ArticlePDF Available

Monitoring the use of road-crossing structures by arboreal marsupials: Insights gained from motion-triggered cameras and passive integrated transponder (PIT) tags

Authors:

Abstract and Figures

Context: Wildlife crossing structures are installed to mitigate the impacts of roads on animal populations, yet little is known about some aspects of their success. Many studies have monitored the use of structures by wildlife, but studies that also incorporate individual identification methods can offer additional insights into their effectiveness. Aims: We monitored the use of wildlife crossing structures by arboreal marsupials along the Hume Freeway in south-eastern Australia to (1) determine the species using these structures and their frequency of crossing, (2) determine the number and demographic characteristics of individuals crossing, and (3) use the rate of crossing by individuals to infer the types of movement that occurred. Methods: We used motion-triggered cameras to monitor five canopy bridges and 15 glider pole arrays installed at 13 sites along the Hume Freeway. The five canopy bridges were also monitored with passive integrated transponder (PIT)-tag readers to identify the rate of use by individuals. Key results: Five species of arboreal marsupial were detected using canopy bridges and glider poles at 11 sites. Our analysis suggested that increasing the number and the distance between poles in a glider pole array reduced the rate of use by squirrel gliders. The PIT tag and camera footage revealed that the structures were used by adult males, adult females and juveniles, suggesting that all demographic groups are capable of using canopy bridges and glider poles. At two canopy bridges, multiple squirrel gliders and common brushtail possums crossed more than once per night. Conclusions: Given that previous studies have shown that the freeway is a barrier to movement, and that many of the species detected crossing are subject to road mortality, we conclude that canopy bridges and glider poles benefit arboreal marsupials by providing safe access to resources that would otherwise be inaccessible. Implications: Although the factors influencing crossing rate require further study, our analysis suggests that glider pole arrays with fewer poles placed closer together are likely to be more successful for squirrel gliders. The individual identification methods applied here offer insights that are not possible from measuring the rate of use alone and should be adopted in future monitoring studies.
Content may be subject to copyright.
Page 1 of 42
Published version available http://www.publish.csiro.au/paper/WR14067.htm 1
Wildlife Research - http://dx.doi.org/10.1071/WR14067 2 Submitted: 9 April 2014 Accepted: 10 May 2015 Published online: 26 June 2015 3
Monitoring the use of road-crossing structures by arboreal marsupials: insights gained 4
from motion-triggered cameras and passive integrated transponder (PIT) tags 5
Kylie Soanesab*, Peter A. Veskb and Rodney van der Reea
6
a Australian Research Centre for Urban Ecology, Royal Botanic Gardens, Melbourne, VIC 3010, 7
Australia 8
b School of BioSciences, University of Melbourne, VIC 3010, Australia 9
* corresponding author: Phone +61 (03) 8344 0146, Fax +61 (03) 9347 9123, e-mail: 10
ksoanes@unimelb.edu.au.soanes@pgrad.unimelb.edu.au (K. Soanes) 11
Running head: Monitoring arboreal mammal crossing structures 12
Key words: Canopy bridge, glider pole, habitat fragmentation, barrier effect, road mitigation, 13
connectivity, monitoring, individual identification 14
Abstract 15
Context: Wildlife crossing structures are installed to mitigate the impacts of roads on animal 16
populations, yet little is known about some aspects of their success. Many studies monitor the 17
use of structures by wildlife, but studies that also incorporate individual identification methods 18
can offer additional insights into their effectiveness. 19
Aims: We monitored the use of wildlife crossing structures by arboreal marsupials along the 20
Hume Freeway in south-east Australia to: 1) determine the species using these structures and 21
their frequency of crossing; 2) determine the number and demographic characteristics of 22
Page 2 of 42
individuals crossing; and 3) use the rate of crossing by individuals to infer the types of 23
movement that occurred. 24
Methods: We used motion-triggered cameras to monitor five canopy bridges and fifteen glider 25
pole arrays installed at thirteen sites along the Hume Freeway. The five canopy bridges were 26
also monitored with PIT tag readers to identify the rate of use by individuals. 27
Key results: Five species of arboreal marsupial were detected using canopy bridges and glider 28
poles at eleven sites. Our analysis suggested that increasing the number and the distance 29
between poles in a glider pole array reduced the rate of use by squirrel gliders. The PIT tag and 30
camera footage revealed that the structures were used by adult males, adult females and 31
juveniles, suggesting that all demographic groups are capable of using canopy bridges and 32
glider poles. At two canopy bridges multiple squirrel gliders and common brushtail possums 33
crossed more than once per night. 34
Conclusion: Given that previous studies showed that the freeway was a barrier to movement, 35
and that many of the species detected crossing are subject to road mortality, we conclude that 36
canopy bridges and glider poles benefit arboreal marsupials by providing safe access to 37
resources that would otherwise have been inaccessible. 38
Implications: While the factors influencing crossing rate require further study, our analysis 39
suggests that glider pole arrays with fewer poles placed closer together are likely to be more 40
successful for squirrel gliders. The individual identification methods used in this study offer 41
insights that are not possible from measuring the rate of use alone and should be adopted in 42
future monitoring studies. 43
44
Page 3 of 42
Introduction 45
Roads and traffic form barriers to animal movement when animals are unable to cross due to 46
physical or behavioural limitations, or when those attempting to cross are killed by vehicles 47
(Bennett 1991; Burnett 1992; Goosem 2001; Oxley et al. 1974). This process sub-divides habitat 48
creating smaller, isolated populations with a higher risk of local extinction (Bennett 1991; 49
Fahrig and Rytwinski 2009; Forman et al. 2003; van der Ree et al. 2015b). Road agencies often 50
install wildlife crossing structures (e.g. culverts, tunnels, land bridges) to mitigate these effects 51
and conserve roadside populations (Forman et al. 2003; Taylor and Goldingay 2010; van der 52
Grift et al. 2013; van der Ree et al. 2007). Given the financial investment in crossing structures 53
worldwide, and their use to meet environmental regulatory requirements, it is critical that we 54
understand how well they work. 55
Many studies use cameras to monitor the use of crossing structures by wildlife and infer the 56
effectiveness of the structures, or particular design features, based on how frequently animals 57
use them (e.g. Clevenger and Waltho 2000; Grilo et al. 2008; Ng et al. 2004; van der Ree et al. 58
2007). However, few studies identify and record the number of individuals that use a structure 59
(but see Baxter-Gilbert et al. 2013; Boarman et al. 1998; Chambers and Bencini 2015; Clevenger 60
and Sawaya 2010; Dodd et al. 2007; Harris et al. 2010). Knowing the number and type of 61
individuals crossing adds value to a monitoring program for two reasons. First, it can show 62
whether a structure is used by all types of individuals or just a small subset of the population 63
and whether demographic connectivity (i.e. affecting recruitment and reproductive rates) is 64
facilitated. For example, if a crossing structure only benefits a small proportion of the local 65
population, or use is dominated by one gender or age-class, then the structure is unlikely to 66
facilitate demographic connectivity (Clevenger 2005; Herrod 2005; Olsson et al. 2008). Second, 67
crossing structures, like corridors, may facilitate multiple types of movement, which can be 68
categorised by different patterns of use. Movements to access resources within an animal's 69
home-range are generally short and frequent, while dispersal movements over long distances 70
Page 4 of 42
tend to be infrequent or seasonal (Bennett 1999; Bissonette and Adair 2008; Van Dyck and 71
Baguette 2005). These different types of movement can potentially be determined by 72
identifying the individuals that use the crossing structure. For example, a camera may detect 20 73
crossings in a night, but only methods that identify individuals can reveal if the structure was 74
used by 20 individuals crossing once each or a single individual crossing 20 times. This 75
information can give valuable insights into how a crossing structure might benefit a population, 76
e.g. through occasional dispersal or regular habitat access (Clevenger 2005; Foster and 77
Humphrey 1995; van der Ree et al. 2007; van der Ree et al. 2009). Monitoring use is an 78
important first step in evaluating the effectiveness of wildlife crossing structures and by making 79
these initial studies more informative, they can better guide and complement future population-80
level studies (Soanes 2014; van der Grift et al. 2013; van der Grift and van der Ree 2015). 81
Canopy bridges and glider poles (hereafter referred to as 'bridges' and 'poles') are crossing 82
structures installed around the world to mitigate the impacts of roads on arboreal mammals 83
(Goldingay et al. 2013; Kelly et al. 2013; Mass et al. 2011; Soanes et al. 2013; Soanes and van der 84
Ree 2015; Taylor and Goldingay 2012b; Teixeira 2013; Weston et al. 2011). In eastern Australia, 85
road agencies increasingly rely on bridges and poles to mitigate the impacts of major roads on 86
threatened arboreal marsupials, including the squirrel glider (Petaurus norfolcensis). In the 87
absence of crossing structures, major roads create a barrier to glider movement and reduce the 88
survival rate and viability of squirrel glider populations (McCall et al. 2010; Taylor and 89
Goldingay 2012a; van der Ree et al. 2010). Canopy bridges and glider poles were installed along 90
a freeway in south-east Australia to provide safe passage for squirrel gliders and other arboreal 91
marsupial species. The effectiveness of these structures has yet to be determined. We 92
monitored the use of canopy bridges and glider poles by arboreal marsupials with the following 93
aims: 94
1) to determine the species using these structures and their frequency of crossing; 95
Page 5 of 42
2) to determine the number and demographic characteristics of individuals crossing 96
canopy bridges; and 97
3) to use the rate of crossing by individuals to infer the types of movement that occurred 98
across canopy bridges. 99
100
Page 6 of 42
Methods 101
Study area 102
The Hume Freeway is a major interstate freeway in south-east Australia that has been 103
progressively upgraded to a four-lane, divided freeway over the past 50 years. The width of the 104
freeway from road edge to road edge varies from 40 to 80 m, depending on the width of the 105
centre median, which ranges from 21 to 38 m. Each carriageway is 12 m wide (including 106
emergency and travel lanes). The treeless gap across the Hume Freeway can exceed 100 m 107
where mature trees have been cleared from the centre median and roadsides, presenting a 108
barrier to the movement of arboreal marsupials. We studied two sections of freeway located 109
approximately 200 km apart; a length of 63 km in the state of Victoria between the towns of 110
Avenel and Benalla that was upgraded in the 1970s-80s, and a length of 70 km in the state of 111
New South Wales between the towns of Albury and Tarcutta that was upgraded in 2009 (Figure 112
1). The traffic volume averages 10,000 vehicles per day at a maximum legal speed of 110 km/h 113
(DOTARS 2007). The landscape surrounding the Hume Freeway is predominantly agricultural 114
land and rural townships, with remnant and regrowth woodland occurring only in small 115
patches, travelling stock reserves (TSRs) and linear reserves as described in van der Ree (2002) 116
and Gibbons and Boak (2002). 117
[Figure 1] 118
Study species 119
The patches of Eucalyptus spp. woodland along the Hume Freeway are a critical resource for 120
arboreal marsupials including: the squirrel glider; the sugar glider (Petaurus breviceps); the 121
common brushtail possum (Trichosurus vulpecula); the common ringtail possum (Pseudocheirus 122
peregrinus); the brush-tailed phascogale (Phascogale tapoatafa); the yellow-footed antechinus 123
(Antechnius flavipes), and; the koala (Phascolarctos cinereus). While all of these species can be 124
negatively affected by roads (e.g. Dique et al. 2003; Gulle 2006; Herrod 2005; Russell et al. 125
Page 7 of 42
2009) the squirrel glider was the primary target of mitigation and monitoring. Listed as 126
Threatened in Victoria and Vulnerable in New South Wales (Claridge and van der Ree 2004; DSE 127
2007), squirrel gliders move by gliding from tree to tree with an average glide length of 3040 128
m, although glides of up to 70 m have been recorded (Jackson 2000; van der Ree et al. 2003; van 129
der Ree et al. 2010. 130
Wildlife crossing structures 131
Crossing structures were installed where the Hume Freeway intersected mature woodland 132
habitat and was likely to limit squirrel glider movement. That is, at sites where squirrel gliders 133
were not expected to be able to safely cross the freeway using existing roadside trees (1225 m 134
tall) based on a glide ratio of 2.5:1 (Jackson 2000; van der Ree et al. 2010). The width of the 135
treeless gap created by the freeway ranged from 60380 m (Table 1). Previous studies show 136
major roads with a treeless gap of > 50 m wide restrict squirrel glider movement (Soanes et al. 137
2013; van der Ree 2006; van der Ree et al. 2010). In 2007, crossing structures were retrofitted 138
