Content uploaded by Lora Smith
Author content
All content in this area was uploaded by Lora Smith on Dec 11, 2018
Content may be subject to copyright.
Effectiveness of a barrier wall and culverts in reducing
wildlife mortality on a heavily traveled highway in Florida
C. Kenneth Dodd Jr.
a,*
, William J. Barichivich
a
, Lora L. Smith
a,1
a
Florida Integrated Science Centers, US Geological Survey 7920 NW 71st Street, Gainesville, FL 32653, USA
Received 25 March 2003; received in revised form 5 October 2003; accepted 12 October 2003
Abstract
Because of high numbers of animals killed on Paynes Prairie State Preserve, Alachua County, Florida, the Florida Department of
Transportation constructed a barrier wall-culvert system to reduce wildlife mortality yet allow for passage of some animals across
the highway. During a one year study following construction, we counted only 158 animals, excluding hylid treefrogs, killed in the
same area where 2411 road kills were recorded in the 12 months prior to the construction of the barrier wall-culvert system. Within
the survey area lying directly in Paynes Prairie basin, mortality was reduced 65% if hylid treefrogs are included, and 93.5% with hylid
treefrogs excluded. Sixty-four percent of the wildlife kills observed along the barrier wall-culvert system occurred at a maintenance
road access point and along 300 m of type-A fence bordering private property. The 24 h kill rate during the post-construction survey
was 4.9 compared with 13.5 during the pre-construction survey. We counted 1891 dead vertebrates within the entire area surveyed,
including the ecotone between the surrounding uplands and prairie basin which did not include the barrier wall and culverts.
Approximately 73% of the nonhylid road kills occurred in the 400 m section of road beyond the extent of the barrier wall-culvert
system. We detected 51 vertebrate species, including 9 fish, using the 8 culverts after the construction of the barrier wall-culvert
system, compared with 28 vertebrate species in the 4 existing culverts prior to construction. Capture success in culverts increased 10-
fold from the pre-construction survey to the post-construction survey. Barrier wall trespass was facilitated by overhanging vege-
tation, maintenance road access, and by the use of the type-A fence. Additional problems resulted from siltation, water holes, and
human access. These problems could be corrected using design modifications and by routine, periodic maintenance.
Published by Elsevier Ltd.
Keywords: Barrier wall; Wildlife mortality; Roads; Highway mitigation; Culverts; Amphibians; Reptiles; Mammals
1. Introduction
Roads are an important feature in an animalÕs land-
scape, especially with regard to movement. Roads serve
both as facilitators and barriers to dispersal, fragment
habitats, and result in significant mortality (Case, 1978;
Heine, 1987; Andrews, 1990; Fahrig et al., 1995; Ashley
and Robinson, 1996; Forman and Alexander, 1998;
Trombulak and Frissell, 2000; Gibbs and Shriver, 2002;
Forman et al., 2003; Smith and Dodd, 2003). Numerous
factors influence the number and species killed on a
highway, including vehicle speed, volume, and traffic
pulses, local topography, accessibility of cover, and
structural features of a road, such as whether the
roadbed is raised or level with the surrounding envi-
ronment (Clevenger et al., 2003). Certain behavioral
traits also may affect the probability of mortality on
roads, such as active foraging (Bonnet et al., 1999),
vagility (Carr and Fahrig, 2001) and inclination to cross
open habitats (Gibbs, 1998; de Maynadier and Hunter,
2000). Most studies of the effects of roads on wildlife
have focused on largely terrestrial or avian species, or
semi-aquatic species moving between a breeding pond
and terrestrial habitats (reviewed by Forman et al.,
2003). Few studies, however, have examined highway
effects on animal communities or potential mitigation
Biological Conservation 118 (2004) 619–631
www.elsevier.com/locate/biocon
BIOLOGICAL
CONSERVATION
*
Corresponding author. Tel.: 1-352-378-8181; fax: 1-352-378-4956.
E-mail address: ken_dodd@usgs.gov (C. Kenneth Dodd Jr.).
1
Present address: Joseph W. Jones Ecological Research Center
Route 2, Box 2324, Newton, GA 31770, USA.
0006-3207/$ - see front matter. Published by Elsevier Ltd.
doi:10.1016/j.biocon.2003.10.011
options when highway corridors cross large wetlands
(Forman et al., 2003).
For the last few decades, biologists and engineers
have grappled with ways to facilitate animal movement
across transportation corridors, and developed a num-
ber of potential solutions (Langton, 1989; ALASV,
1994; Percsy, 1995; Duguet and Melki, 2003; Forman
et al., 2003). A common mitigation measure provides for
passage over or under a road or highway, usually with a
type of fence channeling movement towards a tunnel or
culvert. While culverts may be important in facilitating
cross-road movements (Yanes et al., 1995; Clevenger
and Waltho, 1999; Clevenger et al., 2001), follow-up
studies have not systematically determined the effects of
such mitigation efforts on the mortality of small verte-
brates, particularly amphibians and reptiles.
Large numbers of animals have been killed on the
section of US Highway 441 where it crosses Paynes
Prairie State Preserve, Alachua County, Florida, since
the highway was constructed in the early 1920s (Beck,
1938; Carr, 1940, 1974; Hellman and Telford, 1956;
Kauffeld, 1957; Franz and Scudder, 1977; Smith, 1996;
Smith and Dodd, 2003). High levels of wildlife mortality
along the highway corridor bisecting this large, ecolog-
ically diverse wetland for nearly eight decades may have
adversely impacted animal populations adjacent to the
roadway by serving as a continuous drain on numbers.
Collateral impacts also may extend far beyond the im-
mediate vicinity of the actual roadway (Andrews, 1990;
Fahrig et al., 1995; Findlay and Houlahan, 1997; For-
man and Deblinger, 2000; Hels and Buchwald, 2001).
In addition, the many animals attempting to cross the
highway (particularly alligators, large turtles, and me-
dium-sized mammals such as Raccoons and Opossums)
created human safety concerns as motorists collided
with crossing animals or attempted to avoid them. An-
imal carcasses detracted from the beauty of the prairie
and, because they were so numerous, sometimes caused
the road surface to become slippery, creating an addi-
tional motorist safety hazard.
In 1996, the Florida Department of Transportation
(FDOT) proposed constructing a wildlife barrier wall-
culvert system (which FDOT terms an ecopassage) across
Paynes Prairie to ameliorate the effects of the highway on
wildlife populations. The barrier wall-culvert system
consists of a concrete wall located on both sides of the
highway parallel to, and 9–11 m from, the roadway. The
wall is 1.1 m high with a 15.2 cm overhanging lip
(Fig. 1(a)), and interconnects with eight prefabricated
concrete culverts (two 2.4 2.4 44 m partially sub-
merged box culverts; two 1.8 1.8 44 m usually dry box
culverts; four cylindrical culverts 0.9 m in diameter 44
m) which allow passage of water and wildlife underneath
US Highway 441. The wall extends across the entire 2.8
km prairie wetland basin on the east side of the highway,
but only for 2.5 km on the west side of the highway. In the
northwest portion of the study area (300 m in length), a
‘‘Type-A’’ fence was constructed. The Type-A fence
consists of two guard rails (one on top of the other) with a
hardware cloth barrier sunk below ground (Fig. 1(b)).
