SPATIAL ORGANIZATION OF A REINTRODUCED POPULATION OF BOBCATS

Journal of Mammalogy, 87(2):394 401, 2006 SPATIAL ORGANIZATION OF A REINTRODUCED POPULATION OF BOBCATS DUANE R. DIEFENBACH,* LESLIE A. HANSEN, ROBERT J. WARREN, AND MICHAEL J. CONROY D. B. Warnell School of Forest Resources, University of Georgia, Athens, GA 30602, USA (DRD, LAH, RJW) United States Geological Survey, Georgia Cooperative Fish and Wildlife Research Unit, D. B. Warnell School of Forest Resources, University of Georgia, Athens, GA 30602, USA (MJC) Present address of DRD: United States Geological Survey, Pennsylvania Cooperative Fish and Wildlife Research Unit, Pennsylvania State University, 113 Merkle Lab, University Park, PA 16802, USA Present address of LAH: Los Alamos National Laboratory, P.O. Box 1663, MS M887, Los Alamos, NM 87544, USA The spacing patterns and mating systems of solitary carnivores have important implications for social behavior and for the survival and reproduction of individuals. Over 2 years, we reintroduced 32 (15 males and 17 females) bobcats (Lynx rufus) to a barrier island off the coast of Georgia and studied patterns of bobcat spatial distribution. Population density increased to 3.1 bobcats/10 km 2. We found overlap of the home range for all females on the island increased during 1989 1991 such that, on average, each female shared a home-range area with the equivalent of.2 other females, and for core areas overlap was equivalent to sharing a core area with nearly 1 other female. Reproduction and home-range overlap were related inversely and food resources did not seem to be limiting. Our results were consistent with the land tenure concept in that the initial reintroduced bobcats established home ranges that changed little in size and location. However, bobcats resident on the island for 1 year did not successfully exclude newcomers from their home ranges or core areas and no bobcats retained areas of exclusive use from conspecifics of the same sex. We suggest that the propensity of female bobcats to reproduce successfully may be related to their access to exclusive use areas even under conditions of adequate or good food availability. Under the conditions in this study (moderate bobcat density, adequate food availability, and limited dispersal) bobcats exhibited no evidence of an ability to exclude other adult individuals from their home ranges or core areas. Key words: bobcat, home range, land tenure, Lynx rufus, population regulation, reintroduction, social organization Bobcats (Lynx rufus), like most felids, are essentially solitary with direct social interactions only occurring among females with kittens and between males and females during the breeding season (Packer 1986). Anderson and Lovallo (2003) reviewed published literature on bobcats and reported that all studies found significant intersexual overlap of home ranges and varying degrees of intrasexual overlap. Generally, females are more likely to have nonoverlapping home ranges, whereas home ranges of males may have substantial intrasexual overlap and encompass the home ranges of multiple females. The degree to which bobcats are territorial is not well understood, because documented incidences of aggression are lacking (Benson et al. 2004) and movements of bobcats only can be monitored indirectly via radiotelemetry. We define a * Correspondent: ddiefenbach@psu.edu Ó 2006 American Society of Mammalogists www.mammalogy.org home range to be the area traversed by an animal in the course of its normal activities (sensu Burt 1943), and define a territory to be a defended area to the exclusion of conspecifics (Begon et al. 1990). It is believed that bobcats passively exclude conspecifics via deposition of scats and scent marking, and rarely rely on physical confrontation (Anderson and Lovallo 2003; Bailey 1974). The pattern in which bobcats secure and retain home-range areas has been described as a system of prior rights and land tenure (Bailey 1974, 1981), whereby the 1st bobcat to occupy an area is therein successful in excluding other bobcats (e.g., neighboring conspecifics or dispersers) and maintains the rights to this area. This pattern has been observed in males and females (Anderson and Lovallo 2003). Anderson (1988) confirmed this relationship by experimentally removing a freeranging male bobcat from its home range and observed a neighboring male bobcat shift its home range into the vacancy created by the experimental removal. Nielsen and Woolf (2001) and Chamberlain and Leopold (2001) reported that core areas of bobcat home ranges, especially those of females, were 394

April 2006 DIEFENBACH ET AL. SOCIAL ORGANIZATION OF BOBCATS 395 nearly exclusive to conspecifics of the same sex. These observations suggest that bobcats can be territorial in the sense that they generally maintain areas to exclude conspecifics of the same sex. Studies of bobcats have reported increases in home-range size and intrasexual overlap with reduced prey density (Buie et al. 1979; Knick 1990; Litvaitis et al. 1986). Others have suggested that only low to moderate population densities would result in little overlap among conspecifics of the same sex (Buie et al. 1979; Rolley 1983; Zezulak and Schwab 1979), although studies with greater densities have reported exclusive male or female home ranges (Lembeck and Gould 1979; Miller and Speake 1979). If territoriality permits individuals to control food resources to the exclusion of conspecifics, then overlap in home ranges is predicted to be least when it is possible to defend food resources, and greatest when either food is superabundant or scarce (Maher and Lott 2000; McLoughlin et al. 2000). If food resources limit reproduction, then bobcats should reproduce only when food resources are abundant, regardless of their ability to defend territories. In contrast, the infanticide hypothesis (Wolff 1997) predicts that reproduction will occur only when females can maintain areas of exclusive use