INTRODUCTION

Ground-dwelling squirrels vary greatly among species in their degree of sociality, ranging from asocial to highly social (Armitage, 1981; Michener, 1983). Social squirrels tend to be large-bodied species, in which delayed maturity (>1 y old) leads to delayed dispersal (al y old) , thereby facilitating formation of social groups through philopatry of female offspring (Armitage, 1981). In contrast, small-bodied squirrels exhibit rapid development and early dispersal, during the summer of birth, hence these species are typically asocial (Armitage, 1981).

The golden-mantled ground squirrel (Callospermophilus lateralis) is a small-bodied species (230-240 g at onset of hibernation) that is considered asocial (Armitage, 1981; Michener, 1983), a classification characterized by nonsharing of territories by males and females, agonistic interactions and early dispersal, with juveniles dispersing during the summer of birth (Michener, 1983). However, this classification is considered provisional because behavior of the species is poorly known; asocial species attract little attention from researchers (Michener, 1983; Ferron, 1985). In the only study to address the social behavior of golden-mantled ground squirrels under natural conditions, Ferron (1985) reported high levels of agonistic interactions, consistent with an asocial classification. However, Ferron (1985) documented some sharing of space by adults and suggested that instead of being territorial, golden-mantled ground squirrels might maintain exclusive use of only a core area of their home ranges and share more peripheral areas with other conspecifics. No studies of dispersal in golden-mantled ground squirrels have been published.

We studied the spatial organization of adult female golden-mantied ground squirrels. We quantified home range size and overlap to determine if exclusive use is expressed for the entire home range or only the core area, and we studied dispersal to determine if juveniles emigrate during their first summer.

MATERIALS AND METHODS

The golden-mantled ground squirrel is a hibernating species that is distributed across the mountainous regions of western North America, including the northern Sierra Nevada, from 1200 m elevation to above tree line (McKeever, 1964). Sexual maturity is reached at 1 y of age, mating occurs soon after emergence from hibernation, which typically occurs at the time of snow melt, and young usually emerge from the natal burrow in Jul. (Bronson, 1980; Bartels and Thompson, 1993).

Our study was conducted May through Oct. 2003-2005 in the Plumas National Forest (UTM 10S E 686864, N 4430498), near Quincy, California, USA, at an elevation of -2100 m. The habitat in the 1 6-ha study area was a conifer forest dominated by red-fir (Abies magnifica) and western white pine (Pinus monticola), with some lodgepole pine (Pinus contorta). The understory consisted of pinemat manzanita (Arctostaphylus nevadensis), herbaceous vegetation, rocky outcrops and coarse woody debris.

Squirrels were captured with Tomahawk live traps (Model 201, 40.6 X 12.7 X 12.7 cm, Tomahawk Live Trap, Tomahawk, Wisconsin, USA) . Traps were baited with rolled oats and black oil sunflower seeds coated with peanut butter, set in the early morning and checked at mid-morning and noon. Captured squirrels were permanently identified with a tag in each ear (#1005-1, National Band and Tag, Newport, Kentucky, USA) and sex was determined. Squirrels were chemically immobilized with an intramuscular injection of ketamine hydrochloride (100 mg/ml HCl) and fitted with 4-g radiocollars (Holohil Systems, Carp, Ontario, Canada; Model PD-2C) with an expected life of 6 mo, then allowed to fully recover before release at the site of capture. Squirrels were recaptured periodically to adjust collar size, and collars were removed shortly before entry into hibernation. All handling protocols were approved by the Animal Care and Use Committee at die University of California, Davis and met guidelines recommended by the American Society of Mammalogists (Gannon et al., 2007).

We determined home ranges of adult (si y old) females during summer 2003. During Jul. we captured 12 adult females, fitted them with radiocollars and radiotracked them until shortly before they entered hibernation. Hence, our results reflect home ranges during the post-weaning period. Reproductive status of the 12 females was not known. We believe these females constituted most, but not all, of the adult females in the study area. All females survived and entered hibernation. Diurnal locations of females were determined by triangulation using Locate II and maximum likelihood estimation (Nams, 2000). Triangulations were conducted three to four times daily for ≥5 d per month from Jul. through entry into hibernation in Oct., with at least 2 h between locations for each squirrel (Swihart et al, 1988), using a radiotelemetry receiver (Model R-1000, Communications Specialists, Orange, California, USA) and antenna (AF Antronics Inc., Urbana, Illinois, USA) . Triangulation was conducted by two technicians working in a synchronized manner to collect three to six signal bearings over a short period of time (<10 min).