to five sites in Victoria. A further eight sites in New South Wales were mitigated during the 139
freeway upgrade in 2009. The number of structures required at each site varied (between one 140
and three) depending on the extent and configuration of roadside habitat (Table 1). 141
[Figure 2] 142
Canopy bridges were made of UV-stabilised marine-grade rope (~15 mm diameter) woven into 143
a flat net 50 cm wide, resembling a long, narrow strip of cargo net. Two steel cables suspended 144
the rope between hardwood timber poles on either side of the freeway. Four bridges, ranging 145
from approximately 60 to 85 m long, were erected at a minimum of 6 m above the road (Figure 146
2). A fifth bridge (Yarra Yarra Creek), 170 m long (zig-zagged across a gap of ~150 m), was 147
installed under an open-span road bridge at a minimum height of 4 m from the ground and 2 m 148
from the underside of the bridge and was supported by six poles along its length. Canopy 149
bridges were installed as close as possible to the existing roadside woodland, usually within 5 m 150
Page 8 of 42
of overhanging tree branches (up to 15 m from tree trunk). Additional ropes connected the ends 151
of each canopy bridge to the branches of roadside trees to facilitate access by arboreal 152
marsupials (1 to 3 ropes per end). 153
Glider poles (round, hardwood timber poles ~1318 m tall, 4050 cm in diameter) provide 154
gliders with an alternative glide site to reduce the width of the treeless gap across the road 155
(Soanes and van der Ree 2015). A timber cross-beam (10 cm x 10 cm x 2.4 m) was fixed 156
horizontally 50 cm from the top of each pole (oriented parallel to the road edge) providing a 157
'branch-like' launch site. The number and height of poles required depended on the width of the 158
gap across the freeway and the height of roadside trees relative to the height of the road (i.e. if 159
road was in a cutting or raised). As such, each pole crossing could include a single pole in the 160
centre median, or an array of multiple poles placed in the median and roadsides (Figure 2, Table 161
1). Glide paths were calculated in detailed schematics to ensure that glides in each direction 162
were achievable within a recommended glide ratio of 2.5:1, allowing animals to pass safely 163
above the maximum expected height of traffic. These designs took into account the terrain and 164
the height and location of roadside trees to determine the appropriate height and spacing of 165
glider poles in each array (e.g. Soanes and van der Ree 2015). Squirrel gliders and sugar gliders 166
were the only species within the study area capable of using glider poles to cross the freeway. 167
Monitoring crossing structures 168
We used two methods to monitor the use of crossing structures by arboreal marsupials; motion 169
triggered, infrared cameras (Orion, Buckeye Pty Ltd) and PIT tag reading systems (Trovan, 170
Microchips Australia Pty Ltd). Cameras were installed on all crossing structures. PIT tag readers 171
were only placed on canopy bridges, as a suitable design for glider poles was not available. 172
Equipment was installed between April and June 2012 and the structures were monitored until 173
February 2013, providing 911 months of monitoring per structure. This period occurred 174
approximately five years after the crossing structures were installed in Victoria and three years 175
after crossing structures were installed in New South Wales. The number of possible monitoring 176
Page 9 of 42
nights per structure ranged from 258 to 315, however, due to equipment malfunctions, the 177
actual monitoring effort ranged from 75 to 279 nights per structure (Table 2). Detail on the 178
monitoring set-up for each method is described below. 179
[Table 2] 180
Motion triggered cameras 181
We placed one camera at each end of each bridge (i.e. two cameras per bridge), and one on the 182
centre median pole of each glider-pole array. At bridges, cameras were triggered by animal 183
movement past a pair of active infrared sensors placed approximately 1 and 4 m from each 184
camera. At poles, cameras were mounted on one end of the cross-beam, so that animals moving 185
along the length of the beam or to the top of the pole would trigger the passive infrared sensor. 186
Each time a sensor was triggered the cameras recorded 920 seconds of video. Cameras were 187
powered by an appropriately sized lead-acid battery and solar panel. 188
Videos from the Victorian structures were transmitted wirelessly to the memory card of an on-189
ground unit from where they were downloaded during fortnightly field visits. Cameras on the 190
New South Wales structures used an internal modem to transmit videos to the office computer 191
each morning via the 3G mobile phone network. We inspected all videos for the presence of 192
animals on bridges and poles. Each time an animal was detected, the date, time and species 193
were recorded, as well as the number of videos the animal appeared in and direction of travel. 194
Where observed, distinctive markings such as male scent glands, ear notches, or differences in 195
body size were noted to determine the animal’s approximate age and sex. 196
At bridges, the placement of a camera at each end of the bridge allowed us to confirm crossings. 197
A crossing was confirmed when an animal was viewed moving out onto the bridge by one 198
camera and away from the camera until it was out of sight, then detected by the camera on the 199
opposite side of the bridge and viewed exiting the structure. When only one camera was 200
operational, crossings were inferred based on the animal's behaviour and direction of travel. 201
Page 10 of 42
For example, if an animal moved out onto the bridge (i.e. approaching the other side) until it 202
was no longer visible and did not turn around and return within ten minutes we classified it as a 203
crossing. This is based on the fact that more than 90% of crossings confirmed by both cameras 204
were completed in less than ten minutes. In this way, we distinguished crossings from 'visits', in 205
which an animal was detected moving on to the bridge before turning around and exiting the 206
same side without completing the crossing. 207
Crossings could not be confirmed at glider poles. Animals can glide to the pole in the centre 208
median from any number of roadside trees (or roadside poles) and land on the pole below the 209
cross-arm, out of view of the camera. This means that it is impossible to determine the direction 210
of travel. While we acknowledge that some animals may glide to the centre median pole and 211
then return to the same side, we expect this to represent only a small proportion of detections, 212
as there is no habitat in the centre median and therefore no reason for animals to repeatedly 213
glide to the centre pole without completing the crossing. This is supported by previous 214
radiotracking data, which showed that individuals detected on the centre median pole went on 215
to cross the freeway (Soanes et al. 2013). Therefore, we refer to all detections on the centre 216
median pole as crossings for the purposes of this study. However this may overestimate the true 217
number of complete crossings where squirrel gliders glide to the centre median pole and return 218
without crossing. 219
Preliminary analysis of factors affecting crossing rate 220
Site covariates and design attributes of the road and crossing structures can influence the use of 221
crossing structures by wildlife. For each site, the road width (width of the treeless gap across 222
the road as measured from trunk to trunk), the distance of the structure to roadside trees, 223
bridge length, number of poles in a glider pole array and maximum glide distance (between two 224
successive poles or pole and tree in an array of poles) were obtained from schematic diagrams, 225
satellite imagery and site visits (Table 1). We also calculated an activity index for squirrel 226
gliders (the target species) using data from a concurrent mark-recapture study to estimate the 227
Page 11 of 42
number of individuals that were active in the habitat adjacent to the mark-recapture study. The 228
activity index was determined at the site level by dividing the average minimum number of 229
squirrel gliders known to be alive (MNKTBA) by the area of habitat trapped at each site (details 230
provided in Appendix 1). 231
Upon inspection of the data it became clear that it would not be possible to investigate all 232
potential factors influencing crossing rates within the scope of this study. There were only five 233
canopy bridge sites available, and of these only three were used by arboreal marsupials. This 234
sample size was not suited to further statistical analysis, and so we limited our analysis to glider 235
poles. As non-gliding species were restricted to using canopy bridges, and sugar gliders were 236
detected at only three sites, we further restricted our analysis to squirrel gliders. At the time of 237
this study, two glider pole arrays at Kyeamba TSR were >100 m from the nearest roadside 238
squirrel glider habitat, as the habitat restoration works activities surrounding the structures 239
have not yet successfully linked the crossing structure with the existing roadside trees. This 240
distance is well beyond the maximum glide range of a squirrel glider, and therefore we excluded 241
these two structures from further analysis. This left us with a sample size of 13 glider poles to 242
include in statistical analysis. 243
Due to the limited dataset we restricted our analysis to two factors likely to influence the use of 244
glider poles by squirrel gliders: the number of poles in an array (hereafter ‘number of poles’), 245
and the maximum glide distance required to cross the glider pole array (hereafter ‘glide 246
distance’). We chose to focus on these factors because they are both biologically meaningful and 247
also relevant to managers as these design features are relatively easily to manipulate when 248
installing crossing structures. We did not include the activity index as a parameter, as 249
preliminary analysis revealed that its effect on crossing rate was highly uncertain. Inspection of 250
the data showed that a high activity index did not correspond with a high crossing rate at that 251
site. Further, multiple structures from the same site and hence the same activity index 252
commonly varied widely in crossing rates. 253
Page 12 of 42
We used a Poisson regression to investigate the effect of glide distance and the number of poles 254
on the crossing rate. The data used to estimate the model parameters were the number of 255
crossings observed within each of 26 time periods of varied length during which the cameras 256
were operating. The expected number of crossings per night was modelled as the exponent of a 257
linear model with a grand mean, and fixed effects for the number of poles and glide distance. 258
Error terms were included to account for residual error and a random effect of each individual 259
crossing structure on the crossing rate. The model was estimated using Bayesian inference in 260
the program OpenBugs 3.2.1 (Spiegelhalter et al. 2011). We used vague priors for all 261
parameters, using a normal distribution with a mean of 0 and a standard deviation of 1000 for 262
the fixed effects and half-Cauchy distributions with scale 25 for the random effects (Gelman 263
2006). The model code and data are provided in Appendix 2. The model was run for 50,000 264
iterations after discarding a burn-in of 20,000. Convergence was assessed through the visual 265
inspection of three independent chains. 266
PIT tag scanners on canopy bridges 267
We installed one flat panel antenna (ANT-612, Trovan, Microchips Australia Pty Ltd, Figure 3.2) 268
at each end of each canopy bridge (i.e. two antennae per bridge). Antennae were approximately 269
the same width as the bridge (40 cm) with a read distance of 35 cm. Data were stored in an 270
attached decoder unit (LID650, Trovan, Microchips Australia Pty. Ltd.) fixed to the support pole. 271
The antennae were not continuously operational due to constraints on the size of the battery 272
and solar panel that could be installed on canopy bridges. Antennae on bridges in New South 273
Wales were integrated with the camera's active infrared sensors and only operated when a 274
sensor was triggered. In Victoria, we could not integrate antennae with the existing camera 275
sensors and therefore we scheduled them to operate for half of the night. At these bridges one 276
antenna was operational from 17:00 h until 23:59 h and the opposite from 00:01 h until 07:00 277
h. A full night of monitoring was achieved if both antennae were operational on the same night. 278
Unfortunately, the units frequently malfunctioned and a full night of monitoring only occurred 279
Page 13 of 42
on 32 nights at the Longwood bridge and four nights at Violet Town. On all other nights the 280
Victorian bridges were only monitored with one antenna (67 hours per night), leaving 281
approximately 36 hours of the night unmonitored depending on the season, therefore our 282
study is likely to underestimate the true number of individuals that used the canopy bridges. 283
Data from the Victorian PIT tag readers were downloaded from the decoder via a USB cable 284
during fortnightly site visits. Decoders on bridges in New South Wales contained modems 285
programmed to transmit data to the office computer each morning via the 3G mobile phone 286
network. We recorded the time and date of each PIT tag reading and matched it with the 287
corresponding video where possible. PIT tag readings were also cross-checked against mark-288
recapture records to identify each individual. Unless both PIT tag antennae at a bridge were 289
simultaneously operational, it was not possible to confirm crossings using PIT tags alone. 290
Therefore we cross-checked the time and date of all tag reads with the videos recorded by 291