and defend it from conspecific females. Nearly all details regarding bobcat social organization and spatial distribution have been obtained by studying patterns of home-range use by capturing free-ranging bobcats and monitoring their movements via radiotelemetry. Consequently, all studies have been fraught with the problem that an unknown proportion of the population was monitored. Anderson (1987:15) noted that the true relationship of bobcat density to degree of overlap may be obscured by sampling problems. Thus, inferences regarding the spatial distribution of bobcats may be incorrect in studies that failed to monitor all individuals of the population. We reintroduced bobcats to Cumberland Island, Georgia, from which bobcats had been extirpated in the early 1900s (Diefenbach et al. 1993). Cumberland Island has a warm, temperate to subtropical climate that should be able to sustain a bobcat population because there were no terrestrial mammalian predators on the island (except the raccoon [Procyon lotor]) before our reintroduction and populations of preferred bobcat prey were present. We monitored nearly all bobcats in the population to determine reproduction and intrasexual overlap with minimal confounding effects of monitoring an unknown proportion of the population. We used the opportunity of having nearly all bobcats in the population radiocollared to observe space-use patterns and assess if they were consistent with the concepts of prior rights and land tenure. If these concepts accurately describe bobcat social organization, we predicted that bobcats would have very limited intrasexual overlap of core-use areas, and that bobcats released during the 2nd year of the reintroduction would be excluded from areas already containing bobcats of the same sex released during the 1st year of the reintroduction. MATERIALS AND METHODS Study area. Cumberland Island, a coastal barrier island 0.5 km north of the Georgia Florida border (308489160N, 818279360W), was designated a National Seashore in 1972 (Public Law 92-536) to be administered by the United States Department of the Interior, National Park Service. The island is 25 km long, varies in width from 1 to 6 km, and is separated from the Georgia mainland by 2 4 km of salt marsh and open water. To the north, separated,0.25 km by salt marsh and a tidal creek, is Little Cumberland Island, which is relatively undeveloped and contains the same habitats. The eastern shores of the islands face the Atlantic Ocean. Behind the primary dune is an interdune meadow, 200 m in width and dominated by waxmyrtle (Myrica cerifera). The interior forests of the islands are composed of live oak (Quercus virginianus) and pines (Pinus) with much of the understory dominated by monoculture stands of sawtooth palmetto (Serenoa repens). Freshwater wetlands generally follow natural depressions between former dune ridges in the interior of the islands. The western edges of the islands are salt marsh. Cumberland Island contains 5 major vegetation associations covering 84.5 km 2 : sandy beach and interdune meadow (14.7 km 2 ); maritime forest, including lowland hardwoods (38.7 km 2 ); scrub shrub thickets that developed after natural fires (7.0 km 2 ); freshwater wetlands (6.7 km 2 ); and salt marsh (17.4 km 2 ). Little Cumberland Island is 9.0 km 2 and contains sandy beach and interdune meadow (1.3 km 2 ), maritime forest (4.8 km 2 ), and salt marsh (4.8 km 2 ). Hereafter, we refer to the 2 islands together as Cumberland Island. The climate is warm temperate to subtropical, with normal mean temperatures ranging from 128C in January to 288C in July (Johnson et al. 1974). The average annual rainfall is 134 cm, with the wettest months being June through September (X ¼ 16 cm/month) and the driest months being October, November, and April (X ¼ 7 cm/month). Hillestad et al. (1975) described the study area in detail. Historical records indicated that bobcats were likely indigenous to the island and were extirpated around 1907 (Diefenbach et al. 1993). The National Park Service allows limited public hunts on the island for white-tailed deer (Odocoileus virginianus) and feral swine (Sus scrofa). Potential prey species on the island were feral swine, whitetailed deer, hispid cotton rats (Sigmodon hispidus), cotton deermice (Peromyscus gossypinus), eastern gray squirrels (Sciurus carolinensis), marsh rabbits (Sylvilagus palustris), and 9-banded armadillos (Dasypus novemcinctus). Reintroduction and monitoring of bobcats. Bobcats reintroduced to the island were adults captured on the coastal plain of Georgia, and the translocation methods were described in detail by Diefenbach et al. (1993). During late September through December of 1988 and 1989 we released 3 groups of 4 6 bobcats approximately every 30 days at locations throughout the island, for a total of 32 bobcats. Details on the release procedure were described in Diefenbach et al. (1993). All bobcats were adults (.1 year old) and each was fitted with a radiocollar before its release on the island. We conducted trapping on the island to recapture bobcats to assess their physical condition and to capture and radiocollar bobcats born on the island. We followed guidelines established by the American Society of Mammalogists in the care and handling of bobcats (Animal Care and Use Committee 1998). We monitored the survival and movements of bobcats via triangulation of radiosignals from the ground or by determining the location of signals from fixed-wing aircraft. We located bobcats throughout the year, recorded x y locations via Universal Transverse Mercator meter coordinates, and attempted to obtain locations throughout the 24-h day (James 1992). We analyzed only locations obtained at times separated by 24 h because analysis of autocorrelation among locations (Swihart and Slade 1985) indicated locations separated by 24 h were reasonably independent. We monitored reproduction by conducting intensive radiotelemetry monitoring of female bobcats during the denning season. Once a

