Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
1
SUMMARY
Evaluation of the Use and Effectiveness
of Wildlife Crossings
Introduction
Efforts at concerted and purposeful activity towards linking transportation services and
ecological services into a context-sensitive planning, construction, and monitoring process
have increased dramatically during the past few years. As a result, once piecemeal and hap-
hazard mitigation approaches have been replaced with much more integrated efforts that
have provided useful data to highway planners and engineers. The research team for this proj-
ect, NCHRP 25-27, "Evaluation of the Use and Effectiveness of Wildlife Crossings," was
charged to provide guidance in the form of clearly written guidelines for the selection,
configuration, and location of crossing types, as well as suggestions for the monitoring and
evaluation of crossing effectiveness, and the maintenance of crossings. Providing guidance
on the use and effectiveness of wildlife crossings to mitigate habitat fragmentation and reduce
the number of wildlifevehicle collisions involves thinking in a large-scale, context-sensitive
framework that is based on sound ecological principles. Landscape permeability (i.e., the abil-
ity for species to move freely across the landscape) is the guiding principle for this work, and
the foundation for effective mitigation. The goal for this research project was to develop
sound guidelines based on the premise that understanding and establishing landscape per-
meability leads to effective landscape connectivity and the restoration of ecosystem integrity.
At the same time, the guidelines should allow for efficient and effective transportation infra-
structure mitigation in a cost-effective, economic manner. The guidelines were developed in
the form of a final report and a web-based interactive decision guide.
As a convention, the term "wildlifevehicle collision" (WVC) is used rather than "animal
vehicle collision" (AVC), because this report is specifically dealing with only wildlife species
and not domestic animals or livestock that may be hit by vehicles on the road. All other per-
mutations, including for example, "wildlife crashes," "WVC carcass collection," and "wildlife
collisions," are used rather than the more generic word "animal." In Section 3.2, the term
"ungulatevehicle collision" is used because the data involved only hooved wildlife.
The research to accomplish the charge was carried out in two phases. In Phase 1, two
research efforts were completed. The first involved a North American telephone survey on
the state of the practice and research of wildlife crossings. The research team was able to doc-
ument virtually all of the wildlife and aquatic crossings in the United States and Canada and
to assemble information on them. The research team also reviewed studies that assess the
efficacy of crossings and in doing so, learned what was working well. The second study of
Phase 1 was aimed at creating a continent-wide list of priority actions needed for practice
and research. The final list of top-ranked priorities was the result of the participation of
approximately 444 professionals from across North America. In Phase 2, the research team
conducted five research efforts: (1) a safety research analysis of WVCs that included the
OCR for page 2
2
development of Safety Performance Functions and an analysis of differences obtained when
using WVC data versus deer carcass data, (2) an accuracy modeling effort that involved the
relative importance of spatially accurate data, (3) an analysis that investigated the usefulness
of different kinds of clustering techniques to detect hotspots of wildlife killed on roads, (4) a
field study of small mammals conducted in Utah and British Columbia that investigated the
putative habitat degradation effects of roads, and (5) an investigation into allometric meth-
ods to effectively place wildlife crossings to increase habitat permeability. Both Phase 1 and
2 efforts provide linked and important data that were used to develop the web-based inter-
active guidelines to inform decisions concerning wildlife crossings.
Clearly, transportation departments need reliable methods to identify WVC locations and to
identify potential mitigation measures, their placement, and their efficacy. The research devel-
oped in this NCHRP project addressed all of these. There are serious methodological problems
associated with current WVC research, so creating solutions requires the use of state-of-the-art
methods, such as predictive negative binomial models and empirical Bayes procedures. These
statistical methods can help to produce a widely accepted and usable guide that can be readily
applied by Departments of Transportation. However, the choice of which database to use (e.g.,
WVCs or carcass collection of wildlife road kills) to evaluate the WVC problem almost always
leads to the identification of different "hotspot" locations and ultimately different counter-
measure improvements because (1) reported WVC data may represent only a small portion of
the larger number of WVCs that occur 61,201 and (2) the spatial location accuracy of the datasets
can influence the validity of WVC models. The identification of collision-prone locations from
model results is one step in the location of appropriate wildlife crossings.