cameras to collect information on the crossing behaviour of tagged individuals. 292
Mark-recapture surveys were also conducted at all five canopy bridge sites (as described in 293
Appendix 1). All squirrel gliders and common brushtail possums captured were implanted with 294
a PIT tag under the skin between the shoulder blades (ID 100, Trovan, Microchips Australia Pty 295
Ltd). Data from the existing mark-recapture data set suggests that the rate of tag loss is <0.05% 296
(unpublished data), which corresponds to other studies on small mammals (Schooley et al. 297
1993). Other species were not marked with PIT tags, as they are rarely captured in wire cage 298
traps. The average MNKTBA for common brushtail possums and squirrel gliders in habitat 299
adjacent to each canopy bridge site was calculated as described for glider poles (Appendix 1, 300
Table 1, Table 6). 301
Results 302
Rate of use by arboreal marsupials 303
Page 14 of 42
Using motion-triggered cameras, we detected five species of arboreal marsupial using both 304
types of crossing structure over 3,929 nights of camera monitoring: squirrel gliders (n=1,317 305
detections), common ringtail possums (n=394), common brushtail possums (n=241), sugar 306
gliders (n=258) and brush-tailed phascogales (n=4) (Tables 3 and 4). Only 42 detections were 307
classed as visits and excluded from further analysis. No koalas or yellow-footed antechinus 308
were detected, despite occurring in the general area. Arboreal marsupials used crossing 309
structures to cross the freeway at 11 of the 13 monitoring sites. Two canopy bridges and four 310
glider pole arrays were not used by any species. Squirrel gliders were detected crossing the 311
road using canopy bridges and glider poles. Both species of possum and phascogales were only 312
detected crossing bridges. Sugar gliders were only detected using poles to cross the road. 313
[Table 3, Table 4] 314
The number of crossings per night varied widely among sites, with some structures used 315
frequently and others not at all (Tables 3 and 4). For example, squirrel gliders were detected 316
crossing the Longwood bridge, Warrenbayne pole, Sages pole 1 and Kyeamba Creek pole 1 more 317
than once each night, while the Violet Town bridge and Sages pole 2 were used less than once 318
every two months. Seven structures were not used by squirrel gliders, even though the species 319
was present at all sites. Therefore while the mean crossing rate for squirrel gliders was 0.84 320
crossings per night at canopy bridges (± SE 0.84) and 0.38 at glider poles (± SE 0.17), the 321
crossing rate at any one site ranged from 0.00 to 4.19 crossings per night. Crossing rates by 322
other species ranged from 0.00 to 1.14 crossings per night for common brushtail possums 323
(mean 0.23, ± SE 0.23); 0.00 to 1.36 crossings per night for common ringtail possums (mean 324
0.46, ± SE 0.28); and 0.00 to 0.93 crossings per night for sugar gliders (mean 0.07, ± SE 0.06). 325
Brush-tailed phascogales were detected infrequently on one bridge, with 1 crossing per 50 326
nights of monitoring (0.02 crossings per night). 327
The regression analysis revealed that the number of poles in a glider pole array and the 328
maximum glide distance required to cross had a negative effect on the use of glider poles by 329
Page 15 of 42
squirrel gliders. The point estimates for the effect of number of poles and glide distance were 330
both negative (-1.26 and -0.12, respectively), although the credible intervals for both 331
parameters overlapped zero reflecting uncertainty in the estimates, and rather more so for the 332
distance than number of poles (Table 5). The wide credible intervals are not surprising given 333
the low sample size available for analysis (n=13). Based on these data, the predicted mean rate 334
of use for a glider pole array consisting of two poles with a maximum distance of 30 m apart 335
was 0.11 crossings per night (95% CI 0.000.53). Note that this does not predict beyond the 336
observed data and does not incorporate possible sources of uncertainty. 337
[Table 5] 338
The video footage showed that multiple individuals of both sexes used the crossing structures, 339
including adults and juveniles. This finding was based on observed differences in body size, the 340
presence of active scent glands and ear notches visible in 445 of the 2,291 of the occasions 341
during which animals were detected. Both species of possum were observed carrying 342
dependent young across the Longwood and Violet Town canopy bridges and adult squirrel 343
gliders were seen carrying pouch young or accompanied by smaller individuals, presumably 344
juveniles, on poles (Figure 3). Independent juveniles (i.e. not carried by parents) were also 345
observed, with juvenile ringtail possums at the Violet Town bridge, and juvenile squirrel gliders 346
at Kyeamba Creek pole 2, Sages pole 2, and the Warrenbayne and Blue Metal poles. 347
Rate of use by individuals 348
Three squirrel gliders and five common brushtail possums with PIT tags were detected crossing 349
two canopy bridges over the 11 months during which PIT tag readers were in use. Three 350
squirrel gliders and a common brushtail possum were detected crossing the Longwood bridge 351
and four common brushtail possums were detected crossing the Violet Town bridge. A fourth 352
squirrel glider was detected on the canopy bridge at Sages TSR but did not complete the 353
crossing. All individuals were reproductively active adults. Due to equipment malfunctions, only 354
Page 16 of 42
527% of detections by PIT tag reader could be confirmed as crossings. However, given that 355
only 42 of the 1,112 camera detections at bridges were identified as visits, we are confident that 356
a large portion of the detections by the PIT tag reader also represent complete crossings. The 357
number of individuals observed using the canopy bridges was generally lower than the number 358
present in the surrounding habitat (as indicated by the average MNKTBA), with the exception of 359
the bridge at Longwood, where the number of squirrel gliders observed crossing was 360
approximately equal to the average MNTKBA (Table 4). Canopy bridges at Yarra Yarra Creek, 361
Blue Metal TSR and Sages TSR were not crossed by any tagged squirrel gliders or common 362
brushtail possums despite the presence of tagged individuals in the adjacent habitat (Table 4). 363
Discussion 364
Use of structures by arboreal marsupials 365
Five species of arboreal marsupial were detected using crossing structures to cross the Hume 366
Freeway, including structures that were retrofitted to existing sections of the freeway and those 367
that were installed during freeway construction. Squirrel gliders, common brushtail possums, 368
common ringtail possums, sugar gliders and brush-tailed phascogales are all affected by road 369
mortality (pers. obs.; McCall et al. 2010; Russell et al. 2009; Taylor and Goldingay 2004) and 370
structures that increase their safe passage across the road may help reduce rates of road-kill. 371
There is now mounting evidence that canopy bridges and glider poles of various designs are 372
used by a wide range of arboreal mammals, including monkeys, lemurs, and squirrels 373
(Donaldson and Cunneyworth 2015; Goldingay et al. 2013; Kelly et al. 2013; Mass et al. 2011; 374
Soanes et al. 2013; Taylor and Goldingay 2012a; Teixeira 2013; Weston et al. 2011) and these 375
structures are likely to be useful for other species with similar behaviours and ecological 376
requirements, including arboreal reptiles and amphibians. For example, mahogany gliders 377
(Petaurus gracilis), mountain brushtail possums (Trichosurus cunninghami) and western ringtail 378
possums (Pseudocheirus peregrinus occidentalis) have similar ecology and behaviour to species 379
Page 17 of 42
detected in this study and are negatively affected by roads and fragmentation (Asari et al. 2010; 380
Taylor and Goldingay 2004; Trimming et al. 2009). 381
While small sample sizes limited the statistical analysis possible within this study, we were able 382
to identify some general principles for the design of crossings structures for arboreal 383
marsupials. Arboreal marsupials used all structure types monitored in this study, including 384
bridges up to 85 m long and roadside glider pole arrays consisting of up to four poles. While no 385
animals were detected using the under-road canopy bridge in our study, previous work by 386
Goldingay et al. (2013) detected common brushtail possums, common ringtail possums and 387
feathertail gliders (Acrobates pygmaeus) using an under-road bridge, suggesting that this design 388
is also useful. Our analysis suggests that the use of glider pole arrays by squirrel gliders 389
decreases as the number of poles and the distance between the poles increases. This is 390
supported by studies of glide capacity, which show that glide success decreases as glide distance 391
increases (Ball and Goldingay 2008; Goldingay and Taylor 2009; Jackson 2000; van der Ree et 392
al. 2003). Though we could not analyse it in our study, a similar principle is likely to apply to 393
canopy bridges, with shorter spans likely to be more successful than longer ones, particularly 394
for species that prefer closed canopy (e.g. Goosem et al. 2006; Weston et al. 2011). Crossing 395
structures for arboreal mammals are therefore more likely to be successful if the width of the 396
treeless gap across the road is minimised during construction, or structures are retrofitted to 397
locations where the existing gap is narrowest. This will allow glider pole arrays to consist of 398
fewer poles that can be placed closer together, or shorter-span canopy bridges. 399
Identifying the number and type of individuals 400
Collecting data on the age and sex of the animals using crossing structures provides valuable 401
information on the ability of crossing structures to benefit the wider population (Clevenger 402
2005; Olsson et al. 2008; Sawaya et al. 2013). For example, Sawaya et al. (2013) identified eight 403
female and nine male black bears (Ursus americanus) using crossing structures in Banff National 404
Park, Canada and concluded that these structures provided demographic connectivity. 405
Page 18 of 42
Conversely, Olsson et al. (2008) concluded a highway overpass was unlikely to affect the 406
population demographic rates of moose (Alces alces) populations based on the low number of 407
crossings made predominantly by males. In our study, PIT tags and body markings observed in 408
video footage revealed that males and females, including females carrying dependent young, 409
used canopy bridges and glider poles to cross the freeway. That is, no demographic group was 410
excluded from using either structure type to cross the freeway. A comparison of the crossing 411
rates obtained through video and PIT tag readers suggests that the crossing rate of common 412
brushtail possums and squirrel gliders at the Violet Town and Longwood canopy bridges was 413
almost entirely due to the repeated movements of a few individuals (14 individuals of each 414
species per structure). Population-level analyses are required to assess whether or not the 415
number of individuals crossing is enough to maintain demographic and genetic connectivity for 416
the wider population. However based on our findings it is unlikely that there are any features of 417
the current crossing structure designs that would prevent crossing structures from facilitating 418
demographic and genetic connectivity. 419
Inferring types of movement 420
Identifying patterns of use by individuals allowed us to infer the types of movement that 421
crossing structures facilitated. For example, daily movements are generally associated with 422
foraging and accessing resources within an individual’s home range (Bennett 1999; Bissonette 423
and Adair 2008; Van Dyck and Baguette 2005). The high average crossing rates (more than one 424
crossing per night) of squirrel gliders and common brushtail possums at canopy bridges as 425
detected by cameras, were revealed by the PIT tag reader to be 34 individual animals crossing 426
repeatedly (often multiple times each night). Prior to the installation of the Longwood and 427
Violet Town canopy bridges, squirrel gliders and common brushtail possums did not cross the 428
freeway regularly at these sites (Gulle 2006; Soanes et al. 2013; van der Ree et al. 2010). In our 429
study, two squirrel gliders at the Longwood bridge and one common brushtail possum at the 430
Violet Town bridge crossed almost every night, gaining nightly access to habitat on both sides of 431
Page 19 of 42
the freeway. Such regular use suggests that these individuals incorporate the crossing structure 432
within their home range. Therefore it is possible that the high crossing rate at these sites is 433
driven by a few individuals frequently crossing in order to access portions of their home range 434
on either side of the road. It is likely that the high frequency of crossings (~ 1 crossing/night) by 435
squirrel gliders at glider poles such as Warrenbayne, Little Billabong, Sages TSR pole 1 and 436
Kyeamba Creek pole 1, common ringtail possums at the Violet Town bridge, and sugar gliders at 437
the Little Billabong pole also reflect repeated crossings by a few individuals. Radiotracking 438
studies that investigate home-range movements could be used to confirm this theory. Canopy 439
connectivity is critical to maintain daily movements of arboreal species that are gap-limited or 440
unwilling to cross open spaces (e.g. Anzures-Dadda and Manson 2007; Asari et al. 2010; 441
Laurance and Laurance 1999; van der Ree et al. 2010; Wilson et al. 2007). This supports the 442
idea that crossing structures can benefit wildlife by providing access to resources that would 443
have been unavailable, or dangerous to reach, if crossing structures were not present (Bennett 444
1999; Eigenbrod et al. 2008). 445
Despite the advantages of using PIT tags to identify individuals, it did not help us infer other 446
