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 35
35
Table 10. Total WVC and deer carcass removals by roadway
characteristic (20012003).
Number and Number and
Number and
Percentage of Percentage of
Percentage of
WildlifeVehicle Deer Carcass
Roadway Milesa
Collisions Removals
Roadway System
1,020.46 1,892 6,382
Interstate
(0.9%) (8.2%) (25.3%)
3,635.25 6,042 10,205
U.S. Highway
(3.2%) (26.2%) (40.4%)
5,039.19 5,722 8,075
Iowa State Route
(4.4%) (24.8%) (32.0%)
30,843.84 6,826 119
Farm to Market Route
(27.3%) (29.6%) (0.4%)
Area Type
97,885.5 20,222 22,155
Rural
(86.6%) (87.6%) (87.7%)
15,172.75 2,872 3,103
Urban
(13.4%) (12.4%) (12.3%)
Number of Lanesb
109,471.10 16,429 13,393
Two
(96.8%) (71.1%) (53.0%)
2,033.43 4,898 9,650
Four
(1.8%) (21.2%) (38.2%)
a
Roadway mileage changes each year. Number and percentage of roadway miles in table represents
average annual mileage that existed from 2001 to 2003.
b
Number includes through, turn, and two-way left-turn lanes.
ferent spatial accuracies of the data and the plots on the re- The length of the segments evaluated and modeled was pri-
sults of this work are noted where appropriate. Table 10 marily defined by the changes in roadway cross section design
shows the number and percentage of Iowa roadway mileage, (e.g., number of lanes). Only those rural roadway segments with
reported WVCs, and deer carcass removals along roadways a length of <
_ 0.1 mi were used in the development of the model.
with varying characteristics.
The traffic volume and cross section attribute data collected Findings and Results
were also used with the WVC and deer carcass removal data
Aspect 1: Application of Reported WildlifeVehicle
to develop prediction models. Descriptive statistics for the
Collision Data
2001 to 2003 roadway length, AADT, WVC, and deer carcass
removal data used in the model development are summarized Tables 12 through 14 provide details of the SPFs. For each of
in Table 11. the four states, three levels of SPFs were developed with varying
Table 11. Modeling database summary (rural segments > 0.1 mi).
Roadway Two-Lane Rural Roadway Multilane Rural Roadway
Category Total Mean Min Max Total Mean Min Max
Length (Miles) 6,529 0.49 0.10 1.78 1,317 0.35 0.10 1.39
Average Annual
Daily Traffic NAa 2,433 103 13,000 NA1 12,659 180 77,433
(AADT)
WildlifeVehicle
Collisions/ 6,721 0.39 0.00 16.32 3,438 0.87 0.00 14.23
Mile-Year
Carcass
Removals/ 11,640 0.64 0.00 75.85 8,288 1.97 0.00 93.33
Mile-Year
a
NA = Not Applicable
OCR for page 36
36
Table 12. SPFs for rural two-lane roadways.
Model Form: Total wildlifevehicle collisions/mile-year
State/ = AADT 1
exp 2 SURFWID 3 HI 4 SPEED 5 LANEWID
Terrain
Model ln Dispersion
1 2 3 4 5
(s.e.) (s.e.) (s.e.) (s.e.) (s.e.) (s.e.) parameter
-7.8290 0.6123
CA 1 All 1.6098
(0.1868) (0.225)
-8.7034
Flat
(0.2005)
-8.1810 0.6636
CA 2 Rolling 1.4831
(0.1930) (0.0228)
-8.0343
Mountainous
(0.1989)
-8.5357
Flat Design
(0.2046)
55
-7.9275 0.6518
CA 3 Rolling -0.3310 1.4493
(0.1968) (0.0230)
(0.0449)
-7.7157 Else = 0
Mountainous
(0.2029)
-4.5625 0.3743
NC 1 All 0.9222
(0.0576) (0.0078)
Flat -4.3984
Rolling (0.0745) 0.3637
NC 2 0.8142
-5.5363 (0.0077)
Mountainous
(0.0653)
Flat -4.3805 Posted
Rolling (0.0773) < 55
0.4447 -0.0122
NC 3 -0.7165 0.7353
-5.7195 (0.0087) (0.0022)
Mountainous (0.0248)
(0.0685)
Else = 0
-9.1135 1.0237
UT 1 All 1.7610
(0.1423) (0.0205)
-9.3123
Flat
(0.3385)
-9.0528 1.0092
UT 2 Rolling 1.6123
(0.3393) (0.0410)
-8.7728
Mountainous
(0.3006)
-12.987
Flat Posted
(0.9608)
55
-12.803 0.8073 0.4751
UT 3 Rolling -0.6646 1.3985
(0.9613) (0.0455) (0.0838)
(0.1344)
-12.408 Else = 0
Mountainous
(0.9485)
-8.6850 0.7802
WA 1 All 1.3825
(0.3020) (0.0367)
-8.5319 0.8034 -0.0584
WA 2 All 1.0237
(0.3552) (0.0426) (0.0117)
Posted
55
-8.5161 0.7622 -0.0696
WA 3 All 0.4358 0.9528
(0.3493) (0.0426) (0.0124)
(0.0964)
Else = 0
data requirements. The first level required only the length and research. These applications are illustrated in the "Interpreta-
AADT of a segment. The second level included the requirement tion, Appraisals, and Applications" section.
