2012 SOUTHEASTERN NATURALIST 11(3):507–516 Observations of Gray Foxes (Urocyon cinereoargenteus) in a Suburban Landscape in the Piedmont of North Carolina Joshua M. Kapfer1,2,* and Ryan W. Kirk3 Abstract - Studies of Urocyon cinereoargenteus (Gray Fox) in suburban landscapes are rare. Past work has suggested that this species will only tolerate urbanization to a certain density of residences (50–125 residences/km2). To test this, we employed visual observations and camera traps to monitor Gray Fox activity within a suburban and adjacent rural property from January to July 2011. We also used a geographical information system (GIS) to calculate the density of buildings associated with both properties. We observed Gray Foxes and detected them with camera traps in our properties on numerous occasions. GIS analyses revealed an estimated suburban density of 237–347 residences/km2 (depending on spatial scale) and rural density of 50 residences/km2. The number of Gray Fox observations did not differ greatly between rural and suburban properties, although the peak periods of Gray Fox observations varied by site. We propose that a tolerance for high suburban building density exhibited by Gray Foxes at our site is related to the large amount of mature wooded buffers located adjacently. Introduction Research focused on Canids in urban and suburban landscapes has recently become more common (see review by Gehrt et al. 2010). For example, a respectable amount of research has been conducted in North America on urban and suburban Canis latrans Say (Coyote; see review by Gehrt and Riley 2010). In contrast, the ecology of Urocyon cinereoargenteus Schreber (Gray Fox) inhabiting anthropogenic landscapes has received little attention (Riley and White 2010). The bulk of research on urban Gray Foxes has been carried out in the western US (e.g., California and New Mexico; Harrison 1993, 1997; Riley 2006; Riley et al. 2004). Although work has been completed on this species in the eastern US, only a handful of these studies have dealt with individuals that were associated with urban or suburban habitat (e.g., Rountree 2004, Temple et al. 2010). This past research has suggested that Gray Foxes will use anthropogenically disturbed landscapes, but do not prefer them. For example, radio telemetry research conducted on Gray Foxes within the Golden Gate National Recreation Area (Marin County, CA) found that the core home ranges of 17 tracked individuals rarely involved anthropogenic habitat present within 1Departments of Environmental Studies and Biology, Elon University, Campus Box 2015, Elon, NC 27244. 2Current address - Department of Biological Sciences, University of Wisconsin-Whitewater, Whitewater, WI 53190. 3Departments of History and Geography, and Environmental Studies, Elon University, Campus Box 2335, Elon, NC 27244. *Corresponding author - kapferj@uww.edu. 508 Southeastern Naturalist Vol. 11, No. 3 the park (Riley 2006). Of the foxes tracked, only two had core home ranges that included “stables and residences” located in park boundaries. A third individual had a core home range that included some urban habitat outside of park boundaries. Furthermore, Rountree (2004) reported that Gray Foxes tracked with radio telemetry (1 M, 3 F) used areas close to or within “human habitation”, but were found in “fields” and “mixed forests” more often than expected. Temple et al. (2010) also used radio telemetry to monitor 23 Gray Foxes (15 M, 8 F) with access to residential areas. They found that Gray Foxes showed a preference for hardwood forests, although individuals made use of residential habitat. Harrison (1997) suggested that Gray Foxes may only be tolerant of urban or suburban landscapes to a density of 50–125 residences/ km2. Use of suburban landscapes may also be limited by factors such as interactions with larger carnivores. For example, past studies have indicated that Gray Foxes select habitats to reduce the potential for interaction with Coyotes and avoid mortality due to competitive exclusion or predation (Fedriani et al. 2000, Temple et al. 2010). Numerous studies exist that show Coyotes will frequently use urban or suburban landscapes (see review by Gehrt and Riley 2010), which may influence Gray Fox use of associated habitats. Herein we report numerous observations of Gray Foxes active within a parcel of residential property over a