1	
 
 
 
 
 
 
 
The Effects of Red Fox (Vulpes vulpes) Predator Scent on Winter 
Burrow Use by Eastern Cottontail Rabbits (Sylvilagus floridanus) 
 
 
 
Honors Thesis 
Presented to the College of Agriculture and Life Sciences,  
Cornell University 
in Partial Fulfillment of the Requirements for the Biological Sciences 
Honors Program 
 
by 
Jeremy Pustilnik 
May 2018 
 
Advisors: 
Paul Curtis, Department of Natural Resources, Cornell University 
Jeremy Searle, Department of Ecology and Evolutionary Biology, 
Cornell University 
 
 
 
 
 
 
 
 
 
 
 
	
	 2	
Abstract 
During the winter months, the Eastern cottontail (Sylvilagus floridanus) must 
simultaneously cope with freezing temperatures that impose physiological stress, while 
remaining vigilant for a variety of predators. Because of this, rabbits frequently use burrows 
around man-made structures to escape the cold and predation. To enhance their detection of 
predators, Eastern cottontails are also sensitive to a variety of olfactory cues. However, despite 
the importance of Eastern cottontails in red fox (Vulpes vulpes) diets, little is known about scent 
cues that rabbits use to detect red foxes. As such, I designed an experiment to determine whether 
winter burrow use by Eastern cottontails is affected by the application of red fox urine near 
burrows. After identifying active rabbit burrows near abandoned buildings in Ithaca, New York 
during winter 2017-2018, camera traps were installed to monitor their use before and after 
setting up scent wicks dipped in red fox urine or a water control near burrow entrances. Although 
no statistically significant effect was observed between burrow use before and after treatment, a 
variety of mammalian and avian species were seen in and around burrows over the course of the 
winter. This indicates that burrows may be important to a range of North American mammals 
and birds, knowledge which may be helpful in creating future conservation management plans 
for these species. This project has also provided continued evidence for the usefulness of camera 
traps in documenting predator-prey behavior and studying wildlife ecology. 
 
 
 
 
 
 
 
 
 
 
	
	 3	
Introduction 
Prey are faced with many constraints as they attempt to maximize survival and 
reproduction. In the temperate northeast of North America, seasonal changes are a large source 
of these constraints, especially those associated with winter. Winter temperatures usually dip 
below freezing in the northeast (Campbell et al. 2005), which imposes significant stress on the 
homeothermic physiological balance of endothermic mammals, and snowfall obscures vegetation 
to herbivores, among a whole host of other challenges associated with the season. To cope with 
these physical stressors, many organisms (Watts and Jonkel 1988, Yacoe 1983, Lyman and 
Chatfield 1955, Fishman and Lyman 1961, Lyman and Blinks 2005, Patil et al. 2013) go into a 
state of dormancy to reduce metabolic energy usage. Hibernating species have become poster 
children for winter, which has led to the season being inaccurately labeled the “’dormant season’, 
based on the belief that biological activity ceases during this period” (Campbell et al. 2005). This 
misconception has resulted in a lack of enthusiasm for studying ecological systems in winter 
except for in the arctic (Campbell et al. 2005), which precisely necessitates more research on 
winter ecology. In reality, many common northeastern prey species do not hibernate or become 
dormant, such as the white-tailed deer (Odocoileus virginianus) (Ozoga and Harger 1966), 
Eastern grey squirrel (Sciurus carolinensis) (Lyman and Blinks 2005, Brown and Yeager 1945), 
and Eastern cottontail (Sylvilagus floridanus), which remain active throughout the winter 
(Chapman and Litvaitis 2003). These species continue foraging or otherwise draw from hoarded 
food caches as they cope with constraints imposed by climate and other stressors. 
One way these species are able to cope with the snow and freezing temperatures is to find 
temporary shelter in spaces that are shielded from the elements. Snow itself has insulating 
properties that allow it to function as a temporary seasonal refuge (Palm and Tveitereid 2012, 
	
	 4	
Colbeck 2012, Pomeroy and Brun 2001, Gouttevin et al. 2012). However, in areas with less 
snowfall or short-lived snow, less ephemeral dwellings are preferred. Favorite choices for many 
northeastern mammals are previously constructed woodchuck (Marmota monax) burrows and 
holes up against the exterior bases of man-made suburban structures. Eastern cottontails are 
among the species that use burrows (Godin 1977, Linduska 1947), though this is largely a 
statement of expert opinion. This has led many biologists to doubt that Eastern cottontails take 
shelter in places apart from small indentations in the ground used for nesting (Beule and 
Studholme 1942, Casteel 1966). One aim of this project is to clarify the contested extent of 
burrow use by Eastern cottontails. 
Because prey species often remain active throughout the winter, many North American 
predators are similarly active such as wolves (Canis lupus) (Johnson et al. 2017), red foxes 
(Vulpes vulpes) (Ables 1969), gray foxes (Urocyon cinereoargenteus) (Harrison 1997), coyotes 
(Canis latrans) (Ozoga and Harger 1966, Gese and Grothe 1995, Neale and Sacks 2003), 
American martens (Martes americana) (Drew and Bissonette 1997), fishers (Pekania pennanti) 
(Leonard 1981), and bobcats (Lynx rufus) (Neale and Sacks 2003). The presence of predators 
presents an added challenge to the survival of prey species in winter, who must now consider 
tradeoffs in foraging due to the constraints of both the physical environment and predation 
pressure. There is considerable interest in understanding the decision-making process of prey 
species as they attempt to maximize foraging and minimize multiple risks to their survival 
(Villen-Perez et al. 2013, Hilton et al. 1999). Prey that may use burrows to escape both predators 
and the weather in winter have been insufficiently studied in this context. Being able to piece 
apart when and why prey use burrows is valuable for understanding their behavioral ecology. 
 One way to solve this problem is to try and understand how prey specifically respond to 
	
	 5	
predators. In the co-evolutionary game between predators and prey, as predators increase their 
efficiency of capturing prey, prey respond by increasing their ability to detect and discriminate 
among predators (Atkins et al. 2016). By detecting cues from predators, prey can modulate their 
behavior to avoid capture. This manifests itself in many prey species through a fear response to 
fight, flee, or freeze in place (Bracha 2004, Apfelbach et al. 2005). Many mammalian fear-
inducing cues are olfactory, which are sensed from predator odors. It has been shown that these 
scent cues can powerfully affect the behavior of prey in a myriad of predator-prey systems 
(Rosen et al. 2015, Lindgren et al. 1995, Apfelbach et al. 2015, Osada et al. 2014, Sullivan 
1986). Widespread predators may induce fear responses in a variety of prey species, which may 
be an indicator of how importantly integrated they have become in an ecosystem, as fear 
responses to novel predators are slow to evolve (Atkins et al. 2016). One such predator is the red 
fox, which is known to induce fear responses in mice (Apodemus sylvaticus, Peromyscus 
leucopus, and P. polionotus), rats (Rattus norvegicus), voles (Clethrionomys glareolus and 
Microtus agrestis), shrews (Sorex spp.), and snowshoe hares (Lepus americanus) (Dickman and 
Doncaster 1984, Fanson 2010, Jedrzejewski et al. 1993, Laska et al. 2005, Lindgren et al. 1995, 
Navarro-Castilla and Barja 2014a, Navarro-Castilla and Barja 2014b, Orrock et al. 2004, 
Sullivan and Crump 1986, Takahashi et al. 2005). It is also known from the literature that “the 
most important prey for red foxes in the eastern US are cottontails (Sylvilagus spp.)” (Frey 
2013). However, despite the importance of cottontails to the diet of red foxes and the prevalence 
of research on how predator scents induce fear responses in prey, not a single study exists that 
demonstrates the olfactory relationship between the red fox and the Eastern cottontail. Coupled 
with the facts that evidence for burrow use by Eastern cottontails remains largely based on expert 
opinion, winter ecology is comparatively understudied in relation to the ecology of other 
	
	 6	
seasons, and that tradeoffs prey make to cope with the multiple constraints of the physical 
environment and predation pressure are of interest, this presented an exceptional opportunity to 
study the predator-prey interaction between the red fox and Eastern cottontail.  
 I designed this project with three main goals. The first was to document animal 
behavioral ecology at winter burrows in the suburban American northeast. This included 
characterizing the diversity of species that occur in and around burrows and their behavior in and 
around the burrows. This also included describing the physical characteristics of burrows 
themselves. The second goal was to experimentally test the hypothesis that Eastern cottontails 
exhibit a fear response to the scent of red foxes. The third goal was to experimentally determine 
whether this fear response could cause the frequency of burrow use to decrease despite winter 
cold temperatures and snowfall that originally necessitate burrow use. These goals were 
accomplished through an experimental design employing camera traps at burrow entrances 
around old buildings near Ithaca, New York in the winter of 2017-2018. For the observational 
part of this project, camera traps merely observed the “natural” activity at the burrows. For the 
other parts of this project however, a controlled experiment was designed in which red fox urine 
was applied to wicking devices at burrow entrances over a two week period, and the camera 
traps functioned to document changes (or lack thereof) in patterns of animal activity at the 
burrows before and after treatment, and between control and treatment burrows.  
 