types of movement, such as dispersal, or allow us to detect effects at the population level. 447
Dispersing individuals may be sourced from a large number of source populations, and tagging 448
all potential dispersers that might use the structure would require an intensive and extensive 449
survey effort. Furthermore, movement across a structure does not necessarily result in 450
population-level effects (van der Grift et al. 2013; van der Ree et al. 2007; van der Ree et al. 451
2011). For example, territorial behaviour by resident individuals in roadside habitat can limit 452
the reproductive success of dispersing individuals, creating a social barrier to gene flow despite 453
movement (e.g. Riley et al. 2006, Corlatti et al. 2009, Simmons et al. 2010). Similarly, while the 454
detection of animals crossing suggests that those individuals crossed safely and that roadkill is 455
less likely to occur, we cannot say that population-level survival rates have improved. 456
Ultimately, mark-recapture surveys collecting information on populations size, gene flow and 457
survival rates before and after crossing structures are installed are required to determine 458
Page 20 of 42
whether the patterns of crossing behaviour yield population-level benefits (Corlatti et al. 2009; 459
Sawaya et al. 2014; Simmons et al. 2010; van der Grift et al. 2013; van der Ree et al. 2007; van 460
der Ree et al. 2011). 461
Limitations of studying arboreal crossing structures 462
Investigating the optimal dimensions of crossing structures and factors influencing use (e.g. 463
structure placement, habitat quality and needs of the local population) guide best-practice 464
mitigation and are a critical aspect of road ecology (van der Grift and van der Ree 2015). 465
However, our study had limited statistical power due to the low replication and low crossing 466
rates by some species. Our study almost certainly underestimated the true number of crossings 467
and longer-term surveys may reveal that more of these structures are used by wildlife over time 468
(e.g. Bond and Jones 2008; Clevenger and Waltho 2003; Gagnon et al. 2011; Soanes et al. 2013). 469
While longer survey periods are likely to increase the number of crossings observed, increasing 470
the number of canopy bridges and glider poles is more difficult. It is unlikely that a study along a 471
single stretch of road, or mandated monitoring from a single construction project will have 472
sufficient replication to investigate how structural and site attributes affect rates of use for 473
different arboreal species (Rytwinski et al. 2015; van der Ree et al. 2015a). A promising avenue 474
to improve replication in future studies is for road agencies and researchers to combine 475
multiple projects into a coordinated monitoring program, thus enabling a thorough 476
investigation of the factors that influence the effectiveness of crossing structures for arboreal 477
mammals. Alternatively, the experimental manipulation of structural attributes in field trials 478
could be used to explore the optimal structure designs (Rytwinski et al. 2015; van der Ree et al. 479
2015a). 480
Monitoring arboreal crossing structures also presents a unique set of challenges when 481
compared to culverts or land-bridges (Gregory et al. 2014; Taylor and Goldingay 2014). 482
Cameras, sensors, batteries and solar panels must be securely installed 620 m high above an 483
active roadway (storing power sources at height reduces risk of vandalism and theft). In our 484
Page 21 of 42
study, access to the monitoring equipment was only possible with an elevated work platform, 485
which requiring a licensed operator, and traffic management to close highway lanes at 486
approximately $5,000$10,000AUD per day. To our knowledge, this is the first study to use 487
wireless download technology and PIT tag readers on arboreal crossing structures. 488
Unfortunately, though we tested the equipment extensively prior to its installation there were 489
several unexpected technical issues once the monitoring program was in place. Due to the 490
difficulties and costs associated with accessing the equipment, it was not feasible to conduct 491
regular repairs, leading to a loss of data at some sites. These issues should be carefully 492
considered when planning a monitoring study of arboreal structures, and additional 493
maintenance costs should be factored into monitoring budgets. 494
Conclusions 495
Improving the quality of information gained from monitoring studies is crucial if we are to 496
understand the conservation value of wildlife crossing structures. We found that individual 497
identification provided insights that would not have been possible through monitoring the rate 498
of use alone. While we used PIT tag readers, other methods such as non-invasive genetic 499
sampling (e.g. Sawaya et al. 2013), radio- or GPS-tracking (e.g. Dodd et al. 2007; Olsson et al. 500
2008) or coat pattern recognition (e.g. Mendoza et al. 2011; Trolle and Kery 2003) could be 501
used to obtain similar information. The most appropriate method will depend on the species 502
(e.g. not all species can be recognised by coat pattern). Canopy bridges and glider poles were 503
used by several species of arboreal marsupial, allowing safe access to resources that would have 504
been otherwise unavailable. In our study crossing structures were primarily used by a small 505
proportion of the population to regularly access habitat on both sides of the freeway. However, 506
while this benefits the small number of individuals that use the structure, it is unknown 507
whether this is enough to provide demographic or genetic connectivity for the wider 508
population. Still, if even a few dispersing individuals successfully cross and reproduce, and the 509
structures reduce rates of roadkill, then they are likely to benefit populations (Taylor and 510
Page 22 of 42
Goldingay 2012a). Based on this and previous studies, these structures are likely to have a 511
positive effect on a wide range of arboreal mammals, particularly gap-limited species or those 512
that are frequent victims of road mortality. Canopy bridges can benefit a wider range of species 513
than glider poles, and should be the preferred mitigation method where feasible. Methods such 514
as PIT tag readers are underutilised and should be widely adopted in studies monitoring the use 515
of structures by wildlife (Gibbons and Andrews 2004; van der Ree et al. 2007). 516
Acknowledgements 517
We thank The Baker Foundation, the Australian Research Centre for Urban Ecology, the 518
Australian Research Council (grant LP0560443), The Australian Research Council Centre of 519
Excellence for Environmental Decisions, the Holsworth Wildlife Research Endowment (ANZ 520
Trustees Foundation), the M.A. Ingram Trust, the RSPCA Alan White Scholarship and the 521
Wildlife Preservation Society of Australia for supporting this research. VicRoads and the New 522
South Wales Roads and Maritime Services provided invaluable financial and institutional 523
support for this project. Thanks to Doug Black, Varun Uthappa and Ross Meggs for help in 524
developing the monitoring systems. All animals in Victoria were trapped and tagged under the 525
approval of The University of Melbourne Animal Ethics Committee (1112269.1) and the 526
Department of Sustainability and Environment wildlife research permit (1006094). All animals 527
in New South Wales were trapped and tagged under the NSW Animal Research Authority (TRIM 528
09/5853) and the NSW National Parks and Wildlife Service Scientific License (S12782 and 529
SL100739). W. Sowersby, R. Soanes, L. Harrison, B. Mitchell and R. Bull provided help with data 530
collection. We also thank M.A. McCarthy, J. Stokes and several anonymous reviewers for their 531
valuable comments on earlier versions of this manuscript. 532
533
534
Page 23 of 42
535
Figure 1. Location of the study area along the Hume Freeway in south-east Australia (inset). 536
Major towns (black squares) and the position of retrofitted and newly-installed crossing 537
structures (circles) are indicated. At some sites, multiple structures are present as identified in 538
Table 1. 539
Page 24 of 42
540
Figure 2. Canopy bridges (top left) and glider poles (top right) along the Hume Freeway in 541
south-east Australia. Monitoring equipment (bottom left and right) included motion-triggered 542
cameras (1), active-infrared sensors (2), PIT tag antennae (3) and power supply (4). 543
544
Page 25 of 42
545
546
Figure 3. Video screen-shots of squirrel gliders on the cross-beam of the glider pole at Little 547
Billabong with male head gland (left) and ear notch (right) visible. 548
549
Page 26 of 42
Table 1. A description of the crossing structures present at each site. Road width measures the width of the treeless gap created by the road corridor 550
(trunk to trunk). Squirrel glider activity index refers to average number of individuals per hectare based on concurrent mark-recapture surveys 551
(described in Appendix 1). Max. glide distance refers to the maximum distance between glider poles and/or roadside trees. 552
Structures
present
Road
width (m)
Max.
distance to
roadside
trees (m)
Bridge
length
(m)
No. of
glider
poles
Pole
heights
(m)
Max.
glide
distance
(m)
Squirrel
glider
activity
index
General description
Victoria retrofitted to existing freeway
Bridge 1
80
9
73
-
-
-
0.33
Above road canopy bridge.
Bridge 2
94
9
86
-
-
-
0.23
Above road canopy bridge.
Pole 1
80
-
-
2
1416
47
0.34
One pole in centre median,
one on roadside.
Pole 2
62
-
-
1
16
31
NA
Pole in centre median
Pole 3
75
-
-
1
15
39
0.13
Pole in centre median
New South Wales installed during freeway construction
Bridge 3
70
12
63
-
-
-
0.18
Above-road canopy bridge.
Pole 4
76
-
-
2
1518
30
One pole in centre median,
one on roadside.
Pole 5
88
-
-
2
1618
35
One pole in centre median,
one on roadside.
Bridge 4
75
5
58
-
-
-
0.26
Above-road canopy bridge.
Pole 6
103
-
-
4
1517
28
Two poles in centre median,
two on roadsides.
Bridge 5
151
4
140
-
-
-
1.05
Under-road canopy bridge.
Pole 7
60
-
-
1
1418
29
0.10
Pole in centre median.
Page 27 of 42
Structures
present
Road
width (m)
Max.
distance to
roadside
trees (m)
Bridge
length
(m)
No. of
glider
poles
Pole
heights
(m)
Max.
glide
distance
(m)
Squirrel
glider
activity
index
General description
Pole 8
66
-
-
3
1318
26
0.05
One pole in centre median,
two on roadsides.
Pole 9
66
-
-
3
1318
30
One pole in centre median,
two on roadsides.
Pole 10
56
-
-
2
1318
28
0.59
One pole in centre median,
one on roadside.
Pole 11
68
-
-
2
16
29
One pole in centre median,
one on roadside.
Pole 12
69
-
-
1
1518
38
0.23
Pole in centre median.
Pole 13
93
-
-
4
1518
28.5
0.57
Two poles in centre median,
three on roadsides.
Pole 14
382
-
-
5
1417
184
Two poles in centre median,
three on roadsides.
Pole 15
244
-
-
5
1517
110
One pole in centre median,
three on roadsides.
553
Page 28 of 42
Table 2. Monitoring effort at each site with motion-triggered cameras and PIT tag readers. At 554
canopy bridges, camera nights shows the number of nights at least one camera was operational, 555
and PIT reader nights shows the number of nights at least one PIT tag reader was operational. 556
The number of nights that units at each end of the bridge were simultaneously operating is 557
given in parentheses. Hyphen indicates no monitoring effort. 558
Site
Structures
present
Camera
nights
PIT-tag
reader nights
Longwood
Bridge
108 (102)
*151 (0)
Violet Town
Bridge
199 (79)
*99 (0)
Balmattum
Pole
234
-
Baddaginnie
Pole
270
-
Warrenbayne
Pole
107
-
Sages TSR
Bridge
265 (102)
240 (30)
Pole 1
231
-
Pole 2
128
-
Blue Metal TSR
Bridge
235 (43)
271 (73)
Pole
133
-
Yarra Yarra Creek
Bridge
161 (81)
272 (125)
Little Billabong
Pole
258
-
Westby
Pole 1
152
-
Pole 2
152
-
Kyeamba Creek
Pole 1
75
-
Pole 2
279
-
MR1
Pole
268
-
Kyeamba TSR
Pole 1
258
-
Pole 2
268
-
Pole 13
148
-
*PIT tag readers were operational for 614 hours per night559
Page 29 of 42
Table 3. The number of crossings detected by cameras at each glider pole, ranked in order of nightly crossing rate (number of crossings per night of 560
camera operation) by squirrel gliders and sugar gliders. No animals were detected at Westby Lane (2), MR1, Kyeamba TSR (2) or Kyeamba TSR (3). 561
Hyphen indicates ‘not applicable’. 562
Kyeamba
Creek
pole 1
Warren
-bayne
Sages TSR
pole 1
Little
Billabong
Blue Metal
TSR
Baddaginnie
Kyeamba
Creek
pole 2
Balmattum
Sages
TSR
pole 2
Kyeamba
TSR pole 1
Westby
Lane
pole 1
Squirrel glider
Number of crossings
144
160
269
224
10
18
10
4
2
1
0
Nightly rate
1.92
1.50
1.16
0.87
0.08
0.07
0.04
0.02
0.02
<0.01
-
Sugar glider
Number of crossings
0
16
0
240
0
0
0
0
0
0
2
Nightly rate
-
-
0.15
0.93
-
-
-
-
-
-
0.01
Page 30 of 42
Table 4. Monitoring results from cameras and PIT tag readers at each canopy bridge for each 563
species, including: the nightly rate (number of crossings per night of operation); the percentage 564
of crossings that were confirmed by units at each end of the bridge; the average number of 565
tagged squirrel gliders and common brushtail possums present in the habitat surrounding each 566
canopy bridge (detail in Appendix 1); and the identity (ID) of individuals detected by the PIT tag 567
reader (♂=male, ♀=female). *indicates visit only. Hyphen indicates ‘not applicable’ 568
Site
Longwood
Violet Town
Sages
TSR
Blue Metal
TSR
Yarra Yarra
Creek
Squirrel glider
Average MNKTBA
2.3
3.0
4.3
8.3
4.0
Crossings detected by camera
Total number
% Confirmed
453
82%
2
0%
0
0
0
Nightly rate
4.19
0.01
-
-
-
Crossings detected by PIT reader
Total number
% confirmed
244
9%
-
0*
-
-
Nightly rate
1.62
-
-
-
-
No. indiv.
3
0
1
0
0
ID: nightly rate
H1: 0.67
P3: 0.79
Z1: 0.15
-
J5: 0*
-
-
Common brushtail possum
Average MNKTBA
11.7
14.3
1
7.8
4
Crossings detected by camera
Total number
% Confirmed
2
100%
227
35%
0
-
0
-
0
-
Nightly rate
0.02
1.14
-
-
-
Crossings detected by PIT reader
Total number
% Confirmed
19
5%
120
27%
0
-
0
-
0
-
Nightly rate
0.13
1.21
-
-
-
No. indiv.