period of roughly 6 months, which we generally compare to an exurban/rural habitat monitored nearby during the same period. We also compare our results, including estimates of housing density at the suburban property, to Harrison’s (1997) estimate of the housing density that this species will tolerate. Our observations give further insight into the habits of urban/suburban Gray Foxes, which is a topic in need of investigation (Riley and White 2010). Methods We monitored the activity of Gray Foxes within a single legal real-estate parcel of residential property (27.4 x 29 m) located in a suburban landscape within the town of Elon (Alamance County), which is in the Piedmont Region of North Carolina. Monitoring included both visual observations and camera-trap surveys. This property was 1.56 km from the nearest rural/suburban fringe, and was located on the corner of a road intersection (one road of moderate vehicular traffic, and the other with low vehicular traffic). The property in question included a mostly open tree canopy and possessed a manicured lawn, with a small ornamental fruit tree (Malus sp.) located in the center. Mature trees existed along the property boundaries (e.g., Pinus sp., Carya sp.) and strips or parcels of mature wooded habitat were present in several of the adjacent properties (Fig. 1). For relative comparison, we also monitored an exurban/rural site located 3.02 km to the northeast of the suburban site. This property consisted primarily of deciduous woodland (65% of area based on aerial photograph interpretation), with smaller components of old field habitat. Deciduous forests contained a mix of mesic, 2012 J.M. Kapfer and R.W. Kirk 509 alluvial, and oak-hickory plant communities, whereas old fields were dominated by sedge species associated with disturbance. Although this property was 9.8 ha, in total, we monitored only the small portion associated with our camera trap, and areas we were able to visually survey. Visual observations of foxes occurred opportunistically by either the lead author or other residents of the suburban property being monitored. Two observations of Gray Foxes adjacent to and within the suburban property boundaries were made in May (ca. 1000 hrs) and December (ca. 2100 hrs) of 2010. In response to these sightings, more intense monitoring of the property was conducted via a motion-triggered camera trap with an infrared flash (Bushnell Trophy Cam XLT, Bushnell, Olathe, KS). This camera trap was deployed at various locations throughout the property in question from 6 January 2011 to 18 July 2011. The camera was deployed nightly between 1800–2200 h and removed each morning between 0600–0900 h to avoid recording excessive pictures of vehicular traffic and human activity. Although this camera was not active for several brief time periods due to extrinsic factors beyond our control (e.g., researcher illness), a total of 178 camera nights occurred (92.2% of the primary monitoring period). A similar camera was deployed at the rural property during the same period. This camera remained at a fixed location during this time and was operational constantly (i.e., not removed during the daylight hours). We also recorded time and moon phase associated with all photographs to determine if any discernible relationship existed among these variables and fox activity. Different individuals could not be definitively identified, therefore we recorded photographs and observations as Gray Fox “encounters” and made no attempt to quantify the abundances of foxes present in our study sites. We did not attempt to lure Gray Foxes to the camera traps with artificial scent or food bait. Domestic dog scat/urine had been deposited throughout the suburban property periodically over a roughly 1.5-y period prior to initiation of monitoring, although scat was typically removed by the property owner within 24 hr of deposition. Furthermore, a domestic dog often accompanied the lead investigator when he deployed the suburban camera each night. No wildlife feeders, bird seed, brush piles, or any other cover or food that may attract wildlife were present on the suburban site. Fruit is reported as important in the diets of Gray Foxes (see review by Riley 2006), and fruit from a tree on the suburban site (Malus sp.) may have provided a food source. This fruit was only available from late May through July of our monitoring