396 JOURNAL OF MAMMALOGY Vol. 87, No. 2 female exhibited restricted movements and repeated visits to the same location, 3 people would search for the den. Females with kittens usually remained at the den site until approached to within a few meters. A more detailed description of methods is provided by Ragsdale (1993). We calculated population density as the number of bobcats on the island divided by area of the island. In 1989, adult population size (density) was known, but by 1990 kittens born on the island in 1989 were 1 year old and adult population size had to be estimated. Therefore, for 1990 and 1991 we used estimates of reproduction and survival to estimate population size (see Diefenbach 1992). We analyzed locations for each bobcat on an annual basis. Although we reintroduced bobcats late in the years 1988 and 1989 (October December), we refer to them as the 1989 reintroduction and 1990 reintroduction, respectively, because we used only bobcat locations collected after 1 January 1989 from individuals reintroduced in 1988, and similarly, we used only locations collected after 1 January 1990 from bobcats reintroduced in 1989. We included bobcats born in 1989 in our analyses beginning in 1991. Stability of home ranges. To assess the stability of home ranges established by bobcats on the island, we assessed changes in homerange size, changes in overlap of home-range boundaries among years, and shifts in the centroid of locations among years. To test for changes in home-range size of bobcats over years, we calculated the average distance between all possible pairs of locations (d) for each individual bobcat for each year (Cade and Richards 2001). We used this metric of home-range size rather than a traditional home-range estimator (e.g., minimum convex polygon) because statistics based on Euclidean distances have greater power to detect changes in distribution patterns (Cade and Richards 2001). Moreover, analyses based on home-range areas disregard information about the nonuniform use of area within the home-range boundary. We used d in a repeated-measures analysis of variance (ANOVA; PROC GLM, REPEATED statement, SAS Institute Inc. 1999) to test for changes in home-range dispersion, with sex of bobcats as a group effect, over time. A sphericity test was used to test if the covariance matrix satisfied the Huynh Feldt condition for univariate tests. If the sphericity test was rejected, we used probabilities for F-statistics adjusted by Huynh Feldt epsilon (a ¼ 0.10). We tested for changes in annual home-range size over 1989, 1990, and 1991 for the 1989 reintroduction, and changes in annual home-range size between 1990 and 1991 for the 1990 reintroduction. Because PROC GLM requires data for all years from each bobcat, we used data only from bobcats in which we had data for all 3 years for the 1989 reintroduction and both years for the 1990 reintroduction. We calculated a 95% utilization distribution (UD) using a kernel home-range estimator for each bobcat. The UD for each bobcat was calculated using program KERNELHR (Seaman et al. 1998) and analyzed all bobcats of each sex on a single grid (160 160 m) using least-squares cross-validation to select the level of smoothing. Each grid cell within the home range contained a z value that represented relative use by the bobcat (Seaman et al. 1998, 1999). To assess homerange stability, we then calculated the percentage of the UD volume that occurred outside the home-range area of the previous year. To further assess home-range stability, we examined plots of x y locations for evidence of shifts in the center of home ranges within and between years. For locations within a year, we plotted locations sequentially and examined these plots for evidence that a given bobcat s home range may have shifted, which could bias positively our measures of home-range size. We examined the data for shifts in home-range location among years by calculating changes in the mean x y coordinate. Intrasexual overlap. To examine intrasexual overlap in the spatial distribution of bobcats, we calculated a 95% and 50% UD for each bobcat. The UD for each bobcat was calculated using program KERNELHR, in which each grid cell within the home range contained a z value that represented relative use by the bobcat (Seaman et al. 1998, 1999). We analyzed all bobcats of each sex on a single grid using least-squares cross-validation to select the level of smoothing. After excluding 5% (for 95% UD) or 50% (for 50% UD) of the grid cell volume by removing cells with the smallest z values, we transformed the z values so the 95% and 50% UDs summed to 1.0. For each bobcat for every year the bobcat was on the island, and for 95% and 50% UDs, we calculated an intrasexual overlap index (OI) OI ij ¼ Xn ðud ij kþ; k¼1 where i indexes each bobcat, j is the year, UD ij is the proportion of the utilization distribution for bobcat i in year j that is overlapped by k bobcats of the same sex. Therefore, the OI for each bobcat can range from zero (no overlap) to n (all bobcats of the same sex encompassed the same home range). We calculated the OI for each bobcat based on all other bobcats present on the island. We used the OI in a repeatedmeasures ANOVA (a ¼ 0.10) to test for changes in overlap over time and for differences between sexes. Whenever we calculated mean values of d and OI for a given sex and year, we calculated a 95% confidence interval (CI) using the following: where h represents d or OI and 95%CI ¼ð h=c; h CÞ; " sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi C ¼ exp 1:96 ln 1 þ varðhþ # h 2 : RESULTS Reintroduction and monitoring of bobcats. In 1988 we reintroduced 3 males and 11 females, of which 1 female died in January 1989, 1 female returned to the mainland in February 1989, the radiotransmitter failed on a female in March 1990, and a male died in June 1991. In September December 1989 we reintroduced an additional 