Construction of the concrete barrier wall was prohibited
in this section because the roadway abutted private land
and because of concerns by FDOT about drainage.
In mid-1998, we began a study to assess the effec-
tiveness of the barrier wall-culvert system. The first
phase of the project (18 August 1998 to 13 August 1999)
determined the level of pre-construction wildlife mor-
tality, and how many and what types of animals used
the existing four box culverts (Smith and Dodd, 2003).
The results confirmed significant mortality on the high-
way and adjacent right-of-way, although some animals
used the existing culverts to successfully traverse the
highway. Construction of the barrier wall-culvert system
was completed in February 2001. The post-construction
Fig. 1. (a). The barrier wall-culvert system on the northwest side of
Paynes Prairie at culvert No. 2. The prefabricated concrete box culvert
is 2.7 2.7 m. (b). Type A fence bordering 300 m on the northwest side
of US 441 across the prairie basin. This fence was not effective in
eliminating trespass.
620 C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631
phase of the study began on 14 March 2001 and con-
tinued until 5 March 2002. Our study objectives were to
assess the effectiveness of the barrier wall-culvert system
by measuring post-construction levels of wildlife mor-
tality and the extent of culvert use. Based on our find-
ings, we provide suggestions concerning modifications in
the construction design and the need for regular main-
tenance of the barrier wall-culvert system.
2. Methods
2.1. Study area
Paynes Prairie is a large, highland freshwater marsh
(18.3 m above mean sea level) on the central Florida
Ridge in Alachua County. The prairie basin encom-
passes an area nearly 5000 ha. It is bordered by uplands
ca. 4.5–7.5 m higher in elevation than the basin proper,
and extends 13 km east to west and from 1.5 to 7 km
from north to south. Highland marshes are shallow
wetlands characterized by unstable drainage patterns
(Kushlan, 1980). Depending on rainfall and drainage,
Paynes Prairie may be a dry prairie, marsh, or shallow
lake. Water on the prairie flows east where it drains
into Alachua Sink and enters the Floridan Aquifer.
Water levels are normally highest during the summer,
and lowest in April, September, October and Novem-
ber. Drought conditions prevailed from April 2001
throughout the remainder of the year. In June 2001,
water-table levels were 0.58–1.2 m below ground level
(Jacobs et al., 2002). More details on the prairieÕs veg-
etation and hydrology are provided by Jacobs et al.
(2002).
Paynes Prairie was designated as a State preserve in
1970. The prairie basin is transected by two major
highways, Interstate 75 (I-75) and US Highway 441 (US
441), both of which are 4-lane divided highways. US 441
was built in 1923 and was expanded from two to four
lanes in 1957. Fill for the roadway was taken from the
adjacent marsh, which created shallow canals that par-
allel the highway. Prior to construction of the barrier
wall-culvert system, the road sat on a gently sloping
raised bed ca. 1.75 m above the surrounding wetlands.
The highway corridor is 44 m wide (including a grassy
right-of-way, paved highway and narrow bicycle lanes,
and grassy median), and it traverses 2.8 km of the prairie
basin. The current speed limit on US 441 is 97 km/h.
Daily traffic volumes exceed 11,000 vehicles per day.
2.2. Road kill survey
The highway survey methods were similar to those
employed during the pre-construction phase of data
collection (Smith and Dodd, 2003). The highway was
divided into 32 100 m sections for a total oneway length
of 3.2 km. Section 1 was located on the north rim of the
Paynes Prairie basin at the first private driveway
(29.5874 °N, 82.3382 °W), and Section 32 was located
on the south rim at the first private drive (29.5579 °N,
82.3305 °W). There were no barriers to wildlife entry
onto the highway right-of-way in Sections 1, 2, 31, and
32 on the prairie rim, which were outside the boundaries
of the state preserve. These four sections are in an eco-
tone between the prairie basin and surrounding uplands.
They were monitored in order to assess whether animals
used these ecotones as movement corridors. In addition,
an increase in post-construction mortality in these eco-
tonal areas might reflect an attempt by wildlife to cross
the highway at the first available area not fronted by the
barrier wall. The type-A fence bordered the highway
adjacent to the southbound lanes of Sections 3–6.
Surveys consisted of from one to four observers
walking the 3.2 km survey area one time in each direc-
tion on each sampling occasion, for a total of 6.4 km.
One observer walked along the highway in the grassy
median, whereas the others walked in the east or west
right-of-way. If only one observer was present, separate
passes were made along the median and the north and
south bound lanes to ensure complete coverage. In this
manner, both the north and southbound lanes of the
entire paved highway surface, extending 3–4 m onto the
grassy shoulders (the right-of-way), and the entire grassy
median between the north and southbound lanes, were
surveyed.
A sampling period consisted of 3 consecutive 24 h
sampling units, with one sampling period scheduled
each week. The actual start day was chosen randomly
using Julian calendar days. On day 1 (first sampling
unit), researchers marked all dead animals found
throughout the study site. On days 2 and 3 (sampling
units 2 and 3), all road kills that had accumulated during
the previous 24 h were recorded.
Road kill surveys began at first light and all live and
dead animals were recorded. Dead animals were marked
with Day-GloÒorange spray paint so that they were not
counted more than once. The paint was free of lead and
toluene. Locations of all animals were recorded, that is,
whether the animal was in the north or southbound
lane, the right-of-way, or the grassy median, and in
which 100 m section it was found. Freshly killed, un-
damaged specimens were collected and deposited in the
Florida Museum of Natural History at the University of
Florida.
Beginning 5 September 2001, our sampling protocol
was adjusted so that treefrogs (Family Hylidae) were
counted only in the southbound lanes of three randomly
selected Sections (3, 14, and 23). We made this change
because of the large volume of morning commuter traffic
and associated safety concerns. The large volume of
traffic quickly obliterated hylid carcasses, and we chose
to rigorously sample three highway sections rather than
C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631 621
to underestimate hylid mortality over the entire study
length. In the results that follow, hylid totals are thus for
the entire 3.2 km survey area from 14 March to 4 Sep-
tember 2001, but only for the 3 100 m sections from 5
September to the end of the survey.
2.3. Culvert survey
The eight culverts that underlie US 441 also were
monitored for wildlife use. Culverts were numbered one
to eight from north to south (Fig. 2). Culverts 1 and 8
were usually dry with an earthen substrate, whereas
culverts 2 and 7 were inundated throughout the study.