Home ranges (95%) and core areas (50%) of adult females were estimated in Ranges6 (Kenward et al, 2003) using the fixed kernel method (FK) (Worton, 1989); in addition, we estimated home ranges using the 95% minimum convex polygon method (MCP) (Mohr, 1947). MCP and FK were estimated using the arithmetic mean (Hayne, 1949) and least squares cross validation (Seaman and Powell, 1996; Seaman et al, 1999) methods, respectively. Incremental area analyses were conducted to determine if locations for individual squirrels reached an asymptotic plateau (Kenward, 2001); all 12 squirrels met this criterion, with a minimum of 23 locations needed to reach an asymptote (mean 32.6 ± 7.5 sd) . A median smoothing parameter of 0.79 was used to estimate FK home ranges (Kenward, 2001).

We assessed home range overlap and dynamic interaction between members of all dyads of adjacent adult females. To evaluate home range overlap we used the spatial overlap index of Minta (1992), which ranges from 0 (no overlap) to 1 (total overlap) and was calculated using both the 95% home range and the 50% core area. We used FK results for these analyses because FK is considered to be less biased than MCP (Jennrich and Turner, 1969; Kenward, 2001). We evaluated dynamic interaction using Jacob's index, which assesses attraction or avoidance by comparing synchronous observed distances with all possible observed distances between dyad member locations. Jacob's index ranges -1 to +1, with -1 indicating avoidance and +1 indicating attraction (Macdonald et al, 1980; Walls and Kenward, 2001). For dynamic interactions we used paired locations for each dyad that were recorded within a 2 h time span. We expected that the territoriality and agonistic behavior hypothesized for golden-mantled ground squirrels (Armitage, 1981; Michener, 1983) would result in little or no overlap in home ranges and evidence of avoidance by neighboring squirrels.

To study dispersal, we used radiotelemetry to locate juveniles immediately after emergence from the natal burrow and also to track them thereafter. We recaptured seven adult females during early Jun. of 2004 that were studied during 2003 and captured an additional nine adult females during 2005. We fitted adult females with radiocollars and used homing to identify the burrows of those that were reproductive. Homing involved walking in the direction where the radio signal was loudest until the source of the signal was identified. We captured 21 newly weaned juveniles (12 male, 9 female) from 10 litters by placing traps around the entrances to natal burrows, and fitted them with radiocollars. All juveniles were radiotracked a5 d per month using homing; one female was killed by a predator and hence excluded from analysis, whereas the remainder survived the summer and entered hibernation in Oct.. We defined dispersal as a one-way movement away from the natal burrow that exceeded one home range radius, which was calculated by averaging the FK home range size of adult females and assuming a circular shape. Dispersal distance was calculated as the linear distance between the natal burrow and the final location (hibernacula) for juveniles that dispersed. Some juveniles made exploratory excursions, defined as a brief, round-trip movement more than one home range radius from the natal burrow. All statistical analyses were performed using JMP version 7 (JMP, 2007).

RESULTS

We obtained a mean of 42.5 locations (range = 28-57) for 12 adult female goldenmanded ground squirrels during 2003. Mean home ranges based on 95% MCP (Sc= 3.57 ha, range = 1.05-9.32) were similar to those based on 95% FK (x = 3.75 ha, range = 1.1413.52). Core areas based on 50% FK averaged 1.03 ha (range = 0.25-4.24). Analysis of spatial overlap indicated considerable overlap in 95% FK home ranges among females (n = 12, x= 0.31 ± 0.20 sd, range = 0.00-0.69; Fig. 1). However, we found very little overlap in 50% FK core areas (n = 12, ? = 0.03 ± 0.08 SD, range 0.00-0.43; Fig. 1). Analysis of dynamic interactions revealed Jacob's index scores that did not differ from zero (Wilcoxon sign-rank test, P > 0.05) for either the 95% FK home range (n = 108, S = 0.003 ± 0.052 sd) or the 50% FK core area (n = 36, ? = -0.001 ± 0.067 sd), indicating that dynamic interactions were neutral.

Some juvenile females (two of eight) remained philopatric and hibernated within the 109 m distance we used to define dispersal (median = 75 m, range = 70-79 m), and the remaining six individuals dispersed a median of 204 m (range = 124-283 m). Most juvenile males dispersed (10 of 12) and setded a median of 236 m (range = 153-1060 m) from their natal burrow; the two philopatric males setded a median of 34 m (range = 17-52 m) from their natal burrows. Although some males moved longer distances than females, dispersal distances did not differ between males and females (Mann-Whitney test, P = 0.25) . Most juveniles (14 of 20), including both dispersers and those that remained philopatric, made exploratory excursions, averaging 1.9 per juvenile (range = 0-9) and extending as far as 1142 m. Juveniles often made exploratory excursions to locations to which they later dispersed.

