E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
748
Activity Patterns of Allegheny Woodrats in Tennessee
Elizabeth A. Stovall1 and Steven E. Hayslette2,*
Abstract - Neotoma magister (Allegheny Woodrat) is a species of management concern
throughout its range. Better knowledge regarding timing of activity may improve monitoring
and conservation of the species. We used remote cameras to examine activity patterns of
Allegheny Woodrats at 9 rock outcrops on Catoosa Wildlife Management Area, TN, February–
October 2010. Woodrat activity was documented during all months of study. Highest
levels of activity occurred during May and October, during the hours of 02:00–03:00, at
<40% full moon illumination, and at minimum nightly temperatures of 5–10 ºC. Focusing
on periods of greatest activity (late spring and early fall during darker phases of the lunar
cycle) may increase efficiency of monitoring ef forts for the species.
Introduction
Neotoma magister Baird (Allegheny Woodrat) inhabits rocky areas in the central/
southern Appalachian Mountains and Interior Low Plateaus of the eastern United
States (Castleberry et al. 2002). Allegheny Woodrat populations have declined markedly
during the last 30 years, particularly in the northern portion of their range. To date,
Allegheny Woodrats have been extirpated from 3 states and are considered threatened
or endangered in 4 others (LoGiudice 2006). Although few studies have been conducted
in the southern portion of their range, including Tennessee, populations appear
stable (Wright 2008). Definitive causes of Allegheny Woodrat population declines are
unclear, but habitat loss/fragmentation, declines in food availability, and/or mortality
due to parasites have been implicated (LoGiudice 2006).
To date, only general information is available regarding daily and seasonal
timing of Allegheny Woodrat activity. Allegheny Woodrats are active nocturnally;
activity is thought to be highest for several hours following sunset and just before
dawn (Castleberry et al. 2006), although an activity peak around midnight als had
been reported (Poole 1940). Seasonal activity is highest in spring and fall, when
dispersal, reproduction, active foraging, and food caching takes place (Castleberry
and Castleberry 2008, Wood 2008). During winter, home ranges are smaller
and movements are reduced (Castleberry 2008). Additionally, temperature and
precipitation patterns are known to affect capture rates (Wood 2008), perhaps because
of their influence on activity patterns. More complete information regarding
Allegheny Woodrat activity patterns may improve cost-efficiency of surveys and
monitoring efforts. The objectives of our study were to document activity patterns
(nightly and seasonal) of Allegheny Woodrats, and to evaluate effects of ambient
temperature and moon illumination on these patterns. We hypothesized that woodrat
activity would decrease with decreasing ambient temperature and increasing
moon illumination because of greater energetic costs during colder temperatures
and increased risk of predation on brighter nights.
2013 SOUTHEASTERN NATURALIST 12(4):748–756
12478 Airport Road, Bridgewater, VA 22812. 2Department of Biology, Box 5063, Tennessee
Technological University, Cookeville, TN 38505. *Corresponding author - shayslette@
tntech.edu.
749
E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
Methods
Our study of activity patterns was part of a larger study of Allegheny Woodrat
habitat use conducted at Catoosa Wildlife Management Area (CWMA), located
on the Cumberland Plateau in Cumberland, Morgan, and Fentress counties, TN.
Fieldwork was conducted during February–October 2010, excluding April, during
which public hunting on CWMA precluded access to study sites. The area
ranges 335–700 m in elevation and is 95% forested; topography includes upland
ridges dominated by Quercus (oak)-Carya (hickory) and Quercus-Pinus (pine)
communities separated by deep gorges dominated by mixed mesophytic vegetation
(Tennessee Wildlife Resources Agency 2001).
In our larger habitat-use study, we surveyed 34 rock outcrops in 12 drainages for
signs of Allegheny Woodrat activity (e.g., food items, sticks or other nest material,
and feces; see Stovall 2011 for details). Of these, 16 (47%) showed signs of woodrat
activity and were monitored with infrared-triggered digital cameras (Reconyx
models PC85 and PC90). This monitoring formed the basis for activity-pattern
analyses. At each outcrop where woodrat sign was observed, we attached a camera
to a tree near the sign at a height such that the potential area of woodrat activity
was within the camera field of view. We programmed the cameras to take three
photographs at each triggered event. Each photo included date, time of day, and
ambient temperature stamps. Each camera remained at an individual rock outcrop
for a minimum of two weeks, after which it was relocated to another outcrop. At the
time of relocation, all pictures were downloaded from media cards taken from the
cameras and uploaded onto a laptop computer for analysis and storage, and camera
batteries were replaced, if necessary.