Culverts 3–6 were installed as part of the new barrier
wall-culvert system. Culvert 4 was usually wet, whereas
the other culverts (3, 5, and 6) were dry or wet de-
pending on prairie water levels. These culverts also had
earthen substrates.
Wire screen-mesh funnel traps (see Fig. 15 in Karns,
1986) were installed in the four box culverts to sample
amphibians, reptiles, and small mammals. In both
1.8 1.8 m box culverts (1 and 8), 10 square hardware-
cloth funnel traps were placed flush with the sides of the
culverts. We installed 10 floating screen funnel traps in
the remaining two 2.4 2.4 m box culverts (2 and 7) in
the center of the culvert under the roadway. The latter
two culverts were monitored until they became unsuit-
able for sampling because of high water levels or the
presence of alligators.
Each of the four recently-constructed 0.9 m diameter
culverts (3, 4, 5, and 6) were sampled using two com-
mercial crayfish traps (Lee Fisher, Inc., Tampa, FL;
Darby et al., 2001) placed in each of two light boxes
located in the right-of-way, for a total of four traps per
culvert. The trapping schedule coincided with the road
surveys, although it was adjusted to include two addi-
tional sampling units; thus, all culverts were sampled
five nights per week.
A sand track station (1.8 m long by 1.0 m wide) oc-
casionally was set up in the center of the northern, dry
culvert (number 1). TrailMaster TM1500ÒActive In-
frared monitors and cameras (Goodson and Associates,
Width (m)
West ROW 10
South travel lanes 9
Median 6
North travel lanes 9
East ROW 10
Total 44
Symbol
Distance
from 0 (m)
Culvert
size(m)
a. Beginning of survey area 0
b.
Start of type-A fence, start of west
concrete barrier, North Rim 200
c.
End of type-A fence, beginning of
east concrete barrier 650
d. Cattle gate 675
1 Culvert 1 700 1.8 x 1.8
2 Culvert 2 1200 2.7 x 2.7
3 Culvert 3 1600 0.9 dia
4 Culvert 4 2000 0.9 dia
5 Culvert 5 2300 0.9 dia
6 Culvert 6 2600 0.9 dia
7 Culvert 7 2800 2.7 x 2.7
8 Culvert 8 3000 1.8 x 1.8
e.
End of east west concrete barriers
South Rim 3015
f. End of survey area 3200
N
3 2 1
d.
c. b.
54
6
78
e.
f.
Fig. 2. Schematic representation of US Highway 441 across Paynes Prairie State Preserve. The road is bordered by a prefabricated concrete barrier
wall and underlain by 8 culverts. Light boxes (squares) occur across the road only in the new small culverts (g, h, i, j) to allow light; a small grate is in
culvert l. An access road enters on the southbound lane near the northern prairie rim (d), and a visitor turn out is located between culverts g and h. A
type-A fence borders private property along the southbound lanes on the north prairie rim (b to c).
622 C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631
Inc., Lenexa, KS) were installed at the center of the
north and south box culverts (1 and 8) to record the use
of these culverts by larger vertebrates. To be recorded,
an animal had to pass through the infrared light beam;
thus, animals less than ca. 30 cm in height were not
recorded. The track station and cameras were moni-
tored five days per week, weather and water conditions
permitting.
3. Results
3.1. Live animals on highway
Only 13 live vertebrates were observed on the high-
way during the year-long post-construction survey: 2
Eastern Narrow-mouthed Toads (Gastrophryne carolin-
ensis), 1 unidentified Treefrog (Hyla sp.), 2 Green
Treefrogs (Hyla cinerea), 1 Squirrel Treefrog (H. squi-
rella), 1 Southern Leopard Frog (Rana sphenocephala), 1
Eastern Mud Turtle (Kinosternon bauri), 3 Brown An-
oles (Anolis sagrei), 1 Broad-headed Skink (Eumeces
laticeps), 1 Eastern Glass Lizard (Ophisaurus ventralis),
1 Southern Watersnake (Nerodia fasciata), 1 Coyote
(Canis latrans).
3.2. Species killed
We counted 1891 dead vertebrates (1647 frogs; 1
alligator; 7 turtles; 4 lizards; 149 snakes; 83 mammals)
during the post-construction phase of the study
(Table 1). The total includes all of the dead animals
observed during 152 road kill surveys over the entire 3.2
km study area within the prairie basin and on the ad-
jacent rim. With the exception of hylid treefrogs, which
easily trespass the barrier system by climbing up the
barrier wall and adjacent vegetation, the mean number
of vertebrate kills per 24-h sampling period was 4.9
(SD ¼17.4; range ¼0–162). Monthly means ranged
from 0.75 to 33.2 vertebrates killed per 24-h period
(Fig. 3). The mean number of animals killed per month
was less than 10 throughout the study except for Sep-
tember 2001 when large numbers of G. carolinensis were
killed (see below).
Hylid treefrogs (n¼1301) accounted for 68.8% of all
road kills counted throughout the post-construction
survey. In survey Sections 3, 14, and 23 where the
sampling protocol was the same between pre- and post-
construction surveys, 149 hylids were counted during the
pre-construction survey versus 194 during the post-
construction survey. Most treefrogs could not be iden-
tified to species because of the extent of body damage,
although H. cinerea (n¼135) accounted for at least
10.4% of the hylids. The second most abundant species
found dead on the highway, also an amphibian, was
G. carolinensis (10.9%; n¼218). One hundred sixty-two
Eastern Narrow-mouthed Toads were counted on a
single day (6 September 2001) in the northbound lane in
Section 1 (on the prairie rim) beyond the barrier wall
(represented by the extended bar in Fig. 3).
DeKayÕs Brownsnakes (Storeria dekayi;n¼54) and
Southern Watersnakes (N. fasciata;n¼21) were the
most commonly killed snakes (Table 1). Most S. dekayi
also were killed on the north and south prairie rims
(n¼28). However, this small, semifossorial species also
may have colonized the highway right-of-way, based on
the number killed (n¼26) that were rather evenly dis-
tributed across the prairie basin. Most N. fasciata were
killed within the prairie basin (81%), but the relatively
large percentages killed in front of the prairie access gate
(33%; Section 7) and type-A fence (19%; Sections 3–6)
suggest that trespass within the basin was opportunistic
rather than a failure of the wildlife barrier to deter
movement. Only 12 other individual reptiles were
counted dead during the study.
The Rice Rat (Orozomys palustris) was the most
common mammal observed dead on the highway, with
Opossums and Armadillos next in abundance. Few in-
dividuals of other mammal species were found (Table 1).
Rice Rats trespassed the fence by climbing adjacent veg-
etation, whereas Opossums, Armadillos and other large
mammals were usually found dead on the prairie rims.
