Radio-collared squirrel glider (Petaurus norfolcensis) struck by vehicle and transported 500 km along freeway

Abstract

Roadkill (the mortality of animals through wildlife–vehicle collisions) is one of the main impacts of roads on wildlife. Studies quantifying the location and rate of roadkill to identify ‘hot spots’ are often used to guide the location of mitigation efforts, such as fencing or wildlife crossing structures. However, sometimes quantifying rates of roadkill can be challenging, particularly for species that are small and difficult to detect. In our study, a squirrel glider that was trapped and radio-collared in north-east Victoria was found as roadkill more than 500 km away, suggesting that a vehicle struck the animal and carried the carcass away from the site of impact. Our observation is the first evidence that this occurs for squirrel gliders.

Full text

!
1!
Published in Australian Mammalogy on 23 October 2015 1!
http://dx.doi.org/10.1071/AM15013 2!
Radio-collared squirrel glider (Petaurus norfolcensis) struck by vehicle and 3!
transported 500 km along freeway 4!
Kylie Soanesab*, Melissa Carmody Loboc and Rodney van der Reeab
5!
aAustralian Research Centre for Urban Ecology, Royal Botanic Gardens, Melbourne, 6!
VIC 3010, Australia 7!
bSchool of Biosciences, University of Melbourne, Parkville, VIC 8!
cWildlife Conservation Society, Santiago, Chile 9!
*corresponding author: Phone +61 (03) 8344 0146, Fax +61 (03) 9347 9123, e-mail: 10!
ksoanes@unimelb.edu.au (K. Soanes) 11!
Abstract. Roadkill (the mortality of animals through wildlife-vehicle collisions) is 12!
one of the main impacts of roads on wildlife. Studies quantifying the location and rate 13!
of roadkill to identify ‘hot spots’ are often used to guide the location of mitigation 14!
efforts, such as fencing or wildlife crossing structures. However sometimes 15!
quantifying rates of roadkill can be challenging, particularly for species that are small 16!
and difficult to detect. In our study, a squirrel glider that was trapped and radio-17!
collared in northeast Victoria was found as roadkill more than 500 km away, 18!
suggesting that a vehicle struck the animal and carried the carcass away from the site 19!
of impact. Our observation is the first evidence that this occurs for squirrel gliders. 20!
Introduction 21!
Roadkill – the mortality of animals through wildlife-vehicle collisions – is one of the 22!
main ways that roads affect wildlife populations (Forman et al. 2003; van der Ree et 23!
al. 2015). Quantifying the rates and spatio-temporal patterns of roadkill are common 24!
approaches to assessing the impacts of roads and traffic on wildlife and guiding the 25!
placement of mitigation (Forman et al. 2003; Taylor and Goldingay 2010). However, 26!
for some species quantifying the number of animals killed on roads can be 27!
complicated. For example, small animals are difficult to identify, particularly after 28!
carcasses have been repeatedly run-over by vehicles (e.g. Taylor and Goldingay 2004; 29!
Gonzalez-Gallina et al. 2013). Smaller carcasses are often more difficult to detect, as 30!
they can be removed by scavengers before surveys are conducted, or the impact of the 31!
collision may throw the carcass into roadside vegetation (e.g. Santos et al. 2011). It is 32!
also possible that carcasses may remain lodged on the vehicle itself and transported 33!
away from the site of impact; however records of this occurring are anecdotal. All of 34!
these factors can influence the accuracy of roadkill counts. 35!
We present evidence that a squirrel glider (Petaurus norfolcensis), a threatened 36!
gliding marsupial, was killed by a vehicle and its carcass transported approximately 37!