DISCUSSION

Our study was constrained by small sample sizes and limited duration; we studied 12 adult females during a portion of the active season of 1 y. Nonetheless, our results indicate that adult female golden-mantled ground squirrels did not defend their entire home range as a territory, since we found substantial spatial overlap with neighboring females. Further, our mean overlap value may have been an underestimate because we did not radiocollar all adult females in our study area. However, core areas showed litde or no overlap among females, supporting Ferron's (1985) suggestion that territoriality in this species is expressed only for die inner portion of the home range. We did not know relatedness of adults in our study, but sharing of space by adjacent females may reflect kinship, a pattern that has been shown for both asocial (Maher, 2009) and social (Armitage, 1996) species of grounddwelling squirrels. Analysis of dynamic interactions indicated no evidence of attraction among adult females, consistent with Ferron's (1985) finding that cohesive behaviors among adults were restricted almost exclusively to male-female interactions. It was somewhat surprising that dynamic interactions also provided no evidence of avoidance, since agonistic and avoidance behaviors dominate interactions among adult females (Ferron, 1985). Kinship may have played a role in the lack of avoidance; close kin in other species show reduced agonistic interactions compared with nonrelatives (e.g., Armitage, 1989), and some of our dyads may have been close relatives.

Similar patterns of space sharing were shown by the other two species of ground-dwelling squirrels classified as asocial, Franklin's ground squirrel (Poliocitellus franklinii) and woodchucks (Marmota monax) (Armitage, 1981; Michener, 1983). Chormanski-Norris et al (1989) found that home ranges of Frankin's ground squirrels showed some overlap both within and between sexes. Woodchuck spatial organization is variable, with evidence of territoriality but also some overlap in home ranges that varies temporally and spatially (Ferron and Ouellet, 1989; Meier, 1992; Swihart, 1992; Maher, 2004).

As expected, most golden-mantled ground squirrels dispersed during their first summer. However, some juveniles, including both males and females, did not disperse until at least their yearling summer. Our estimate of the frequency of philopatry (20%) may have been conservative, since we used a more restrictive definition (home range radius) than the typical criterion of one home range diameter used in other studies (e.g., Maher, 2009). Martin and Heske (2005) found a similar pattern for Franklin's ground squirrels; dispersal activity was most pronounced late in the summer of birth, but some juveniles hibernated close to their natal area. Like golden-mantled ground squirrels, Franklin's ground squirrels made extensive exploratory excursions during the dispersal period (Martin and Heske, 2005), likely to allow assessment of prospects outside the natal area (Johnson, 1989). For woodchucks, the pattern of early dispersal is not as strong. Many juveniles dispersed during their first summer, but a substantial portion (35-55%) hibernated in their natal area (Meier, 1992; Maher, 2009) with some females becoming philopatric residents as adults (Maher, 2009).

Our results, along with those of Ferron (1985), provide support for the classification of golden-mantled ground squirrels as asocial: interactions are mostly agonistic, females appear to maintain areas of exclusive use and most dispersal occurs during the juvenile summer. However, results for all three species of asocial squirrels suggest that the expression of territoriality and early dispersal may be variable. Moreover, evidence of philopatry in females, combined with some overlap in home ranges, may indicate opportunities for kin associations and the benefits that result from them (Waser and Jones, 1983; Maher, 2009).

Acknowledgments. - We thank R. LeChalk, J. Csakany, A. Goldman, M. Gilbart, C. Morcos, H. Robertson and D. Haggerty for help with fieldwork, J. Smith for technical support and P. Stine for his administrative and moral support of the study. This work was funded by the Joint Fire Sciences Program and the United States Department of Agriculture, Forest Service (Region 5).

LITERATURE CITED

ARMITAGE, K. B. 1981. Sociality as a life history tactic of ground squirrels. Oecologia, 48:36-49.

_____. 1989. The function of kin discrimination. Ethol. Ecol. Evol., 1:111-121.

_____. 1996. Resource sharing and kinship in yellow-bellied marmots, p. 129-134. In: M. Le Berre, R. Ramousse and L. Le Guelte (eds.). Biodiversity in marmots. International Network on Marmots, Lyon, France.

BARTELS, M. A. AND D. P. THOMPSON. 1993. Spermophilus lateralis. Mammal. Spec, 440:1-8.

BRONSON, M. T. 1980. Altitudinal variation in emergence time of golden-manded ground squirrels (Spermophilus lateralis) . J. Mammal, 61:124-126.

CHORMANSKI-NORRIS, J., E. K. FRITZELL AND A. B. SARGEANT. 1989. Movements and habitat use of Franklin's ground squirrels in duck-nesting habitat/. WMIf Manage., 53:324-331.

FERRON, J. 1985. Social behavior of the golden-manded ground squirrel (Spermophilus lateralis). Can. J. Zool, 63:2529-2533.

_____ AND J. P. OUELLET. 1989. Temporal and sexual variations in the use of space with regard to social organization in die woodchuck (Marmota monax). Can. J. Zool, 67:1642-1649.