18 bobcats (12 males and 6 females), of which 1 male died upon release, 1 male died in October 1990, another male died in November 1990, and a female died in July 1991. We monitored all bobcats via radiotelemetry through August 1991. Of the reintroduced bobcats that remained on the island, we monitored all but 1 female during the study. In addition, we included in our analyses data from 1991 for 2 adult males and 1 female that were born on the island in 1989. In 1989 we documented 10 kittens born in 4 litters, of which we monitored 3 from 4 to 10 months of age, captured and radiocollared 3 as adults in 1990, and recovered the carcass of 1 that died at 2 years of age. In 1990, we located 1 den with 2 kittens, and in 1991 we found no evidence that any females denned, although later in the year we observed two 3- to 4- month-old kittens that were born on the island. Recaptures of females for which we did not find dens with kittens indicated

April 2006 DIEFENBACH ET AL. SOCIAL ORGANIZATION OF BOBCATS 397 FIG. 1. Measure of home-range dispersion (d, average distance between all possible pairs of locations) and 95% confidence intervals for female bobcats reintroduced to Cumberland Island, Georgia. that they were not lactating and were unlikely to have produced young (Ragsdale 1993). Population density in 1989, after the 1st reintroduction, was 1.2 adult bobcats/10 km 2, of which male densities were 0.3 bobcats/10 km 2 and female densities were 1.0 bobcats/10 km 2.In 1990, bobcat density was 3.1 bobcats/10 km 2, of which male density was 1.5 bobcats/10 km 2 and female density was 1.6 bobcats/10 km 2. In 1991, we did not know the exact density of adult bobcats because of uncertainty regarding number of kittens born in 1989 that entered the adult population. However, bobcat density was approximately the same as in 1990 because known mortality of adults and estimated recruitment to the population were similar (Diefenbach 1992). Stability of home ranges. We failed to detect any trend over time in home-range dispersion (d) for the 1989 reintroduction (F ¼ 2.55, d.f. ¼ 2, 18, P ¼ 0.116; Figs. 1 and 2). Males had greater d values than did females (F ¼ 13.64, d.f. ¼ 1, 9, P ¼ 0.005), and no interaction was found between time and sex of bobcats (F ¼ 2.20, d.f. ¼ 2,18, P ¼ 0.149). For the 1990 reintroduction, we found a decline in d between years (F ¼ 3.98, d.f. ¼ 1, 14, P ¼ 0.066), but no difference between sexes (F ¼ 0.10, d.f. ¼ 1, 9, P ¼ 0.754) and no interaction between year and sex of bobcats (F ¼ 0.06, d.f. ¼ 1, 9, P ¼ 0.811). Mean number of locations per bobcat per year was 54.2 (SD ¼ 19.85, range ¼ 12 98); 85% of annual bobcat locations consisted of n. 30. We found no evidence for a relationship between sample size and d (r ¼ 0.006, n ¼ 53, P ¼ 0.962). Our review of locations for each bobcat each year did not suggest any shifts in their distribution within a calendar year. Nearly half (47%, n ¼ 36) of shifts in home-range centers between years were shifts of 600 m (Table 1). Males of the 1989 and 1990 reintroductions exhibited greater shifts between years in home-range centers than did females, especially between the 1st and 2nd year on the island. No females shifted their home-range center.2.9 km, and only 4 of 21 shifted their home-range center.1 km for all female bobcats and years. In no instance did a bobcat completely abandon a homerange area between years, although bobcats shifted the distribution of locations within a home-range area and may have FIG. 2. Measure of home-range dispersion (d, average distance between all possible pairs of locations) and 95% confidence intervals for male bobcats reintroduced to Cumberland Island, Georgia. expanded or reduced their home-range area. Between 1989 and 1990, the average proportion of the 1990 UD outside the 1989 home-range boundary was 18.4% (range 1.4 52.1%) for females and 36.7% for males (range 12.1 53.1%). Between 1990 and 1991, the average proportion of the 1991 UD outside the 1990 home-range boundary was 17.7% (range 0.3 71.7%) for females and 34.9% (range 9.8 57.3%) for males. Intrasexual home-range overlap for 95% UD. We found that home-range overlap increased over years for bobcats reintroduced in 1989 (F ¼ 22.19, d.f. ¼ 2, 18, P, 0.001; Table 2) but declined for the 1990 reintroduction (F ¼ 13.16, d.f. ¼ 1, 14, P ¼ 0.003; Table 2). By 1991, average overlap among bobcats was equivalent to each male bobcat sharing its home range with 3 other males, and each female sharing its home range with 2 other females (Table 2). For males and females from the 1989 reintroduction there was an interaction between year and sex (F ¼ 7.25, d.f. ¼ 2, 18, P ¼ 0.005) in overlap; overlap among males was less than among females in 1989 but was greater in 1990 1991 (Table 2). Males exhibited greater overlap than did females for the 1990 reintroduction (F ¼ 7.26, d.f. ¼ 1, 14, P ¼ 0.017) and the interaction between year and sex was significant (F ¼ 6.62, d.f. ¼ 1, 14, P ¼ 0.022) because males from the 1990 reintroduction exhibited a greater decline in overlap than did females (Table 2). Intrasexual home-range overlap for 50% UD. For the 1989 reintroduction, we found no difference in overlap between sexes (F, 0.01, d.f. ¼ 1, 9, P ¼ 0.949), an increase in overlap among years (F ¼ 4.20, d.f. ¼ 2, 18, P ¼ 0.048), but no interaction between year and sex (F ¼ 1.13, d.f. ¼ 2, 18, P ¼ 0.333; Table 2). Home-range overlap for the 1990 reintroduction indicated no difference between males and females (F ¼ 2.21, d.f. ¼ 1, 12, P ¼ 0.163), a decline in overlap between 1990 and 1991 (F ¼ 4.97, d.f. ¼ 2, 12, P ¼ 0.046), and no interaction between years and sex (F ¼ 0.25, d.f ¼ 2, 12, P ¼ 0.627; Table 2). At the end of our study, the average overlap among males and females was equivalent to each bobcat sharing a core area with approximately 1 other bobcat for the 1989 reintroduction and approximately 0.75 bobcats for the 1990 reintroduction (Table 2; Figs. 3 and 4).