Materials and Methods 
Field design 
 Beginning in late November 2017, large animal burrows, likely old woodchuck burrows, 
around old abandoned buildings on Cornell University-owned lands near Ithaca, New York were 
	
	 7	
identified. Camera traps were deployed facing these burrows either screwed into a tree or on a 
wooden stake between 30 and 60 cm off the ground and at a distance of 120 to 180 cm from each 
burrow. At sites where multiple burrows were identified, only burrows that were at least 3 m 
apart were monitored to increase the likelihood of each burrow being independent, rather than 
connected underground to other burrow entrances. The camera traps used were Cuddeback brand 
(De Pere, WI) Model C1 or Model 11339, and were set to take 3 consecutive photo bursts for 
both day and night. A few days after initial set up, cameras were checked to see if any photos 
had been taken of eastern cottontails. If not, the camera was removed and placed at another 
identified burrow. In this way, 21 separate burrows were used over the course of this project, at 6 
different sites (Figure 1). Characteristics of each burrow were also documented, which included 
the greatest width and height (or length) of the entrance holes, the coordinates of each burrow, 
and a site description. All work was approved by Cornell University IACUC protocol number 
2017-0123. 
 
Site descriptions 
 The six sites used were: (1) East Hill barn, (2) Game Farm shack, (3) Turkey Hill beehive 
building, (4) Lydell Lab, (5) Stevenson Road barns, and the (6) Cornell Ecology and 
Evolutionary Biology (EEB) Research Pond facility. 
 (1) East Hill barn – Two cameras were set up at this site. It is characterized by an old 
abandoned farm animal barn that is surrounded by a broken concrete patio. Sparse vegetation, 
weeds, and thorny brambles had sprouted between the cracks of the concrete or around the base 
of the building. The first burrow (cam 40) was identified as a hole in the base of the wall of the 
building, underneath some brambles and a caving in roof. The second burrow (cam 54) was 
	
	 8	
identified under a slab of broken concrete near, but not connected to the base of the building, 
underneath a bush. 
 (2) Game Farm shack – Three cameras were set up at this site. It is characterized by an 
old wooden shack set in an empty field along Game Farm Road. Old slabs of broken concrete 
were strewn about the shack, remnants of a small patio, but grass, weeds, bushes, and brambles 
had also sprung up all around. This made the concrete unnoticeable until closer examination of 
the substrate when I was driving wooden stakes into the ground. The first burrow (cam 42) was 
identified as a hole near the base of the shack surrounded by a few small concrete slab pieces and 
grass. The second burrow (cam 43) was identified as a hole in the side of the wooden shack that 
was less covered by grass and brambles, but was still partially surrounded by concrete slab 
pieces. The third burrow (cam 45) was identified under a slab of broken concrete and dirt farther 
away from the shack, but also covered by grass and tall brambles. 
 (3) Turkey Hill beehive building – Six cameras were set up at this site. It is characterized 
by an old metal building about 100 m west of Turkey Hill Road. Beehives are actively kept along 
one side of this building during the growing season, but were absent during this project. The 
building itself is surrounded by tall, thick grass, shorter grass, small coniferous trees, and old 
metal equipment debris. The first burrow (cam 48) was identified as a hole up against one corner 
of the building, surrounded by thick grass. The second burrow (cam 49) was identified as a hole 
up against another corner of the building. It was surrounded by grass and old debris like plywood 
and a tire, and was partially shaded by a coniferous tree. The third and fourth burrows (cams 50 
and 51) were identified as holes underneath one side of the building. They were more exposed, 
though some small, thin trees and short grass had sprouted nearby. The fifth burrow (cam 55) 
was identified as a hole up against the third corner of the building, covered by old wooden pallets 
	
	 9	
and some short grass. The sixth burrow (cam 58) was identified as a hole underneath another 
side of the building, and was covered by some thick grass and small, thin trees.  
 (4) Lydell Lab – Three cameras were set up at this site. It is characterized by an old 
wooden barn-like building in a field a few meters away from an actively maintained and used 
concrete research building and greenhouse. The first burrow (cam 39) was identified as a hole 
underneath one corner of the wooden building. The burrow was only surrounded by short lawn 
grass and was not covered by overhanging brambles or other forms of cover. The second burrow 
(cam 60) was identified as a hole underneath one side of the wooden building, similarly only 
surrounded by short lawn grass and no overhanging brambles or cover. The third burrow (cam 
59) was identified as a hole underneath another side of the building, but this burrow was heavily 
covered by a dense thicket of brambles that stretched along the entire side of the building. 
 (5) Stevenson Road barns – Four cameras were set up at this site. It is characterized by a 
series of old barns along Stevenson Road that are still somewhat maintained by staff who tend to 
horses that are kept behind the barns. However, many of these buildings are empty, and a nearby 
tall, empty, cylindrical corn silo towers over these buildings. The first burrow (cam 61) was 
identified as a hole underneath the long side of a rectangular building. It was partially but thinly 
covered by small burdock bushes. The second burrow (cam 57) was identified as a hole 
underneath the short side of this building, near the silo, and was only slightly shaded by a few 
thin, dead brambles. The third burrow (cam 52) was identified as a hole underneath the same 
building as the first and second burrows, right near the silo. It was surrounded primarily by rocky 
gravel and a few small trees which had grown up through cracks in the concrete. The fourth 
burrow (cam 56) was identified as a hole underneath the side of another building, and was 
surrounded only by short grass and patches of old hay.  
	
	 10	
 (6) Cornell EEB Research Pond facility – Three cameras were set up at this site. It is 
characterized by a large metal garage-like building and a nearby smaller temporary building at 
the end of a gravel road that are both still maintained by staff. The first burrow (cam 62) was 
identified as a hole in the vinyl siding of the temporary building. There was no immediate cover 
above or around this hole, merely short lawn-type grass. The second burrow (cam 38) was 
identified as a hole underneath the larger building near the pathway between the larger building 
and the temporary building. Some wooden siding on the large building overhung this hole, but 
there were no other forms of cover. The third burrow (cam 53) was identified as a hole 
underneath another side of the large building. It was surrounded, but not covered by some thick 
grass, weeds, and some short bramble bushes.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
	
	 11	
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1. A Google Earth map of Ithaca, NY and surrounding towns with each of the six 
study sites plotted 
 
Design of red fox urine repellent trial 
 During the final two weeks of this 8-week project, from 1 January to 16 January 2018, 
each of the 21 identified burrows was treated either with red fox urine or a water control. This 
was accomplished using Quik-wiks brand scent dispensers (Wildlife Research Center, Inc., 
Ramsey, MN), which are small plastic tubes that hang upside down as a felt tip dipped in liquid 
hangs out from the bottom of the tubes. The wicks were spray-painted a dark brown to make 
	
	 12	
them more inconspicuous from their original neon-orange color. These wicks were hung from 
the loops of green 45 cm metal plant props with the help of thin galvanized steel wire and duct 
tape, because the hooks at the tops of the tubes would not fully close around the plant prop loops. 
The plant props with the hanging wicks were then placed within 30 cm of the burrow entrances 
for each of the 21 identified burrows, and the felt wick tips were dipped in either red fox urine, 
which was purchased in 473 mL bottles from the Trap Shack Company (Arcadia, WI), or a bottle 
of Poland Spring water with no added minerals. A total of 14 burrows received red fox urine 
treatment (cams 38, 43, 45, 49, 50, 51, 52, 54, 56, 57, 58, 59, 60, and 62) and 7 burrows received 
the water control (cams 39, 40, 42, 53, 55, and 61). Each of the 6 sites had at least one burrow 
treated with the control, and at least one burrow treated with urine, to help eliminate site-specific 
interactions. The control and treatment burrows were usually spaced far apart at each site to try 
and eliminate potential wafting of the urine scent to the control burrows. 
 Because temperatures during the experimental period often dipped below freezing, the 
liquid-treated wicks often froze. Therefore, wicks were frequently re-dipped in their respective 
urine or water control treatment every three days (January 1, 4, 7, 10, and 13).  
 Snow also piled up on the burrows and in front of the cameras during this 2-week period. 
Therefore on the days that the wicks were being treated, snow was brushed off in the path in 
front of the camera, and off the camera itself, so as not to obscure each camera’s ability to sense 
movement and take photos. However, brushing off the snow was not perfect and there may have 
been days when the snow buried the camera and burrow, even after snow had just been cleared. 
This severe weather was one major limitation of the field component of this experiment. 
Cameras were checked once per week for battery levels and to replace SD cards. 
 