To better identify potential mitigation measures for wildlife along transportation corridors,
it is necessary to identify not only collision-prone zones, but also areas where landscape
permeability can be addressed for suites of species. Although crossings may be constructed
based in part on the WVC models and provide some measure of connectivity, landscape per-
meability as experienced by the animal may not be achieved because of differences in move-
ment ability among species. The allometric relationship between dispersal distances and
home range size of mammalian species can assist in deciding on the placement of wildlife
crossings that will help restore landscape permeability across fragmented habitat networks.
The placement of wildlife crossings--in accordance with the movement needs of suites of
species, when used with additional information regarding hotspots of animalvehicle colli-
sions as well as dead animal counts on roads, along with appropriate auxiliary mitigation such
as exclusion fences and right-of-way escape structures--should significantly improve road
safety as well as provide for easier movement of wildlife across the roaded landscape.
Even when wildlife crossings are appropriately placed, it is possible that road effects may
include habitat loss or degradation at some distance from the road, even though the roaded
landscape is permeable. It is necessary that mitigation efforts be evaluated for not only their
efficacy in reducing WVCs but also their ability in allowing multiple species to move across
the roaded landscape, thus promoting permeability.
The seven research efforts conducted in Phases 1 and 2 as part of NCHRP 25-27 addressed
these issues and provided usable data that helped in the development of the decision guide.
The following sections provide the essential findings from each of the research efforts.
Phase 1 Results
Literature Review
The research team searched the literature pertaining to wildlife and roads and wildlife
vehicle collisions. The references were entered into the online database of literature for this
project. The majority of references are annotated with key words and a description of the
OCR for page 3
3
research methods and results. There are more than 370 references in the database. URL ad-
dresses for papers and reports that are posted on the internet are provided with the citations
in order to provide users maximum access to the literature. These references are accessible
from the search engine page of the Wildlife and Roads website (www.wildlifeandroads.org).
Wildlife Crossings Telephone Survey
The wildlife crossings research reported here is a summation of the North American tele-
phone survey conducted to document as many known wildlife passages as possible in the United
States and Canada. The telephone survey included participants employed by state/provincial
and federal agencies, private organizations and companies, and academic institutions. More
than 410 respondents answered questions concerning wildlife crossings, planning for wildlife
and ecosystems; WVC information; and past, current, and future research activities related to
roads and wildlife.
The survey revealed 684 (663 U.S., 121 Canada) terrestrial and more than 10,692 (692+
U.S., 10,000+ Canada) aquatic crossings in North America. These passages are found in 43
of the United States and in 10 Canadian provinces and two territories. Trends found in the
practice of wildlife crossings included an increase in the number of target species in mitiga-
tion projects, increasing numbers of endangered species as target species for mitigation,
increasing involvement of municipal and state agencies, increasing placement of accompa-
nying structures such as fencing and escape jump-out ramps, and a continent-wide neglect
in maintenance of these structures. The trends in the science related to wildlife passages
included a greater tendency to monitor new passages for efficacy, a broadening of the num-
ber of species studied, an increase in the length of monitoring time, increases in the number
of scientific partners conducting wildlife passage research, and increasingly sophisticated
research technology. The research team documented several projects in North America where
a series of crossings have been, or will be, installed to accommodate a suite of species and their
movement needs, thus promoting permeability. A list of recommendations is presented to
assist in the research, design, placement, monitoring, and maintenance of crossings. As an
extension of the evaluation of the state of the science of wildlife crossings, the research team
reviewed studies that evaluated the use of wildlife passages. Approximately 25 scientific stud-
ies assessed the efficacy of 70 terrestrial wildlife passages across North America and found that
all crossings passed wildlife; 68 passed the target wildlife species.