!
2!
500 km away. This information may help better understand the limitations and 38!
uncertainties of roadkill studies for this and similar species. 39!
Method 40!
We have undertaken a series of comprehensive studies to quantify the impacts of 41!
roads, traffic and wildlife crossing structures on arboreal mammal populations 42!
(McCall et al. 2010; Soanes 2014; Soanes et al. 2013; Soanes et al. 2015; van der Ree 43!
et al. 2010). Our study area encompassed a 330 km stretch of the Hume Freeway in 44!
southeast Australia, between the towns of Avenel in northeast Victoria (36°54'2.54"S, 45!
145°14'0.01"E) and Tarcutta in southeast New South Wales (35°16'34.94"S, 46!
147°44'18.94"E). Within this region the traffic volume averaged 10,000 vehicles per 47!
day, approximately 25% of which occurs at night when many native mammals are 48!
active. The width of the freeway was approximately 40- 100 m (measured as the 49!
distance between woodland habitat on opposite sides of the road) with a centre 50!
median up to 40 m wide. The surrounding area was primarily agricultural fields and 51!
rural townships. Woodland vegetation (Eucalyptus spp, Box-Ironbark and Box-Gum 52!
woodland) predominantly occurs as linear strips of remnant vegetation along 53!
roadsides and waterways, although some larger reserves are present (Gibbons and 54!
Boak 2002; van der Ree 2002). 55!
Our primary study species, the squirrel glider, is a small (~250 g) gliding marsupial in 56!
the family Petauridae. Glide distance is largely dependent on launch height and the 57!
average glide distance ranges from 20 to 40 m though longer glides of 70–90 m have 58!
been recorded (Goldingay and Taylor 2009; van der Ree 2006; van der Ree et al. 59!
2003). Squirrel gliders are occasionally observed as roadkill (pers. obs), but the 60!
carcasses are likely to be difficult to detect due to their small size and grey colouring. 61!
We captured and radio-collared squirrel gliders along the Hume Freeway in northeast 62!
Victoria as part of a project to measure the impacts of the freeway and subsequent 63!
mitigation on squirrel glider movement. The trapping and radio-tracking methods for 64!
the project are described in detail in Soanes et al. (2013). An adult male squirrel 65!
glider (SQB1M) was captured at the study site at Longwood (Fig. 1) on 17 November 66!
2010, at which time it was fitted with a VHF radio-collar, tattooed and implanted with 67!
a passive integrated transponder (PIT) tag. The site was a linear remnant strip of 68!
mature Eucalyptus woodland ~20 m wide along a secondary road (~10 m wide, <100 69!
vehicles per day) that intersected the Hume Freeway. A 70 m long canopy bridge 70!
connected the woodland habitat on either side of the freeway (described in Soanes et 71!
al. 2013). 72!
Results and Discussion 73!
We radio-tracked SQB1M over 36 nights from 17 November 2010 to 3 May 2011. 74!
Three fixes were obtained, on 23 November, 24 November and 3 December 2010, 75!
after which the signal disappeared. We conducted extensive searches on foot and in 76!
vehicles within a 5 km radius of the study site, however, SQB1M was not detected 77!