1
4
0
0
0
ID: nightly rate
B1: 0.13
98: 0.22
92: 0.95
M1: 0.02
Z6: 0.03
-
-
-
Common ringtail possum
Crossings detected by camera
Total number
% Confirmed
94
85%
270
26%
0
-
22
45%
0
-
Nightly rate
0.87
1.36
-
0.09
-
Brush-tailed phascogale
Crossings detected by camera
Total number
% Confirmed
2
0%
0
-
0
-
0
-
0
-
Nightly rate
0.02
-
-
-
-
Page 31 of 42
Table 5. Parameter estimates from the Poisson regression of use of glider poles by squirrel 569
gliders. SD is standard deviation. 570
Parameter
Mean
Standard
error
95% credible
interval
Grand mean
-3.11
1.45
[-6.23, -0.63]
Effect of number of poles
-1.26
1.33
[-3.92, 1.45]
Effect of gap width
-0.12
0.23
[-0.58, 0.38]
SD Residual error
1.10
0.49
[0.42, 2.35]
SD Structure error
3.89
1.41
[1.97, 7.39]
571
572
Page 32 of 42
References 573
Anzures-Dadda, A., Manson, R.H. (2007). Patch- and landscape-scale effects on howler monkey 574
distribution and abundance in rainforest fragments. Animal Conservation 10(1), 6976. [In 575
English] 576
Asari, Y., Johnson, C.N., Parsons, M. and Larson, J. (2010). Gap-crossing in fragmented habitats by 577
mahogany gliders (Petaurus gracilis). Do they cross roads and powerline corridors? Australian 578
Mammalogy 32(1), 1015. 579
Ball, T.M. and Goldingay, R.L. (2008). Can wooden poles be used to reconnect habitat for a 580
gliding mammal? Landscape and Urban Planning 87, 140146. 581
Baxter-Gilbert, J., Lesbarreres, D. and Litzgus, J.D. (2013) On the road again: Measuring the 582
effectiveness of mitigation structures for reducing reptile mortality and maintaining population 583
connectivity. In ‘Proceedings of the 2013 International Conference on Ecology and 584
Transportation. Retrieved from http://www.icoet.net/ICOET_2013/proceedings.asp. 585
Bennett, A.F. (1991). Roads, roadsides and wildlife conservation: a review. In 'Nature 586
Conservation 2: The role of corridors.' (Eds DA Saunders and JH Hobbs). (Surrey Beatty 587
Chipping Norton, NSW) 588
Bennett, A.F. (1999). 'Linkages in the landscape: The role of corridors and connectivity in 589
wildlife conservation.' (IUCN: Gland, Switzerland & Cambridge) 590
Bissonette, J.A. and Adair, W. (2008). Restoring habitat permeability to roaded landscapes with 591
isometrically-scaled wildlife crossings. Biological Conservation 141(2), 482488. 592
Boarman, W.I., Beigel, M.L., Goodlett, G.C. and Sazaki, M. (1998). A passive integrated 593
transponder system for tracking animal movements. Wildlife Society Bulletin 26(4), 886891. 594
Bond, A.R. and Jones, D.N. (2008). Temporal trends in use of fauna-friendly underpasses and 595
overpasses. Wildlife Research 35, 103112. 596
Burnett, S.E. (1992). Effects of a rain-forest road on movements of small mammals - 597
mechanisms and implications. Wildlife Research 19(1), 95104. 598
Chambers, B. and Bencini, R. (2015) Factors affecting the use of fauna underpasses by 599
bandicoots and bobtail lizards. Animal Conservation doi: 10.1111/acv.12189. 600
Claridge, A.W. and van der Ree, R. (2004). Recovering endangered populations in fragmented 601
landscapes: the squirrel glider Petaurus norfolcensis on the south-west slopes of New South 602
Wales. In ‘Conservation of Australia's Forest Fauna, Second Edition.’ (Ed. D Lunney). pp. 678603
687. (CSIRO). 604
Clevenger, A.P. (2005). Conservation value of wildlife crossings: Measures of performance and 605
research directions. GAIA 14(2), 124129. 606
Page 33 of 42
Clevenger, A.P. and Sawaya, M.A. (2010). Piloting a non-invasive genetic sampling method for 607
evaluating population-level benefits of wildlife crossing structures. Ecology and Society 15(1). 608
Clevenger, A.P. and Waltho, N. (2000). Factors influencing the effectiveness of wildlife 609
underpasses in Banff National Park, Alberta, Canada. Conservation Biology 14(1), 4756. 610
Clevenger, A.P. and Waltho, N. (2003). Long-term, year-round monitoring of wildlife crossing 611
structures and the importance of temporal and spatial variability in performance studies. In 612
‘Proceedings of the 2003 International Conference on Ecology and Transportation. (Eds C.L. 613
Irwin, P. Garrett, K.P. McDermott.) pp. 239302, (North Carolina State University: Raleigh, NC.) 614
Corlatti, L., Hacklander, K. and Frey-Roos, F. (2009). Ability of wildlife overpasses to provide 615
connectivity and prevent genetic isolation. Conservation Biology 23(3), 548556. 616
Dique, D.S., Thompson, J., Preece, H.J., Penfold, G.C., de Villiers, D.L. and Leslie, R.S. (2003). Koala 617
mortality on roads in south-east Queensland: the koala speed-zone trial. Wildlife Research 618
30(4), 419426. 619
Dodd, N.L., Gagnon, J.W., Boe, S. and Schweinsburg, R.E. (2007). Assessment of elk highway 620
permeability by using Global Positioning System telemetry. Journal of Wildlife Management 621
71(4), 11071117. 622
Donaldson, A. and Cunneyworth, P. (2015) Canopy bridges for primate conservation. In 623
Handbook of Road Ecology.’ (Eds R. van der Ree, D.J. Smith, C. Grilo.) pp. 341343. (Wiley 624
Blackwell: London). 625
DOTARS (2007) Sydney-Melbourne Corridor Strategy - building our national transport future. 626
DSE (2007) Advisory list of threatened vertebrate fauna in Victoria - 2007. 627
Eigenbrod F, Hecnar SJ, Fahrig L (2008) Accessible habitat: An improved measure of the effects 628
of habitat loss and roads on wildlife populations. Landscape Ecology 23(2), 159168. 629
Fahrig, L. and Rytwinski, T. (2009). Effects of roads on animal abundance: an empirical review 630
and synthesis. Ecology and Society 14(1). 631
Forman, R.T.T., Sperling, D., et al. (2003). 'Road ecology: Science and solutions.' (Island Press: 632
Washington) 633
Foster, M.L. and Humphrey, S.R. (1995). Use of highway underpasses by Florida panthers and 634
other wildlife. Wildlife Society Bulletin 23(1), 95100. 635
Gagnon, J.W., Dodd, N.L., Ogren, K.S. and Schweinsburg, R.E. (2011). Factors associated with use 636
of wildlife underpasses and importance of long-term monitoring. Journal of Wildlife 637
Management 75, 14771487. 638
Gelman, A. (2006). Prior distributions for variance parameters in hierarchical models. Bayesian 639
Analysis 1, 515533. 640
Page 34 of 42
Gibbons, J.W. and Andrews, K.M. (2004). PIT tagging: Simple technology at its best. BioScience 641
54(5), 447454. 642
Gibbons, P. and Boak, M. (2002). The value of paddock trees for regional conservation in an 643
agricultural landscape. Ecological Management & Restoration 3(3), 205210. 644
Goldingay, R.L., Rohweder, D. and Taylor, B.D. (2013) Will arboreal mammals use rope-bridges 645
across a highway in eastern Australia? Australian Mammalogy 35(1), 3038. 646
Goldingay, R.L. and Taylor, B.D. (2009). Gliding performance and its relevance to gap crossing by 647
the squirrel glider (Petaurus norfolcensis). Australian Journal of Zoology 57, 99104. 648
Goosem, M. (2001). Effects of tropical rainforest roads on small mammals: inhibition of crossing 649
movements. Wildlife Research 28(4), 351364. 650
Goosem, M., Weston, N. and Bushnell, S. (2006). Effectiveness of rope bridge arboreal 651
overpasses and faunal underpasses in providing connectivity for rainforest fauna. In 652
‘Proceedings of the 2005 International Conference on Ecology and Transportation. (Eds. C. 653
Irwin, P. Garrett, K. McDermott). (North Carolina State University: Raleigh, NC.) 654
Gregory, T., Rueda, F.C., Deichmann, J., Kolowski, J. and Alonso, A. (2014) Arboreal camera 655
trapping: taking a proven method to new heights. Methods in Ecology and Evolution 5, 443451. 656
Grilo, C., Bissonette, J.A. and Santos-Reis, M. (2008). Response of carnivores to existing highway 657
culverts and underpasses: Implications for road planning and mitigation. Biodiversity and 658
Conservation 17(7), 16851699. 659
Gulle, N. (2006). Barrier effect of roads on common brushtail possums. Honours Thesis, School 660
of Botany, The University of Melbourne. 661
Harris, I.M., Mills, H.R. and Bencini, R. (2010). Multiple individual southern brown bandicoots 662
(Isoodon obesulus fusciventer) and foxes (Vulpes vulpes) use underpasses installed at a new 663
highway in Perth, Western Australia. Wildlife Research 37(2), 127133. 664
Herrod, A. (2005). Quantifying a barrier effect of a major freeway to Yellow-footed Antechinus 665
(Antechinus flavipes) occurring in roadside habitat fragments in northern Victoria, using 666
genotypic analyses. Honours Thesis, School of Biological Sciences, Monash University. 667
Jackson, S.M. (2000). Glide angle in the genus Petaurus and a review of gliding in mammals. 668
Mammal Review 30(1), 930. 669
Kelly, C.A., Diggins, C.A. and Lawrence, A.J. (2013). Crossing structures reconnect federally 670
endangered flying squirrel populations divided for 20 years by a road barrier. Wildlife Society 671
Bulletin DOI: 10.1002/wsb.249. 672
Laurance, S.G. and Laurance, W.F. (1999). Tropical wildlife corridors: use of linear rainforest 673
remnants by arboreal mammals. Biological Conservation 91(2-3), 231239. 674
Page 35 of 42
Mass, V., Rakotomanga, B., Rakotondratsimba, G., Razafindramisa, S., Andrianaivomahefa, P., 675
Dickinson, S., Berner, P.O. and Cooke, A. (2011). Lemur bridges provide crossing structures over 676
roads within a forested mining concession near Moramanga, Toamasina Province, Madagascar. 677
Conservation Evidence 8, 1118. 678
McCall, S.C., McCarthy, M.A., van der Ree, R., Harper, M.J., Cesarini, S. and Soanes, K. (2010). 679
Evidence that a highway reduces apparent survival rates of squirrel gliders. Ecology and Society 680
15(3). 681
Mendoza, E., Martineau, P.R., Brenner, E. and Dirzo, R. (2011). A novel method to improve 682
individual animal identification based on camera‐trapping data. The Journal of Wildlife 683
Management 75, 973979. 684
Ng, S.J., Dole, J.W., Sauvajot, R.M., Riley, S.P.D. and Valone, T.J. (2004). Use of highway 685
undercrossings by wildlife in southern California. Biological Conservation 115(3), 499507. 686
Olsson, M.P.O., Widen, P. and Larkin, J.L. (2008). Effectiveness of a highway overpass to promote 687
landscape connectivity and movement of moose and roe deer in Sweden. Landscape and Urban 688
Planning 85(2), 133139. 689
Oxley, D.J., Fenton, M.B. and Carmody, G.R. (1974). Effects of roads on populations of small 690
mammals. Journal of Applied Ecology 11(1), 51-59. 691
Riley, S.P.D., Pollinger, J.P., Sauvajot, R.M., York, E.C., Bromley, C., Fuller, T.K. and Wayne, R.K. 692
(2006). A southern California freeway is a physical and social barrier to gene flow in carnivores. 693
Molecular Ecology 15(7), 17331741. 694
Russell, T.C., Herbert, C.A. and Kohen, J.L. (2009). High possum mortality on urban roads: 695