that segments be classified as flat, rolling, or mountainous ter- In general, the calibrated SPFs make good intuitive sense in
rain. The third level of SPFs added additional roadway variables that the sign, and to some extent the magnitude, of the esti-
such as average lane width. All variables were from state HSIS mated coefficients and exponents accord with expectations.
data. Segments were defined as sections of roads, generally be- Surprisingly, the exponent of the AADT term, although rea-
tween significant intersections and having essentially common sonably consistent for the three levels of models in a state, varied
geometric characteristics. Illustration of the application of the considerably across states. This exponent varied significantly
SPFs developed is a key component of this aspect of the safety across facility types, reflecting differences in traffic operating
OCR for page 37
37
Table 13. SPFs for rural multilane roadways.
Model Form: Total wildlifevehicle collisions/mile-year
= AADT 1
exp 2 MEDWID 3 HI 4 SPEED
State/Model Terrain
ln 1 2 3
(s.e.) (s.e.) (s.e.) (s.e.)
-5.2576 0.3290
CA 1 All
(0.4397) (0.0470)
-6.4592
Flat
(0.4523)
-5.7615 0.3926
CA 2 Rolling
(0.4398) (0.0464)
-5.5220
Mountainous
(0.4498)
-6.4885
Flat
(0.4485)
-5.8372 0.4145 -0.0057
CA 3 Rolling
(0.4360) (0.0464) (0.0015)
-5.6577
Mountainous
(0.4462)
-3.3660 0.2501
NC 1 All
(0.6314) (0.0684)
Flat -2.5310
Rolling (0.6063) 0.1736
NC 2
-4.1844 (0.0641)
Mountainous
(0.5934)
Flat -2.4303
Rolling (0.5871) 0.1858
NC 3
-4.0785 (0.0621)
Mountainous
(0.5741)
-4.1217 0.4414
UT 1 All
(0.6231) (0.0742)
-4.4878
Flat
(1.5295) 0.3900
UT 2
Rolling -3.4508 (0.1754)
Mountainous (1.5013)
-12.7417 1.2066
WA 1 All
(1.9219) (0.2028)
-12.9945
Flat
(1.9091) 1.1398
WA 2
Rolling -11.8326 (0.1987)
Mountainous (1.8894)
-14.1608
Flat
(2.1029) 1.2721 0.1244
WA 3
Rolling -13.2591 (0.2153) (0.0775)
Mountainous (2.0800)
conditions. The variables found to be significant at the 10% For application in another state, or even for application
level varied by state were: in the same four states for different years from those in the
calibration data, the models should be recalibrated to re-
AADT: Annual average daily traffic flect differences across time and space in factors such as
SURFWID: Total surface width (feet) collision reporting practices, weather, driver demograph-
LANEWID: Average lane width (feet) ics, and wildlife movements. In essence, recalibration
HI: Average degree of curvature involves using a multiplier, which is estimated to reflect
SPEED: Posted speed in North Carolina & these differences by first using the models to predict the
design speed in California (mph) number of collisions for a sample of sites for the new state
MEDWID: Median width (feet) or time period. The sum of the collisions for those sites is
MEDTYPE: Positive barrier or unprotected divided by the sum of the model predictions to derive the
OCR for page 38
38
Table 14. SPFs for rural freeways.