!
3!
again. Further mark-recapture surveys at the site in December 2010 (41 trap nights) 78!
and March 2011 (39 trap nights) also failed to detect this animal, suggesting either 79!
that it had left the site or that the radio-collar had stopped working. 80!
In July 2011, we were contacted by an ecologist (D. Engel) who had found a tattooed 81!
and collared squirrel glider carcass on the side of the Hume Freeway near the town of 82!
Goulburn in New South Wales (34°48'57.08"S,!149°26'19.97"E, accuracy: +/- 13 m). 83!
The carcass was described as ‘fresh’ and in good condition with no signs of decay or 84!
scavenging (eyes still present); consistent with an animal being killed by a vehicle the 85!
previous night. The carcass was stored in a freezer until we were able to collect and 86!
inspect it in July 2012. The radio-collar number, tattoo and PIT tag confirmed that 87!
this individual was SQB1M. We could not determine whether the collar had 88!
malfunctioned or if the battery had gone flat, as the collection time was beyond the 89!
expected battery life (>18 months). 90!
The collection site was approximately 500 km away from the site in Victoria where 91!
the glider was originally marked. The roadside vegetation at the collection site 92!
consisted of shrubs 2–4 m high with no tall trees. Remote inspection using Google 93!
Earth and Google Street View shows that the collection site was within a highly 94!
agricultural landscape with only small (< 4 ha) isolated patches of mature woodland 95!
apparent within a 2 km radius. More than six months had passed since the carcass was 96!
found and the radio-collar signal first disappeared from the study area (Fig. 1). 97!
Given the long distance and extreme level of habitat fragmentation between the site of 98!
capture and the site of collection, it is very unlikely that the animal travelled from 99!
Longwood to Goulburn unassisted. The journey would have required crossing many 100!
large treeless gaps, urban areas and cleared agricultural land. Little is known about the 101!
dispersal distances of squirrel gliders, but given the species body size and typical 102!
home-range size (3–6 ha, vand der Ree and Bennett 2003; Sharpe and Goldingay 103!
2007 ) it is likely to be far less than 500 km (Bowman et al. 2002). 104!
We believe that the animal was killed by a vehicle while crossing the Hume Freeway, 105!
became lodged on the vehicle and was then carried along the freeway. We cannot 106!
determine where the animal was killed, except to say that it is extremely unlikely that 107!
it was killed at the site it was found. Interestingly, it does not appear that SQB1M 108!
used the canopy bridge that was present at the Longwood site to cross the freeway. 109!
The canopy bridge was monitored with a PIT tag reader for 46 nights from November 110!
2010 to April 2011 (Soanes et al. 2013) and the PIT tag of SQB1M was not detected. 111!
Based on the average tree height of 20 m, if the animal attempted to cross the freeway 112!
at a location where the distance between trees on opposite sides of the road exceeded 113!
40 m, the glide path would have intersected with the path of traffic (Goldingay and 114!
Taylor 2009; Soanes and van der Ree 2015). 115!
Our result illustrates the potential for wildlife carcasses to be transported away from 116!
the site of impact. Although our observation represents only a single, and likely 117!

!
4!
extreme case, the possibility that roadkilled carcasses were killed at locations other 118!
than where they were found should be considered in future surveys of small flying 119!
and gliding species. 120!
Acknowledgements 121!
Thanks to Deryk Engel for going over and above the call of duty by collecting and 122!
storing the squirrel glider carcass that was so critical to this observation. Thanks to 123!
Amy Evans for posting the radio-collar back to us, and to the New South Wales 124!
Animal Ethics Authority for connecting us all together. All animals were trapped and 125!
radio-tracked under The University of Melbourne Animal Ethics Committee Permit 126!
0810924 and the DSE Permit 10004763. We thank the Holsworth Wildlife Research 127!
Endowment, VicRoads and the Road and Maritime Service New South Wales for 128!
their support of this project. 129!
References 130!
131!
Bowman, J., Jaeger, J.A., and Fahrig, L. (2002) Dispersal distance of mammals is 132!
proportional to home range size. Ecology 83(7), 2049–2055. 133!
Forman, R.T.T., Sperling, D., Bissonette, J.A., Clevenger, A.P., Cutshall, C.D., Dale, 134!
V.H., Fahrig, L., France, R., Goldman, C.R., Heanue, K., Jones, J.A., Swanson, F.J., 135!
Turrentine, T., and Winter, T.C. (2003) 'Road ecology: Science and solutions.' (Island 136!
Press: Washington). 137!
Gibbons, P., and Boak, M. (2002) The value of paddock trees for regional 138!
conservation in an agricultural landscape. Ecological Management & Restoration 139!
3(3), 205–210. 140!
Goldingay, R.L., and Taylor, B.D. (2009) Gliding performance and its relevance to 141!
gap crossing by the squirrel glider (Petaurus norfolcensis). Australian Journal of 142!
Zoology 57(2), 99–104. 143!
Gonzalez-Gallina, A., Benitez-Badillo, G., Rojas-Soto, O.R., and Hidalgo-Mihart, 144!
M.G. (2013) The small, the forgotten and the dead: highway impact on vertebrates 145!
and its implications for mitigation strategies. Biodiversity and Conservation 22(2), 146!
325–342. [In English]. 147!
McCall, S.C., McCarthy, M.A., van der Ree, R., Harper, M.J., Cesarini, S., and 148!
Soanes, K. (2010) Evidence that a highway reduces apparent survival rates of squirrel 149 !
gliders. Ecology and Society 15(3). 150!
Santos, S.M., Carvalho, F., and Mira, A. (2011) How long do the dead survive on the 151!
road? Carcass persistence probability and implications for road-kill monitoring 152!
surveys. Plos One 6(9). 153!
Sharpe, D. J., and Goldingay, R.L. (2007) Home range of the Australian squirrel 154!
glider Petaurus norfolcensis (Diprodontia). Journal of Mammalogy 88, 1515–1522. 155!