implications for the population viability of the common brushtail and the common ringtail 696
possum. Australian Journal of Zoology 57(6), 391397. 697
Rytwinski, T., van der Ree, R., Cunnington, G.M., Fahrig, L., Findlay, C.S., Houlahan, J., Jaeger, 698
J.A.G., Soanes, K. (2015). Experimental Study Designs to Improve the Evaluation of Road 699
Mitigation Measures for Wildlife. Journal of Environmental Management 154, 4864. 700
Sawaya, M.A., Clevenger, A.P. and Kalinowski, S.T. (2013). Demographic connectivity for ursid 701
populations at wildlife crossing structures in Banff National Park. Conservation Biology DOI: 702
10.1111/cobi.12075. 703
Sawaya, M.A., Kalinowski, S.T. and Clevenger, A.P. (2014). Genetic connectivity for two bear 704
species at wildlife crossing structures in Banff National Park. Proceedings of the Royal Society B-705
Biological Sciences 281 (1780). 706
Schooley, R.L., Vanhorne, B. and Burnham, K.P. (1993). Passive integrated transponders for 707
marking free-ranging Townsend ground-squirrels. Journal of Mammalogy 74(2), 480484. 708
Simmons, J.M., Sunnucks, P., Taylor, A.C. and van der Ree, R. (2010). Beyond roadkill, 709
radiotracking, recapture and F-ST - A review of some genetic methods to improve understanding 710
of the influence of roads on wildlife. Ecology and Society 15(1). 711
Page 36 of 42
Soanes, K. (2014) Evaluating the success of road-crossing structures for arboreal mammals: 712
How does monitoring effort influence the detection of population level effects? PhD Thesis, 713
School of Botany, The University of Melbourne. 714
Soanes, K., Lobo, M.C., Vesk, P.A., McCarthy, M.A., Moore, J.L. and van der Ree, R. (2013). 715
Movement re-established but not restored: Inferring the effectiveness of road-crossing 716
mitigation for a gliding mammal by monitoring use. Biological Conservation 159, 434441. 717
Soanes, K., and van der Ree, R. (2015). Reducing road impacts on tree-dwelling animals. In 718
‘Handbook of Road Ecology’. (Eds. R. van der Ree, D. Smith, C. Grilo). pp. 334–340. (Wiley 719
Blackwell:London). 720
Spiegelhalter, D., Thomas, A., Best, N. and Lunn, D. (2011). Open BUGS User Manual, Version 721
3.2.1. (MRC Biostatistics Unit: Cambridge, UK). 722
Taylor, B.D. and Goldingay, R.L. (2004). Wildlife road-kills on three major roads in north-eastern 723
New South Wales. Wildlife Research 31(1), 8391. 724
Taylor, B.D. and Goldingay, R.L. (2010). Roads and wildlife: impacts, mitigation and implications 725
for wildlife management in Australia. Wildlife Research 37(4), 320331. 726
Taylor, B.D. and Goldingay, R.L. (2012a). Facilitated movement over major roads is required to 727
minimise extinction risk in an urban metapopulation of a gliding mammal. Wildlife Research 728
39(8), 685695. 729
Taylor, B.D. and Goldingay, R.L. (2012b). Restoring connectivity in landscapes fragmented by 730
major roads: A case study using wooden poles as "stepping stones" for gliding mammals. 731
Restoration Ecology 20(6), 671678. 732
Taylor, B.D., Goldingay, R.L. (2014). Using camera traps to monitor use of roadside glide poles 733
and rope canopy-bridges by Australian gliding mammals. In ‘Camera Trapping: Wildlife 734
Management and Research.’ (Eds P. Meek and P. Fleming) pp. 245–252. (CSIRO: Melbourne). 735
Teixeira, F.Z. (2013) Canopy bridges as road overpasses for wildlife in urban fragmented 736
landscapes. Biota Neotropica 13(1). 737
Trimming, E., Chambers, B., Grillo, D., de Tores, P. and Bencini, R. (2009). Road kills of the 738
western ringtail possum (Pseudocheirus occidentalis) occur at specific hotspots. In 'Australian 739
Mammal Society Semi-Centenary and 55th Meeting in Perth July 59', 2009, Perth. 740
Trolle, M. and Kéry, M. (2003). Estimation of ocelot density in the Pantanal using capture-741
recapture analysis of camera-trapping data. Journal of Mammalogy 84, 607614. 742
van der Grift, E.A., van der Ree, R., Fahrig, L., Findlay, S., Houlahan, J., Jaeger, J.A.G., Klar, N., 743
Madriñan, L.F. and Olson, L. (2013). Evaluating the effectiveness of road mitigation measures. 744
Biodiversity and Conservation 22(2), 425448. 745
Page 37 of 42
van der Grift, E.A. and van der Ree, R. (2015). Guidelines for evaluating use of wildlife crossing 746
structures. In ‘Handbook of road ecology.’ (Eds R. van der Ree, D. Smith and C. Grilo) pp. 119–747
128. (Wiley Blackwell: London). 748
van der Ree, R. (2002). The population ecology of the squirrel glider (Petaurus norfolcensis) 749
within a network of remnant linear habitats. Wildlife Research 29(4), 329340. 750
van der Ree, R. (2006). Road upgrade in Victoria a filter to the movement of the endangered 751
squirrel glider (Petaurus norfolcensis): Results of a pilot study. Ecological Management & 752
Restoration 7(3), 226228. 753
van der Ree, R., Bennett, A.F. and Gilmore, D.C. (2003). Gap-crossing by gliding marsupials: 754
thresholds for use of isolated woodland patches in an agricultural landscape. Biological 755
Conservation 115, 241249. 756
van der Ree, R., Gulle, N., Holland, K., van der Grift, E., Mata, C. and Suarez, F. (2007). Overcoming 757
the barrier effect of roads - how effective are mitigation strategies? In 'Proceedings of the 2007 758
International Conference on Ecology and Transportation'. (Eds C.L. Irwin, D. Nelson and K.P. 759
McDermott.) pp. 423431. (Centre of Transportation and The Environment, North Carolina 760
State University, Raleigh, North Carolina and Little Rock, Akansas, USA.) 761
van der Ree, R., Heinze, D., McCarthy, M. and Mansergh, I. (2009). Wildlife tunnel enhances 762
population viability. Ecology and Society 14(2). 763
van der Ree, R., Cesarini, S., Sunnucks, P., Moore, J.L. and Taylor, A. (2010). Large gaps in canopy 764
reduce road crossing by a gliding mammal. Ecology and Society 15(4). 765
van der Ree, R., Jaeger, J.A.G., van der Grift, E.A. and Clevenger, A.P. (2011). Effects of roads and 766
traffic on wildlife populations and landscape function: Road ecology is moving toward larger 767
scales. Ecology and Society 16(1). 768
van der Ree, R., Jaeger, J.A.G., Rytwinksi, T. and van der Grift, E. (2015a). Good science and 769
experimentation are needed in road ecology. In ‘Handbook of road ecology.’ (Eds R. van der Ree, 770
D. Smith and C. Grilo) pp. 7181. (Wiley Blackwell: London). 771
van der Ree, R., Smith, D.J. and Grilo, C. (2015b) ‘Handbook of road ecology’. (Wiley Blackwell: 772
London). 773
Van Dyck, H. and Baguette, M. (2005). Dispersal behaviour in fragmented landscapes: Routine or 774
special movements? Basic and Applied Ecology 6(6), 535545. 775
Weston, N., Goosem, M., Marsh, H., Cohen, M. and Wilson, R. (2011). Using canopy bridges to link 776
habitat for arboreal mammals: successful trials in the Wet Tropics of Queensland. Australian 777
Mammalogy 33, 93105. 778
Wilson, R.F., Marsh, H. and Winter, J. (2007). Importance of canopy connectivity for home range 779
and movements of the rainforest arboreal ringtail possum (Hemibelideus lemuroides). Wildlife 780
Research 34(3), 177184. 781
Page 38 of 42
782
Page 39 of 42
Appendix 1 Mark-recapture surveys at crossing structure sites 783
We obtained data on the local squirrel glider and common brushtail possum populations using 784
data from a concurrent mark-recapture study. Data were available for 12 of the 13 sites. Each 785
site was surveyed between one and four times between September 2011 and February 2013, 786
with the exception of Baddaginnie, where no surveys were conducted. The trapping surveys 787
covered the majority of the mature woodland within ~600 m of the crossing structure (or 788
crossing zone, where multiple structures were present at a single site) on both sides of the 789
freeway. 790
Wire-cage traps (17 cm x 20 cm x 50cm, Wiretainers Pty Ltd) were baited with a mixture of 791
honey, oats and peanut butter and placed 35 m high on tree trunks. The mean number of traps 792
set was 26 (range 838) per site per survey depending on the shape and extent of available 793
habitat surrounding the crossing structure. Traps were spaced at 50100 m intervals and 794
arranged in a grid or transect extending up to 600 m away from the crossing structure. Traps 795
were set for an average of six nights per survey (range 48). The trap effort at each site varied 796
from 125635 nights (mean 406). 797
Traps were checked each morning and all animals were weighed and their tooth-wear, gender 798
and reproductive status recorded. Animals were marked with a unique tattoo in the ear flap and 799
two 2 mm tissue biopsies were taken from the ear margin for use in genetic analysis for another 800
study. Each squirrel glider and common brushtail possum captured was implanted with a PIT 801
tag under the skin between the shoulder blades (ID 100, Trovan, Microchips Australia Pty Ltd). 802
We did not implant PIT tags in other species during these surveys (e.g. common ringtail 803
possums) because they are rarely detected in surveys using wire-cage traps. 804
Mark-recapture detected 267 animals during 4,883 trap nights. Species detected included 805
squirrel gliders (n=88), common brushtail possums (n=147), common ringtail possums (n=1), 806
sugar gliders (n=12), brush-tailed phascogales (n=3) and yellow-footed antechinus (n=16). We 807
calculated the Minimum Number of Animals Known To Be Alive (MNKTBA) for squirrel gliders 808
at each site and for common brushtail possums at canopy bridge sites. Where multiple surveys 809
were conducted, we used the average MNKTBA across all surveys at that site (Table A1). This 810
information was used to calculate the activity index for squirrel gliders presented in Table 1 of 811
the main text. The activity index was determined by dividing the average MNKTBA by the area 812
of habitat trapped at each site. The area of habitat trapped (in hectares) was calculated as the 813
area in which traps were set plus a buffer zone of 50 m to include any habitat immediately 814
adjacent to the trapping area (Table A1). The average MNKTBA for both squirrel gliders and 815
common brushtail possums at canopy bridge sites was used to compare the number of 816
Page 40 of 42
individuals present at each site (i.e. the number of animals likely to have access to the canopy 817
bridge at any one time) with the number that were detected crossing the canopy bridge by the 818
PIT tag reader. This data is presented in Table 4 of the main text. 819
Table A1. Mark-recapture trapping effort at each site. The average MNKTBA for squirrel gliders 820
and common brushtail possums is shown (+/- 1 SE in parentheses). 821
Average MNKTBA
Site
Trapping
area (ha.)
Number
of
surveys
Total
number
of trap
nights
Squirrel
glider
activity
index
(indiv./ha.)