Total camera-nights at active sites (sites at which woodrats were photographed)
and total nights during which one or more woodrats was photographed (active
camera-nights) were tallied for each month of study. We defined a camera-night as
an individual camera surveying one site for one night. For active sites, we compared
proportion of active camera-nights across all months using a G- (log-likelihood)
test of independence with Williams’ correction (Sokal and Rohlf 1995). February
was excluded because of a low (n = 3) number of active camera-nights. We used an
alpha level of 0.05 for all analyses.
For each active camera-night, we calculated two indices of activity: number
of visits by woodrats and number of hours of activity. A visit was defined
as a collection of photographs of an individual woodrat separated from other
collections by ≥5 minutes. We calculated number of hours of activity for a
camera-night as the number of hour intervals during which at least one visit occurred.
These two indices were used as dependent variables in activity analyses.
We derived two independent variables, midnight moon-illumination fraction and
minimum nightly temperature, for each active camera-night using the web-based
astronomical moon-illumination calculator JSkyCalc v1.2.1 (ESO 2001) and
lowest temperature recorded on a woodrat photo from the camera-night, respectively.
We did not visit sites at night, so we could not quantify or consider cloud
cover effects on moon illumination.
E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
750
Simple linear regression was used to examine the relationship between each
of the two activity-related dependent variables and each of the two independent
variables, using active camera-nights as experimental units. To examine temporal
patterns of activity within active camera-nights, we tallied visits and frequency of
activity (number of times one or more visits occurred) by hour intervals across all
active camera-nights. To examine duration of activity within months, we tallied
frequency of activity by hour intervals within months.
Results
Allegheny Woodrats were photographed at nine (56%) of the 16 outcrops surveyed.
A total of 572 camera-nights was logged at the active sites during the eight
months of survey. Woodrat activity was documented during 106 of these cameranights
(19% of active site total). Activity was documented during all survey months,
but proportion of active camera-nights varied among months (G = 48.6, P < 0.001;
Table 1). Activity levels were highest during May and October and lowest during July
and August. Activity occurred during 13 one-hour intervals (Fig. 1); earliest activity
occurred 18:00–19:00, and latest activity occurred 07:00–08:00. Both number of
visits (Fig. 1a) and frequency of activity (number of camera-nights with activity in an
interval; Fig.1b) peaked 02:00–03:00. Nightly duration of activity was greatest during
October, when activity occurred during 11 time intervals (Fig. 2). Activity prior
to 20:00 was restricted to March, September, and October. Activity generally ceased
at 05:00; activity after 05:00 was restricted to August, September, and October, and
activity after 07:00 only occurred during one camera-night in August. Duration of
activity was shortest in February, when activity occurred only during three time intervals,
and during July, when activity began after 22:00 and ended by 05:00.
Both numbers of visits (F1,104 = 8.3, P = 0.005; Fig. 3a) and hours of activity
(F1,104 = 8.5, P = 0.004; Fig. 3b) were negatively related to moon-illumination
fractions across active camera-nights, although regression models explained <8%
Table 1. Number of active Allegheny Woodrat sites surveyed with remote cameras in each month, total
number of camera-nights, number of active camera-nights (nights a woodrat was documented), and
proportion of active camera-nights at Catoosa Wildlife Management Area, TN, February–October 2010.
Camera-nights
Month Active sites surveyed Total No. active Proportion of total
February 2 20 4 0.20
March 3 70 18 0.26
May 2 42 20 0.48
June 4 114 20 0.18
July 5 152 11 0.07
August 6 73 7 0.10
September 4 79 16 0.20
October 2 22 10 0.45
Total 9* 572 106 0.19
*Some active sites were surveyed in >1 month.
751
E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
of the overall variability. Activity occurred at moon-illumination values ranging
0.0–1.0, but relatively high levels of activity (>5 visits, >3 hours of activity) only
occurred at illumination values <0.4. Neither activity index was linearly related
to minimum nightly temperature (F1,104 ≤ 2.4, P ≥ 0.126; Fig. 4a, b) across active
camera-nights, although highest levels of activity (>7 visits, >6 hours of activity)
occurred at minimum nightly temperature values of 5–10 ºC.
Figure 1. (a) Number of visits and (b) frequency of Allegheny Woodrat activity (number of
camera-nights with activity in a particular hour interval, by hour interval at Catoosa Wildlife
Management Area, TN, February–October 2010.