period. Human and domestic dog recreational activity occurred throughout the monitored suburban property almost daily from mid-March 2011 through July 2011. Recreational activity varied considerably in length on any given day. Extremely limited human activity occurred on the rural property, and was mostly the result of researchers visiting the site to check camera traps. A geographic information system (GIS; ArcMap 9.3; ESRI, Redlands, CA) was used to estimate building density at two different scales: first within a 2.59-km2 (1-mi2) circle (radius of 908 m) centered on the monitored suburban 510 Southeastern Naturalist Vol. 11, No. 3 property, and second for the entire area of the town (Elon, NC) in which the property was located (9.92 km2). We also estimated the building density within a 2.59-km2 circle centered on the camera at the rural site. Buildings within these areas were almost entirely residential, although several commercial properties and larger private properties (i.e., religious sanctuaries, etc.) existed. Given this residential dominance, we refer to our building density as “residential density” when comparing it to other studies. Within each of these boundaries, we counted the number of structures present to estimate buildings per unit area. We also delineated contiguous suburban forest stands at least 0.25 ha in size within the smaller suburban landscape scale analyzed (2.59 km2) to estimate the percent forest cover. Results The most effective method to observe Gray Foxes was via camera trap. A total of 15 distinct encounters of Gray Fox were recorded by the camera trap on the suburban property during the monitoring period. These occurred over 13 dates, with one of these survey dates (14 May 2011) resulting in 3 distinct passes (Table 1). Only 11 camera-trap encounters of Gray Foxes were recorded on the rural property during this time (Table 2). Visual observations of Gray Foxes associated with the suburban property in 2011 occurred on 7 occasions over 5 dates (Table 1). No visual observations of Gray Foxes were recorded at the rural site. Two general peaks in the frequency of Gray Fox photographs occurred at the Table 1. Summary of camera-trap encounters and visual observations of Gray Foxes (Urocyon cinereoargenteus) monitored on a suburban property from 6 January–18 July 2011 (Elon, Alamance County, NC). Associated date, type of observation, time, hours before sunrise of last observation (for nocturnal encounters), and moon phase are reported. Date Type Time Hours before sunrise Moon phase 1/24/2011 Camera trap 0522 h 1.98 Waning gibbous 1/25/2011 Camera trap 0537 h 1.21 Waning gibbous 1/26/2011 Camera trap 2309 h 8.16 Third quarter 2/8/2011 Camera trap 0220 h 4.83 Waxing crescent 2/18/2011 Visual 1800 h NA Full moon 3/9/2011 Camera trap 0251 h 4.03 Waxing crescent 3/21/2011 Camera trap 0121 h 5.96 Waning gibbous 5/14/2011 Camera trap 0245, 0258, 0505 h 1.10 Waxing gibbous 5/18/2011 Camera trap 2251 h 7.20 Full moon 5/20/2011 Camera trap 2354 h 6.18 Waning gibbous 5/26/2011 Visual 1700, 2030 h 9.53 Waning crescent 5/29/2011 Visual 1100, 2200 h 8.01 Waning crescent 6/6/2011 Camera trap 0214 h 3.75 Waxing crescent 6/14/2011 Camera trap 0447 h 1.18 Waxing gibbous 6/16/2011 Camera trap 0401 h 1.95 Full moon 6/24/2011 Camera trap 0051 h 5.13 Waning crescent 6/27/2011 Visual* 1000 h NA Waning crescent 7/7/2011 Visual* 1001 h NA Waxing crescent *Indicates juveniles and adults observed. 2012 J.M. Kapfer and R.W. Kirk 511 suburban property: a small peak from 24 January–21 March 2011 (which may correspond with the breeding season) and an additional peak occurring from 14 May– 7 July 2011 (Table 1). In contrast, almost all photographs of Gray Foxes at the rural site were taken prior to 1 April 2011 (Table 2). All Gray Fox captures on camera traps, regardless of location, were nocturnal. At the suburban site, camera trap captures occurred between 0051–0600 h (n = 12) and between 2200–2400 h (n = 3). Similarly, rural Gray Foxes were most often captured on camera traps between 0034–0616 h (n = 7) and 1949–2100 h (n = 4, Table 2). Several (n = 5) visual observations at the suburban site were made during daylight hours (Table 1). Our sample sizes are too small to make significant inferences in regards to moon phase and fox activity, although we saw interesting correlations. For example, we recorded no Gray Fox