Squirrel
gliders
Common
brushtail
possums
Longwood
6.9
3
532
0.33
2.3 (0.3)
11.7 (2.0)
Violet Town
10. 3
4
460
0.23
2.3 (0.5)
10.8 (1.0)
Balmattum
10.4
2
125
0.34
3.5 (1.5)
-
Warrenbayne
15.7
1
140
0.13
2 (NA)
-
Sages TSR
23.4
3
575
0.18
4.3 (0.7)
1 (0.4)
Blue Metal TSR
32.3
3
633
0.26
8.3 (0.3)
7.7 (0.9)
Yarra Yarra Creek
3.8
3
166
1.05
4.0 (0.6)
0.3 (0.3)
Little Billabong
19.1
3
456
0.10
2 (0.6)
-
Westby Lane
19.2
3
517
0.05
1 (0.6)
-
Kyeamba Creek
9.6
3
439
0.59
5.7 (0.7)
-
MR1
8.8
3
205
0.23
2 (0.6)
-
Kyeamba TSR
18.3
3
635
0.57
5 (0.6)
-
822
823
Page 41 of 42
Appendix 2 - Model code 824
825
model{ 826
for (i in 1:Nobs) { 827
x[i] ~ dpois(lambda[i]) 828
lambda[i] <- theta[i] * t[i] 829
theta[i] <- exp(mu[i]) 830
mu[i] ~ dnorm(phi[i], prec[1]) 831
832
phi[i] <- max(min(GM + npoles.eff * (npoles[i] -2) + gap.eff * (gap[i] - 30) + str.eff[str[i]],99),-833
99) 834
} 835
836
for (k in 1:Nstr) { str.eff[k] ~ dnorm (0, prec[2]) } 837
838
GM ~dnorm(0,0.0001) 839
npoles.eff ~dnorm(0,0.0001) 840
gap.eff ~dnorm(0,0.0001) 841
842
for (m in 1:2) { 843
prec[m] <- 1/sd[m]/sd[m] 844
sd[m] ~ dt(0, .0016, 1)T(0,) 845
} 846
847
pred.mean.rate <- exp(GM) 848
849
} 850
851
} 852
END MODEL 853
854
DATA 855
list(Nobs=26, Nstr=13) 856
t[] x[] str[] npoles[] gap[] 857
228 3 1 2 47 858
6 1 1 2 47 859
30 36 2 1 39 860
77 124 2 1 39 861
206 7 3 1 31 862
59 10 3 1 31 863
5 1 3 1 31 864
217 246 4 2 30 865
14 23 4 2 30 866
114 2 5 2 35 867
114 0 5 2 35 868
119 4 6 4 28 869
14 6 6 4 28 870
258 224 7 2 29 871
138 0 8 3 26 872
14 0 8 3 26 873
138 0 9 3 30 874
14 0 9 3 30 875
61 118 10 2 28 876
14 26 10 2 28 877
Page 42 of 42
279 10 11 2 29 878
60 0 12 1 38 879
208 0 12 1 38 880
70 0 13 4 28.5 881
20 1 13 4 28.5 882
168 0 13 4 28.5 883
END 884
885
... Despite recent arboreal bridge studies from Peru (Gregory et al., 2017) and China (Chan et al., 2020), most arboreal wildlife bridge studies are from Australia, with multiple projects trialling a variety of designs throughout different regions (Abson, and Lawrence, 2003;Weston, 2003;Taylor and Goldingay, 2009). Collectively, over 10 arboreal bridge designs and variants have been trialled and tested including single ropes, rope tunnels (with and without square cross-sections), rope ladders, rope bridges with glider pole intervals, rope and mesh combination bridges, and woven rope bridges (Goosem et al., 2005;Taylor and Goldingay, 2009;Soanes and van der Ree, 2010;Soanes et al., 2013;Soanes et al., 2015;Goldingay and Taylor, 2017). This has allowed for effective designs and techniques across different regions and various species in Australia to be increasingly well-defined and have influence on projects in other countries, such as South Africa (Linden et al., 2020), the United Kingdom (White and Hughes, 2019), and Japan (Minato et al., 2012). ...
... The efficient testing and design of arboreal wildlife bridges is key to understanding the needs of specific species (Soanes and van der Ree, 2015;Taylor, 2017). Large-bodied species such as the Hainan gibbon (Nomascus hainanus) for example have been observed using a double rope canopy bridge in Asia (Chan et al., 2020). ...
Article
Full-text available
Linear infrastructures, especially roads, affect the integrity of natural habitats worldwide. Roads act as a barrier to animal movement, cause mortality, decrease gene flow and increase the probability of local extinctions, particularly for arboreal species. Arboreal wildlife bridges increase connectivity of fragmented forests by allowing wildlife to safely traverse roads. However, the majority of studies about such infrastructure are from Australia, while information on lowland tropical rainforest systems in Meso and South America remains sparse. To better facilitate potential movement between forest areas for the arboreal wildlife community of Costa Rica’s Osa Peninsula, we installed and monitored the early use of 12 arboreal wildlife bridges of three different designs (single rope, double rope, and ladder bridges). We show that during the first 6 months of monitoring via camera traps, 7 of the 12 bridges were used, and all bridge designs experienced wildlife activity (mammals crossing and birds perching). A total of 5 mammal species crossing and 3 bird species perching were observed. In addition to preliminary results of wildlife usage, we also provide technical information on the bridge site selection process, bridge construction steps, installation time, and overall associated costs of each design. Finally, we highlight aspects to be tested in the future, including additional bridge designs, monitoring approaches, and the use of wildlife attractants.
... Cameras that have the capacity for wireless data transfer allow data access without the need to access cameras directly (e.g. Soanes et al., 2015). The status and function of cameras can also be remotely assessed, either through an online diagnostic or by programming the cameras to record regular 'test' images (i.e. ...
... Soanes pers. comm., regardingSoanes et al., 2015). Therefore, it is important to consider additional costs and equipment necessary Arboreal camera traps may be placed using climbing or non-climbing methods. ...
Article
Full-text available
Arboreal camera trapping is a burgeoning method providing a novel and effective technique to answer research questions across a variety of ecosystems, and it has the capacity to improve our understanding of a wide range of taxa. However, while terrestrial camera trapping has received much attention, there is little guidance for dealing with the unique challenges of working in the arboreal realm. Our review draws on the expertise of researchers from six continents and the broader literature to investigate the advantages and disadvantages of arboreal camera trapping, and challenges to consider when using this technology. We also include mini-guides with detailed information on the current arboreal camera trap literature, mounts used to install arboreal cameras, tree climbing pointers and safety tips, methods for deploying cameras without climbing, and tips for managing interference with camera function. We find that arboreal camera traps have been most commonly used in the study of mammals in forests; however, there is potential for this method to be applied to a broad range of habitats including urban areas, and taxa such as birds, amphibians, invertebrates, and plants. Methods in arboreal camera trapping could be improved by developing a greater understanding of the factors affecting detection of species. The most common challenges of arboreal camera trapping are camera placement and camera site access. These can be overcome by understanding correct camera orientation, managing potential sources of interference in front of cameras, utilizing appropriate cameras mounts and training researchers properly. Given the benefits and opportunities presented by arboreal camera trapping, it is likely to become an ever-more popular method of studying arboreal species and systems. The information synthesized in this review provides guidance for future studies to help direct more reliable and robust ecological inferences from arboreal camera trapping.
... Additionally, rope bridges and glider poles have successfully encouraged squirrel gliders and sugar gliders to cross highways across the east coast of Australia. These species were observed using the man-made crossing structures through radiotracking, hair-trapping and camera trapping (Ball and Goldingay 2008;Taylor and Goldingay 2012;Soanes et al. 2015Soanes et al. , 2018Goldingay et al. 2019). Research in Port Macquarie concluded that glider poles were preferred over rope bridges for squirrel gliders and sugar gliders, while rope bridges were utilised by non-gliding mammals (Goldingay et al. 2019). ...
... Even after installing gates, surveys should continue to ensure that turtle movement throughout the landscape is not impeded by fences. Motion triggered cameras and passive transponder trackers (Soanes et al. 2015) could be installed at gates while routine inspections along fences (as part of general maintenance) ensuring that gates are in the most effective location. Further modifications could be administered retrospectively after gates are installed. ...
Article
Installing conservation fences to prohibit feral animal access to wetlands can become a barrier for non‐target species of interest. We collected 161 turtles (Chelodina rugosa, Emydura subglobosa worrelli, Myuchelys latisternum) from twenty floodplain and riverine wetlands during post‐wet (June–August) and late‐dry season (November–December) surveys (2015–2018) in northern Australia. Wetlands were fenced (150 × 150 mm square, 1.05 m high wire mesh) or unfenced around the wet perimeter. Ninety‐seven percent of individuals caught in either fenced or unfenced wetlands had a shell carapace width greater than mesh width, of these 44 (46%) were captured inside fenced wetlands, while 50 were caught in unfenced wetlands. The remaining 35 turtles were smaller than 150 mm and would likely pass easily through fence mesh. Sixty‐five turtles partook in a fencing manipulative experiment. Turtles with carapace widths wider than mesh often successfully escaped through fences by lifting one side of their shell and passing diagonally through the mesh. In a second experiment where a piece of vertical wire (1500 × 300 mm) was removed, turtles located ‘gates' after prospecting and fitting through meshing areas that were too small to pass. Ninety‐two percent of turtles were able to locate and pass through gates, while 8% failed to locate a gate after 2 h. Gates applied every 4 m showed an 83% passage rate, every 2 m was 91%, and every 1 m was 100%. Combing field and manipulative experiments revealed that large turtles will prospect and move along a fence until they find suitable passage, which has important consequences when considering that gates could be easily retrofitted to existing sites, as well in new fencing programs, which has enormous positive conservation benefits for turtles in an already challenging and changing floodplain environment.
Article
Although artificial crossing structures are increasingly implemented by conservationists and wildlife managers to connect fragmented wildlife habitat, the study of artificial crossing structure design, particularly of canopy bridges, is an emerging field of study in primatology. We address this issue by evaluating five competing bridge models with varying width, material stiffness, and substrate spacing. We assessed bridge preference and performance by sampling the behavior of three species of Costa Rican monkeys ( Alouatta palliata : n = 4, Ateles geoffroyi : n = 3, Cebus imitator : n = 3). In a semi-wild setting, we used focal individual sampling with instantaneous recording once every minute for ten-minute intervals and all occurrences sampling whenever study subjects used the bridge. We hypothesized that monkeys prefer bridges that are more stable, and that are made of materials that resemble tree branches. During nearly 37 sampling hours we observed 119 crossing events. We found that study subjects prefer bridge models that are built using more rigid materials, such as the bamboo pole bridge, or wider bridges that offer more stability than narrower bridges. The bridge type and the materials used to build the bridges are both significant predictors of bridge use. While preference for bridges and their performance varies by species, the bamboo pole bridge model and the horizontal mesh bridge were preferred and performed best in our study. The simple liana bridge model was the least preferred by all species and performed poorly in comparison to the other models. Our findings will help us better understand how design and materials impact the use of canopy bridges by monkeys, which can help improve biological corridors and offer new information for the management and conservation of primates living near infrastructure corridors and other kinds of dangerous matrix.
Article
Although artificial crossing structures are increasingly implemented by conservationists and wildlife managers to connect fragmented wildlife habitat, the study of artificial crossing structure design, particularly of canopy bridges, is an emerging field of study in primatology.We address this issue by evaluating five competing bridge models with varying width, material stiffness, and substrate spacing. We assessed bridge preference and performance by sampling the behavior of three species of Costa Rican monkeys (Alouatta palliata: n = 4, Ateles geoffroyi: n = 3, Cebus imitator: n = 3). In a semi-wild setting, we used focal individual sampling with instantaneous recording once every minute for ten-minute intervals and all occurrences sampling whenever study subjects used the bridge. We hypothesized that monkeys prefer bridges that are more stable, and that are made of materials that resemble tree branches. During nearly 37 sampling hours we observed 119 crossing events. We found that study subjects prefer bridge models that are built using more rigid materials, such as the bamboo pole bridge, or wider bridges that offer more stability than narrower bridges. The bridge type and the materials used to build the bridges are both significant predictors of bridge use. While preference for bridges and their performance varies by species, the bamboo pole bridge model and the horizontal mesh bridge were preferred and performed best in our study. The simple liana bridge model was the least preferred by all species and performed poorly in comparison to the other models. Our findings will help us better understand how design and materials impact the use of canopy bridges by monkeys, which can help improve biological corridors and offer new information for the management and conservation of primates living near infrastructure corridors and other kinds of dangerous matrix.
Article
Full-text available
Australian arboreal mammals are experiencing significant population declines, particularly due to land clearing and resulting habitat fragmentation. The squirrel glider, Petaurus norfolcensis , is a threatened species in New South Wales, with a stronghold population in the Lake Macquarie Local Government Area (LGA) where fragmentation due to urbanization is an ongoing problem for the species conservation. Here we report on the use of squirrel glider mitochondrial (385 bp cytochrome b gene, 70 individuals) and nuclear DNA (6,834 SNPs, 87 individuals) markers to assess their population genetic structure and connectivity across 14 locations sampled in the Lake Macquarie LGA. The mitochondrial DNA sequences detected evidence of a historical genetic bottleneck, while the genome-wide SNPs detected significant population structure in the Lake Macquarie squirrel glider populations at scales as fine as one kilometer. There was no evidence of inbreeding within patches, however there were clear effects of habitat fragmentation and biogeographical barriers on gene flow. A least cost path analysis identified thin linear corridors that have high priority for conservation. These areas should be protected to avoid further isolation of squirrel glider populations and the loss of genetic diversity through genetic drift.