Model Form: Total wildlifevehicle collisions/mile-year
= AADT 1
exp 2 MEDWID 3 HI 4 SURFWID 5 MEDTYPE
State/Model Terrain
ln 1 2 3 4 5 Dispersion
(s.e.) (s.e.) (s.e.) (s.e.) (s.e.) (s.e.) parameter
-6.2814
Flat
(0.7166) 0.2810
CA 1 1.5885
Rolling -4.7526 (0.0726)
Mountainous (0.7098)
-5.6746
Flat
(0.6925) 0.3050 -0.0126
CA 2 1.3543
Rolling -4.3198 (0.0700) (0.0014)
Mountainous (0.6857)
-4.3930 0.4356
UT 1 All 1.9966
(1.4121) (0.1550)
-7.8707
Flat
(1.4831)
-6.9760 0.7272
UT 2 Rolling 1.5641
(1.4811) (0.1632)
-6.0374
Mountainous
(1.4516)
8.0592 Median
Flat
(1.4808) Type
-7.1234
Rolling
(1.4773) Positive
0.7472 barrier
UT 3 1.5277
(0.1630) -1.0633
-6.0651 (0.4623)
Mountainous
(1.4465)
Unprotected
0.0000
-15.5153 1.3969
WA 1 All 0.8816
(1.7866) (0.1809)
-16.8612
Flat
(1.7977)
-15.8572 1.4355
WA 2 Rolling 0.7807
(1.7634) (0.1784)
-15.4443
Mountainous
(1.7846)
-9.9014
Flat
(3.9034)
-8.8909 1.4507 -0.1483
WA 3 Rolling 0.7867
(3.8877) (0.1793) (0.0765)
-8.4610
Mountainous
(3.8975)
multiplier. Further details of this procedure are provided in removal data on a roadway network within a GIS platform.
Appendix B. This information is useful because it helps define where the
In deciding which among available competing models is WVC and deer carcass removal data were reported or
best to adopt for another state for which a similar model may collected, and whether the occurrence of either is actually over-
not be available, goodness-of-fit tests must be conducted. or under-represented along roadways with particular charac-
Choosing the most appropriate model is especially important teristics. In addition, the results of visual and quantitative
because the exponents for AADT, by far the most dominant WVCs, and deer carcass removal comparisons (statewide,
variable, differ so much between states. A discussion of these example corridor, and model content) are described.
tests is provided in a recent FHWA report.241 A summary is In general, the amount of two-lane roadway mileage used
presented as part of Appendix B. in the modeling was almost 5 times greater than the multilane
roadway mileage (See Table 11). Two-lane roadways with
medians were not included. The multilane database included
Aspect 2: Comparison of WildlifeVehicle Collision
all State Routes, U.S. Highways, or Interstate highways with
and Carcass Removal Data
more than two through lanes. Overall, despite the propor-
The findings from this aspect of the safety analysis focused tions of roadway mileage in the database, approximately two
on the challenges related to combining WVC and deer carcass WVCs were reported along the two-lane roadways for every
OCR for page 39
39
WVC reported along the multilane roadways. Similarly, the modeling activities in this research are noted below. The
number of deer carcasses removed from two-lane roadways statistics in Table 10 might also be used for gross compari-
was about 1.4 times that removed from the multilane road- son purposes to roadway segments of interest with similar
ways. The mean number of WVCs and carcass removals per characteristics. A review of the percentages by roadway sys-
mile-year, however, along the multilane roadways in the tem reveals that the deer carcass removal data are primarily
database are much greater than those along the two-lane from the interstates, U.S. Highways, and State Routes. This
roadways. Additionally, the AADT along the multilane rural trend is not surprising because the data provided was from
roadways was also greater than the two-lane roadways. the IaDOT. About 80% of the WVC reported, on the other
hand, occurred on U.S. Highways, State Routes, and farm to
WVC and deer carcass removal GIS activities. There are market roadways. The percentage of WVCs and carcasses
a number of advantages when information is incorporated removed along interstates, U.S. Highways, and State Routes
into a GIS platform, including an increased ability to organ- are much greater than their statewide roadway mileage
ize and integrate spatial data, the relatively easy presentation would suggest. For every reported WVC along the interstate,
of the data, and the capability to quickly analyze and/or com- there were more than three carcasses collected. Table 10
pare one or more datasets. Visual patterns are also easier to shows that the percentage of urban and rural roadway
discern, and data can be assembled from multiple sources and mileage is essentially the same as the percentage of WVCs
formats to produce broader and more rigorous evaluation ac- and deer carcass removals in these areas. From a roadway
tivities. The GIS process in a safety data project is typically mileage point of view, the number of WVCs and deer carcass
composed of three steps: (1) data acquisition and importa- removals also appears to be over-represented along four-
tion, (2) data management, and (3) spatial analysis. The first lane roadways. More than 90% of the WVCs and deer carcass
steps are often the most difficult. removals from 2001 to 2003 occurred along two- and four-
The general objective of the GIS activities in this aspect of the lane roadways.