398 JOURNAL OF MAMMALOGY Vol. 87, No. 2 TABLE 1. Mean changes in home-range dispersion (d, units ¼ m), shifts in the home-range center (X of x y locations), and volume of the 95% fixed-kernel utilization distribution (UD) outside the home-range boundary of the previous year for bobcats reintroduced to Cumberland Island, Georgia, 1989 1990. 95% CI ¼ 95% confidence interval. Year reintroduced Years n Change in home-range dispersion (m) Shift in home-range center (km) Shift in UD (%) d 95% CI X Range X Range Males 1989 1989 1990 3 1,553 7,728 4,622 2.54 0.74 5.55 18.9 2.0 28.4 1990 1991 3 661 994 2,316 0.66 0.32 0.98 21.4 19.2 25.5 1990 1990 1991 10 1,024 2,333 285 2.40 0.12 5.42 18.6 3.4 40.7 Females 1989 1989 1990 8 32 529 465 0.69 0.07 2.13 10.5 0.01 30.0 1990 1991 8 284 793 226 0.66 0.17 2.44 7.8 0.0 35.7 1990 1990 1991 6 801 2,541 939 1.24 0.19 2.84 10.5 0.0 29.8 DISCUSSION We believe our observations of the spatial patterns of bobcats are consistent with some aspects of the concepts of prior rights and land tenure as proposed by Bailey (1974). Because of the substantial overlap in home ranges, our results do not support a hypothesis that land tenure is related to securing exclusive access to food resources at the food resource densities observed in our study, which we judged to be adequate to abundant. Baker et al. (2001) monitored relative abundance of prey and bobcat food habits on the island. Marsh rabbits, a preferred bobcat prey item, exhibited a 14-fold decline in abundance between July and November 1989, which coincided with extensive flooding on the island. However, marsh rabbits were restricted primarily to the interdune meadow and wetland habitats, and many bobcats established permanent home ranges in areas where marsh rabbits were rare. Baker et al. (2001) reported that with the decline in marsh rabbits, bobcats shifted their diet to other prey items, especially white-tailed deer, but they found no evidence food was limiting. Bobcats recaptured during 1989 1991 increased their body mass, on average, 12% (n ¼ 17), of which females increased an average 17% (n ¼ 7 Diefenbach et al. 1993). Moreover, few bobcats established home ranges encompassing the interdune meadow of the eastern shore, the most extensive habitat for marsh rabbits, yet they still exhibited weight gains. Consequently, the observed weight gains and the lack of change, or slight decline, in homerange areas further suggest that bobcats were not limited by food availability. Bobcat populations whose prey species experience population crashes exhibit large increases in home-range size of individuals (e.g., Knick 1990). In addition, we have no evidence that other environmental factors identified by Bailey (1981) as important to reproduction in bobcats might have limited the population. Den sites were not a limiting factor because most females placed their dens on the ground under dense vegetation (Ragsdale 1993), which, unlike denning sites in the western United States (Bailey 1974), was widespread on the island. Also, annual survival of adult bobcats was 93% (Diefenbach 1992), which indicated mortality factors were not limiting the population; annual survival is about 75% in harvested populations (Chamberlain et al. 1999). As we increased population density, which was associated with increased overlap of home ranges and core areas (Table 2), we observed a significant decline in reproduction. In each of the 2 breeding seasons after the 1990 reintroduction to the island, Ragsdale (1993) documented only 1 litter born on the island. Despite the lack of females producing young, we detected pairing of males and females during the breeding season. Knick (1990) reported evidence of placental scars in reproductive tracts of nonterritorial females (e.g., transients and juveniles that remained in their natal home range), but they failed to raise kittens. Knick (1990) concluded that possession TABLE 2. Mean of the overlap index (OI) for the 95% and 50% kernel home-range utilization distribution (UD), by year of reintroduction, for male and female bobcats reintroduced to Cumberland Island, Georgia. 95% CI ¼ 95% confidence interval. Males 95% UD 50% UD Females 95% UD 50% UD Year n OI 95% CI OI 95% CI n OI 95% CI OI 95% CI 1989 reintroduction 1989 3 0.28 0.14 0.54 0.00 9 1.44 0.97 2.14 0.35 0.23 0.52 1990 3 3.40 2.38 4.85 0.89 0.45 1.76 8 2.14 1.51 3.03 0.37 0.15 0.93 1991 3 3.02 1.89 4.82 0.96 0.59 1.56 8 2.57 1.58 4.14 1.08 0.48 2.38 1990 reintroduction 1990 11 3.75 3.25 4.23 1.12 0.87 1.44 6 2.19 1.43 3.32 0.71 0.38 1.32 1991 8 3.01 2.57 3.52 0.81 0.60 1.07 7 2.28 1.42 3.63 0.73 0.32 1.66