GANNON, W. L., R. S. SIKES, THE ANIMAI. CARE AND USE COMMITTEE OF THE AMERICAN SOCIETY OF MAMMALOGISTS. 2007. Guidelines of the American Society of Mammalogists for the use of wild mammals in research./. Mammal, 88:809-823.

HAYNE, D. W. 1949. Calculation of size of home range./ Mammal, 30:1-18.

JENNRICH, R I. AND F. B. TURNER. 1969. Measurement of non-circular home range terrestrial vertebrates./ Theor. Biol, 22:227-237.

JMP. 2007. Version 7. SAS Institute Inc., Cary, North Carolina.

JOHNSON, M. L. 1989. Exploratory behavior and dispersal: a graphical model. Can. J. Zool, 67:2325-2328.

KENWARD, R. E. 2001. A manual for wildlife radio tagging, 2nd ed. Academic Press, London, U.K.

_____, A. B. SOUTH AND S. S. WALLS. 2003. Ranges6. Wareham, U. K

MACDONALD, D. W., F. G. BALL AND N. G. HOUGH. 1980. The evaluation of home range size and configuration using radio tracking data, p. 405-424. In: C. Amlaner and D. Macdonald (eds.). A Handbook on Biotelemetry and Radio Tracking. Pergamon Press, Oxford, U.K

MAHER, C. R. 2004. Intrasexual territoriality in woodchucks (Marmota monax). J. Mammal, 85:1087-1094.

_____. 2009. Genetic relatedness and space use in a behaviorally flexible species of marmot, the woodchuck (Marmota monax). Behav. Ecol. Soaobiol., 63:857-868.

MARTIN, J. M. AND E. J. HESKE. 2005. Juvenile dispersal of Franklin's ground squirrel (Spermophilus franklinii) from a prairie "island." Am. Midi Nat, 153:444-449.

MCKEEVER, S. 1964. The biology of the golden-manded ground squirrel (Citellus lateralis). Ecol. Monogr., 34:383-406.

MEIER, P. T. 1992. Social organization of woodchucks (Marmota monax). Behav. Ecol. Soaobiol, 31:393-400.

MICHENER, G. R. 1983. Kin identification, matriarchies, and die evolution of sociality in ground-dwelling sciurids, p. 528-572. In: J. Eisenberg and D. Kleiman (eds.). Advances in the Study of Mammalian Behavior. Special publication No. 7. The American Society of Mammalogists. Shippensburg, Pennsylvania.

MINTA, S. C. 1992. Tests of spatial and temporal interaction among animals. Ecol. Appi, 2:178-188.

MOHR, C. O. 1947. Table of equivalent populations of North American small mammals. Am. Midi. Nat, 37:223-249.

NAMS, V. O. 2000. Locate II user's guide. Pacer Computer Software, Tatamagouche, Nova Scotia, Canada.

SEAMAN, D. E. AND R. A. POWELL. 1996. An evaluation of the accuracy of kernel density estimators for home range analysis. Ecology, 77:2075-2085.

_____,J.J. MILLSPAUGH, B. J. KERNOHAN, G. C. BRUNDIGE, RJ. RAEDEKE AND R. A. GITZEN. 1999. Effects of sample size on kernel home range estimates./ WMl. Manage., 63:739-747.

SWIHART, R. K., N. A. SLADE AND B.J. BERGSTROM. 1988. Relating body size to the rate of home range use in mammals. Ecology, 69:393-399.

_____. 1992. Home-range attributes and spatial structure of woodchuck populations. / Mammal, 73:604-618.

WALLS, S. S. AND R. E. KENWARD. 2001. Spatial consequences of relatedness and age in buzzards. Anim. Behav., 61:1069-1078.

WASER, P. M. AND W. T.JONES. 1983. Natal philopatry among solitary mammals. Q. Rev. Biol, 58:355-390.

WORTON, B. J. 1989. Kernel methods for estimating the utilization distribution in home-range studies. Ecology, 70:164-168.

SUBMITTED 8 MARCH 2010 ACCEPTED 21 AUGUST 2010

Author affiliation:

BRETT R. JESMER1 and DIRK H. VAN VUREN

Department of Wildlife, Fish, and Conservation Biology, University of California,

One Shields Avenue, Davis, California 95616

JAMES A. WILSON

Department of Biology, University of Nebraska-Omaha, 60th and Dodge Streets, Omaha, Nebraska 68182

DOUGLAS A. KELT

Department of Wildlife, Fish, and Conservation Biology, University of California,

One Shields Avenue, Davis, California 95616

AND

MICHAEL L.JOHNSON

Watershed Sdences Center, University of California, One Shields Avenue, Davis, California 95616

Author affiliation:

1 Corresponding author: e-mail: brjesmer@ucdavis.edu