.g., number of lanes, median type and/or barriers, etc.), and the height, length, and location of bridges. Some jurisdictions have also developed and begun to use roadway planning/programming tools that assist in the general AVC impact estimation of roadway alignment locations. Maintenance activities (e.g., roadside vegetation and ice removal) also have potential AVC impacts, and are discussed in this paper. Overall, however, little quantitative knowledge exists about the individual or combined AVC impacts of roadway planning, design, and maintenance decisions. This is a gap in safety research that should be addressed.

INTRODUCTION

It has been estimated that more than a million deer-vehicle crashes (DVCs) occur each year in the United States (US), but that less than half of them are reported (1). These collisions are believed to cause more than one billion US dollars in property damage (1). It has also been suggested that the same number of vertebrate are killed along US roadways in a day. Animal-vehicle crashes (AVCs) are a high-cost safety problem that continues to increase in most areas of the US and throughout the world.

A more focused consideration of the wildlife habitat and movement impacts during the operation, design, and planning of roadway or roadway systems could lead to a reduction in the roadway animal mortality problem. However, this level of consideration in the US would generally need to involve an approach that builds upon the current activities that focus on roadway impacts to endangered or otherwise threatened species. This new approach would attempt to address the overall ecological impact of roadways on wildlife.

The most direct and obvious interaction between roadways and wildlife is animal mortality or AVCs. The purpose of this paper is to introduce and discuss a sample of the roadway maintenance, design, and planning decisions/choices/policies that might have an individual or combined impact on the number of AVCs that occur. The current state-of-the-knowledge with respect to the safety impacts of DVC or large animal collision countermeasures, however, and many of the studies referenced in this paper are also discussed in much more detailin the Deer-Vehicle Crash Countermeasures Toolbox: A Decision and Choice Resource at (2). A discussion of the wider range of the ecological impacts due to roadways (e.g., habitat fragmentation, reduced air quality, and increased noise and water runoff) is also found in the recently published Road Ecology: Science and Solutions (3). Road ecology is defined as the science that “…explores and addresses the relationship between the natural environment and the road system (3).”

There are a number of roadway maintenance, design, and planning choices that could have an impact on the number of AVCs. The focus of the choices discussed in this paper are listed below:

  1. Roadway Maintenance

Winter Maintenance

Roadside Vegetation Installation and Maintenance

Carcass Removal

  1. Roadway Design

Posted Speed Limit

Curvature

Cross Section

Bridge Height and Length

  1. Roadway Planning

Roadway Alignment Location

Project Programming

The “points of wildlife consideration” listed above range from very specific maintenance operational procedures and design choices to more general planning-level alignment and project programming decisions. The final section of this paper also includes some general suggestions about how agencies might develop more AVC sensitive roadways and roadway systems. In general, however, it should be recognized that the quantitative safety impacts of the specific activities and choices listed above have rarely been studied. This is a gap in the knowledge that needs to be addressed.

ROADWAY MAINTENANCE

Winter Maintenance

Once a roadway has been constructed there are several maintenance-related decisions that could impact the number of AVCs. It has been generally suggested, for example, that the deicing and/or anti-icing salt mixtures used to keep roadways clear of ice and snow may also attract animals and subsequently increase collisions. Only one study was found that attempted to consider the quantitative impacts of roadway salt on AVCs, and it focused on the patterns of moose-vehicle collisions near roadside pools with significant concentrations of salt (2, 4). This study found that approximately 43 percent of the moose-vehicle collisions in the study area occurred within 328.1 feet (100 meters) of a saltwater pool. However, about the same amount occurred more than 984.3 feet (300 meters) away from the pools (2, 4). The researchers concluded that the distribution of the observed moose-vehicle crashes near the roadside pools was much higher than what might randomly be expected. The assumption used in this comparison (i.e., all locations have an equal chance for a crash), however, is questionable and no comparisons were completed about how many moose-vehicle crashes might not have occurred if the saltwater pools (or the use of roadway salt) were eliminated or reduced. The potential AVC impacts could be one consideration in winter maintenance decisions.