April 2006 DIEFENBACH ET AL. SOCIAL ORGANIZATION OF BOBCATS 399 FIG. 3. Contours of 50th percentile of the kernel home-range estimates for female bobcats reintroduced in 1989 (thick lines), and females reintroduced in 1990 or born on the island in 1989 (thin lines), Cumberland Island, Georgia. of a territory seemed essential for females to raise young. Lembeck and Gould (1979) observed a negative relationship between population density and reproduction, in which they reported that 100% of females produced young when population densities were least and only 50% produced young when densities were greatest. Wolff (1997) proposed that females seek exclusive use of home-range areas to prevent infanticide from conspecific females. Given that we observed no denning behavior, perhaps the inability to avoid conspecifics triggers a physiological response that reduces the fertility of females (Archer 1979); however, infanticide is difficult to document in bobcats. The home ranges of female bobcats reintroduced in 1989 were smaller than those of females reintroduced in 1990 (Fig. 1), and exhibited smaller changes in size and shifts in location (Table 1). This pattern of home-range establishment is consistent with the hypothesis of land tenure proposed by Bailey (1974) and with the observations of Anderson (1988), who experimentally removed a male bobcat and observed an adjacent male occupy the vacant home range. However, we observed substantial intrasexual overlap of female home ranges (95% UD) and core areas (50% UD; Table 2), which generally has not been reported for bobcats (Anderson 1987; Chamberlain and Leopold 2001; Nielsen and Woolf 2001) and is not consistent with the concept of socially controlled prior rights. FIG. 4. Contours of the 50th percentile of the kernel home-range estimates for male bobcats reintroduced in 1989 (shaded polygons) and reintroduced in 1990 or born on the island in 1989 (thick lines), Cumberland Island, Georgia. The opportunity to reintroduce bobcats to an island allowed us to monitor nearly all the bobcats on the island for 3 years. However, of concern is whether these reintroduced bobcats behaved differently than mainland populations. We increased population density on the island to 3.1 bobcats/10 km 2, which was within the range reported for bobcats in South Carolina and Florida (1.0 10.0 bobcats/10 km 2 Conner et al. 1992; Guenther 1980; Kight 1962; Wassmer et al. 1988) and less than one-third of the greatest population densities reported in the literature (10 15 bobcats/10 km 2 Anderson 1987). Emigration was clearly limited on Cumberland Island, although 1 reintroduced female bobcat returned to the mainland from the 1st reintroduction group of bobcats, and immigration was not known to have occurred since bobcats were extirpated in the early 1900s (Diefenbach et al. 1993). Home ranges were remarkably stable, especially for female bobcats on the island (Table 1). Nielsen and Woolf (2001) defined a stable home range as one in which home-range boundaries overlapped between years and they reported 8% (n ¼ 52) of bobcats shifted their home range; none of the bobcats on Cumberland Island shifted their home range according to this definition. Moreover, bobcats on Cumberland Island had normal home-range sizes. We found that estimates of average annual home-range area

400 JOURNAL OF MAMMALOGY Vol. 87, No. 2 (females ¼ 13.6 km 2, SD ¼ 6.7, n ¼ 38; males ¼ 22.1 km 2, SD ¼ 14.8, n ¼ 32) were similar to what other researchers reported in southeastern United States using various homerange estimators (females ¼ 1.0 16.1 km 2, males ¼ 2.6 60.4 km 2 Anderson 1987; Chamberlain and Leopold 2001). Consequently, we have no evidence that the reintroduced population differed from mainland populations in terms of population density, home-range size, or home-range stability. Evaluating overlap using a 95% UD could have resulted in substantial overlap observed among bobcats, although core areas of the home range could have provided exclusive home-range areas where food resources were defended (Chamberlain and Leopold 2001; Jonsson et al. 2002; Nielsen and Woolf 2001). Consequently, large OI values for the 95% UD may not necessarily indicate a failure to maintain an area of exclusive access (e.g., Jonsson et al. 2002). However, the overlap among core areas (50% UD) indicated that, on average, bobcats shared their core area with almost 1 conspecific of the same sex (Table 2). We agree with Benson et al. (2004) that bobcat social organization may be more complex than previously reported. Research on bobcats and mountain lions suggested that juveniles delay reproduction until a vacant home range is sequestered (Litvaitis et al. 1987; Seidensticker et al. 1973). We found little evidence for defended territories in male or female bobcats at greater population densities because we observed