activity during the “new moon” phase on the suburban property (Table 1). In addition, only one photograph of Gray Foxes in the rural habitat was taken during this moon phase (Table 2). At the smaller suburban landscape scale (i.e., within 2.59-km2 area centered on the suburban site), we counted a total of 901 buildings which were primarily single-family residences. This resulted in an estimate of 347 residences/km2 (Fig. 1). At the larger landscape scale (i.e., the entire town in which our observations were made), we identified 2359 buildings with primary functions including individual residences, multi-unit residences, and commercial uses. This resulted in a building density estimate of 237 buildings/km2. In contrast, we estimated only 52 residences/km2 associated with the rural site. Discussion We have recorded conclusive evidence that Gray Foxes will repeatedly use, or migrate through, anthropogenically disturbed areas. In some cases, individuals were photographed less than two meters from the residential building on-site, where substantial outdoor activity by humans had occurred several Table 2. Summary of camera trap encounters of Gray Foxes (Urocyon cinereoargenteus) monitored on a rural property from 6 January–18 July 2011 (Elon, Alamance County, NC). Associated date, type of observation, time, hours before sunrise of last observation, and moon phase are reported. Date Type Time Hours before sunrise Moon phase 1/13/2011 Camera trap 1941 h 11.85 First quarter 1/14/2011 Camera trap 2020 h 11.06 First quarter 1/21/2011 Camera trap 2100 h 10.36 Waning gibbous 2/4/2011 Camera trap 0034 h 6.65 New moon 2/15/2011 Camera trap 0125 h 5.8 Waxing gibbous 2/26/2011 Camera trap 0420 h 2.51 Waning crescent 3/9/2011 Camera trap 0616 h 0.31 Waxing crescent 3/10/2011 Camera trap 0605 h 0.48 Waxing crescent 3/15/2011 Camera trap 1954 h 11.51 Waxing gibbous 3/23/2011 Camera trap 0323 h 3.86 Waning gibbous 6/16/2011 Camera trap 0103 h 4.91 Full moon 512 Southeastern Naturalist Vol. 11, No. 3 hours prior (Fig. 1). Furthermore, this property was located in an area with a housing density of two to three times greater than suggested as a tolerance threshold by Harrison (1997; i.e., 50–125 residences/km2). The reasons why Figure 1. Top: map of buildings (points), roads (lines) and wooded area (gray polygons) within a 2.59-km2 (1-mi2) area centered on the monitored suburban property (indicated by star). Elon, Alamance County, NC. Bottom (2 photos): Example of photographic evidence of Gray Fox (Urocyon cinereoargenteus) visiting a suburban property in 2011. 2012 J.M. Kapfer and R.W. Kirk 513 Gray Foxes exist within higher building densities here than what Harrison suggested are not fully understood. Harrison (1993) reported that the extent of “original” habitat present within a rural residential area had a significant influence on Gray Fox density. Gray Foxes prefer habitats with some element of tree and brush cover (reviewed by Cypher 2003). The presence of this species at the suburban site we monitored may be related to the prevalence of mature wooded buffers and wooded parcels located adjacently. For example, 12.1% of the 2.59-km2 suburban area analyzed was forested (Fig. 1). In addition, this property was also located less than 1.6 km from the nearest rural/suburban fringe, which may have given Gray Foxes easy access. The greatest activity of Gray Foxes monitored at the rural site differed seasonally from the suburban site. The largest number of suburban fox encounters occurred in May and June 2011. In contrast, Gray Foxes at the rural site were photographed most often in January and February 2011, and only one cameratrap photograph was taken there from 1 April–18 July 2011. We believe that the frequency of Gray Fox encounters at the suburban property suggests that these were residents of the general area. Although we cannot say if the property that we monitored was part of their core home range, it at least appeared to be part of the home range they traversed during the active season. On one occasion (26 May at 1700 h), an individual was witnessed moving across the adjacent road with a small mammal of unknown species in its mouth. This observation, coupled with the peak in activity observed during late May, may be related to parturition and nursing by females, during which males are known to forage alone and bring food to the nursing