	
	 13	
Data Selection Criteria 
 After the field component of this project concluded, all photos were collected into a 
single database and examined. Of the 6,466 total photos, those not containing animals were 
deleted. For photos that did contain animals, it was possible that the camera traps had taken 
multiple pictures of the same animal(s) within a short period of time. I therefore decided to 
document photos by event, rather than just document every single photo collected, which would 
have led to overrepresentation of the amount of activity at some burrows, especially during the 
day. Due to apparent constraints with warming up the bulb for camera flash, more photos were 
taken per burst during the day than at night. This delay in flash for evening pictures resulted in 
approximately one-minute delays between many of the night photos. Because a photo burst could 
have started in the last few seconds of 10:00 am and continued through 10:01 am for example, I 
therefore decided that each event would span a period of two minutes, and any animals 
documented within the same two-minute period would only be counted once. For example, if a 
series of photo bursts contained pictures from 10:02 through 10:09, I only counted 4 events 
(10:02-10:03, 10:04-10:05, 10:06-10:07, and 10:08-10:09). This eliminated overrepresentation of 
daytime photos which sometimes had upwards of 5 to 10 photos of the same animal within a 
one-minute interval, activity that cameras would not have been able to capture at night. This 
ultimately led to a total of 1,734 usable photo events. 
 Because cameras were initially set up at burrows on different dates all throughout 
November and December 2017, it was not possible to compare activity at each burrow from day 
to day. Instead, burrow activity was averaged over periods of one week. For ease of analysis, all 
data were eliminated before the start of the first consecutive week on 20 November 2017 at 
17:01 and after the last consecutive week on 15 January 2018 at 17:00 so that individual weeks 
	
	 14	
could be compared to one another. This resulted in 8 total weeks that were used for this project 
(Table 1). 
 
Table 1. Start and end dates and times for each week during the study from 20 November 
2017 through 15 January 2018. 
 
Week Start Date  End Date 
1 20 November 2017 at 17:01 27 November 2017 at 17:00 
2 27 November 2017 at 17:01 4 December 2017 at 17:00 
3 4 December 2017 at 17:01 11 December 2017 at 17:00 
4 11 December 2017 at 17:01 18 December 2017 at 17:00 
5 18 December 2017 at 17:01 25 December 2017 at 17:00 
6 25 December 2017 at 17:01 1 January 2018 at 17:00 
7 (treatment) 1 January 2018 at 17:01 8 January 2018 at 17:00 
8 (treatment) 8 January 2018 at 17:01 15 January 2018 at 17:00 
 
All cameras were assigned to a starting week that corresponded to the first day of a 
consecutive week, and any previous days of data were eliminated for that camera. All cameras 
had at least 4 total weeks of data, including the 2 treatment weeks, and at least 2 weeks before 
treatment. The cameras were assigned to starting weeks based on their initial deployment date 
(Table 2). 
 
Table 2. Week of deployment for each camera used during the study from 20 November 
2017 through 15 January 2018. 
 
Starting week Cameras 
1 40, 42, 43, 45, 49, 51 
2 48, 50, 55 
3 39, 54, 58, 59, 60 
4 38, 53, 62 
5 52, 56, 57, 61 
 
Cameras that started in week 1 therefore had the most weeks of data averaged, and 
cameras that started in week 5 had the fewest. These limitations were due to time constraints on 
placing cameras, and confirming whether a burrow had Eastern cottontail activity.  
	
	 15	
Weather data for each day was compiled from the Northeast Regional Climate Center’s 
web page for Ithaca, New York using their Game Farm Road Weather Station 
(http://www.nrcc.cornell.edu/wxstation/ithaca/ithaca.html). This included maximum temperature 
in degrees Fahrenheit, minimum temperature in degrees Fahrenheit, average temperature in 
degrees Fahrenheit, snowfall in inches, and snow depth in inches. Fahrenheit and inches were 
converted to Celsius and centimeters, respectively. Snowfall and snow depth that were marked as 
“Trace” were assumed to be 0 cm. Unlike the camera trap weekly dates I assigned to each of the 
8 weeks of the project (which ended at 17:00 and began at 17:01), weather data were only 
available per day. Therefore, I assigned weather to another set of weekly dates, for which one 
day of the week did not completely overlap with the assigned camera trap weeks (Table 3).  
 
Table 3. Weeks to which weather and climate data were assigned during 21 November 2017 
to 15 January 2018 
 
Weather week Dates 
1 21 November 2017 to 27 November 2017 
2 28 November 2017 to 4 December 2017 
3 5 December 2017 to 11 December 2017 
4 12 December 2017 to 18 December 2017 
5 19 December 2017 to 25 December 2017 
6 26 December 2017 to 1 January 2018 
7 2 January 2018 to 8 January 2018 
8 9 January 2018 to 15 January 2018 
 
Statistical Analysis 
All statistical analyses were performed in program R (R Core Team 2017). 
 
Burrow characteristics 
 The burrow dimensions for all 21 burrows were recorded (Table 4), which included the 
burrow hole size (calculated by multiplying the greatest burrow width by the greatest burrow 
	
	 16	
height, or length). A plot was then created with all the individual burrow sizes, along with the 
mean burrow size (Figure 2). 
 
Red fox urine repellent trial 
 A series of Welch Two Sample t-tests were performed to compare the means in pre-
treatment and post-treatment weekly burrow visitation for the 5 most abundant mammals found 
at the 21 camera traps, for both control and treatment burrows. Burrow visitation was defined as 
the sum total number of photo events from all the cameras per week. Weeks with high burrow 
visitation had more photo events than weeks with low burrow visitation. The five most abundant 
species included the Eastern cottontail (Sylvilagus floridanus), mouse (Peromyscus spp.), striped 
skunk (Mephitis mephitis), house cat (Felis sylvestris catus), and Virginia opossum (Didelphis 
virginiana). T-tests were performed that included all 21 burrows, or excluded burrows for 
species (other than Eastern cottontail, which was observed at every burrow at least once during 
both pre-treatment and post-treatment weeks) that were neither observed in the pre-treatment nor 
the post-treatment weeks.  
 
Analysis of variance for significant results 
Because t-tests revealed statistically significant differences in mean weekly burrow 
visitation between pre- and post-treatment weeks for 1 of the 5 species, analysis of variance 
(ANOVA) was conducted in R comparing the species’ weekly burrow visitation against urine 
treatment, maximum weekly temperature, minimum weekly temperature, average weekly 
temperature, average weekly snowfall, total weekly snowfall, or average weekly snow depth.  
 
	
	 17	
Climate variable graphs 
 Plots (Figures 3 and 4) were created for both Eastern cottontails and striped skunks that 
compared weekly burrow visitation against maximum weekly temperature, minimum weekly 
temperature, average weekly temperature, average weekly snowfall, total weekly snowfall, and 
average weekly snow depth based on the information contained in Tables 8 and 9. 
 
Results 
Observed species 
Table 4. All species encountered at burrows from 17 November 2017 to 16 January 2018 
near Ithaca, New York, and the number of photo events documented per species. Mice, 
weasel, shrew, and North American sparrow species were only identifiable to the genus 
level or higher. 
 
Common name Scientific name Number of events  
Eastern cottontail Sylvilagus floridanus 789 
Mouse Peromyscus spp. 317 
Striped skunk Mephitis mephitis 181 
House cat Felis silvestris catus 125 
Virginia opossum Didelphis virginiana 56 
Eastern gray squirrel Sciurus carolinensis 28 
American red squirrel Tamiasciurus hudsonicus 13 
Unknown rodent Various 10 
Weasel Mustela spp. 5 
White-tailed deer Odocoileus virginianus 3 
Eastern chipmunk Tamias striatus 3 
Shrew Blarina spp. 3 
Raccoon Procyon lotor 2 
Bobcat Lynx rufus 2 
Mink Neovison vison 2 
Coyote Canis latrans 1 
Gray fox Urocyon cinereoargenteus 1 
Sparrow Various 12 
Unknown bird Various 12 
Blue Jay Cyanocitta cristata 11 
Ring-necked Pheasant Phasianus colchicus 8 
Dark-eyed Junco Junco hyemalis 7 
American Crow Corvus brachyrhynchos 2 
Northern Cardinal Cardinalis cardinalis 1 
	
	 18	
Over the course of this project, at least 22 mammal and bird species visited the burrows 
(Table 4). This list includes one species that was observed on 16 January 2018, the American 
Crow, despite the fact that I omitted photos from that date for quantitative analysis. However, I 
still believe it is relevant to mention all species observed during the winter season as part of the 
qualitative analysis. 
Of the 1,752 total animals identified from the 1,734 total individual photo events (since 
some photo events contained multiple individuals) from 17 November 2017 to 16 January 2018, 
158 contained pictures of unknown animals (not including unknown rodents and unknown birds) 
that I was unable to classify. This makes my identification rate approximately 91%.  
 
Burrow characteristics 
Most of the 21 evaluated burrows were relatively rectangular so I multiplied the widest 
width by the highest height, or longest length, to get the total opening area (Table 5). The 
average burrow size is 440.5 cm2. Burrow size ranges from 156 cm2 to 1,148 cm2 (Figure 2).  
 