Preprint
Full-text available
Australian arboreal mammals are experiencing significant population declines, particularly due to land clearing and resulting habitat fragmentation. The squirrel glider, Petaurus norfolcensis, is a threatened species in New South Wales, with a stronghold population in the Lake Macquarie Local Government Area (LGA) where fragmentation due to urbanization is an ongoing problem for the species conservation. Here we report on the use of squirrel glider mitochondrial (385 bp cytochrome b gene, 70 individuals) and nuclear DNA (6,812 SNPs, 87 individuals) markers to assess their population genetic structure and connectivity across 14 locations sampled in the Lake Macquarie LGA. The mitochondrial DNA sequences detected evidence of a historical genetic bottleneck, while the genome-wide SNPs detected significant population structure in the Lake Macquarie squirrel glider populations at scales as fine as one kilometer. There was no evidence of inbreeding within patches, however there were clear effects of habitat fragmentation and biogeographical barriers on gene flow. A least cost path analysis identified thin linear corridors that have high priority for conservation. These areas should be protected to avoid further isolation of squirrel glider populations and the loss of genetic diversity through genetic drift.
Preprint
Full-text available
Arboreal gliders are vulnerable to habitat fragmentation and to barriers that extend their glide distance threshold. Habitat fragmentation through deforestation can cause population isolation and genetic drift in gliding mammals which in turn can result in a loss of genetic diversity and population long-term persistence. This study utilised next generation sequencing technology to call 11, 292 genome-wide SNPs from 90 adult sugar gliders ( Petaurus breviceps ). Samples were collected from 12 locations in the Lake Macquarie Local Government Area (New South Wales), with two of these locations west of the Pacific Motorway, a potential major barrier to their dispersal. Overall, Lake Macquarie sugar gliders appeared to have high levels of gene flow and little genetic differentiation, however spatial least cost path analyses identified the Pacific Motorway as a barrier to their dispersal. This Motorway is still relatively new (< 40 years old), so man-made crossing structures should be erected as a management priority to mitigate any long-term effects of population isolation by assisting in the dispersal and gene flow of the species. This study provides further insight into the sugar glider after it was classed as three separate species in 2020 and could potentially be used as a model for its threatened congener in the area, the squirrel glider ( Petaurus norfolcensis ).
Thesis
Full-text available
Resumen: La fragmentación de hábitats producidos por actividades antropogénicas como la construcción de carreteras es perjudicial para las poblaciones de fauna que se encuentran alrededor de ellas pues intervienen en la dispersión y migración de individuos, causando atropellos de fauna, así como inseguridad vial. La correcta instalación de pasos de fauna para la mitigación de este impacto, resulta ser de mucha utilidad, debido a que son estructuras que ayudan a que las poblaciones se mantengan conectadas por ambos lados de la carretera. En esta investigación, se analizó la efectividad del uso correcto y seguro de los nueve pasos arbóreos ya instalados. Se realizó por medio del uso de cámaras trampa colocadas en uno de los extremos de cada paso para observar si los animales logran cruzarlas. Se obtuvo un sólo registro de 4 individuos de la especie Cebus capucinus para el paso de fauna cinco de las 1296 horas trampa conectadas en el tiempo de evaluación. Se realizó el conteo de animales arbóreos atropellados como complemento de la información del uso de los pasos para tener conocimiento de la fauna aledaña, donde se registraron nueve individuos atropellados en cercanías los pasos. Se instalaron cámaras trampa en el bosque, continuas a los pasos para tener conocimiento de la fauna arbórea existente en la zona boscosa. Se determinó que los pasos de fauna no están siendo utilizados por la fauna debido a que algunas de estas se encuentran mal conectadas a la zona boscosa, otras se encuentran dañadas, otras muy cercanas a zonas urbanas. Se recomienda el realizar un monitoreo exhaustivo para determinar los sitios estratégicos del paso de fauna, así como la revisión del tipo de estructuras instaladas. La información obtenida, será favorable para la mejora de la instalación de pasos arbóreos así como para futuras investigaciones. Palabras clave: aislamiento, atropello de fauna, cámara trampa, fragmentación de hábitat, pasos de fauna, conexión Abstract: The fragmentation of habitats produced by anthropogenic activities like construction of highways, are harmful for the fauna populations who lives around of it, causing hitting and road insecurity. The correct installation of canopy bridges for the mitigation of the impact, are very useful because they are structures who helps the populations to keep connectivity by both sides of the road. In this investigation the connectivity of the correct use of nine already installed bridges canopy was analyzed by cameras trap placed in one extreme of the bridges to observe if the arboreal animals are using them. The result was just one species of four individuals of Cebus capucinus in the canopy bridge number five in the 1296 hours of connected camera. The roadkills was registrated like an extra of the evaluation and the result was a total of nine individuals who were close of the canopy bridges. Also, cameras trap were installed close to the
Article
Full-text available
We attempted a complete review of the empirical literature on effects of roads and traffic on animal abundance and distribution. We found 79 studies, with results for 131 species and 30 species groups. Overall, the number of documented negative effects of roads on animal abundance outnumbered the number of positive effects by a factor of 5; 114 responses were negative, 22 were positive, and 56 showed no effect. Amphibians and reptiles tended to show negative effects. Birds showed mainly negative or no effects, with a few positive effects for some small birds and for vultures. Small mammals generally showed either positive effects or no effect, mid-sized mammals showed either negative effects or no effect, and large mammals showed predominantly negative effects. We synthesized this information, along with information on species attributes, to develop a set of predictions of the conditions that lead to negative or positive effects or no effect of roads on animal abundance. Four species types are predicted to respond negatively to roads: (i) species that are attracted to roads and are unable to avoid individual cars; (ii) species with large movement ranges, low reproductive rates, and low natural densities; and (iii and iv) small animals whose populations are not limited by road-affected predators and either (a) avoid habitat near roads due to traffic disturbance or (b) show no avoidance of roads or traffic disturbance and are unable to avoid oncoming cars. Two species types are predicted to respond positively to roads: (i) species that are attracted to roads for an important resource (e.g., food) and are able to avoid oncoming cars, and (ii) species that do not avoid traffic disturbance but do avoid roads, and whose main predators show negative population-level responses to roads. Other conditions lead to weak or non-existent effects of roads and traffic on animal abundance. We identify areas where further research is needed, but we also argue that the evidence for population- level effects of roads and traffic is already strong enough to merit routine consideration of mitigation of these effects in all road construction and maintenance projects.
Article
Full-text available
Roads and traffic reduce landscape connectivity and increase rates of mortality for many species of wildlife. Species that glide from tree to tree may be strongly affected by roads and traffic if the size of the gap between trees exceeds their gliding capability. Not only are wide roads likely to reduce crossing rates, but mortality may also be increased if gliders that do cross have poor landing opportunities. The road-crossing behavior of 47 squirrel gliders (Petaurus norfolcensis) was investigated in southeast Australia using radio-tracking. The proportion of gliders crossing one or both roadways of a freeway where trees were present or absent from the center median was compared to that at single-lane country roads (control). The proportion of gliders crossing the road at control sites (77%) was similar to the proportion that crossed one or both roadways at the freeway with trees in the median (67%), whereas only a single male (6%) crossed the freeway where trees were absent from the median. The frequency of crossing for each individual was also similar at control sites and freeway sites with trees in the median. The almost complete lack of crossing at sites where trees were absent from the median was attributed to the wider gap in canopy (50 - 64 m vs. 5 - 13 m at sites with trees in the median). This suggests that traffic volume, up to 5,000 vehicles per day on each roadway, and the other characteristics of the freeway we studied are not in themselves complete deterrents to road crossing by squirrel gliders. This study demonstrates that retaining and facilitating the growth of tall trees in the center median of two-way roads may mitigate the barrier effect of roads on gliders, thus contributing positively to mobility and potentially to connectivity. This information will be essential for the assessment of road impacts on gliding species using population viability models.
Article
Full-text available
Road ecology has developed into a significant branch of ecology with steady growth in the number of refereed journal articles, books, conferences, symposia, and “best practice” guidelines being produced each year. The main objective of this special issue of Ecology and Society is to highlight the need for studies that document the population, community, and ecosystem-level effects of roads and traffic by publishing studies that document these effects. It became apparent when compiling this special issue that there is a paucity of studies that explicitly examined higher order effects of roads and traffic. No papers on landscape function or ecosystem-level effects were submitted, despite being highlighted as a priority for publication. The 17 papers in this issue, from Australia, Canada, the Netherlands, and USA, all deal to some extent with either population or community-level effects of roads and traffic. Nevertheless, many higher order effects remain unquantified, and must become the focus of future studies because the complexity and interactions among the effects of roads and traffic are large and potentially unexpected. An analysis of these complex interrelations requires systematic research, and it is necessary to further establish collaborative links between ecologists and transportation agencies. Many road agencies have “environmental sustainability” as one of their goals and the only way to achieve such goals is for them to support and foster long-term and credible scientific research. The current situation, with numerous small-scale projects being undertaken independently of each other, cannot provide the information required to quantify and mitigate the negative effects of roads and traffic on higher levels. The future of road ecology research will be best enhanced when multiple road projects in different states or countries are combined and studied as part of integrated, well-replicated research projects.
Article
We describe an automated system that uses passive integrated transponder (PIT) tags to track movements of animals past specific locations. The system was designed to operate maintenance free for several months, be secure from vandalism and environmental damage, and record the identity, date, and time of passage of animals past a 2.4-m wide area. We used the system to monitor effectively the movements of 172 desert tortoises (Gopherus agassizii) through 2 storm drain culverts that pass beneath a state highway in the Mojave Desert, California. Four tortoises entered or passed through the culverts on 60 occasions. The system can be easily adapted to other species.
Article
The Ambatovy Project includes a large, open-pit nickel mine located in Madagascar's eastern humid forest, and an associated pipeline to remove laterite slurry off site. The area is recognized for its high biodiversity exemplified by the presence of at least 13 lemur species in forests surrounding the mine site. In order to reduce potential habitat fragmentation impacts on the lemur populations as a consequence of recent access road construction, seven crossing structures (referred to as 'lemur bridges') were erected within the mine footprint area and along the slurry pipeline that will remain in place until rehabilitated forest allows for movement over roads via the forest canopy. Two bridge designs were used due to differences in road width and vehicle traffic type. Lemur bridges have been monitored since their construction in January-February 2009. To date (10 August 2010), bridges have been used by six lemur species. Mine footprint type bridges (suspension bridge design) have been used more frequently than slurry pipeline bridges (plank bridge design) and, overall, there has been an increase in bridge use in 2010 when compared to 2009 (from 8% to 24% of total observations where lemurs are present in proximity to bridges). These results suggest that although a certain time period may be required for lemurs to locate and habituate to bridges, these crossing structures offer an effective mitigation measure to assist in reducing the impacts of habitat fragmentation.
Article
Intuitively, wildlife crossing structures should enhance the viability of wildlife populations. Previous research has demonstrated that a broad range of species will use crossing structures, however, questions remain as to whether these measures actually provide benefits to populations. To assess this, studies will need to determine the number of individuals using crossings, their sex, and their genetic relationships. Obtaining empirical data demonstrating population-level benefits for some species can be problematic and challenging at best. Molecular techniques now make it possible to identify species, individuals, their sex, and their genetic relatedness from hair samples collected through non-invasive genetic sampling (NGS). We describe efforts to pilot a method to assess potential population-level benefits of wildlife crossing structures. We tested the feasibility of a prototype NGS system designed to sample hair from black bears (Ursus americanus) and grizzly bears (U. arctos) at two wildlife underpasses. The piloted hair-sampling method did not deter animal use of the trial underpasses and was effective at sampling hair from more than 90% of the bear crossing events at the underpasses. Hair samples were also obtained from non-target carnivore species, including three out of five (60%) cougar (Puma concolor) crossing events. Individual identification analysis revealed that three female and two male grizzly bears used one wildlife underpass, whereas two female and three male black bears were identified as using the other underpass. Of the 36 hair samples from bears analyzed, five failed, resulting in an 87% extraction success rate, and six more were only identified to species. Overall, 70% of the hair samples from bears collected in the field had sufficient DNA for extraction purposes. Preliminary data from our NGS suggest the technique can be a reliable method to assess the population-level benefits of Banff wildlife crossings. Furthermore, NGS can be an important tool for the conservation value of wildlife crossings for other taxa, and we urge others to carry out evaluations of this emerging methodology.
Chapter
Global road length, number of vehicles and rate of per capita travel are high and predicted to increase significantly over the next few decades.2The ‘road-effect zone’ is a useful conceptual framework to quantify the negative ecological and environmental impacts of roads and traffic.3The effects of roads and traffic on wildlife are numerous, varied and typically deleterious.4The density and configuration of road networks are important considerations in road planning.5The costs to society of wildlife-vehicle collisions can be high.6The strategies of avoidance, minimisation, mitigation and offsetting are increasingly being adopted around the world – but it must be recognised that some impacts are unavoidable and unmitigable.7Road ecology is an applied science which underpins the quantification and mitigation of road impacts.