Gaps and Priorities
The research team developed a list of priorities related to wildlife and roadways and
ranked them based on the results of a web-based survey of U.S. and Canadian professionals
involved in transportation ecology. Initially, the research team developed a list of priorities
based on its knowledge of current research and practices in road safety and ecology. The pri-
orities were developed and ranked to help direct research, policy, and management actions
across North America that addressed the issue of reducing the impacts of the roaded land-
scape on wildlife and ecosystem processes. The research team asked ecologists, engineers,
and road-related professionals across North America to rank these priorities. The objective
was to determine where additional research, field evaluations, and policy actions were
needed to help maintain or restore landscape connectivity and permeability for wildlife
across transportation corridors, while also minimizing wildlifevehicle collisions. The list of
priorities was initially reviewed and annotated by dozens of practitioners and researchers in
North America and then ranked and annotated in surveys by persons attending two work-
shops. The survey was refined and posted on the internet in April 2006, and potential
OCR for page 4
4
participants were invited to complete the survey by rating priorities. They were also asked
to notify other qualified transportation and ecology professionals and invite them to take
the survey. The final list of ranked priorities was the result of the participation of 444
professionals from across North America. The top five priorities were:
1. Incorporate wildlife mitigation needs early in the Department of Transportation
(DOT)/Ministry of Transportation (MoT) programming, planning, and design process;
2. Better understand the dynamics of animal use of mitigation structures (e.g., what works
and what does not) and disseminate this information;
3. Combine several integrated animal-friendly mitigation methods such as wildlife cross-
ings, fences, and escape ramps rather than relying on just one method;
4. Use conservation plans and connectivity analyses to inform the transportation pro-
gramming/planning/design process on where mitigation is needed and how it may be
carried out; and
5. Develop alternative cost-effective wildlife crossing designs and the principles upon which
they are based.
Phase 2 Research Studies
Safety
The safety research involved analyses of WVC and road environment data from state DOT
sources. Data were analyzed in two parts. In the first part, safety performance functions (SPFs)
were calibrated for data on AVCs and road and traffic variables from four states; SPFs are
predictive models for WVCs that relate police-reported WVCs to traffic volume and road en-
vironment data (geometrics) usually available in DOT databases. For each state, three levels of
SPFs were developed with varying data requirements. The first level required only the length
and annual average daily traffic volume (AADT) of a road segment (a section of road, gener-
ally between significant intersections and having essentially common geometric characteris-
tics). The second level required road segments to be classified as flat, rolling, or mountainous
terrain. The third level SPFs included additional roadway variables such as average lane width.
SPF functions relate police-reported AVCs to traffic volume and road environment data
usually available in DOT databases. (Police-reported AVCs include domestic animals as well
as wildlife, hence the use of the term "AVC" to characterize these reports. Only WVCs were
used in the analyses.)
Three SPF applications most relevant to the development of the desired guidelines for this
project are included in this report: (1) network screening to identify roadway segments that
may be good candidates for WVC countermeasures, (2) the evaluation of the effectiveness
of implemented countermeasures, and (3) methodology for estimating the effectiveness of
potential countermeasures. In general, the calibrated SPFs make good intuitive sense in that
the sign, and to some extent the magnitude, of the estimated coefficients and exponents are
in accord with expectations.
Surprisingly, the exponent of the AADT term, although reasonably consistent for the
three levels of models in a state, varied considerably across states and across facility types,
reflecting differences in traffic operating conditions. The most significant variable found was
AADT. For application in another state, or even for application in the same four states for
different years to those in the calibration data, model recalibration is necessary to reflect dif-
ferences across time and space for factors such as collision reporting practices, weather,
driver demographics, and wildlife movements. In essence, a multiplier is estimated to reflect
these differences by first using the models to predict the number of collisions for a sample
OCR for page 5
5
of sites for the new state or time period. The sum of the collisions for those sites is divided
by the sum of the model predictions to derive the multiplier.
In deciding which of the four models is best to adopt for another state, it is necessary to
conduct goodness-of-fit tests. Choosing the most appropriate model is especially important
because the exponents for AADT, by far the most dominant variable, differ so much between
states. A discussion of these tests is provided in a recent FHWA report.61 Additional sup-
porting information is presented in the appendices.
The second part of the research effort involved an evaluation of the hypothesis that the
magnitude and patterns of reported WVC data differed from the magnitude and patterns of
deer carcass removal data as they typically exist at a DOT. These two types of data have been
used in the past, but their differences could lead to varying and possibly ineffective/ineffi-
cient WVC-related policy and countermeasure decision making. Reported AVCs (which
typically are provided by state highway safety enforcement agencies in crash reports) and
deer carcass removal locations (which are provided by highway maintenance crews in their
daily activity reports) were acquired from Iowa and plotted within a geographic informa-
tion system (GIS) platform. The spatial patterns of the two types of data were clearly differ-
ent, and their calculated safety measures (e.g., average frequencies) varied. The use of the
GIS plots, safety measures, or predictive models developed as part of this project could,
therefore, lead to different WVC-related policies and countermeasure implementation and
evaluation decisions. The choice of the database used to define and evaluate the WVC prob-
lem and its potential countermeasures should be considered carefully. Recommendations
are provided regarding how the databases might be used appropriately and how the data
would be most profitably collected.