E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
752
Discussion
Our results indicate that activity by Allegheny Woodrats peaked in May and
October on our study area. High levels of activity during May might reflect breeding
activity and/or foraging on newly available foods. The peak in activity observed
in October may reflect the onset of food-caching behavior, which generally
begins in September and October (Castleberry and Castleberry 2008, Castleberry
et al. 2002). During winter, it is believed that Allegheny Woodrats primarily rely
on cached foods and seldom leave their dens (Castleberry and Castleberry 2008,
Hornsby et al. 2005), which likely explains lower activity levels observed in our
study during February. Longer duration of activity early in our study in March and
later on in October likely reflects a reduced photoperiod (longer dark period) and,
perhaps, lower nightly temperatures, which occur during spring and fall compared
to summer.
Activity pattern data for Allegheny Woodrats in our study generally indicate
activity peaking between 02:00–03:00. This peak in activity occurred somewhat
later than the peak at “about midnight” reported for Allegheny Woodrats by Poole
(1940), and later than activity of Neotoma floridana Ord (Eastern Woodrat), which
peaks at or before midnight (Wiley 1971). Using similar camera-trap methods
during a similar seasonal period (June–August) in central Kansas, Wiley (1971)
documented that >50% of all Eastern Woodrat activity took place between 20:30–
22:30, and 75% occurred before midnight. In that same study, activity by subadult
Eastern Woodrats peaked later (01:00–02:00), but this was attributed to dominance
by adults that forced younger woodrats to forage later. Apparent differences in the
timing of peak activity between Eastern and Allegheny Woodrats may be due to
Figure 2. Duration of activity (shaded boxes) of Allegheny Woodrats by months and hour
intervals at Catoosa Wildlife Management Area, TN, February–October 2010.
753
E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
basic behavioral differences between species, differences in resource abundance/
distribution, and/or differences in the community of competitors/predators. Few
existing data speak to any of these possibilities; future research is needed to better
understand interspecific differences in activity timing.
Although Castleberry et al. (2001) reported no influence of moon illumination
on distance traveled from dens by Allegheny Woodrats, we found that moon
Figure 3. Relationships between moon illumination and (a) number of visits by Allegheny
Woodrats and (b) hours of activity at Catoosa Wildlife Management Area, TN, February–
October 2010.
E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
754
illumination affected activity of Allegheny Woodrats at CWMA, as it does other
Neotoma species. Wiley (1971) reported highest levels of activity by Eastern
Woodrats during new or quarter moon phases, and lowest levels of activity during
a full moon. Finley (1959) reported that Eastern Woodrats tended to be
timid in the presence of moonlight. In our study, activity by Allegheny Woodrats
took place at all moon-illumination levels, but relatively high activity levels
Figure 4. Relationships between (a) number of visits by Allegheny Woodrats and (b) hours
of activity, versus low nightly temperature documented on photographs at Catoosa Wildlife
Management Area, TN, February–October 2010.
755
E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
(>5 visits, >3 hours of activity) only occurred at illumination levels of less than 40% of
full moon levels. Decreased activity under brighter illumination may reflect increased
vulnerability to predation during bright nights. No evidence of predation
on Allegheny Woodrats was found during fieldwork for our study, but visits by
Spilogale putorius L. (Eastern Spotted Skunk) and Mustela frenata Lichtenstein
(Long-tailed Weasel), both known predators of Allegheny Woodrats (Hassinger et
al. 2001), were documented in photographs at active sites.
Our results indicated no linear relationship between Allegheny Woodrat activity
levels and ambient temperature, but we did observe temporal variability
in activity across the range of temperatures. Activity was documented at nightly
temperatures between -3 ºC and 25 ºC, with peak activity between 5 ºC and 10
ºC. The lower end of this temperature range (-3 ºC) may represent the limit below
which Allegheny Woodrats do not leave their dens at CWMA, but instead rely on
cached food. Nightly temperature in the region fell below -3º C during 20 nights
in February and three nights in March (NOAA 2011), and no activity was documented
on those nights. In contrast, Allegheny Woodrat activity does not appear
limited by high temperatures during summer months, as the high end of this active
temperature range (25 ºC) equaled the highest minimum nightly temperatures recorded
during 2010 (NOAA 2011).
Conservation concerns regarding Allegheny Woodrat populations throughout
their range make development of efficient monitoring protocols a high priority.
Results of our study indicate that monitoring efforts may be more efficient in May
and October, and on relatively dark nights (moon illumination < 40% of full) due to
increased woodrat activity levels during those times. Additional studies elsewhere
are needed to determine if the activity patterns documented at Catoosa WMA, TN,
are characteristic across the species range.