safety data analysis was to combine and document spatial rep-
resentations of the WVC and deer carcass removal locations. Statewide and sample corridor comparisons. The avail-
Deer carcass removal data and locations are not normally avail- ability of WVC and deer carcass removal data in Iowa within a
able in any consistent manner across jurisdictions. In this study, GIS platform that contains information about the Iowa roadway
the carcass reports included route and milepost to reference network allowed a relatively easy comparison and calculation of
locations of deer carcasses to the road network. To geo-code various safety measures related to each dataset. Statewide WVC
these records, the research team obtained the location of the and deer carcass removal frequencies and rates are shown in
mileposts from the Iowa State University Center for Trans- Table 15 for the 3-year analysis time period as are the combined
portation and Education (CTRE). This information was devel- number of deer carcasses removed by the IaDOT and those sal-
oped from different DOT data sources and combined with a vaged through the Iowa Department of National Resources
GIS data set. The WVC data were relatively easy to incorporate (IaDNR). About 34% of roadside deer carcasses are salvaged
into the GIS platform because latitude and longitude coordinate under permit from the state. Sixty-six percent of the roadside
positions for each incident were available. The spatial accuracy deer carcasses are removed by IaDOT and their location noted
of the carcass removal locations was different; they were esti- (these are the removals plotted in Figures 4 and 5). According to
mated to the nearest 0.1 milepost. In addition, the individual the IaDNR, the roadway locations for the deer carcasses it
whole milepost locations (e.g., 1.0, 2.0, etc.) on the Iowa road- permits for salvage are not consistently collected and should
way GIS map were the only spatial data connection that would therefore not be used for analysis.
allow the plotting of the deer carcass removal locations. For The numbers in Table 15 are general statewide measures
schedule and budget reasons, therefore, the estimated locations and when recalculated for individual roadway segments are
of the deer carcass removals were rounded to the nearest often different (Table 16). The data in Table 15 illustrate three
milepost, summed, and plotted. statewide databases that provide different values for the WVC
The total number of deer carcass removals in 2002 is plot- data in Iowa. The number of deer carcasses removed by
ted in Figure 4 at each milepost (with scaled and shaded IaDOT, for example, is approximately 1.09 times greater than
circles to represent the different number at each location). the number of WVCs reported to the police. The number of
This spatial modification was considered appropriate given salvaged and unsalvaged deer carcasses, on the other hand, is
the accuracy of the datasets provided, the objective of this approximately 1.66 times greater. The other safety measures
work (i.e., a comparison of data as they might be available show a similar trend. However, only the WVCs and deer car-
to a decision maker), and the WVC and carcass removal data cass removals in Table 15 are related to roadway location in
likely to be available within other states. The impact of Iowa, and typically the location of the latter is not known. The
this spatial alteration on the results of the comparisons and plots in Figures 4 and 5 show that the spatial patterns of the
OCR for page 40
40
Table 15. Statewide wildlifevehicle collision and deer carcass removal
metrics (2001 to 2003).
Salvaged and
Carcass
Metricsa WVC Unsalvaged Deer
Removalsb
Carcassesc
Total 23,094 25,258 38,283
Rate per Year 7,698 8,419 12,761
Rate per Roadway Mile 0.20 0.22 0.34
Rate per Hundred Million
25.3 27.6 41.9
Vehicle-Miles-of-Travel
a
Statewide roadway mileage and vehicle-miles-of-travel used in all calculations.
b
Deer carcass removals are those recorded and summarized by the Iowa DOT by location.
c
Salvaged and unsalvaged deer carcasses are summarized by the Iowa Department of National Resources.