!
5!
Soanes, K. (2014) Evaluating the success of road-crossing structures for arboreal 156!
mammals: How does monitoring effort influence the detection of population-level 157!
effects? PhD Thesis, School of Botany, The University of Melbourne. 158!
Soanes, K., Lobo, M.C., Vesk, P.A., McCarthy, M.A., Moore, J.L., and van der Ree, 159!
R. (2013) Movement re-established but not restored: Inferring the effectiveness of 160!
road-crossing mitigation for a gliding mammal by monitoring use. Biological 161!
Conservation 159, 434–441. 162!
Soanes, K., and van der Ree, R. (2015) Reducing road impacts on tree-dwelling 163!
animals. In 'Handbook of road ecology.' (Eds. R van der Ree, DJ Smith and C Grilo). 164!
(Wiley Blackwell). 165!
Soanes, K., Vesk, P., and van der Ree, R. (2015) Monitoring the use of road-crossing 166!
structures by arboreal marsupials: insights gained from motion-triggered cameras and 167!
passive integrated transponder (PIT) tags. Wildlife Research 42, 241–256. 168!
Taylor, A.C., Walker, F.M., Goldingay, R.L., Ball, T., and van der Ree, R. (2011) 169!
Degree of landscape fragmentation influences genetic isolation among populations of 170!
a gliding mammal. Plos One 6(10). 171!
Taylor, B.D., and Goldingay, R.L. (2004) Widlife road-kills on three major roads in 172!
north-eastern New South Wales. Wildlife Research 31, 83–91. 173!
Taylor, B.D., and Goldingay, R.L. (2010) Roads and wildlife: impacts, mitigation and 174!
implications for wildlife management in Australia. Wildlife Research 37(4), 320–331. 175!
van der Ree (2006) Road upgrade in Victoria a filter to the movement of the 176!
endangered Squirrel glider (Petaurus norfolcensis): Results of a pilot study. 177!
Ecological Management & Restoration 7(3), 226–228. 178!
van der Ree, R. (2002) The population ecology of the squirrel glider (Petaurus 179!
norfolcensis) within a network of remnant linear habitats. Wildlife Research 29(4), 180!
329–340. 181!
van der Ree, R., Bennett, A.F., and Gilmore, D.C. (2003) Gap-crossing by gliding 182!
marsupials: thresholds for use of isolated woodland patches in an agricultural 183!
landscape. Biological Conservation 115, 241–249. 184!
van der Ree, R., Cesarini, S., Sunnucks, P., Moore, J.L., and Taylor, A. (2010) Large 185!
gaps in canopy reduce road crossing by a gliding mammal. Ecology and Society 186!
15(4). 187!
van der Ree, R., Smith, D.J., and Grilo, C. (2015) 'Handbook of road ecology.' (Wiley 188!
Blackwell: London) 189!
Figures 190!

!
6!
191!
Figure 1. Map showing the capture location and roadkill location of a radio-collared 192 !
squirrel glider along the Hume Freeway in southeast Australia. Dark shading indicates 193!
woodland and lighter areas indicate cleared land. 194!
195!
!196!