3.3. Time and location of kills
Most post-construction mortality was recorded in
August and September (729 and 727, respectively, or 132
and 224 if hylid treefrogs are excluded; Fig. 4). Ap-
proximately 73% of road kills, excluding hylid treefrogs,
occurred in the ecotonal areas on the prairie rim beyond
the extent of the barrier wall-culvert system (Table 2,
Fig. 5). A total of only 158 dead amphibians, reptiles,
and mammals was counted within the prairie basin.
Most of the dead animals on the roadway within the
prairie basin were found in sections adjacent to the type-
A fence (Sections 3–6; n¼62; 39%) and in front of the
prairie access gate (Section 7; n¼39; 25%).
Mortality was higher on the north rim than the south
rim of the prairie beyond the barrier wall during the
post-construction phase of the survey (Table 3). The
majority of kills were G. carolinensis during September
2001 (203 of 302 [67%] recorded kills). Mortality on the
south rim of the prairie beyond the barrier wall also
increased during the post-construction phase of the
survey. Unlike the north rim, mortality on the south rim
resulted mostly from kills of Southern Toads (Bufo
terrestris) and was more evenly distributed throughout
the year.
Most carcasses were located in the outside lanes
(46.5%) and contiguous bicycle lanes (28.5%), followed
by the inside lane (11.1%) and grassy right-of-way
(9.9%, Table 3). Very few animals were found in the
C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631 623
Table 1
Vertebrate roadkills on US 441 at Paynes Prairie State Preserve from 14 March 2001 through 5 March 2002
Scientific name Common name Within prairie basin Rim Total
Frogs
Bufo terrestris Southern Toad 7 71 78
Gastrophryne carolinensis Eastern Narrow-mouthed Toad 4 214 218
Hyla cinerea Green Treefrog 101 34 135
Hyla squirella Squirrel Treefrog 7 6 13
Hyla sp. Unidentified Treefrog 763 390 1153
Rana sphenocephala Southern Leopard Frog 6 6 12
Rana sp. Unidentified Ranid 5 7 12
Scaphiopus holbrookii Eastern Spadefoot 7 7
Unidentified frog 6 13 19
Crocodilian
Alligator mississippiensis American Alligator 1 1
Turtles
Chelydra serpentina Snapping Turtle 2 2
Kinosternon bauri Striped Mud Turtle 2 2
Kinosternon sp. Unidentified Mud Turtle 1 1
Pseudemys nelsoni Florida Red-bellied Turtle 1 1 2
Lizards
Anolis sp. Unidentified Anole 1 1
Ophisaurus sp. Unidentified Glass Lizard 2 2
Ophisaurus ventralis Eastern Glass Lizard 1 1
Snakes
Agkistrodon piscivorus Cottonmouth 1 1
Coluber constrictor Eastern Racer 4 4
Diadophis punctatus Ring-necked Snake 1 1 2
Elaphe guttata Cornsnake 2 2
Elaphe obsoleta Yellow Ratsnake 5 11 16
Farancia abacura Red-bellied Mudsnake 7 1 8
Lampropeltis triangulum Milksnake 1 1
Nerodia fasciata Southern Watersnake 18 3 21
Nerodia floridana Florida Green Watersnake 2 1 3
Opheodrys aestivus Rough Greensnake 1 1
Seminatrix pygaea Black Swampsnake 10 2 12
Storeria dekayi DeKayÕs Brownsnake 26 28 54
Thamnophis sauritus Eastern Ribbonsnake 2 1 3
Thamnophis sirtalis Common Gartersnake 7 5 12
Unidentified snake 5 4 9
Mammals
Blarina carolinensis Southeastern Short-tailed
Shrew
11
Canis familiaris Domestic Dog 1 1
Canis latrans Coyote 2 2
Dasypus novemcinctus Nine-banded Armadillo 2 8 10
Didelphis virginianus Virginia Opossum 1 14 15
Lutra canadensis River Otter 1 1
Odocoileus virginianus White-tailed Deer 1 2 3
Oryzomys palustris Rice Rat 17 8 25
Peromyscus gossypinus Cotton Mouse 1 1 2
Procyon lotor Raccoon 5 5
Sigmodon hispidus Hispid Cotton Rat 1 1 2
Sylvilagus palustris Marsh Rabbit 1 1 2
Sylvilagus sp. Unidentified rabbit 1 1
Urocyon cinereoargenteus Gray Fox 1 1
Unidentified bat 2 3 5
Unidentified mammal 3 4 7
Total 1028 863 1891
Nomenclature of amphibians and reptiles follows Crother (2000).
624 C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631
median (3.6%) and directly on the centerline between
lanes (0.4%) (Table 3). Wildlife mortality in the north-
bound lanes was 1.75 times that of southbound lanes
(Table 3).
3.4. Culvert use
We recorded 51 vertebrate species using the culverts
during post-construction monitoring (Table 4). We re-
corded 1046 captures during 7580 trap-nights (13.8%
capture success) in funnel traps. Most captures occurred
from mid-June to early July, when large numbers of
juvenile Southern Leopard Frogs (40.5%) passed
through culvert 8. Captures of Rice and Hispid Cotton
Rats (Oryzomys palustris and Sigmodon hispidus, re-
spectively) were most numerous in the dry culverts
(culverts 1, 3, 5, 6, 8; Table 4) during the summer.
Tracks of Ninebanded Armadillo (Dasypus novemcinc-
tus), River Otter (Lutra canadensis), Virginia Opossum
(Didelphis virginianus), and Raccoon (Procyon lotor)
often were observed in the dry north culvert (number 1).
These four species also were repeatedly photographed
with the motion sensor cameras. Two species previously
undocumented from the culverts, the Marsh Rabbit
(Sylvilagus palustris) and American Alligator (Alligator
mississippiensis), were photographed in culvert 1.
The newly installed rounded culverts often contained
considerable amounts of water, and appeared to be used
readily by fish, when inundated, and small mammals,
when dry. Amphibians and reptiles were captured or
observed less often (Table 4), perhaps because they were
able to travel through or around the commercial cray-
fish traps. Because of the small diameter of these cul-
verts and their sometimes wet environs, we were unable
to use motion cameras or track stations to monitor
culvert use. Other vertebrate species may have used the
culverts and not been captured in the specialized traps.