The Department of Transportation deer carcass removals are a portion of this total, but they are the only
removals for which roadway location is known.
WVC and deer carcass removal data are also different. It is show that reported WVCs and deer carcass removal data (as
not likely that this conclusion will change if the data were available) likely have different spatial patterns. This lack of
plotted differently. The use of different databases could lead similarity could lead to the implementation of countermea-
to different statewide policy and corridor-level decisions sures along different roadway segments. Table 16 summarizes
related to WVCs. In addition, the choice of the database used the WVC and deer carcass removal data from 2001 to 2003
could lead to different conclusions. for the roadway segments shown in Figure 5. The differences
Figure 5 shows the reported WVCs and deer carcass in the magnitude of the WVCs and deer carcass removals that
removals for sample roadway segments along Interstate 80 occur along these roadway segments are clear. Overall, the
and U.S. Highway 18 in Iowa. Note that no WVCs were number of carcasses removed along the Interstate 80 segment
reported along this segment of U.S. Highway 18 in 2002. was 8.6 times greater than the number of WVCs reported.
A more detailed summary of the WVCs and deer carcass The number of carcasses collected along U.S. Highway 18, on
removals along these two segments is shown in Table 16. the other hand, was 3.8 times greater than the number of re-
These measures could be compared to the statewide results in ported WVCs.
Table 15 and/or those calculated for roadways with similar More than 90% of the Interstate 80 segment length sum-
characteristics (See Table 10). marized in Table 14 (and shown in Figure 5) was classified as a
The results of this type of general comparison can be used four-lane rural freeway. The frequencies and rates in Table 16
as a filter to determine whether a particular roadway segment are all generally greater than the statewide measures for a
needs more detailed consideration. Figures 4 and 5 generally roadway with these characteristics. Only the use of a WVC rate
Figure 5. Deer carcass removal and WVC locations along segments of
Interstate 80 and U.S. Highway 18 (2002).
OCR for page 41
41
Table 16. Comparison of roadway segment WVC and deer carcass removal
measures (2001 to 2003).
I-80 I-80 U.S. Hwy 18
U.S. Hwy 18
Wildlife Deer Wildlife
Deer Carcass
Variable Rates Vehicle Carcass Vehicle
Removals
Collisions Removals Collisions
(9.9 Mi)
(8.4 Mi) (8.4 Mi) (9.9 Mi)
Total Number 19.0 163.0 5.0 19.0
Rate / Year 6.3 54.3 1.7 6.3
Rate / Roadway Mile 2.3 19.3 0.51 1.9
Rate / Hundred Million
10.4 89.6 17.2 65.4
Vehicle-Miles-of-Travel
Note: See Figure 5 for plots of 2002 wildlifevehicle collisions and deer carcass removals along these segments
in Iowa.
might lead to the conclusion that this segment has a typical WVCs or deer carcass removals per mile-year. Details of the
WVC data level. The U.S. Highway 18 segment in Figure 5 is rural two-lane and multilane models are shown in Tables 17
primarily a two-lane rural roadway. Mixed conclusions result and 18. Prediction (not causal) models with only AADT are
when the WVC and deer carcass removal measures for this provided later in this section. Volume-only models were
roadway (See Table 16) are compared to relevant statewide developed for comparison and application purposes. The vari-
measures. The WVCs and deer carcasses removals per mile ables considered for use in each of the models were selected
along the segment are larger than the statewide measures, but from the Iowa roadway cross section database (which included
the rates (based on volume) are both smaller than those cal- more than 90 factors). The following variables, which came
culated for the entire state. Clearly, the choice of the data from the IaDOT database, were considered:
(WVCs or deer carcass removals) and the measures (e.g., per AADT: Annual average daily traffic on roadway
mile or rate) that are used impacts whether a particular road- (vehicles per day in both directions)
way segment might be identified for closer consideration. The AVGSHLD: Average of left- and right-shoulder widths
comparisons described above consider average values, but on two-lane roadways
more critical WVC frequency or rate data could be used as an CRASHES: Number of police-reported animal-vehicle
initial step to identify hotspot roadway segments. collisions (used in one model for deer
carcass removal prediction)
WVC and deer carcass removal model development and LANES: Total number of through lanes present
comparison. Prediction models using WVC, deer carcass re- LSHDWID: Width of the left side or inside shoulder
moval, and roadway cross section data from Iowa were (nearest foot)
developed to assist in the identification of potential hotspot MEDTYPE: Classified as zero (0) if unprotected or 1 if a
roadway segments and are described next. They can be applied positive barrier
in a manner similar to those described previously in this report. MEDWID: Width of the median between the edges of
This section of the safety analysis report focuses on the differ- traffic lanes (nearest foot)
RSHDWID: Width of the right side or outside shoulder
ences between the models developed with the WVC and deer
(nearest foot)
carcass removal data and the potential impact of those differ-
SPEED: Posted speed in miles per hour
ences. A site visit to each potential "high" collision or carcass
SURFWID: Surface width of roadway measured from
segment is necessary for confirmation purposes and the iden-
edge of pavement to edge of pavement
tification of specific countermeasure installation locations.