female (Cypher 2003). In addition, an adult was seen on 27 June crossing a road adjacent to the suburban site followed by four kits. An adult and kit were also seen crossing the same road on 7 July (Table 1). These observations strongly suggest that the suburban site was part of their core home range, due to the apparent likelihood that the natal den was located nearby. The lack of photographs captured at the rural site during the late spring/summer may also be related to parturition. For example, fox photographs at this site would decline in late spring/summer if the camera was not located near the natal den. We did not record Coyote activity as commonly as Gray Fox at the sites we monitored. Coyote activity was observed at the suburban site on two occasions (5 May and 16 June 2011). Interestingly, all photographs of suburban Coyotes were followed soon after by photographs of Gray Foxes (one within 20 min, and the other within 2.25 hr). This would suggest that either these Coyotes were not resident individuals that defended territories and excluded local Gray Foxes, or the presence of these Coyotes did not influence the habitat selection of the suburban Gray Foxes we monitored. In contrast, a Coyote was photographed once on the rural property (26 February 2011), and no further Gray Fox activity was recorded at this location for 10 days. Our observations provide more information on the ecology of Gray Foxes in suburban landscapes. However, we support the recommendation by Riley and 514 Southeastern Naturalist Vol. 11, No. 3 White (2010) that more research is needed on this subject. Because this species prefers brushy cover or wooded habitat, it is possible they will persist in developed areas if mature woody buffers are present. Therefore, we suggest that, in addition to estimates of building density, future research include detailed landscape- level habitat analyses. Such research should include investigations of how the maturity and proportion of vegetation in natural buffers available to suburban Gray Foxes influences their ecology. It would also be valuable to relate data on fox habitat use and movement to fox population estimates for the surrounding landscape. Such population information for our study region does not currently exist. Annual trapping records for Gray Fox and Coyotes in Alamance County show a general increase in numbers harvested over the last five years (Table 3; Colleen Olfenbuttel, North Carolina Wildlife Resources Commission, Pittsboro, NC, pers. comm.). For example, during 2005/2006, 43 Gray Fox and 3 Coyotes were harvested. These numbers steadily increased until 2010/2011, during which 179 Gray Fox and 87 Coyotes were harvested (Table 3). However, this data should be viewed cautiously. Trappers are not required to report harvest totals for these species in North Carolina, and this increase could be indicative of greater survey participation by trappers over time. The legal trapping season for fox and Coyote in Alamance County was also increased by several months in 2008, which likely influenced the total number harvested. However, the numbers of each species harvested showed a general increase prior to 2008 (Table 3). If these data represent an actual increase in Gray Fox and Coyote populations in the region, there may be implications for how these species use suburban/urban habitat. For example, higher fox and Coyote population densities, coupled with loss of habitat due to suburban sprawl, could force individuals from preferred habitat into areas that would otherwise be avoided. Further research on this topic is warranted, particularly in the southeastern US. Camera traps are a relatively new and effective tool for studying wildlife, particularly for medium and large mammals. They provide a non-invasive method for meeting a variety of research needs, such as species inventories, behavioral studies, and population estimates (see multiple chapters in O’Connell et al. 2011). Camera traps are also especially useful for studying Table 3. Annual harvest totals for Gray Foxes (Urocyon cinereoargenteus) and Coyotes (Canis latrans) for Alamance County, NC. Totals based on voluntary trapper reports received from 2005/2006 through 2010/2011 trapping seasons (data provided by Colleen Olfenbuttel, North Carolina Wildlife Resources Commission, Pittsboro, NC). Trapping season Species 2005/2006 2006/2007 