 
 
 
 
 
 
 
 
	
	 19	
Table 5. Dimensions for each of the 21 burrows monitored during 17 November 2017 to 18 
January 2018 near Ithaca, New York.  
 
Cam Location Burrow Burrow Burrow hole size 
width (cm) height (cm) (W x H) (cm2) 
40 East Hill barn 24 14.5 348 
54 East Hill barn 23 13 299 
42 Game Farm shack 28 41 1148 
43 Game Farm shack 9 22 198 
45 Game Farm shack 30 25 750 
48 Turkey Hill beehive building 17 21 357 
49 Turkey Hill beehive building 22 12.5 275 
50 Turkey Hill beehive building 39 10.5 410 
51 Turkey Hill beehive building 30 9 270 
55 Turkey Hill beehive building 21 20 420 
58 Turkey Hill beehive building 28 10 280 
39 Lydell Lab 20 12 240 
59 Lydell Lab 25.5 18 459 
60 Lydell Lab 54 12 648 
52 Stevenson road barns 35 10 350 
56 Stevenson road barns 11.5 30 345 
57 Stevenson road barns 40 17 680 
61 Stevenson road barns 51 13 663 
62 Cornell EEB Research Pond facility 12.5 12.5 156 
38 Cornell EEB Research Pond facility 30 11 330 
53 Cornell EEB Research Pond facility 50 12.5 625 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
	
	 20	
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2. Plot of the calculated entrance hole sizes for the 21 evaluated burrows. The mean 
burrow size of 440.5 cm2 is also marked as a red line. 
 
Red fox urine repellent trial 
Weekly burrow visitation (total number of photo events per week) was determined for 
each of the five most abundant species (Table 6). The eight weeks of the study were then split 
into “pre-treatment” and “post-treatment” groups upon which t-tests could be performed to 
determine statistically significant differences in weekly burrow visitation. Two t-tests were 
conducted for the Eastern cottontail because it was observed at all 21 burrows, while 4 t-tests 
were conducted for each of the other four most abundant mammal species observed (Table 7). 
The only statistically significant (p < 0.05) t-tests were ones including striped skunks observed at 
treatment burrows. This gives strong support that burrow visitation by skunks was different 
during pre- and post-treatment weeks. This warranted conducting analysis of variance to 
determine likely causes of this difference. 
	
	 21	
Choosing to exclude burrows that were neither visited by species during the pre-treatment 
nor post-treatment weeks did not make a difference in terms of finding a statistically significant 
result, but the specific t-test values were different than those that did not exclude any burrows. 
 
Table 6. The average weekly burrow visits for the five most abundant mammal species. 
Week Cottontail Visits Skunk Visits Opossum Visits Mouse Visits Cat Visits 
1 5.0 4.3 0.7 3.3 0.7 
2 5.2 2.8 1.3 6.1 0.8 
3 9.5 0.9 0.7 4.6 0.8 
4 6.6 0.1 0.2 3.7 0.1 
5 3.6 2.4 0.8 1.6 1.3 
6 4.2 0.6 0 0.7 1.3 
7 3.0 0.1 0 1.4 0.4 
8 5.3 1.1 0.3 1.3 1.6 
 
 
Table 7. Results of the Welch Two Sample t-tests performed in R for the five most 
abundant mammal species at burrows. T-tests were performed for both control and 
treatment groups, and using all burrows or only the subset of burrows that had at least one 
photo event instance either during pre-treatment or post-treatment weeks. Eastern 
cottontails were observed at all 21 cameras, so only 2 t-tests were needed. Statistically 
significant results are highlighted in bold. 
 
Species T-test Result 
Eastern cottontail Control: all burrows t = -0.40166, df = 11.186, p = 0.6955 
Treatment: all burrows t = 1.4914, df = 19.529, p = 0.1518 
Striped skunk Control: all burrows t = 0.46038, df = 11.793, p = 0.6536 
Control: 1 burrow excluded t = 0.49331, df = 9.8875, p = 0.6326 
Treatment: all burrows t = 2.2801, df = 16.978, p = 0.0358 
Treatment: 3 burrows excluded t = 2.5635, df = 13.525, p = 0.023 
Virginia opossum Control: all burrows t = 1.342, df = 6.2707, p = 0.2261 
Control: 3 burrows excluded t = 1.4799, df = 3.1654, p = 0.2309 
Treatment: all burrows t = 0.29401, df = 20.666, p = 0.7717 
Treatment: 8 burrows excluded t = 0.35773, df = 6.705, p = 0.7315 
House cat Control: all burrows t = 0.15692, df = 11.79, p = 0.878 
Control: 4 burrows excluded t = 0.3656, df = 3.7514, p = 0.7343 
Treatment: all burrows t = -0.74125, df = 23.408, p = 0.4659 
Treatment: 7 burrows excluded t = -0.99606, df = 11.229, p = 0.3402 
Mouse Control: all burrows t = 0.42988, df = 11.738, p = 0.6751 
Control: 3 burrows excluded t = 0.46245, df = 5.978, p = 0.6601 
Treatment: all burrows t = 0.91024, df = 19.876, p = 0.3736 
Treatment: 7 burrows excluded t = 0.99994, df = 9.4823, p = 0.3422 
	
	 22	
Analysis of variance for significant results 
 The results from Table 5 indicated that there was a statistically significant difference in 
pre-treatment and post-treatment weekly burrow visitation by striped skunks, so analysis of 
variance (ANOVA) was conducted to determine likely causes of this difference. Results from 
ANOVA (Table 8) revealed that weekly skunk visitation to burrows was likely explained by 
minimum weekly temperature and average weekly temperature, because p-values for the two 
predictors were statistically significant (p < 0.05). As temperature decreased, less activity was 
observed around burrow entrances. There was also some evidence that maximum weekly 
temperature influences skunk burrow use, since the p-value from the relevant ANOVA test was 
marginally close to being statistically significant (p = 0.0714). In fact, all other predictors 
including average weekly snowfall, total weekly snowfall, and average weekly snow depth were 
closer to being significant in their ANOVA tests than urine treatment, based on their p-values. 
This provides support for weather and climate, especially temperature, as more important factors 
in determining skunk use of burrows than the presence of red fox urine.  
 
Table 8. Results of Analysis of Variance performed in R for striped skunks weekly burrow 
visitation. Statistically significant results are highlighted in bold. 
 
Model Result 
Skunk visits: Treatment df = 1, F = 1.032, p = 0.349 
Skunk visits: Max Weekly Temperature df = 1, F = 4.78, p = 0.0714 
Skunk visits: Min Weekly Temperature df = 1, F = 11.13, p = 0.0157 
Skunk visits: Average Weekly Temperature df = 1, F = 10.31, p = 0.0183 
Skunk visits: Average Weekly Snowfall df = 1, F = 3.108, p = 0.128 
Skunk visits: Total Weekly Snowfall df = 1, F = 2.83, p = 0.143 
Skunk visits: Average Weekly Snow Depth df = 1, F = 3.873, p = 0.0966 
 
Climate variable graphs 
 Weekly climate variables were obtained from the Game Farm Road weather station 
(Table 9). Little to no apparent correlation can be seen in graphs of Eastern cottontail weekly 
	
	 23	
burrow visitation plotted against other weather and climate variables (Figure 3). This lack of 
observed trends is consistent with the results obtained from the t-tests for Eastern cottontails, in 
which there were no statistically significant differences in weekly burrow visitation during pre-
treatment and post-treatment weeks. Though weekly burrow visitation appears to follow weekly 
snowfall variables as in Figure 3d and Figure 3e, these correlations are likely visually 
misleading. However, visually apparent correlations in Figure 4 for striped skunk weekly burrow 
visitation plotted against weather and climate variables, especially minimum weekly temperature 
(Figure 4b) and average weekly temperature (Figure 4c), may represent real correlations, based 
on the previous t-tests and ANOVAs.  
 
Table 9. Weekly weather and climate variables 
Maximum Minimum Average Average Total Average Snow 
Week Temp (˚C) Temp (˚C) Temp (˚C) Snowfall (cm) Snowfall (cm) Depth (cm) 
1 13.9 -3.3 2.8 0.0 0.0 0.0 
2 12.2 -6.1 1.8 0.0 0.0 0.0 
3 10.6 -7.8 -0.7 0.2 1.0 0.0 
4 1.1 -18.9 -6.7 1.6 10.9 4.4 
5 10.0 -5.0 0.7 0.6 4.1 0.4 
6 -2.2 -19.4 -12.9 1.5 10.7 4.7 
7 -4.4 -19.4 -14.2 0.8 4.6 9.8 
8 16.7 -22.8 -3.6 2.5 17.5 10.5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
	
	 a 24	b 
 
 
 
 
 
 
 
 
 
 
 
 
 c d 
 
 
 
 
 
 
 
 
 
 
 
 
 e 
 f 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 3. (a) Eastern cottontail visits and maximum temperature across weeks. (b) Eastern 
cottontail visits and minimum temperature across weeks. (c) Eastern cottontail visits and 
average temperature across weeks. (d) Eastern cottontail visits and average weekly 
snowfall across weeks. (e) Eastern cottontail visits and total weekly snowfall across weeks. 
(f) Eastern cottontail visits and average weekly snow depth across weeks. 
 