4. Discussion
4.1. Is the barrier wall-culvert system effective?
Judging the effectiveness of a mitigation project of the
magnitude of the Paynes Prairie barrier wall-culvert
system is fraught with interpretive problems. Under
ideal conditions, the pre- and post-construction surveys
should have been carried out under exactly the same
circumstances. The surveys should have begun and
ended on the same dates, carcass detection probabilities
should be equal among taxa, and the environmental
conditions (temperature and precipitation patterns,
Pre-construction
survey
Month
Aug 98
Sep 98
Oct 98
Nov 98
Dec 98
Jan 99
Feb 99
Mar 99
Apr 99
May 99
Jun 99
Jul 99
Aug 99
Mar 01
Apr 01
May 01
Jun 01
July 01
Aug 01
Sep 01
Oct 01
Nov 01
Dec 01
Jan 02
Feb 02
Mar 02
Number of animals killed/24hr
0
15
30
45
60
75 Post-construction
survey
Fig. 3. Monthly roadkill totals on US Highway 441 across Paynes Prairie State Preserve, Alachua County, Florida one-year prior to and after
construction of the barrier wall-culvert system.
C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631 625
water levels) should be controlled experimentally. It
would also have been desirable to avoid possible ob-
server biases, such as by using the same experienced
observers, and to control for human-caused variables,
such as traffic volume and noise. Given the scale of the
project and limitations imposed by construction and
environmental stochasticity, however, this was not
possible. If the magnitude of highway mortality was
similar between surveys, judging the effectiveness of the
barrier wall-culvert system might be extremely difficult.
Forman et al. (2003) discussed six criteria for deter-
mining the success of barrier-related mitigation projects.
If the objectives in constructing the barrier wall-culvert
system were to reduce the mortality of vertebrates, es-
pecially the larger vertebrates, and to allow the passage
of some animals across the highway in order to maintain
habitat connectivity and gene flow between populations
on opposite sides of the roadway corridor, we suggest
that three lines of evidence argue for the success of the
Paynes Prairie project.
4.1.1. Mortality greatly decreased, especially for large
species
In the 12 months prior to construction of the barrier
wall-culvert system, 2411 road kills were recorded in the
study area (Smith and Dodd, 2003), whereas only 158
animals were killed during the post-construction sur-
veys. Within the survey area lying directly in the Paynes
Prairie basin, mortality was reduced 65% if hylid tree-
frogs are included, and 93.5% with hylid treefrogs ex-
Sep 98
Oct 98
Nov 98
Dec 98
Jan 99
Feb 99
Mar 99
Apr 99
May 99
Jun 99
Jul 99
Aug 99
Mar 01
Apr 01
May 01
Jun 01
Jul 01
Aug 01
Sep 01
Oct 01
Nov 01
Dec 01
Jan 02
Feb 02
Mar 02
Number of Roadkills
0
200
400
600
800
Monthly total excluding hylids
Hylids
Pre-construction Post-construction
Fig. 4. Mean number of roadkills per 24 h period, exclusive of hylid treefrogs, on US Highway 441 across Paynes Prairie State Preserve, Alachua
County, Florida one-year prior to and after construction of the barrier wall-culvert system. One standard deviation is expressed by the error bars.
The large number of kills recorded in September 2001 results from an anomalous event whereby 162 Eastern Narrow-mouthed Toads were on a
single day. In contrast, the kills in August 2001 were more evenly distributed throughout the month.
Table 2
Pre- and post-construction highway-related mortality on US Highway 441 in sections bordered and not bordered by the concrete barrier wall
Sections Pre-construction
mortality
Pre-construction
mortality w/o hylids
Post-construction
mortality
Post-construction
mortality w/o hylids
1 and 2 146 85 595 302
3–30 2863 2411 1029 158
31 and 32 92 64 267 130
Total 3101 2560 1891 590
Sections 1, 2, 31 and 32 had no barriers to highway access by wildlife; Sections 3–30 were bordered by the concrete barrier wall.
626 C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631
Table 3
Location of wildlife kills (excluding hylid treefrogs) on the road surface, right-of-way, and median of US 441 at Paynes Prairie State Preserve,
Alachua County, Florida
Location Northbound Southbound Median Total
Amphibians Right-of-way 1 1
Bike lane 89 15 104
Outer lane 136 38 174
Centerline 0
Inner lane 5 8 13
Median 22
Total 230 62 2 294
Reptiles Right-of-way 10 16 26
Bike lane 11 12 23
Outer lane 11 30 41
Centerline 0
Inner lane 5 6 11
Median 33
Total 37 64 3 104
Mammals Right-of-way 4 2 6
Bike lane 3 3
Outer lane 7 4 11
Centerline 1 1
Inner lane 7 8 15
Median 55
Total 22 14 5 41
Lane total 289 140 9 439
Surveys were conducted from 15 March 2001 through 5 March 2002. The data presented represents the two 24 h sampling units (2 and 3) collected
weekly throughout the study.
Number of Pre-construction
Roadkills (n=2560)
050100150200
Road Section (100m)
0
5
10
15
20
25
30
Number of Post-construction
Roadkills (n=590)
0 50 100 150 200
Amphibians
Reptiles
Mammals
North Rim
Beginning of
wildlife barrier
End of wildlife
barrier
South Rim
Fig. 5. Number of roadkills, excluding hylid treefrogs, per 100 m-section of US Highway 441 across Paynes Prairie State Preserve, Alachua County,
Florida. The survey area commenced at the first private drive on the north rim of the prairie (Section 1) and extended 3.2 km to the first private
driveway on the south rim (Section 3.2). Pre-construction data were collected from 18 August 1998 through 13 August 1999, and post-construction
data were collected from 14 March 2001 to 5 March 2002. The concrete barrier wall adjacent to the roadway extends from Section 3 to Section 30.