2007/2008* 2008/2009 2009/2010 2010/2011 Gray Fox 43 62 77 92 106 179 Coyote 3 11 23 34 58 87 *Note: after 2007/2008, the trapping season for fox and Coyote in Alamance County was lengthened from October 1st through January 31st to June 1st through February 28th. 2012 J.M. Kapfer and R.W. Kirk 515 cryptic or wary species that are challenging to survey by other means (O’Connell et al. 2011). As a result, camera traps are often employed in studies focused on rare species (e.g., Janecka et al. 2011, Mohd-Azlan and Sanderson 2007). We found they are equally as effective in studying common species, such as Gray Fox in our study areas. Camera traps can be deployed long-term in a relatively inexpensive fashion, they survey a particular location continuously once deployed, and they reduce the likelihood of injury to animals and researchers that may result from traditional trapping methods. Acknowledgments We thank Elon University for providing the funds that purchased the camera traps used during our study and granting us access to the rural site for monitoring. We also thank J.J. Kapfer and V.E. Kapfer for providing information on visual observations of Gray Foxes from the suburban site. D. Muñoz and M. Forster helped maintain rural camera traps. The University of Wisconsin-Whitewater and Elon University provided funding to support the publication of this manuscript. Literature Cited Cypher, B.L. 2003. Foxes. Pp. 511–546, In G.A. Feldhammer, B.C. Thompson, and J.A. Chapman (Eds.). Wild Mammals of North America: Biology, Management, and Conservation, 2nd Edition. Johns Hopkins University Press, Baltimore, MD. 1216 pp. Fedriani, J.M., T.K. Fuller, R.M. Sauvajot, and E.C. York. 2000. Competition and intraguild predation among three sympatric carnivores. Oecologia 125:258–270. Gehrt, S.D., and S.P.D. Riley. 2010. Coyotes (Canis latrans). Pp. 79–96, In S.D. Gehrt, S.P.D. Riley, and B.L. Cypher (Eds.). Urban Carnivores: Ecology, Conflict, and Conservation. Johns Hopkins University Press, Baltimore, MD. 304 pp. Gehrt, S.D., S.P.D. Riley, and B.L. Cypher (Eds.). 2010. Urban Carnivores: Ecology, Confl ict, and Conservation. Johns Hopkins University Press, Baltimore, MD. 304 pp. Harrison, R.L. 1993. A survey of anthropogenic ecological factors potentially affecting Gray Foxes (Urocyon cinereoargenteus) in a rural residential area. The Southwestern Naturalist 38:352–356. Harrison, R.L. 1997. A comparison of Gray Fox ecology between residential and undeveloped rural landscapes. Journal of Wildlife Management 61:112–122. Janecka, J.E., B. Munkhtsog, R.M. Jackson, G. Naranbaatar, D.P. Mallon, and W.J. Murphy. 2011. Comparison of noninvasive genetic and camera-trapping techniques for surveying Snow Leopards. Journal of Mammalogy 92:771–783. Mohd-Azlan, J., and J. Sanderson. 2007. Geographic distribution and conservation status of the Bay Cat, Catopuma badia, a Bornean endemic. Oryx 41:394–397. O’Connell, A.F. J.D. Nichols, and K.U. Karranth (Eds.). 2011. Camera Traps in Animal Ecology: Methods and Analyses. Springer Publishing, New York, NY. 280 pp. Riley, S.P.D. 2006. Spatial ecology of Bobcats and Gray Foxes in urban and rural zones of a national park. Journal of Wildlife Management 70:1425–1435. Riley, S.P.D., and P.A. White. 2010. Gray Foxes (Urocyon cinereoargenteus). Pp. 197– 200, In S.D. Gehrt, S.P.D. Riley and B.L. Cypher (Eds.). Urban Carnivores: Ecology, Conflict, and Conservation. Johns Hopkins University Press, Baltimore, MD. 304 pp. 516 Southeastern Naturalist Vol. 11, No. 3 Riley, S.P.D., J. Foley, and B. Chomel. 2004. Exposure to feline and canine pathogens in Bobcats (Lynx rufus) and Gray Foxes (Urocyon cinereoargenteus) in urban and rural zones of a national park in California. Journal of Wildlife Diseases 40:11–22. Rountree, G.H., III. 2004. Comparative study of the home range and habitat usage of Red Foxes and Gray Foxes in an urban setting: A preliminary report. Pp. 238–244, In W.W. Shaw, L.K. Harris, and L. VanDruff (Eds.). Proceedings of the 4th International Urban Wildlife Symposium. May 1–5, 1999, Tucson, AZ. College of Agriculture and Life Science, Univeristy of Arizona-Tuscon, Tuscon, AZ. 368 pp. Temple, D.L., M.J. Chamberlain, and L.M. Conner. 2010. Spatial ecology, survival, and cause-specific mortality of Gray Foxes (Urocyon cinereoargenteus) in a Longleaf Pine ecosystem. American Midland Naturalist 163:413–422.