 
	
	 25	
a b 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 d 
 
 
 
 
 
 
 
 
 
 
 
e   f 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 4. (a) Skunk visits and maximum temperature across weeks. (b) Skunk visits and 
minimum temperature across weeks. (c) Skunk visits and average temperature across 
weeks. (d) Skunk visits and average weekly snowfall across weeks. (e) Skunk visits and 
total weekly snowfall across weeks. (f) Skunk visits and average weekly snow depth across 
weeks. 
 
	
	 26	
Discussion 
Observed species 
 The sheer diversity of animal species observed at presumptively abandoned woodchuck 
burrows over the course of one winter season indicates how important these burrows are to a 
variety of organisms, especially mammals. At least 16 mammal species were observed, in 
addition to at least 6 avian species (some species could not be identified to the species level). 
This larger mammalian diversity likely reflects how these mammals are confined to the ground, 
whereas many bird species spend much of their time airborne, and roosting in trees. It is also 
unlikely that birds actually enter the burrows and take refuge from the elements – they were 
merely observed around the entrances. 
Although Eastern cottontails were the primary mammals using old woodchuck burrows 
in winter based on the number of events they were observed, a surprising number of photos 
documented activity of striped skunks, Virginia opossums, house cats, and mice in and around 
burrows. Skunk use of burrows, including those dug by woodchucks (Godin 1977), has been 
previously observed in Ithaca, New York dating back to at least 1937 (Hamilton 1937). Skunks 
are also known to enter a state of dormancy in the winter, though rather than true hibernation, the 
extent of their winter inactivity is better defined as a daily shallow torpor (Geiser 2013, Melvin 
and Andrews 2009). Other studies in Minnesota (Sunquist 1974) and Manitoba, Canada (Mutch 
and Aleksiuk 1977) have not found as much skunk activity around burrows, though this might be 
due to the severity of winters in those places, or limitations in study design, which relied on radio 
telemetry and visual observations. As a contrast, the camera traps used in this project have the 
advantage that they can continue logging data when humans are not out with radio telemetry 
receivers in blizzard conditions, or in the middle of the night, a time when many skunks were 
	
	 27	
observed to be active during this project. Radio collars on study animals also limit observations 
only to those animals, whereas camera traps can uniformly capture the activity of many animals, 
including that of non-target individuals. Radio collars themselves may also affect the behavior of 
study organisms (Gibson et al. 2013), and even those that are motion-sensitive are not able to 
document certain animal behaviors (Gillingham and Bunnell 1985). Camera traps are not without 
their limitations, though. The large amounts of data collected by camera traps can often 
overwhelm researchers in their analysis due to problems with image processing, storage, and 
backup of photos (Newey et al. 2015). Technological failures such as batteries dying, motion 
sensors not picking up movement, or software glitches leading to incorrectly recorded dates and 
times may introduce false negatives and false positives (Newey et al. 2015, Caravaggi et al. 
2017, Pirie et al. 2016). Camera flash or presence may also impact behavior (Meek et al. 2015, 
Caravaggi et al. 2017). However, many of these issues pale in comparison to the value of 
obtaining otherwise unavailable data on extremely rare, cryptic, or critically endangered species 
(Rahman et al. 2016, Jackson et al. 2006).  
Virginia opossums are also known to use burrows (Godin 1977), including those of 
mountain beavers (Aplodontia rufa) (Engeman et al. 1991) and armadillos (Lay 1942). I herein 
document them using burrows likely dug by woodchucks, especially around man-made 
structures. Their activity in and around burrows in winter may be explained by the fact that 
opossums spend almost twice as much time foraging and nest building during this season (Godin 
1977, McManus 1969) in order to survive, compared to other times of the year. Mice are also 
known to use burrows, but are less constrained by the size of existing burrows because they are 
smaller mammals than woodchucks. As such, mice can nest in a variety of places, from under 
old stumps to mere holes in walls (Godin 1977). Their activity around the burrows in this study 
	
	 28	
is less surprising. 
It is of particular note that local house and feral cats frequently visited and entered 
burrows during winter. Although cats are not native to North America, since their introduction 
over 175 years ago (George 1974), the cat population in the United States may now number well 
over 80 million, and they are estimated to take avian and mammalian prey in the billions (Loss et 
al. 2013). Cats are also strong competitors with raptors for prey mice, which may implicate them 
in the declines of many raptor species (George 1974). Due to their abundance and ability to 
capture prey, their competition may exacerbate this decline in winter, when prey are much 
scarcer. Cats may also have significant negative impacts on burrowing species, as they have been 
implicated in reducing the nesting success and burrow densities of seabirds on Marion Island in 
South Africa (Dilley et al. 2017). Cats are apparently knowledgeable of the winter burrow habits 
of many mammalian species, as they were observed entering and leaving burrows many times 
over the course of this project that were also being used by other species, likely as they were 
searching for prey. Though none were captured and marked, by using the unique marking pattern 
observed on many of these cats, particularly the orange and grey tabby varieties, it is evident that 
many of these local suburban cats patrol the same areas repeatedly. In one case, a dark gray 
tabby cat was observed stepping in the exact same spot with the exact same foot hours after it 
had patrolled the burrow.  
Knowledge that prey species use burrows may not be unique to house cats because 
bobcats, gray foxes, mink, weasels, and a coyote were also observed approaching and examining 
burrows during this project. This intelligence is likely characteristic of many northeastern 
carnivores, which provides evidence that they have search images for many different places to 
look for prey. Although bobcats, coyotes, and other larger predators could only examine burrow 
	
	 29	
entrances, mink and weasels were observed exiting burrows, species which are known to enter 
the burrows of many of their prey species (Vaughan 1961, Zielinski 2000). Mink and other 
mustelids may even have knowledge of the circadian rhythms of their prey, as their periods of 
activity coincide with those of their prey (Zielinski 2000, Zielinski 1986, Gerell 1969). The 
finding of the bobcat at a burrow by the East Hill barn may also show that bobcats may live 
closer to suburban areas, and be more integrated in suburban ecosystems, than originally 
believed. This finding may have positive implications for the ability of bobcats to adapt to urban 
sprawl, but may also imply the possibility of human-wildlife conflicts or bobcat-house cat 
interactions.  
Another important finding from this project is the fact that many different species used 
the same burrows within short periods of time. For example, at camera 40 between November 20 
and November 28, I documented use of a single burrow by cottontails, mice, a skunk, an 
opossum, cats, and chipmunks. This has two important implications. First, this implies that 
burrow use is temporary. This means that there may be unique cues that cause specific animals to 
seek out or leave burrows, and that some cues may be different for different species. The usual 
inference that temperature and weather variables cause animals to take shelter may not be the 
biggest factor at play. These factors may aid in building better models of when certain species 
are likely to use burrows. The second implication is that multiple animals may be using the same 
burrow at the same time, which means there may be an increased potential for the coexistence of 
multiple species in a small space. For example, at camera 51 on November 23, 2017, I 
documented a skunk exiting a burrow only 14 minutes after a cottontail had left it (Figure 5). 
Though it is possible the camera failed to take a picture of the skunk or cottontail initially 
entering the burrow, the fact that the photos were taken greater than 1 minute apart rules out the 
	
	 30	
possibility that the camera flash was still warming up for a second picture after it took the first. 
The two species were therefore likely to both be in the burrow at the same time. This may imply 
that burrows are communal havens for many species, perhaps in the same way as communal 
watering holes in Africa may function for the various predators and prey of the Serengeti. 
Alternatively, the presence of multiple chambers within burrows (Grizzell 1955) may allow for 
burrow use by multiple species with little to no interaction between them. 
 It is apparent that burrows are important to the diversity of species observed using them. 
Because most of these burrows were likely dug out by woodchucks, the value of woodchucks as 
suburban ecosystem engineers is likely immense. Although woodchucks may often be seen as 
pests to humans (indeed one monitored burrow in this study had been dug next to a pipe that fed 
water into a horse trough, which allowed cold air in, and subsequently caused the water to freeze, 
restricting water to the horses), they may actually be helping facilitate the coexistence of many 
mammalian species in suburban areas. Additionally, because a majority of the monitored 
burrows in this project were underneath or around older man-made structures often in disrepair, 
humans may indirectly be helping animals to survive the winter. In this way, humans might even 
be considered useful ecosystem engineers. However, humans undoubtedly have also contributed 
many problems, especially as it concerns the introduction of non-native species, as was the case 
with the cats observed at several sites. Also, a non-native Ring-necked Pheasant was seen at 
cameras 49, 50, and 51 near the Turkey Hill beehive building, which likely escaped from the 
nearby New York State Department of Environmental Conservation pheasant farm operation. 
 