Accuracy Modeling
The accuracy modeling involved an investigation into the relative importance of factors
associated with wildlife killed on the road. Two different datasets were used: one based on
high-resolution, spatially accurate location data for carcasses along the roadside (<3 m error)
representing an ideal situation and a second dataset created from the first that was charac-
terized by lower resolution (high spatial error: 0.5 mi or 800 m, i.e., mile-marker data) and
is likely typical of most transportation agency data. The goal of this research was to sum-
marize how well these models identify landscape- and road-geometricsbased causes of
WVCs.
In this research, ungulate carcass datasets were used; the primary results of the analyses were
ungulatevehicle collision (UVC) models. The high-resolution, spatially accurate model had
higher predictive power in identifying factors that contributed to collisions than the lower res-
olution model based on mile-marker locations. Perhaps more noteworthy from this exercise
was the vast difference in predictive ability between the models developed with spatially accu-
rate data versus the less accurate data obtained from referencing UVCs to a mile-marker
system. Besides learning about the parameters that contribute to UVCs in the study area, the
research team discovered that spatially accurate data do make a difference in the ability of
models to provide not just statistically significant results, but more importantly, biologically
meaningful results for transportation and resource managers responsible for reducing UVCs
and improving motorist safety. The results have important implications for transportation
agencies that may be analyzing data that have been referenced to a mile-marker system or
unknowingly analyzing data that are spatially inaccurate. These findings lend support to the
development of a national standard for the recording of WVCs and carcass locations, as well
as further research into new technologies that will enable transportation agencies to collect
data that are more accurate. Use of personal data assistants (PDAs) in combination with a
OCR for page 6
6
global positioning system (GPS) for routine highway maintenance activities126 can help agen-
cies collect more spatially accurate and standardized data that will eventually lead to more in-
formed analyses for transportation decision making.
This project also investigated the types of variables that explain WVCs, in particular
whether they are associated with landscape and habitat characteristics or physical features
of the road itself. In two different types of analyses, the research team identified that vari-
ables related to landscape and habitat were more significant than variables related to road
characteristics. Through this project, the research team demonstrated how WVC data can
be used to aid transportation management decision making and mitigation planning for
wildlife.
Hotspot Modeling
The hotspot analysis used carcass data from wildlife killed on roads to investigate several
hotspot identification clustering techniques within a GIS framework that can be used in a
variety of landscapes. These techniques take into account different scales of application and
transportation management concerns such as motorist safety and endangered species man-
agement. Wildlife carcass datasets were obtained from two locations in North America with
different wildlife communities, landscapes, and transport planning issues. The research team
demonstrated how this information can be used to identify WVC hotspots at different scales
of application, from project-level to state-level analysis. Some clustering techniques that
were tested included Ripley's K-statistic of roadkills, nearest neighbor measurements, and
density measures. An overview of software applications that facilitate these types of analy-
ses is provided.
In summary, data on hotspots of WVCs can aid transportation managers to increase
motorist safety or habitat connectivity for wildlife by providing safe passage across busy
roadways. Knowledge of the geographic location and severity of WVCs is a prerequisite
for devising mitigation schemes that can be incorporated into future infrastructure proj-
ects (bridge reconstruction, highway expansion). Hotspots in proximity to existing below-
grade wildlife passages can help inform construction of structural retrofits that can help
keep wildlife off roadways and increase habitat connectivity.
Influence of Roads on Small Mammals
The small-mammal research in this study involved an assessment of the potential of roads
to affect the abundance and distribution of small mammals by possible habitat degradation.
The research team investigated what influence, if any, highways had on the relative abun-
dance of small mammals and how far any observed effect might extend into adjacent habitat.
Field studies along highways in both Utah and British Columbia were conducted.
In Utah, the research team captured 484 individuals of 13 species. The results showed dif-
ferent trends of species diversity at different distances from the road from one year to the next.