Roadside Vegetation Installation and Maintenance

It has also been generally suggested that choices related to the selection (usually decided during roadway design) and maintenance of roadside vegetation may also impact AVCs. In other words, certain types of vegetation and methods of roadside mowing (e.g., how much and how often) may attract deer and other animals to the roadway. Two studies were found that somewhat show the AVC reduction potential of vegetation clearing (5, 6). In the first study the clearing of low vegetation within 65.6 feet (20 meters) of a roadway appeared to reduce moose-vehicle crashes by almost 20 percent, but this reduction was also close to the natural variability of the data (5). The large cost of this approach was also noted by the researchers (5). The second study evaluated a similar but more extensive removal of vegetation along railroads in Norway, and showed more than a 50 percent reduction in moose-train collisions (6). However, the amount of data in the study was limited and the individual segment AVC reduction results were highly variable (6). It was also recognized by the researchers that their experimental design could have resulted in an overstatement of the crash reductions from vegetation clearing.

A tool to help choose roadside vegetation was recently developed by the Minnesota Department of Transportation (7). Theobjective of the “Plant Selector” program is to help decision-makers make better roadside plant selections. The program allows the user to choose more then 25 site characteristics and about 20 plant characteristics. The criteria and characteristics that it contains which may be relevant to this paper include the ability of the user to indicate a low, medium, or high wildlife ranking as a plant characteristic, and to characterize the site as experiencing animal damage from deer, mice/voles, rabbits/hares, and gophers (7). The “Plant Selector” program is available for use at How a plant within the program was determined to be more or less susceptible to animal browsing does not appear to be documented.

Carcass Removal

Finally, another maintenance-related decision that may have an impact on general roadway safety (versus just AVCs) is the timing of large animal carcass removal. From a safety point-of-view the rapid removal of large animal carcasses from the roadway and roadside is a preferred approach for at least two reasons. One, large animal carcasses on the roadway and/or roadside of a high-speed roadway could (at least temporarily) be hazardous objects if hit by an errant run-off-the-road vehicle. The result of this type of collision could be the vehicle vaulting or rolling over. Two, large animal carcasses on the roadway can attract scavengers to the roadside to feed and this could result in a secondary animal-vehicle collision. Of course, decisions related to carcass removal also need to take into account, among other things, the probability of these types of events occurring and the general costs. In addition, the impact the carcasses may have on roadside maintenance equipment (e.g., mowers) and aesthetics are also considerations.

ROADWAY DESIGN

Several decisions are made during the geometric design and signing of new or reconstructed roadways that could impact AVCs. The decisions summarized in the following paragraphs are related to the AVC impacts of posted speed limit, roadway curvature and cross-section, and bridge design. The AVC impacts of installing (and maintaining) exclusionary ROW fencing, deer crossing warning signs, or other potential AVC countermeasures are discussed in more detail within the previously referenced DVC toolbox (2). Choices or decisions related to new roadway alignment locations and project programming, on the other hand, are discussed in the next section of this paper. Overall, each of these decisions requires a comparison of their potential cost, and the benefits they offer to wildlife and transportation safety.

Posted Speed Limits

Two studies were found that attempted to evaluate the potential AVC impacts of speed limit and/or speed limit reduction (8, 9). In both cases the researchers suggested that there was a relationship between AVCs and posted speed limits. In most instances their research results appear to show a less then expected number of AVCs along roadway segments with lower posted speed limits. To reach this conclusion, however, one study statistically compared the proportion of roadway mileage with a particular posted speed limit to the proportion of animals killed along those segments. The other study, on the other hand, compared the AVC frequency and rate per roadway length before and after a posted speed limit change.

In general, it was concluded that the design of the two “speed limit reduction” studies noted above limits their usefulness (8, 9). For example, a comparison of the proportion of AVCs to the proportion of roadway mileage with a particular posted speed limit assumes a uniform distribution of animal population, and also ignores any positive or negative relationships that might exist between/among roadway design, topography, posted speed limit, operating speed, and animal habitat. Effectively determining and defining a relationship (if any) between reduced posted speed limits (or operating speeds) and the number of AVCs will require additional study. Some of these factors may be considered in the work being completed for National Cooperative Highway Research Program (NCHRP) 25-27: “Evaluation of the Use and Effectiveness of Wildlife Crossings”.