C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631 627
Table 4
Vertebrates observed in culverts under US Highway 441 across Paynes Prairie, Alachua County, Florida, 14 March though 5 March 2002
Scientific name Common name Culvert no. (N)
Fish (n¼9)
Ameiurus nebulosus Brown bullhead 2(1), 4(1)
Elassoma sp. Pygmy sunfish 2(5), 7(21)
Etheostoma fusiforme Swamp darter 7(1)
Fundulus chrysotus Golden topminnnow 7(1)
Gambusia holbrooki Mosquitofish 2(12), 7(88)
Heterandria formosa Least killifish 2(1), 7(19)
Lepisosteus platyrhincus Florida gar 2(3), 3(4), 4(1), 5(2), 6(1)
Lepomis gulosus Warmouth 4(16), 6(1)
Lepomis macrochirus Bluegill 4(2)
Salamanders (n¼2)
Amphiuma means Two-toed Amphiuma 4(1), 6(2)
Siren lacertina Greater Siren 4(2), 5(1)
Frogs (n¼11)
Acris gryllus Southern Cricket Frog 8*
Bufo terrestris Southern Toad 1(9), 8*(36)
Gastrophryne carolinensis Narrow-mouthed Toad 1(3), 6*, 8*(7)
Hyla cinerea Green Treefrog 1(4), 3*, 4*, 5*, 6*, 7*, 8*(34)
Hyla femoralis Pine Woods Treefrog 8(1)
Hyla squirella Squirrel Treefrog 4*, 5*, 8(6)
Rana catesbeiana American Bullfrog 6(1)
Rana clamitans Green Frog 8*(1)
Rana grylio Pig frog 4*, 6(1)
Rana sphenocephala Southern Leopard Frog 1(14), 2(12), 3*, 5*, 7(85), 8*(424)
Scaphiopus holbrooki Eastern Spadefoot 8(1)
Crocodilians (n¼1)
Alligator mississippiensis American Alligator 1*, 2*, 4*(6), 7*, 8*
Turtles (n¼4)
Apalone ferox Florida Softshell 3*, 4(3)
Kinosternon baurii Striped Mud Turtle 8*(2)
Pseudemys nelsoni Florida Red-bellied Turtle 2(1)
Sternotherus odoratus Stinkpot 4(1)
Lizards (n¼1)
Anolis carolinensis Green Anole 1*
Snakes (n¼11)
Agkistrodon piscivorus Cottonmouth 1(4), 3*, 5*, 7*, 8(3)
Coluber constrictor Eastern Racer 1(1)
Diadophis punctatus Ring-necked Snake 4*
Elaphe guttata Cornsnake 1(1)
Elaphe obsoleta Yellow Ratsnake 8(1)
Farancia abacura Eastern Mudsnake 8*
Nerodia fasciata Southern Watersnake 1(4*), 3(1), 4*, 5*, 6*, 7*, 8*
Nerodia floridana Florida Green Watersnake 1(2)
Storeria dekayi DekayÕs Brownsnake 1(1), 3*
Thamnophis sauritus Eastern Ribbonsnake 1(4)
Thamnophis sirtalis Common Gartersnake 1(1)
Mammals (n¼12)
Blarina carolinensis Southeastern Short-tailed Shrew 1(6), 8(4)
Dasypus novemcinctus Nine-banded Armadillo 1*, 8*
Didelphis virginianus Virginia Opossum 1*, 8*
Lutra canadensis River Otter 1*, 6*
Lynx rufus Bobcat 1*
Myotis austroriparius Southeastern Bat 8*
Neofiber alleni Round-tailed Muskrat 1(1), 6(1), 8*(4)
Oryzomys palustris/Sigmodon hispidus Rice Rat/Hispid Cotton Rat 1*(17), 3(29), 5(19), 6(20), 8(19)
Peromyscus gossypinus Cotton Mouse 1(2), 3(3), 5(2)
Procyon lotor Raccoon 1*, 8*
Sylvilagus palustris Marsh Rabbit 1*
*indicates photo, track, or incidental observation.
628 C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631
cluded. Snake (1,291 vs. 149), turtle (374 vs. 7), and
alligator (29 vs. 1) mortality also was notably lower in
the post-construction survey as well. We suggest that
this decrease in highway-related mortality cannot be
attributed solely to changes in water level. The species
killed during the post-construction survey were mostly
small vertebrates that either climbed over the wall or
trespassed the barrier at the maintenance road access
ramp or at the inadequate barrier formed by the type-A
fence. In contrast, more larger vertebrates were killed
during the pre-construction survey, and many were kil-
led throughout the basin on the highway. Snake, turtle,
ranid frog, and alligator mortality declined dramatically
with the construction of the barrier wall-culvert system,
yet treefrog mortality appears to have increased (149 to
194 in the road sections monitored in the same manner
from pre- to post-construction).
The apparent increase is most likely a result of dif-
ferences in water levels on the prairie between pre- and
post-construction surveys. Increased levels of amphibian
mortality observed in the post-construction survey may
be related to the decreasing water levels on the prairie,
which caused animals to move to terrestrial habitats.
During the period of high water in the prairie basin
during part of the pre-construction survey, hylids may
have been less likely to move than they did when water
receded during the drought associated with much of the
post-construction survey. In any case, mortality is un-
doubtedly underestimated for amphibians since small
treefrogs are quickly obliterated by vehicles. The large
number of hylids killed on the highway indicates that
some animals likely will be killed on US 441, despite the
presence of a barrier. At this time, we know of no
practical, effective way to reduce hylid mortality.
4.1.2. Use of culverts increased
In addition to a decrease in highway-related mortality
of most vertebrates, we observed an increase in culvert
use by many species. Capture success increased tenfold
between the pre- and post-construction surveys. The
increase was most pronounced in the number of indi-
vidual amphibians using the culverts (0.006–0.085 cap-
tures per trap-night, respectively). For example, the
huge decline in the number of ranid frogs killed on the
highway, combined with an increase in the use of cul-
verts by these species, provides evidence of the effec-
tiveness of the wall-culvert system to prevent mortality
of certain amphibians while allowing passage under the
highway.
Additionally, the number of species using the culverts
increased from 28 to 42 (excluding fish), and was most
apparent in the number of amphibian species using the
culverts (from 5 to 13 species). Because of sampling
limitations, we were unable to quantify culvert use.
Nonetheless, movements through culverts by at least
some individuals, as observed in this study, should be
sufficient to maintain habitat connectivity and gene flow
between the east and west sides of the prairie. Hypo-
thetically, only a few individuals need to cross a barrier
per generation to maintain heterozygosity (Meffee and
Carroll, 1997; Forman et al., 2003), although roads have
effectively altered the population structure of some small
vertebrates living on different sides of a highway (Reh
and Seitz, 1990; Gerlach and Musolf, 2000).
4.1.3. Use of ecotones increased for some species
Terrestrial vertebrates commonly use topographic
features or ecotones as movement or dispersal corridors
(e.g., Healy, 1975). Vertebrate mortality on US 441,
particularly of the mammals and certain terrestrial am-
phibians, increased at the ecotonal areas on both prairie
rims after construction. This suggests that the barrier
wall directed animals along the bottom of the barrier to
the first place where they could cross, the end of the
wall. The fence also likely directed animals to the cul-
verts, as evidenced by the increase in culvert use post-
construction.
Five of the six criteria outlined by Forman et al.
(2003) to judge success appear to be met along the US
441 barrier wall-culvert system at Paynes Prairie: a re-
duction road-kill rates post-construction, maintenance
of habitat connectivity, maintenance of genetic inter-
change, allowance for dispersal and recolonization (see
4.2), and maintenance of metapopulation processes and
ecosystem function. The Florida Park Service has been
working to re-establish natural water flow regimes on
the prairie since the area became a state preserve. Old
dikes have been removed, ditches have been filled, and
prescribed fires have helped to restore the prairieÕs hy-
drology. However, little is known concerning the pop-
ulation size and structure for most species occurring
within the prairie basin, except for a few large mammals
and alligators. Call surveys and road counts suggest that
populations of many amphibians, reptiles, and small
mammals may be very large on the vast wetland. Until
more precise data are available, however, the last of
Forman et al.Õs criteria, the ability to ensure that bio-
logical requirements are met, cannot be evaluated.