During 2004, the diversity of species was highest further from the road in direct contrast to
2005, when diversity was highest closest to the road. Density and abundance data also differed
between years and species. When the research team compared density in three distinct areas,
sites with higher habitat quality (i.e., with greater forb and grass presence) had significantly
higher small-mammal densities. Overall, it appeared that roads per se had little effect on
small-mammal density. Rather, microhabitat conditions that were most favorable for each
individual species appeared to be most responsible for density responses.
The results were similar for British Columbia, where the research team captured 401 indi-
viduals of 11 species. Our results indicated that highway and transmission-line rights-of-way
OCR for page 7
7
(ROWs) appeared to be negative influences on abundance for most species and potentially
neutral to positive for others. There were no consistent patterns in species abundance as the
distance in a forest increased from the road right-of-way. There was however, a consistent
pattern of lower total species diversity in the road rights-of-way. Microhabitats and local con-
ditions that varied among sites and transects and that remain independent of road or ROW
appeared to be stronger than, or at least mask, any effects related to the road or ROW. For the
most common and most habitat-generalist species, the deer mouse (Peromyscus maniculatus),
there were no strong indications of an effect of distance from the highway or transmission line.
Additionally, there was no evidence of any effect attributable to the highway that was not
evident at the transmission-line sites. Impacts due to the highway itself may exist for
some species, but large samples and highly consistent habitat conditions would be required to
detect them.
Restoring Habitat Networks with Allometrically Scaled
Wildlife Crossings
In the research of allometric placement of crossings, the research team investigated
whether differences in vagility (i.e., the natural ability of mammal species to move across the
landscape) could be used in deciding on the spacing of wildlife crossings that will help re-
store landscape permeability across fragmented habitat networks. Until now, the placement
of crossings has not been grounded in theory but has relied on empirical data to underpin
crossing placement decisions, in part because the idea of landscape permeability has not
been traditionally viewed from an animal perspective. When landscape permeability is
viewed from an animal perspective, inherent species-specific movement capabilities provide
the basis for developing scaling relationships (i.e., allometry) to inform the placement of
crossings. In other words, the animals "tell" us where to place the crossings. There have been
useful developments in allometric scaling laws that have led to important and statistically
sound relationships between home range size and dispersal distance for species. The recently
described implications of the relationship of median dispersal distance (MedDD) to home
range area and the development of a single metric, termed the "linear home range distance"
(LHRD), to represent home range size provide scaling laws that can be related to the con-
cepts of ecological neighborhoods and domains of scale to consider how the movement of
species with similar movement capabilities can be enhanced by effective placement of cross-
ings in roaded landscapes. In turn, this effective placement should reduce barrier effects and
improve permeability across habitat networks. It is possible to use MedDD as the upper
bound and a LHRD as the lower bound to develop alternative domains of scale for groups
of animals to guide the placement of wildlife crossings.
The correct spacing of crossings is perhaps most urgent for large terrestrial mammals that,
when involved in WVCs, tend to cause greater vehicle damage and have greater potential to
cause human injury and death than smaller bodied animals. Large-bodied animals pose a
greater safety risk. It appears that, to achieve the kind of landscape permeability that will help
ensure the health of large-mammal populations (i.e., deer, moose, elk, and bear) and to min-
imize WVCs, placement of wildlife crossings in areas where populations of these animals exist
will entail at least a multistep decision process. The first step involves deciding which allomet-
ric scaling domain is appropriate and feasible. Highest permeability will be obtained when
crossings of appropriate type and design are placed using the LHRD domains. If crossings were
placed according to the MedDD, they would be too far apart to create high permeability of the
landscape. For example, using LHRD domains, wildlife crossings for white-tailed deer and
mule deer would be placed at about 1 mi (1.6 km) apart in areas where these animals cross the
road frequently and are often hit by vehicles, which would certainly improve highway safety
OCR for page 8
8
and help ensure ease of movement, thus improving landscape permeability for these animals.
Using the MedDD values of 6.1 to 7.4 mi to space the crossings is clearly inappropriate and
will do little to facilitate movement, especially if exclusion fencing is part of the mitigation.
Similar arguments are appropriate for all species in general.