Curvature and Cross Section

The relationships found between the location of DVCs or roadside carcasses and the roadway cross section factors have been mixed (8, 10-13). For example, Allen and McCullough found more roadside carcasses on two-lane than four-lane roadways, but Reilly and Green found an increase in roadside carcass after one roadway was widened (10, 11). A study in Canada, however, did not find any difference in the number of DVCs when a roadway was increased from an undivided two-lane to a divided four-lane cross section (12). The number of vehicle collisions with elk and big horn sheep, however, did increase along this same widened roadway segment (12). But, this same study did not find a larger number of elk-vehicle collisions along roadway segments with a concrete median barrier (12). Similarly, a Kansas study did find that roadways with grass medians had higher reported DVC rates than those with median barriers, and those with median barrier had a higher reported DVC rate than two-lane undivided roadways (13). Gunther, et al. believed that curvilinear design and narrow lanes could reduce vehicle speeds and subsequently decrease animal mortality (8).

Overall, it would seem that roadway curvature and cross-section design decisions do have some type of AVC impact, but more evaluation is clearly necessary. The crash impact of these decisions appears to be site and species specific. The variability of where individual animal species live and how they move and interact with roadways must be considered.

Bridge Height and Length

Bridge design decisions can also impact the number of AVCs that occur along a roadway. Animals will sometimes use roadway structures (i.e., underpasses and overpasses) if they are adequately sized, appropriately located, and not heavily used by vehicles and/or humans. In fact, crossings that are specifically designed (and combined with exclusionary fencing) for wildlife are in use throughout the world. The crash and/or roadside carcass reductions from specific fencing and wildlife crossings combinations have ranged from 60 to 97 percent (2). The specifics of these crossing studies are discussed in the toolbox previously referenced, and several additional crossing-focused references are identified (2).

The focus of this discussion is those crossings along existing roadways (over valleys, ravines, and/or watercourses) that might also represent locations where animals naturally travel. During roadway design or reconstruction the height and/or length of these bridges might be altered for their possible use by wildlife. For example, the slope walls or abutments of a bridge over a watercourse are often placed to minimize its length (and cost), but this type of design may also force the animals following a watercourse to cross on the roadway surface rather than under the bridge. A sufficient bridge length and height that provides level ground adjacent to a stream or river might reduce the occurrence of this movement. Of course, all these design decisions will also impact the cost of the bridge, and must be compared against the benefits they provide. No documentation on the AVC impact of these types of decisions was found.

ROADWAY PLANNING

Alignment Location and Project Programming

In some cases a new location must be chosen for all or part of a roadway alignment. At this stage in the development of a roadway or roadway system alignment alternatives are still being evaluated, and the prioritization of improvement projects still being determined. The characteristics of proposed project alignment alternatives are compared, shared with the public, and a preferred location chosen. The prioritization of these improvement projects is then often based on the significance of operational and/or safety concerns. Existing and/or expected roadway animal mortality impacts could be added to these planning-level evaluations, but this type of consideration also requires adequate AVC information and the capability to identify and predict existing and potential problem locations.

Several researchers have investigated and/or modeled the roadway and adjacent land use characteristics that seem to be correlated to a “high” number of AVCs, DVCs, or roadside carcasses (8-18). A sample of the factors contained in the models includes traffic volume, a measure of animal density, and habitat characteristics (e.g., land cover and topography). Some jurisdictions have also electronically combined, among other things, AVC and/or carcass data with landscape descriptors, species data, land uses, highway locations, expert opinion input, and wildlife habitat/connectivity/linkage information within a geographic information system (GIS) (19, 20, 21). These systems can be used to help identify existing or proposed “wildlife critical” roadway segments that may bisect species habitat or habitat linkages. Each state in the US now also has land cover (or habitat) maps electronically available. The general linkability of these maps with safety information within a GIS needs to further investigated and will vary from jurisdiction to jurisdiction. A few states and some countries have also begun or finished wildlife habitat or connectivity plans. All of this information could be combined in some manner and considered in the roadway alignment comparison and project prioritization process.