 
 
	
	 31	
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 5. A skunk was seen leaving a burrow only 14 minutes after an Eastern cottontail 
had exited in photos taken by camera 51 set up at the Turkey Hill beehive building on 23 
November 2017. 
 
Burrow characteristics 
 The size of the burrow is the main limitation of the species that are able to use the 
burrow. However, Eastern cottontails occupied burrows from as small as 156 cm2 to as large as 
	
	 32	
1,148 cm2. This likely reflects the Eastern cottontail’s adaptability as a species to occupying 
unused spaces, rather than requiring spaces of specific dimensions. The 156 cm2 burrow was 
used exclusively by Eastern cottontails, while the burrow that was 1,148 cm2 was used by 
Eastern cottontails, skunks, and mice. It is interesting to note that the largest burrow was not 
used the most by different species. This provides evidence that just because a burrow is large 
enough for a specific species, other characteristics of the burrow may not make it optimal. Other 
such factors might include the amount of cover above and around the burrow, the availability of 
foraging sites nearby, the presence of competitor species already occupying the burrow, or the 
thermoregulatory and insulating properties of the burrow, among others. 
 
Effect of red fox urine on burrow activity 
 The results of t-tests do not give support for an effect of red fox urine on the change in 
Eastern cottontail burrow use in winter. However, because of the importance of Eastern 
cottontails to the red fox diet, and predator scent cues to many mammalian prey species, it is 
unlikely that the possibility of an olfactory relationship between red foxes and Eastern cottontails 
can be ruled out. One reason is because Eastern cottontails may have habituated to the scent of 
red fox urine by their burrows, even within 1 day of treatment (Figure 6).  
Another reason may be that urine may not modulate behavior around previously 
colonized shelters. The scent of a predator may be information an Eastern cottontail uses to avoid 
following a novel path in a forest, but a cottontail may be comfortable with scents around 
familiar areas where it may know of a variety of holes into which it can quickly escape from a 
fox. This might also suggest that the physical characteristics of winter are more of an immediate 
threat to Eastern cottontails than the rare or even occasional encounter with a fox. Cold 
	
	 33	
temperatures can persist from hours to weeks, while an encounter with a fox may last from 
seconds to minutes. Therefore it would be more advantageous to seek shelter from an almost 
certain stressor (e.g., low winter temperatures) than one that has a lower probability of occurring 
(e.g., an encounter with a fox). As a result, olfactory cues from nearby predators may not be paid 
much attention in the winter by Eastern cottontails. The cost of moving to a new burrow and 
risking exposure to the elements or predation may be too high in comparison to any 
inconvenience from a predator’s smell at the current burrow. The lack of a statistically 
significant difference between burrow use before and after urine treatment in this experiment 
provides strong support for this conclusion. 
Lastly, although there was no observed effect of urine treatment on cottontail burrow use, 
there may have been short-term changes in behavior. For example, Eastern cottontails may have 
frozen in place, run away, or had some other alteration in behavior on the first day of treatment, 
when the urine scent was novel. However, this behavior may not have been observed for at least 
three reasons. The first is that 1 January 2018, the date the urine treatment first went out, was a 
particularly cold day, with an average temperature of -16.3 ˚C. This may have possibly 
confounded observing any change in behavior, as little activity in general was observed that day 
(9 total photo events across all burrows, only 5 of which were cottontails). Eastern cottontails 
may have hunkered down in their burrows for the day. The second reason is that the settings of 
the camera traps used in this experiment were not able to capture behavior on short time scales. 
By the time all the urine treatments had been set out that day at 17:00, it was beginning to get 
dark. This meant that camera traps were likely switching to their evening functions that used 
flash. And as discussed earlier, there were often delays greater than 1 minute between photos 
taken at night. To capture and accurately document behaviors that may have lasted only a few 
	
	 34	
seconds or more, camera traps would better have been set on video-taking mode rather than 3-
picture bursts. Regardless, camera traps were still effective at capturing a range of behaviors for 
many species. Thirdly, Eastern cottontails may actually have sensed and responded to the urine 
and stayed in or away from burrows, which would explain the lack of observed activity. 
The small sample sizes in this project and high variability in burrow visits among 
burrows likely contributed to the lack of statistical significance found at urine treatment burrows 
between pre-treatment and post-treatment weeks. However, there was a reduction in Eastern 
cottontail burrow visitation between pre- and post-treatment weeks at urine-treated burrows, 
from a mean of 6.2 to 3.3 weekly visits. There was also less of a difference between the 
visitation means at control burrows (4.7 in pre-treatment weeks and 5.9 visits in post-treatment 
weeks). Taken together, this suggests that a statistically significant pattern might have been 
observed if there had been a larger sample size. Additionally, because the start of some weeks 
did not coincide with the deployment date of camera traps towards the beginning of the project, 
the initial effect of having camera traps set up at burrows was not adequately taken into account. 
However, if Eastern cottontails are able to rapidly habituate to the urine of their predators, they 
may show a similar trend in habituating to foreign objects like camera traps. Indeed if this was 
not the case, it is unlikely that 789 photo instances would have been observed. 
Throughout the entire experiment, no red foxes were observed at burrows, though a gray 
fox was, along with other mammalian carnivores known to use scent marking (e.g., bobcats and 
a coyote). Scent marking by carnivores is an incredibly important behavior; indeed, it is 
speculated that urination alone serves at least 11 distinct functions in mammals including 
territory marking, dominance displays, trail following, and synchronizing the reproductive 
physiology of females (Henry 1977). Therefore, there are a few plausible reasons why no red 
	
	 35	
foxes were observed at burrows, apart from red foxes not responding to scent marking cues, 
which is unlikely (but see Banks et al. 2016). First, there may have been no red foxes denning 
near to, or patrolling the vicinity of the 6 study sites. Indeed the Cornell EEB Research Pond 
facility is gated off with a tall, barbed wire metal fence, so red foxes may not have been 
physically able to enter the facility grounds. Second, red foxes might also not prefer to be around 
more suburban areas such as the 6 sites in this project. Third, the red fox urine treatment may 
actually have been an effective cue which caused red foxes to stay clear of treated burrows to 
avoid intraspecific competitive fights in an already challenging winter environment. However, 
previous research has demonstrated that red foxes are actually attracted to conspecific scent cues, 
rather than avoiding them, which makes this absence of red foxes in this project even more 
perplexing (Arnold et al. 2011, but see Banks et al. 2016). Therefore, more research is needed to 
document the effects of red fox urine on red fox behavior. 
Though there was a statistically significant difference in burrow visitation by striped 
skunks at red fox urine-treated burrows, this is more likely due to the effects of Northeastern 
winter temperatures (specifically, minimum and average weekly temperatures) than urine. Figure 
4b and Figure 4c help reveal how closely skunk appearance at burrows is correlated with 
temperature. Skunks do hibernate to some degree, and this correlation likely reflects how they try 
to conserve energy in particularly cold weeks. Additionally, small sample sizes, especially for 
the water control treatment (only 7 burrows), may be contributing to false statistically significant 
t-test results, or a lack thereof.  
 
 
 
 
 
	
	 36	
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 6. An Eastern cottontail was observed at a burrow being treated with red fox urine 
only one day after the treatment was set up on 1 January 2018. This image was taken by 
camera 38 at the Cornell EEB Research Pond facility on 2 January 2018. The urine-treated 
scent wick can be seen in the background. 
 
Other behavioral observations 
It is generally believed that Eastern cottontails forage on woody plants in winter, girdling 
tree bark and causing damage to both wild and native plants in suburban regions (Baker et al. 
2015, Dalke and Sime 1941, Kellner and Swihart 2017). However, cottontails were observed 
foraging through snow on more herbaceous vegetation in this project (Figure 7).  
 
 
 
 
 
	
	 37	
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 7. An Eastern cottontail was observed eating herbaceous vegetation in the snow on 
17 December 2017 by camera 53 at the Cornell EEB Research Pond facility.  
 
 
Conclusions 
Though this project did not provide evidence that red fox urine can modulate Eastern 
cottontail burrow use in winter, there were many important implications of conducting this quasi-
study-experiment. This project is a testament to the value of camera traps in wildlife ecology 
research, as they can document a range of animal behavior during times that are challenging for 
humans to record data (e.g., during winter, late at night, and during inclement weather). This 
project was also able to clearly document that Eastern cottontails use burrows, likely dug by 
woodchucks, under and around abandoned man-made structures during the winter months. 
Expert opinion on the burrow habits of cottontails is now a proven fact, at least in New York 
	
	 38	
State. North American rabbit burrow use is also known from Michigan (Linduska 1947), and in 
Europe by European rabbits (Oryctolagus cuniculus) who dig their own burrows (Kolb 1985). 
The winter burrow use by Eastern cottontails documented in this project also has implications for 
the conservation of other North American lagomorphs, notably the New England cottontail 
(Sylvilagus transitionalis), which is listed as a species of special concern in New York State 
(New York Natural Heritage Program 2017). Knowledge from this project may aid in future 
conservation management plans aimed at protecting habitat containing potential burrows that 
New England cottontails may use during the winter months. 
 