The use of allometric scaling domains represents only the first step to inform the place-
ment and spacing of wildlife crossings. Additional local information including (1) location
of migration pathways, (2) knowledge of areas of local animal movement across roads, and
(3) hotspots of WVC locations as well as dead animal count locations are needed. When
these data are used in an integrated and context-sensitive mitigation, these measures should
help ensure landscape permeability, providing for easier movement across the roaded land-
scape, and significantly improve highway safety.
Interpretation of Phase 2 Research Results
The sections on safety data analysis (3.1), accuracy modeling (3.2), and hotspot model-
ing (3.3) address different ways to achieve similar purposes and therefore may be confusing
for the reader. The following paragraphs should help guide the reader in understanding the
distinctions.
The safety research (Section 3.1) is most effectively used when the purpose is to assess if
a specific mitigation has been successful in reducing WVCs to improve public safety. The
safety approach has several applications and can be used to:
· Identify collision-prone locations for existing or proposed roads for all collision types com-
bined or for specific target collision types
· Aid in the evaluation, selection, and prioritization of potential mitigation measures; and
· Evaluate the effectiveness of mitigation measures already implemented.
An important caveat is that the safety approach does not address any aspect of wildlife
population response. As the models stand, their primary application is for the safety man-
agement of existing roads as opposed to design or planning applications for new or newly
built roads. Significantly, the before-after analysis may be judged as successful from a road
safety perspective, while at the same time the wildlife population concerned may be signifi-
cantly reduced.
A second aspect of the safety effort clearly showed that the choice of the database used to
define and evaluate the WVC problem impacts whether a particular roadway segment might
be identified for closer consideration and therefore the choice should be made carefully.
Recommendations are provided in this report about how the databases might be used
appropriately and how the data can be most profitably collected.
In the accuracy modeling (Section 3.2), non-road-related variables (i.e., ecological field
variables, distance-to-landscape-feature variables, and GIS-generated buffer variables) were
assessed to determine their relative importance in explaining where ungulates were killed on
the road. Also, spatially accurate data were discovered to make a difference in the ability of
models to provide not just statistically significant results but more importantly, biologically
meaningful results for transportation and resource managers responsible for reducing
WVCs and improving motorist safety. Hence, these models are especially applicable when
it is important to locate hotspot areas of WVCs and hence wildlife crossings during the
design and planning of new roads.
The hotspot analysis (Section 3.3) investigated WVC hotspot identification techniques,
taking into account different scales of application and transportation management con-
cerns. Simple plotting most often results in collision points being tightly packed together, in
OCR for page 9
9
some cases directly overlapping with neighboring WVC carcass locations, thus making it
difficult to identify distinct clusters, i.e., where the real high-risk collision areas occurred.
Modeling or analytical techniques permit a more detailed assessment of where WVCs occur,
their intensity, and the means to begin prioritizing highway segments for potential mitiga-
tion applications.
The Ripley's K analysis clearly shows the spatial distribution of WVCs and the importance
of broad-scale landscape variables (such as elevation and valley bottoms in a mountain en-
vironment). Further, the locations of high-intensity roadkill clustering within each area can
help to focus or prioritize the placement of mitigation activities, such as wildlife crossings
or other countermeasures, on each highway segment. The research team found that the
nearest neighbor (CrimeStat®) approach was useful for identifying key hotspot areas on
highways with many roadkills because it, in essence, filters through the roadkill data to ex-
tract where the most problematic areas lay. The density analysis approach identified more
hotspot clusters on longer sections of highway. Although the density analysis approach
appears to be less useful to management, it may be a preferred option where managers are
interested in taking a broader, more comprehensive view of wildlifevehicle conflicts within
a given area. Such a broader view may be necessary not only to prioritize areas of conflicts
but also to plan a suite of mitigation measures. The location of the larger clusters produced
by the density analysis could be tracked each year to determine how stable they are or
whether there is a notable amount of shifting between years or over longer time periods. This
type of information will be of value to managers in addressing the type of mitigation and
intended duration (e.g., short-term vs. long-term applications).
The WVC data that transportation departments currently possess are suitable for meet-
ing the primary objective of identifying hotspot locations at a range of geographic scales,
from project-level (< 50 km of highway) to larger district-level or state-wide assessments on
larger highway network systems. The spatial accuracy of WVCs is not of critical importance
for the relatively coarse-scale analysis of where hotspots are located. Any of the analytical
clustering techniques can be used when more detailed information is needed.