(feet)
The combination of WVC, deer carcass removal, and road-
way location data in a GIS platform allowed the production of The form and content of the WVC and deer carcass re-
prediction models to describe the relationships between the oc- moval prediction models developed for rural two-lane road-
currence of a WVC or carcass removal and several roadway ways in Iowa are shown in Table 17. Two models were devel-
cross section characteristics typically available through DOT oped for both WVCs and deer carcass removals with different
databases. These analyses applied to rural paved two-lane and sets of independent variables. Both models are provided be-
multilane roadways in Iowa with a State Route, U.S. Highway, cause they produce similar results, but have different input
or Interstate designation. They can be applied within an variables, which may make them useful to different practi-
empirical Bayesian approach. The negative binomial models or tioners. The variables in the models include AADT, SPEED,
SPFs were created from 2001, 2002, and 2003 data to predict and AVGSHLD; for one deer carcass removal model, the num-
OCR for page 42
42
Table 17. Models for rural two-lane roadways (segments > 0.1 mi) in Iowa.
Model Form: Total WVCs or deer carcass removals per mile-year =
Model AADT 1
exp AVGSHLD SPEED CRASHES
2 3 4
Dependent
a
Variable ln( ) 1 2 3 4 Dispersion
(s.e.) (s.e.) (s.e.) (s.e.) (s.e.) Parameter
WVCs/ -5.9203 0.6164 0.0193
1.0179
Mile-Year (0.2088) (0.0283) (0.0067)
WVCs/ -6.4968 0.6429 0.0095
1.0196
Mile-Year (0.3807) (0.0268) (0.0059)
Deer
Carcass -5.4332 0.5784 0.0677
5.2702
Removals/ (0.2957) (0.0403) (0.0096)
Mile-Year
Deer
Carcass -4.9635 0.4890 0.0701 0.2714
5.0062
Removals/ (0.2954) (0.0405) (0.0096) (0.0225)
Mile-Year
a
These symbols represent the parameters estimated in the modeling process and that measure the impact of each
independent variable on the expected crash frequency.
ber of reported WVCs was included. The model coefficients for countermeasure implementation. Of course, some of the dif-
all models are shown in Table 17 along with their standard ferences shown in Table 17 are due to the differences in the spa-
error and the model dispersion parameter. The impact of the tial accuracy of the information provided for the two databases
variables in each model is somewhat different, and the ex- and ultimately plotted in the GIS platform. These accuracies,
planatory value of the WVC model appears to be greater than however, are typical.
the deer carcass removal model. The large dispersion parame- Similar WVC and deer carcass removal prediction models
ter of the deer carcass removal model is high, which should be were also developed for rural multilane roadways in Iowa. The
considered if it is applied. Given that most jurisdictions do not model coefficients for these models are shown in Table 18 as
have deer carcass removal data by location, it is encouraging are their standard errors and the model dispersion parame-
that the CRASHES data may be used as a predictor of carcasses. ters. There are more differences in these models than those
Thus, if carcass data could be collected even for a subset of the produced for the two-lane rural roadways. The models in
roadways in a jurisdiction, a model that included reported col- Table 18 contain different variables. The models include
lisions to predict carcasses could be recalibrated and applied. one or more of the AADT, AVGSHLD, MEDTYPE, and
The differences in these models further support the conclusion MEDWID predictor variables. As with the two-lane models,
that the use of WVC or deer carcass removal data can result in the number of WVCs could also prove to be a useful predic-
the identification of different roadway segments for potential tor of deer carcass removal frequency. The results of this
Table 18. Models for rural multilane roadways (segments >
_ 0.1 mi) in Iowa.