Acknowledgments
We thank C. Simpson and other Tennessee Wildlife Resources Agency personnel for
field assistance and logistical support. We also thank M. Grandstaff, S. Blomquist, and G.
Rhinehart for assistance with GIS and spatial data, and C. Peterson and N. Cantrell for field
assistance. Funding for this project was provided by a Tennessee Technological University
(TTU) faculty research grant and the TTU Department of Biology. T.H. Roberts, S.B. Cook,
and 2 anonymous reviewers provided helpful comments on earlier drafts of this manuscript.
Literature Cited
Castleberry, S.B. 2008. Home range, movements, and habitat selection. Pp. 63–74, In J.D.
Peles and J. Wright (Eds.). The Allegheny Woodrat: Ecology, Conservation, and Management
of a Declining Species. Springer, New York, NY. 231 pp.
Castleberry, N.L, and S.B. Castleberry. 2008. Food selection and caching behavior. Pp.
93–106, In J.D. Peles and J. Wright (Eds.). The Allegheny Woodrat: Ecology, Conservation,
and Management of a Declining Species. Springer, New York, NY. 231 pp.
Castleberry, N.L., S.B. Castleberry, W.M. Ford, P.B. Wood, and M.T. Mengak. 2002. Allegheny
Woodrat (Neotoma magister) food habits in the central Appalachians. American
Midland Naturalist 147:80–92.
E.A. Stovall1 and S.E. Hayslette
2013 Southeastern Naturalist Vol. 12, No. 4
756
Castleberry, S.B., W.M. Ford, P.B. Wood, N.L. Castleberry, and M.L. Mengak. 2001.
Movements of Allegheny Woodrats in relation to timber harvesting. Journal of Wildlife
Management 65:148–156.
Castleberry, S.B., M.T. Mengak, and M.W. Ford. 2006. Mammalian species Neotoma magister.
American Society of Mammalogists 789:1–5.
European Southern Observatory (ESO). 2001. Calendars and calculators: Sky ephemerids.
Available online at http://www.eso.org/sci/observing/tools/skycalc.html. Accessed November
2011.
Finley, R.B., Jr. 1959. Observations of nocturnal animals by red light. Journal of Mammalogy
40:591–594.
Hassinger, J.D., C.M. Butchkoski, and D.R. Diefenbach. 2001. Managing surface rock
communities for Neotoma magister. Pp. 133–152, In J.D. Peles and J. Wright (Eds.). The
Allegheny Woodrat: Ecology, Conservation, and Management of a Declining Species.
Springer, New York, NY. 231 pp.
Hornsby, B.S., A.M. Ruiz, S.B. Castleberry, N.L. Castleberry, P.B. Wood, and W.M. Ford.
2005. Fall movement of Allegheny Woodrats in harvested and intact stands in West Virginia.
Northern Journal of Applied Forestry 22:281–284.
LoGiudice, K. 2006. Toward a synthetic view of extinction: A history lesson from a North
American rodent. Bioscience 56:687–693.
National Oceanic and Atmospheric Administration (NOAA). 2011. National Weather Service
local temperature outlook. Avialable online at http://www.weather.gov/climate/
calendar_outlook.php. Accessed March 2011.
Poole, E.L. 1940. A life-history sketch of the Allegheny Woodrat. Journal of Mammalogy
21:249–270.
Sokal, R.R., and F.J. Rohlf. 1995. Biometry. Third Edition. W.H. Freeman and Company,
New York, NY. 887 pp.
Stovall, E.A. 2011. Habitat characteristics and activity patterns of Allegheny Woodrats
within Catoosa Wildlife Management Area, Tennessee. M.Sc. Thesis. Tennessee Technological
University, Cookeville, TN. 51 pp.
Tennessee Wildlife Resources Agency. 2001. Catoosa Wildlife Management Area map.
Tennessee Wildlife Resources Agency, Nashville, TN.
Wiley, R.L. 1971. Activity patterns and movements of the Eastern Woodrat. Southwestern
Naturalist 16:43–54.
Wood, P.B. 2008. Woodrat population dynamics and movement patterns. Pp. 45–62,
In J.D. Peles and J. Wright (Eds.). The Allegheny Woodrat: Ecology, Conservation, and
Management of a Declining Species. Springer, New York, NY. 231 pp.
Wright, J. 2008. History and current status of the Allegheny Woodrat. Pp. 3–22, In J.D.
Peles and J. Wright (Eds.). The Allegheny Woodrat: Ecology, Conservation, and Management
of a Declining Species. Springer, New York, NY. 231 pp.