4.2. Mortality patterns and season
We observed noticeable differences in wildlife mor-
tality, depending on highway direction. A small number
of carcasses were found in the median suggesting that
few animals made it that far. Differences in mortality
between the north and southbound lanes could result
from the large volume of traffic moving toward
Gainesville during the very early morning hours, when
nocturnal and crepuscular animals were still active. In
the late afternoon during the heat of the day, fewer
animals were likely to be active and thus encountered on
C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631 629
the highway when motorists working in Gainesville re-
turn to their homes south of the Prairie.
Highway-related mortality was greatest during the
late summer months in both surveys, and this pattern is
consistent with the results from the most systematic
previous survey (Franz and Scudder, 1977). Some ani-
mals also may have been moving in a particular direc-
tion at a certain time of the year, such as when juvenile
frogs disperse after metamorphosis. For example, large
numbers of juvenile R. sphenocephala were captured
moving from east to west in culvert funnel traps in July
2002, presumably as the dispersed away from the drying
wetlands on the eastern side of the highway.
5. Recommendations
Patterns of mortality observed in the study revealed
several maintenance and design problems in the Paynes
Prairie barrier wall-culvert system that may be fixed in
this system and avoided in ecological passage projects
elsewhere. Early in the study, we observed small mam-
mals, snakes and treefrogs climbing vegetation along the
barrier wall. The vegetation offered access to the highway
right-of-way and some of the animals observed dead on
the roadway undoubtedly obtained access in this man-
ner. Once we identified the problem, FDOT responded
by applying herbicide along the barrier wall, which
temporarily resolved the problem. However, to minimize
trespass, vegetation must be removed from barriers
regularly, particularly during the growing season.
We documented a peak in the numbers of animals
killed on the roadway adjacent to a maintenance road
access ramp in the northwest part of the study area
(Section 7). A second access road is located in the
southeast part of the prairie. We did not observe high
mortality in this area because it does not ascend to the
level of US 441. In wet years, this access road may allow
animals to trespass onto the roadway by funneling them
toward the south rim beyond the barrier wall. Widely
spaced cattle grates could be used to prevent small an-
imals from accessing the roadway, while allowing service
vehicle access to the prairie.
Significant trespass occurred at the type-A fence on
the northwest edge of our study area. Gaps developed
beneath the barricades as a result of erosion of the top
soil, allowing snakes, turtles, and frogs free access to
the highway. We recommend burying galvanized tin or
aluminum flashing to a depth of at least 20 cm below
barricades to prevent access. However, this method
will require regular maintenance because mammals are
likely to attempt to dig under barriers and over time,
sheet flow associated with heavy rain could still cause
erosion. Since the type-A fence occurs in an area
where many animals were killed in the pre- and post-
construction survey, more permanent structures should
be used, accompanied with ample culverts to act as
underpasses.
Culverts require regular maintenance. During heavy
rain, flow through culverts in our study increased dra-
matically causing silt to accumulate. Although a natural
substrate on the floor of the concrete culverts might
enhance their use by some animals (Yanes et al., 1995), a
heavy build-up of silt would eventually diminish the
area available for wildlife passage. Also, during con-
struction, deep pools were created outside six of the
eight culverts. These pools attracted alligators, which
could discourage movement of animals through the
culvert. Animals exiting a culvert into these pools might
be subject to an increased risk of predation, although we
did not see evidence of this (also see Little et al., 2002).
These pools should be filled in to match the grade of the
surrounding prairie.
Finally, once in place, the barrier wall-culvert system
received an unanticipated amount of public interest.
Modifications to the right-of-way during construction
created space for people to park and observe wildlife on
the prairie. During the post-construction survey, water
levels on the prairie were low and alligators were con-
centrated in the canals adjacent to the wall. Although
illegal, people stopping along the wall regularly fed the
alligators and often climbed down to the prairie to do
so, creating a safety concern. Eventually, the FDOT
erected a fence in the right-of-way restricting public
access to the barrier wall. The fence seems to have
eliminated safety concerns, but is an obstruction to what
was once a scenic view of the prairie. Such issues may
arise in other barrier wall-culvert projects and could
perhaps be resolved in the planning process. Ideally, an
barrier wall-culvert system should be designed to not
only mitigate for wildlife mortality and public safety,
but also to allow public access and appreciation for the
resource being protected.
Acknowledgements
This study was funded by the Florida Department of
Transportation; we thank Gary Evink and Pete Southall
for their support of the project. We are grateful to Jack
Gillen, Dan Pearson, and Jim Weimer of the Florida
Department of Environmental Protection for providing
mortality data for Paynes Prairie. We also thank Dick
Franz and Sylvia Scudder for making their unpublished
data available to us. Kelly Keefe, Audrey Owens,
Kristina Sorensen, Jennifer Staiger, and Maya Zacha-
row assisted in field surveys. Tom Webber, Candace
McCaffery, and Laurie Wilkens helped with the identi-
fication of small mammals. Dick Franz, Russell Hall
and two anonymous reviewers provided helpful com-
ments on the manuscript. This work was carried out
630 C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631
under Florida Fish and Wildlife Conservation Com-
mission Permits WS98348 and WS01511.
References
ALASV (Arbeitsgruppe unter Leitung der Abteilung Straenbau des
Verkehrsministeriums). 1994. Amphibienschutz. Leitfaden f€
ur
Schutzmanahmen an Straen. Schrifttenreihe der Straenbauver-
waltung Baden W€
urttemberg, Heft 4.
Andrews, A., 1990. Fragmentation of habitat by roads and utility
corridors: a review. Australian Zoologist 26, 130–141.
Ashley, E.P., Robinson, J.T., 1996. Road mortality of amphibians,
reptiles and other wildlife on the Long Point Causeway, Lake Erie,
Ontario. Canadian Field-Naturalist 110, 403–412.
Beck, W.M., 1938. Notes on the reptiles of PayneÕs Prairie, Alachua
County, Florida. Florida Naturalist 11, 85–87.
Bonnet, X., Naulleau, G., Shine, R., 1999. The dangers of leaving
home: dispersal and mortality in snakes. Biological Conservation
89, 39–50.
Carr Jr., A.F., 1940. A contribution to the herpetology of Florida. In:
Biological Science Series, 3. University of Florida Publication. 1–
118.
Carr, A.F., 1974. The Reptiles. Time-Life Books, New York.
Carr, L.W., Fahrig, L., 2001. Effect of road traffic on two amphibian
species of differing vagility. Conservation Biology 15, 1071–1078.
Case, R.M., 1978. Interstate highway road-killed animals: a data
source for biologists. Wildlife Society Bulletin 6, 8–13.