Acknowledgements 
 I cannot begin to comprehend what this project would be without the support, 
encouragement, and guidance of my two advisors, Drs. Paul Curtis and Jeremy Searle. There was 
never a day when either of them was not available to talk about any part of the project, from 
experimental design, to comments on the manuscript, to moral support in times of doubt. Jeremy 
Searle’s fierce defense of my study sites and aptitude for compromise allowed all of the burrows 
in this project to be monitored without controversy, and being a part of his lab group and 
mammalogy course have unquestionably been the two greatest events of my undergraduate 
career at Cornell University. Paul Curtis’s immense generosity to lend me 22 of his camera traps, 
despite the fact that I was a total stranger who showed up at his door one day, was only the 
beginning of the incredible amount of respect and reverence I now have for him as both a 
scientist and a person. No obscure shadow or outline of an animal was ever beyond his 
identification abilities and skills as a true naturalist. He made many sacrifices for me over the 
course of this project for which I am truly grateful, from driving me all over town in his personal 
	
	 39	
vehicle, to lending me equipment and his field technician, Michael Ashdown. I am also as much 
indebted to my advisors as I am to Michael Ashdown for many reasons. Along with his similarly 
expert knowledge on animal identification, I can never repay Michael’s willingness to cut and 
hammer stakes into the ground, spray paint scent wick tubes, drive around town with me, 
supervise my smelly bottles of red fox urine in his office, and help with any and all aspects of the 
field work for this project. I am only one of the many people over the course of his decades-long 
service to Cornell University who has had the privilege to know him. I am also very thankful for 
the research funding support that was provided to me by the Morley family and Morley student 
research fund, and the Cornell University CALS Alumni Association, which allowed me to 
purchase all the supplies I needed for this project, from field vehicle gas, to scent wicks and duct 
tape. I am pleased to finally report the source and product of their investment. Finally, I would 
like to thank all of my mentors and friends that have had to suffer me over the past few years of 
my undergraduate education, both in the classroom and outside, especially my friends who 
allowed me to stink up their house over winter break with fox urine after long days out in the 
field. Thank you.  
 
References 
Ables, E. D. 1969. Activity studies of red foxes in southern Wisconsin. Journal of Wildlife 
Management 33:145–153. 
Apfelbach, R., C. D. Blanchard, R. J. Blanchard, R. A. Hayes, and I. S. McGregor. 2005. The 
effects of predator odors in mammalian prey species: A review of field and laboratory 
studies. Neuroscience & Biobehavioral Reviews 29:1123–1144. 
Apfelbach, R., M. H. Parsons, H. A. Soini, and M. V. Novotny. 2015. Are single odorous 
	
	 40	
components of a predator sufficient to elicit defensive behaviors in prey species? 
Frontiers in Neuroscience 9:1-14. 
Arnold, J., C. D. Soulsbury, and S. Harris. 2011. Spatial and behavioral changes by red foxes 
(Vulpes vulpes) in response to artificial territory intrusion. Canadian Journal of Zoology 
89:808–815. 
Atkins, R., D. T. Blumstein, K. E. Moseby, R. West, M. Hyatt, and M. Letnic. 2016. Deep 
evolutionary experience explains mammalian responses to predators. Behavioral Ecology 
and Sociobiology 70:1755–1763. 
Baker, M. A. A., S. E. Emerson, and J. S. Brown. 2015. Foraging and habitat use of eastern 
cottontails (Sylvilagus floridanus) in an urban landscape. Urban Ecosystems 18:977–987. 
Banks, P. B., A. Daly, and J. P. Bytheway. 2016. Predator odours attract other predators, creating 
an olfactory web of information. Biology Letters 12:20151053. 
Beule, J. D., and A. T. Studholme. 1942. Cottontail rabbit nests and nestlings. Journal of Wildlife 
Management 6:133–140. 
Bracha, H. S. 2004. Freeze, flight, fight, fright, faint: adaptationist perspectives on the acute 
stress response spectrum. CNS Spectrums 9:679–685. 
Brown, L. G., and L. E. Yeager. 1945. Fox Squirrels and Gray Squirrels in Illinois. Illinois 
Natural History Survey Bulletin 23:449-536 
Campbell J. L., Mitchell M. J., P. M. Groffmam, L. M. Christenson, and J. P. Hardy. 2005. 
Winter in northeastern North America: a critical period for ecological processes. 
Frontiers in Ecology and the Environment 3:314-322. 
	
	 41	
Caravaggi A., P. B. Banks, A. C. Burton, C. M. V. Finlay, P. M. Haswell, M. W. Hayward, M. J. 
Rowcliffe, and M. D. Wood. 2017. A review of camera trapping for conservation 
behaviour research. Remote Sensing in Ecology and Conservation 3:109–122. 
Casteel, D. A. 1966. Nest building, parturition, and copulation in the cottontail rabbit. American 
Midland Naturalist 75:160–167. 
Chapman, J., and J. Litvaitis. 2003. Eastern Cottontail. Pages 101-125 Wild Mammals of North 
America: Biology, Management, and Conservation. Second edition. 
Colbeck S. C. 1983. Theory of metamorphism of dry snow. Journal of Geophysical Research: 
Oceans 88:5475–5482. 
Dalke, P. D., and P. R. Sime. 1941. Food habits of the eastern and New England cottontails. 
Journal of Wildlife Management 5:216–228. 
Dickman, C. R., and C. P. Doncaster. 1984. Responses of small mammals to red fox (Vulpes 
vulpes) odour. Journal of Zoology 204:521–531. 
Dilley, B., M. Schramm, and P. Ryan. 2017. Modest increases in densities of burrow-nesting 
petrels following the removal of cats (Felis catus) from Marion Island. Polar Biology 
40:625–637. 
Drew, G. S., and J. A. Bissonette. 1997. Winter activity patterns of American martens (Martes 
americana): rejection of the hypothesis of thermal-cost minimization. Canadian Journal 
of Zoology 75:812–816. 
Engeman, R. M., D. L. Campbell, and J. Evans. 1991. An evaluation of 2 activity indicators for 
use in mountain beaver burrow systems. Wildlife Society Bulletin 19:413–416. 
Fanson, B. G. 2010. Effect of direct and indirect cues of predation risk on the foraging behavior 
of the white-footed mouse (Peromyscus leucopus). Northeastern Naturalist 17:19–28. 
	
	 42	
Fishman, A. P., and C. P. Lyman. 1961. Hibernation in mammals. Circulation 24:434–445. 
Frey, J. K. 2013. Re-evaluation of the evidence for the importation of red foxes from Europe to 
colonial America: origins of the southeastern red fox (Vulpes vulpes fulva). Biological 
Conservation 158:74–79. 
Geiser, F. 2013. Hibernation. Current Biology 23:R188–R193. 
George, W. G. 1974. Domestic cats as predators and factors in winter shortages of raptor prey. 
The Wilson Bulletin 86:384–396. 
Gerell, R. 1969. Activity patterns of the mink Mustela vison Schreber in southern Sweden. Oikos 
20:451–460. 
Gese, E. M., and S. Grothe. 1995. Analysis of coyote predation on deer and elk during winter in 
Yellowstone National Park, Wyoming. American Midland Naturalist 133:36–43. 
Gibson, D., E. J. Blomberg, G. L. Patricelli, A. H. Krakauer, M. T. Atamian, and J. S. Sedinger. 
2013. Effects of radio collars on survival and lekking behavior of male greater sage-
grouse. The Condor 115:769–776. 
Gillingham, M. P., and F. L. Bunnell. 1985. Reliability of motion-sensitive radio collars for 
estimating activity of black-tailed deer. Journal of Wildlife Management 49:951–958. 
Godin, A. J. 1977. Order Lagomorpha, Eastern Cottontail. Pages 65–82 Wild Mammals of New 
England. Johns Hopkins University Press, Baltimore. 
Gouttevin I., Menegoz M., Dominé F., Krinner G., Koven C., Ciais P., Tarnocai C., and Boike J. 
2012. How the insulating properties of snow affect soil carbon distribution in the 
continental pan-Arctic area. Journal of Geophysical Research: Biogeosciences 117:1-11. 
Grizzell, R. A. 1955. A study of the southern woodchuck, Marmota monax monax. American 
Midland Naturalist 53:257–293. 
	