high adult survival rates, no physical injuries of recaptured animals, and little shift in home ranges as they were shared with more individuals. Given the inverse relationship we observed between home-range overlap and reproduction, we suggest that establishment of exclusive use areas by females may be important for successful reproduction, and that the social conditions necessary for maintaining these exclusive areas are less likely to occur at greater population densities or in the absence of disperal opportunities, thereby reducing population productivity even without food limitation. ACKNOWLEDGMENTS This study was supported by the University of Georgia, McIntire Stennis Project GEO-MS-0059, the National Park Service (Cooperative Agreement CA-5000-4-8005, Subagreement 24), the United States Fish and Wildlife Service (Cooperative Agreement 14-16-0009-1551, Work Order 13), and the Georgia Department of Natural Resources, Wildlife Resources Division. We thank the following for their assistance with fieldwork: W. E. James, L. L. Ragsdale, J. P. Vaughn, D. K. Abbot, J. C. Cutts, J. M. Schrenkel, R. T. Speer, W. M. Wilcoxin, and numerous volunteers and interns. T. Hon, the personnel at Altamaha Waterfowl Management Area, and the staff of Cumberland Island National Seashore aided us greatly with their cooperation and assistance. We thank B. D. Leopold, M. J. Lovallo, and M. R. Marshall for reviewing drafts of the manuscript. LITERATURE CITED ANDERSON, E. M. 1987. A critical review and annotated bibliography of literature on bobcat. Colorado Division of Wildlife, Special Report 62. ANDERSON, E. M. 1988. Effects of male removal on spatial distribution of bobcats. Journal of Mammalogy 69:637 641. ANDERSON, E. M., AND M. J. LOVALLO. 2003. Bobcat and lynx (Lynx rufus and Lynx canadensis). Pp. 758 786 in Wild mammals of North America: biology, management, and conservation (G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, eds.). 2nd ed. Johns Hopkins University Press, Baltimore, Maryland. ANIMAL CARE AND USE COMMITTEE. 1998. Guidelines for the capture, handling, and care of mammals as approved by the American Society of Mammalogists. Journal of Mammalogy 79:1416 1431. ARCHER, J. 1979. Animals under stress. Studies in Biology 108. University Park Press, Baltimore, Maryland. BAILEY, T. N. 1974. Social organization in a bobcat population. Journal of Wildlife Management 38:435 446. BAILEY, T. N. 1981. Factors of bobcat social organization and some management implications. Pp. 984 1000 in Proceedings of the Worldwide Furbearer Conference (J. A. Chapman and D. Pursley, eds.). Worldwide Furbearer Conference, Frostburg, Maryland. BAKER, L. A., R. J. WARREN, D.R.DIEFENBACH, W.E.JAMES, AND M. J. CONROY. 2001. Prey selection by reintroduced bobcats (Lynx rufus) on Cumberland Island, Georgia. American Midland Naturalist 145:80 93. BEGON, M., J. L. HARPER, AND C. R. TOWNSEND. 1990. Ecology: individuals, populations, communities. Blackwell Scientific Publications, London, United Kingdom. BENSON, J. F., M. J. CHAMBERLAIN, AND B. D. LEOPOLD. 2004. Land tenure and occupation of vacant home ranges by bobcats (Lynx rufus). Journal of Mammalogy 85:983 988. BUIE, D. E., T. T. FENDLEY, AND H. MCNAB. 1979. Fall and winter home ranges of adult bobcats on the Savannah River Plant, South Carolina. Bobcat Research Conference Proceedings, National Wildlife Federation Scientific Technical Series 6:42 46. BURT, W. H. 1943. Territoriality and home range: concepts as applied to mammals. Journal of Mammalogy 24:346 352. CADE, B. S., AND J. D. RICHARDS. 2001. User s manual for BLOSSOM statistical software. United States Geological Survey, Midcontinent Ecological Science Center, Fort Collins, Colorado. CHAMBERLAIN, M. J., AND B. D. LEOPOLD. 2001. Spatio-temporal relationships among adult bobcats in central Mississippi. Pp. 45 50 in Proceedings of a symposium on current bobcat research and implications for management (A. Woolf, C. K. Nielsen, and R. Bluett, eds.). 2000 Conference. Wildlife Society, Nashville, Tennessee. CHAMBERLAIN, M. J., B. D. LEOPOLD, L.W.BURGER, B.W.PLOWMAN, AND L. M. CONNER. 1999. Survival and cause-specific mortality of adult bobcats in central Mississippi. Journal of Wildlife Management 63:613 620. CONNER, L. M., B. D. LEOPOLD, AND K. J. SULLIVAN. 1992. Bobcat home range, density, and habitat use in east-central Mississippi. Proceedings of the Annual Conference of Southeastern Association of Fish and Wildlife Agencies 46:147 158. DIEFENBACH, D. R. 1992. The reintroduction of bobcats to Cumberland Island, Georgia: validation of the scent-station survey technique and analysis of population viablity. Ph.D. dissertation, University of Georgia, Athens. DIEFENBACH, D. R., L. A. BAKER, W.E.JAMES, R.J.WARREN, AND M. J. CONROY. 1993. Reintroducing bobcats to Cumberland Island, Georgia. Restoration Ecology 1:241 247. GUENTHER, D. D. 1980. Home range, social organization and movement patterns of the bobcat, Lynx rufus, from spring to fall in south-central Florida. M.S. thesis, University of South Florida, Tampa. HILLESTAD, H. O., J. R. BOZEMAN, A.S.JOHNSON, C.W.BERISFORD, AND J. I. RICHARDSON. 1975. The ecology of the Cumberland Island