Clevenger, A.P., Waltho, N., 1999. Dry drainage culvert use and
design considerations for small-and medium-sized mammal move-
ment across a major transportation corridor. In: Evink, G.L.,
Garrett, P., Ziegler, D. (Eds.), Proceedings of the Transportation
Related Wildlife Mortality Seminar. Florida Department of
Transportation (FL-ER-58-96), Tallahassee, FL, pp. 263–278.
Clevenger, A.P., Chruszcz, B., Gunnison, K., 2001. Drainage culverts
as habitat linkages and factors affecting passage by mammals.
Journal of Applied Ecology 38, 1340–1349.
Clevenger, A.P., Chruszcz, B., Gunnison, K., 2003. Spatial patterns
and factors influencing small vertebrate fauna road-kill aggrega-
tions. Biological Conservation 109, 15–26.
Crother, B.I. (chair)., 2000. Scientific and standard English names of
amphibians and reptiles of North America north of Mexico, with
comments regarding confidence in our understanding. Society for the
Study of Amphibians and Reptiles, Herpetological Circular No. 29.
Darby, P.C., Valentine-Darby, P.L., Percival, H.F., Kitchens, W.M.,
2001. Collecting Florida applesnails (Pomacea paludosa) from
wetland habitats using funnel traps. Wetlands 21, 308–311.
de Maynadier, P.G., Hunter Jr., M.L., 2000. Road effects on
amphibian movements in a forested landscape. Natural Areas
Journal 20, 56–65.
Duguet, R., Melki, F. (Eds.), 2003. Les Amphibiens de France,
Belgique et Luxembourg. Collection Parth
enope,
editions Biotope,
M
eze, France.
Fahrig, L., Pedlar, J.H., Pope, S.E., Taylor, P.D., Wegner, J.F., 1995.
Effect of road traffic on amphibian density. Biological Conserva-
tion 73, 177–182.
Findlay, C.S., Houlahan, J., 1997. Anthropogenic correlates of species
richness in southeastern Ontario wetlands. Conservation Biology
11, 1000–1009.
Forman, R.T.T., Alexander, L.E., 1998. Roads and their major
ecological effects. Annual Review of Ecology and Systematics 29,
207–231.
Forman, R.T.T., Deblinger, R.D., 2000. The ecological road-effect
zone of a Massachusetts (USA) suburban highway. Conservation
Biology 14, 36–46.
Forman, R.T.T., Sperling, D., Bissonette, J.A., Clevenger, A.P.,
Cutshall, C.D., Dale, V.H., Fahrig, L., France, R., Goldman, C.R.,
Heanue, K., Jones, J.A., Swanson, F.J., Turrentine, T., Winter,
T.C., 2003. Road Ecology Science and Solutions. Island Press,
Washington, DC.
Franz, R., Scudder, S.J., 1977. Observations of snake movements on a
north Florida highway. Unpublished Report. Florida State Mu-
seum. University of Florida, Gainesville, Florida.
Gerlach, G., Musolf, K., 2000. Fragmentation of landscape as a cause
for genetic subdivision in bank voles. Conservation Biology 14,
1066–1074.
Gibbs, J.P., 1998. Amphibian movements in response to forest edges,
roads, and streambeds in southern New England. Journal of
Wildlife Management 62, 584–589.
Gibbs, J.P., Shriver, W.G., 2002. Estimating the effects of road mortality
on turtle populations. Conservation Biology 16, 1647–1652.
Healy, W.R., 1975. Breeding and postlarval migrations of the red-
spotted newt, Notophthalmus viridescens, in Massachusetts. Ecol-
ogy 56, 673–680.
Heine, G., 1987. Einfache Meß- und Rechenmethode zur Ermittlung
der €
Uberlebenschance wandernder Amphibien beim €
Uberqueren
von Straßen. Beihefte zu den Ver€
offentlichungen Naturschutz und
Landschaftspflege in Baden-W€
urttemberg 41, 473–479.
Hellman, R.E., Telford, S.R., 1956. Notes on a large number of red-
bellied mudsnakes, Farancia a. abacura, from northcentral Florida.
Copeia 1956, 257–258.
Hels, T., Buchwald, E., 2001. The effect of road kills on amphibian
populations. Biological Conservation 99, 331–340.
Jacobs, J.M., Mergelsberg, S.L., Lopera, A.F., Myers, D.A., 2002.
Evapotranspiration from a wet prairie wetland under drought
conditions: Paynes Prairie Preserve, Florida, USA. Wetlands 22,
374–385.
Karns, D.R., 1986. Field Herpetology. Methods for the Study of
Amphibians and Reptiles in Minnesota. James Ford Bell Museum
of Natural History, Occasional Paper 18.
Kauffeld, C., 1957. Snakes and Snake Hunting. Hanover House,
Garden City, New York.
Kushlan, J.A., 1980. Freshwater marshes. In: Myers, R.L., Ewel, J.
(Eds.), Ecosystems of Florida. University of Central Florida Press,
Orlando, FL, pp. 324–363.
Langton, T.E.S. (Ed.), 1989. Amphibians and Roads. ACO Polymer
Products, Ltd., Shefford, England.
Little, S.J., Harcourt, R.G., Clevenger, A.P., 2002. Do wildlife
passages act as prey-traps? Biological Conservation 107, 135–145.
Meffee, G.K., Carroll, C.R., 1997. Genetics: conservation of diversity
within species. In: Meffee, G.K., Carroll, C.R. (Eds.), Principles of
Conservation Biology, second ed. Sinauer Associates, Inc., Sun-
derland, Massachusetts, pp. 161–201.
Percsy, C., 1995. Les Batraciens sur nos Routes. Minist
ere de la
R
egion Wallonne. Service de la Conservation de la Nature et des
Espaces verts. Brochure technique No. 1, Wallonne, Belgium.
Reh, W., Seitz, A., 1990. The influence of land use on the genetic
structure of populations of the common frog Rana temporaria.
Biological Conservation 54, 239–249.
Smith, D.J., 1996. The direct and indirect impacts of highways on the
vertebrates of PayneÕs Prairie State Preserve. Florida Department
of Transportation, Environmental Management Office, Tallahas-
see, FL. 91 pp.
Smith, L.L., Dodd Jr., C.K., 2003. Wildlife mortality on highway US
441 across Paynes Prairie, Alachua County, Florida. Florida
Scientist 66, 128–140.
Trombulak, S.C., Frissell, C.A., 2000. Review of ecological effects of
roads on terrestrial and aquatic communities. Conservation Biol-
ogy 14, 18–30.
Yanes, M., Velasco, J.M., Su
arez, F., 1995. Permeability of roads and
railways to vertebrates: the importance of culverts. Biological
Conservation 71, 217–222.
C. Kenneth Dodd Jr. et al. / Biological Conservation 118 (2004) 619–631 631