Model Form: Total WVCs or deer carcass removals per mile-year =
Model AADT 1
exp AVGSHLD MEDWID MEDTYPE CRASHES
2 3 4 5
Dependent
Variable
ln( ) 1 2 3 4 5 Dispersion
(s.e.) (s.e.) (s.e.) (s.e.) (s.e.) (s.e.) Parameter
With Median
Barrier:
-0.2471
WVCs/ -0.9021 0.0527 0.0390
Mile-Year
(0.0851) 0.6360
(0.3905) (0.0391) (0.0205)
Unprotected:
0.0000
Deer
Carcass -4.6677 0.5616 0.0017
Removals/ (0.5972) (0.0660) (0.0011) 7.8601
Mile-Year
Deer
Carcass -4.3118 0.4871 0.3314
Removals/
7.2680
(0.5851) (0.0637) (0.0385)
Mile-Year
OCR for page 43
43
Table 19. Volume-only models (segments > 0.1 mi) in Iowa.
Model Form: Total wildlifevehicle collisions or deer carcass removals
Model per mile-year = AADT 1
Dependent
Variable )ln( 1 Dispersion
(s.e.) (s.e.) Parameter
Rural Two-Lane Roadway
WVCs/ -5.9894 0.6439
1.0204
Mile-Year (0.2077) (0.0268)
Deer Carcass
-5.5973 0.6662
Removals/ 5.3432
(0.2952) (0.0384)
Mile-Year
Rural Multilane Roadways
WVCs/ -1.2494 0.1199
0.6381
Mile-Year (0.2985) (0.0321)
Deer Carcass
-4.8520 0.5919
Removals/ 7.8791
(0.5923) (0.0640)
Mile-Year
model development activity further support the importance the two datasets. The high dispersion parameters of the deer
of choosing the appropriate database to evaluate collision carcass removal models in Table 19 should be noted.
problem locations. The dispersion parameter of the deer car- Figures 6 and 7 plot the AADT (volume-only) deer car-
cass removal model is high, which should be considered in the cass removal and WVC models in Table 19 for two-lane
application of this model. and multilane rural roadways, respectively. Because AADT
Finally, WVC and deer carcass removal models, with is the only independent variable, a simple comparison
AADT as the only input variable, were also developed. These shows that the models diverge as AADT increases, dramat-
models are shown in Table 19. They were created for applica- ically so for multilane roadways. These plots illustrate that
tion if the data for the roadway cross section variables in the the deer carcass removal and WVC frequencies predicted
previous models were not available. In addition, the volume- are different and not strictly linearly correlated. The avail-
only models were compared to the other models to investi- ability of WVC data throughout the United States led the
gate the additional explanatory value offered by the addition research team to ask whether the volume-only WVC mod-
of more road cross section variables. A comparison of the dis- els might be recalibrated to predict deer carcass removals.
persion parameters with those in Tables 17 and 18 reveals that To do so, the volume-only WVC models were applied to
the inclusion of other roadway cross section variables in the the deer carcass removal database. The sum of the observed
models adds little to the predictive strength of the WVC and deer carcass removals was then divided by the sum of the
deer carcass removal models. In other words, the AADT predictions from the WVC model. This factor was applied
measure contains most of the explanatory value of these as a multiplier to the WVC volume-only model and the
models. Overall, the explanatory value of the WVC models deer carcass removal predictions were recalculated and
with only AADT is still better than those developed with deer compared (See Figure 6 and Figure 7). This comparison
carcass removal data. Some of this difference, as previously was completed separately for the two-lane and multilane
stated, is due to the inconsistency in the location accuracy of rural roadway data.
2.50
Removals Per Mile-Year
WVCs or Deer Carcass
2.00
1.50
WVCs
1.00 Deer Carcass
Removals
0.50
0.00
0 2000 4000 6000 8000 10000 12000 14000
Average Annual Daily Traffic
Figure 6. Two-lane rural roadway volume-only model results.