April 2006 DIEFENBACH ET AL. SOCIAL ORGANIZATION OF BOBCATS 401 National Seashore, Camden County, Georgia. University of Georgia, Georgia Marine Science Center, Technical Report Series 75-5. JAMES, W. E. 1992. Bobcat movements and habitat use on Cumberland Island, Georgia during two years of controlled population increases. M.S. thesis, University of Georgia, Athens. JOHNSON, A. S., H. O. HILLESTAD, S. F. SHANHOLTZER, AND G. F. SHANHOLTZER. 1974. An ecological survey of the Coastal Region of Georgia. United States Department of the Interior, National Park Service, Science Monograph Series 3. JONSSON, P., T. HARTIKAINEN, E. KOSKELA, AND T. MAPPES. 2002. Determinants of reproductive success in voles: space use in relation to food and litter size manipulation. Evolutionary Ecology 16: 455 467. KIGHT, J. 1962. An ecological study of the bobcat, Lynx rufus (Schreber), in west-central South Carolina. M.S. thesis, University of Georgia, Athens. KNICK, S. T. 1990. Ecology of bobcats relative to exploitation and a prey decline in southeastern Idaho. Wildlife Monographs 108:1 42. LEMBECK, M., AND G. I. GOULD. 1979. Dynamics of harvested and unharvested bobcat population in California. Bobcat Research Conference Proceedings, National Wildlife Federation Scientific Technical Series 6:53 54. LITVAITIS, J. A., J. T. MAJOR, AND J. A. SHERBURNE. 1987. Influence of season and human-induced mortality on spatial organization of bobcats (Felis rufus) in Maine. Journal of Mammalogy 68: 100 106. LITVAITIS, J. A., J. A. SHERBURNE, AND J. A. BISSONETTE. 1986. Bobcat habitat use and home range size in relation to prey density. Journal of Wildlife Management 50:110 117. MAHER, C. R., AND D. F. LOTT. 2000. A review of ecological determinants of territoriality within vertebrate species. American Midland Naturalist 143:1 29. MCLOUGHLIN, P. D., S. H. FERGUSON, AND F. MESSIER. 2000. Intraspecific variation in home range overlap with habitat quality: a comparison among brown bear populations. Evolutionary Ecology 14:39 60. MILLER, S. D., AND D. W. SPEAKE. 1979. Progress report: demography and home range of the bobcat in south Alabama. Bobcat Research Conference Proceedings, National Wildlife Federation Scientific Technical Series 6:123 124. NIELSEN, C. K., AND A. WOOLF. 2001. Spatial organization of bobcats (Lynx rufus) in southern Illinois. American Midland Naturalist 146:43 52. PACKER, C. 1986. The ecology of sociality in felids. Pp. 429 451 in Ecological aspects of social evolution (D. E. Rubenstein and R. W. Wrangham, eds.). Princeton University Press, Princeton, New Jersey. RAGSDALE, L. L. 1993. Den characteristics, activity patterns, and habitat use of reintroduced female bobcats on Cumberland Island, Georgia. M.S. thesis, University of Georgia, Athens. ROLLEY, R. E. 1983. Behavior and population dynamics of bobcats in Oklahoma. Ph.D. dissertation, Oklahoma State University, Stillwater. SAS INSTITUTE INC. 1999. SAS/STAT user s guide. SAS Institute Inc., Cary, North Carolina. SEAMAN, D. E., B. GRIFFITH, AND R. A. POWELL. 1998. KERNELHR: a program for estimating animal home ranges. Wildlife Society Bulletin 26:95 100. SEAMAN, D. E., J. J. MILLSPAUGH, B.J.KERNOHAN, G.C.BRUNDIGE, K. J. RAEDECKE, AND R. A. GITZEN. 1999. Effects of sample size on kernel home range estimates. Journal of Wildlife Management 63:739 747. SEIDENSTICKER, J. C., M. G. HORNOCKER, W. V. VILES, AND J. P. MESSICK. 1973. Mountain lion social organization in the Idaho Primitive Area. Wildlife Monographs 35:1 60. SWIHART, R. K., AND N. A. SLADE. 1985. Testing for independence of observations in animal movements. Ecology 66:1176 1184. WASSMER, D. A., D. D. GUENTHER, AND J. N. LAYNE. 1988. Ecology of the bobcat in south-central Florida. Bulletin of the Florida State Museum of Biological Sciences 33:159 228. WOLFF, J. O. 1997. Population regulation in mammals: an evolutionary perspective. Journal of Animal Ecology 66:1 13. ZEZULAK, D. S., AND R. G. SCHWAB. 1979. A comparison of density, home range, and habitat utilization of bobcat populations at Lava Beds and Joshua Tree National Monuments, California. Bobcat Research Conference Proceedings, National Wildlife Federation Scientific Technical Series 6:64 79. Submitted 6 April 2005. Accepted 18 August 2005. Associate Editor was Eric C. Hellgren.