	 43	
Hamilton W. J. 1937. Winter activity of the skunk. Ecology 18:326–327. 
Harrison, R. L. 1997. A comparison of gray fox ecology between residential and undeveloped 
rural landscapes. Journal of Wildlife Management 61:112–122. 
Henry, J. D. 1977. The use of urine marking in the scavenging behavior of the red fox (Vulpes 
vulpes). Behaviour 61:82–105. 
Hilton, G. M., G. D. Ruxton, and W. Cresswell. 1999. Choice of foraging area with respect to 
predation risk in redshanks: the effects of weather and predator activity. Oikos 87:295–
302. 
Jackson, R. M., J. D. Roe, R. Wangchuk, and D. O. Hunter. 2006. Estimating snow leopard 
population abundance using photography and capture–recapture techniques. Wildlife 
Society Bulletin 34:772–781. 
Jędrzejewski, W., L. Rychlik, and B. Jędrzejewska. 1993. Responses of bank voles to odours of 
seven species of predators: experimental data and their relevance to natural predator-vole 
relationships. Oikos 68:251–257. 
Johnson, I., T. Brinkman, B. Lake, and C. Brown. 2017. Winter hunting behavior and habitat 
selection of wolves in a low-density prey system. Wildlife Biology 2017:wlb.00290. 
Kellner, K. F., and R. K. Swihart. 2017. Herbivory on planted oak seedlings across a habitat edge 
created by timber harvest. Plant Ecology 218:213–223. 
Kolb H. H. 1985. The burrow structure of the European rabbit (Oryctolagus cuniculus L.). 
Journal of Zoology 206:253–262. 
Laska, M., M. Fendt, A. Wieser, T. Endres, L. T. Hernandez Salazar, and R. Apfelbach. 2005. 
Detecting danger—or just another odorant? Olfactory sensitivity for the fox odor 
component 2,4,5-trimethylthiazoline in four species of mammals. Physiology & Behavior 
	
	 44	
84:211–215. 
Lay, D. W. 1942. Ecology of the opossum in eastern Texas. Journal of Mammalogy 23:147–159. 
Leonard, R. D. 1981. The winter activity and movements, winter diet, and breeding biology of 
the fisher (Martes pennanti) in southeastern Manitoba. The University of Manitoba, 
Winnipeg, Manitoba. 
Lindgren, P. M. F., T. P. Sullivan, and D. R. Crump. 1995. Review of synthetic predator odor 
semiochemicals as repellents for wildlife management in the Pacific Northwest. National 
Wildlife Research Center Repellents Conference 1995. 
Linduska J. P. 1947. Winter den studies of the cottontail in southern Michigan. Ecology 28:448–
454. 
Loss, S. R., T. Will, and P. P. Marra. 2013. The impact of free-ranging domestic cats on wildlife 
of the United States. Nature Communications 4:1396. 
Lyman, C. P., and P. O. Chatfield. 1955. Physiology of hibernation in mammals. Physiological 
Reviews 35:403–425. 
Lyman Charles P., and Blinks Doris C. 2005. The effect of temperature on the isolated hearts of 
closely related hibernators and non-hibernators. Journal of Cellular and Comparative 
Physiology 54:53–63. 
McManus, J. J. 1969. Temperature regulation in the opossum, Didelphis marsupialis virginiana. 
Journal of Mammalogy 50:550–558. 
Meek, P. D., G.-A. Ballard, and P. J. S. Fleming. 2015. The pitfalls of wildlife camera trapping 
as a survey tool in Australia. Australian Mammalogy 37:13–22. 
Melvin, R. G., and M. T. Andrews. 2009. Torpor induction in mammals: Recent discoveries 
fueling new ideas. Trends in endocrinology and metabolism 20:490–498. 
	
	 45	
Mutch, G. R. P., and M. Aleksiuk. 1977. Ecological aspects of winter dormancy in the striped 
skunk (Mephitis mephitis). Canadian Journal of Zoology 55:607–615. 
Navarro-Castilla, Á., and I. Barja. 2014a. Antipredatory response and food intake in wood mice 
(Apodemus sylvaticus) under simulated predation risk by resident and novel carnivorous 
predators. Ethology 120:90–98. 
Navarro-Castilla, Á., and I. Barja. 2014b. Does predation risk, through moon phase and predator 
cues, modulate food intake, antipredatory and physiological responses in wood mice 
(Apodemus sylvaticus)? Behavioral Ecology and Sociobiology 68:1505–1512. 
Neale Jennifer C. C., and Sacks Benjamin N. 2003. Resource utilization and interspecific 
relations of sympatric bobcats and coyotes. Oikos 94:236–249. 
Newey, S., P. Davidson, S. Nazir, G. Fairhurst, F. Verdicchio, R. J. Irvine, and R. van der Wal. 
2015. Limitations of recreational camera traps for wildlife management and conservation 
research: A practitioner’s perspective. Ambio 44:624–635. 
New York Natural Heritage Program. 2017, August 10. New England Cottontail. New York 
Natural Heritage Program. 
Nussberger, B., and P. Ingold. 2006. Effects of radio-collars on behaviour of alpine chamois 
Rupicapra rupicapra rupicapra. Wildlife Biology 12:339–343. 
Orrock, J. L., B. J. Danielson, and R. J. Brinkerhoff. 2004. Rodent foraging is affected by 
indirect, but not by direct, cues of predation risk. Behavioral Ecology 15:433–437. 
Osada, K., S. Miyazono, and M. Kashiwayanagi. 2014. Pyrazine analogs are active components 
of wolf urine that induce avoidance and fear-related behaviors in deer. Frontiers in 
Behavioral Neuroscience 8:1-7. 
Ozoga, J. J., and E. M. Harger. 1966. Winter activities and feeding habits of northern Michigan 
	
	 46	
coyotes. Journal of Wildlife Management 30:809–818. 
Palm, E., and M. Tveitereid. 1979. On heat and mass flux through dry snow. Journal of 
Geophysical Research: Oceans 84:745–749. 
Patil, V. P., S. F. Morrison, T. J. Karels, and D. S. Hik. 2013. Winter weather versus group 
thermoregulation: what determines survival in hibernating mammals? Oecologia 
173:139–149. 
Pirie, T. J., R. L. Thomas, and M. D. E. Fellowes. 2016. Limitations to recording larger 
mammalian predators in savannah using camera traps and spoor. Wildlife Biology 22:13–
21. 
Pomeroy, J. W., and E. Brun. 2001. Physical properties of snow. Pages 45–126 in Snow 
Ecology.Cambridge University Press, Cambridge, UK. 
R Core Team (2017). R: A language and environment for statistical computing. R Foundation for 
Statistical Computing, Vienna, Austria.  
Rahman, D. A., G. Gonzalez, and S. Aulagnier. 2016. Benefit of camera trapping for surveying 
the critically endangered Bawean deer Axis kuhlii (Temminck, 1836). Tropical Zoology 
29:155–172. 
Rosen, J. B., A. Asok, and T. Chakraborty. 2015. The smell of fear: innate threat of 2,5-dihydro-
2,4,5-trimethylthiazoline, a single molecule component of a predator odor. Frontiers in 
Neuroscience 9:1-12. 
Sullivan, T. P. 1986. Influence of wolverine (Gulo gulo) odor on feeding behavior of snowshoe 
hares (Lepus americanus). Journal of Mammalogy 67:385–388. 
Sullivan, T. P., and D. R. Crump. 1986. Feeding responses of snowshoe hares (Lepus 
americanus) to volatile constituents of red fox (Vulpes vulpes) urine. Journal of Chemical 
	
	 47	
Ecology 12:729–739. 
Sunquist, M. E. 1974. Winter activity of striped skunks (Mephitis mephitis) in east-central 
Minnesota. American Midland Naturalist 92:434–446. 
Takahashi, L. K., B. R. Nakashima, H. Hong, and K. Watanabe. 2005. The smell of danger: A 
behavioral and neural analysis of predator odor-induced fear. Neuroscience & 
Biobehavioral Reviews 29:1157–1167. 
Vaughan, T. A. 1961. Vertebrates inhabiting pocket gopher burrows in Colorado. Journal of 
Mammalogy 42:171–174. 
Villén-Pérez, S., L. M. Carrascal, and J. Seoane. 2013. Foraging patch selection in winter: a 
balance between predation risk and thermoregulation benefit. PLOS ONE 8:e68448. 
Watts, P. D., and C. Jonkel. 1988. Energetic cost of winter dormancy in grizzly bear. Journal of 
Wildlife Management 52:654–656. 
Yacoe, M. E. 1983. Protein metabolism in the pectoralis muscle and liver of hibernating bats, 
Eptesicus fuscus. Journal of Comparative Physiology 152:137–144. 
Zielinski, W. 1986. Circadian rhythms of small carnivores and the effect of restricted feeding on 
daily activity. Physiology & Behavior 38:613–620. 
Zielinski, W. J. 2000. Weasels and martens — carnivores in northern latitudes. Pages 95–118 in 
Activity Patterns in Small Mammals. S. Halle and N. C. Stenseth, editors. Springer, 
Berlin, Heidelberg.