SUDDEN INCREASE IN A RARE ENDEMIC CARNIVORE:
ECOLOGY OF THE ISLAND SPOTTED SKUNK
KRISTA L. JONES,* DIRK H. VAN VUREN, AND KEVIN R. CROOKS
Department of Wildlife, Fish, and Conservation Biology, University of California–Davis,
One Shields Avenue, Davis, CA 95616, USA (KLJ, DHVV)
Department of Fish, Wildlife, and Conservation Biology, Colorado State University,
115 Wagar, Fort Collins, CO 80523, USA (KRC)
The island spotted skunk (Spilogale gracilis amphiala), an insular endemic carnivore, recently increased
unexpectedly from rarity to abundance on Santa Cruz Island, California. Two explanations have been proposed
for this striking increase: competitive release due to decline of the island fox (Urocyon littoralis santacruzae) and
vegetative recovery due to removal of feral livestock. To examine the causes and consequences of the increase,
we assessed abundance, body mass, home-range size, spatial resource use (den sites and habitat use), temporal
resource use, and diet of island spotted skunks during abundance in 2003–2004 and compared it with similar
measures during rarity in 1992. Capture success of skunks increased exponentially from 1992 to 2004 (r ¼ 0.38),
leading to extraordinarily high densities (9–19 individuals/km2). Both body mass and home-range size remained
unchanged, suggesting that per capita resource abundance was not higher in 2003–2004 compared to 1992. We
found modest shifts in habitat use, diet, and possibly diurnal activity, providing some support for release from
exploitative competition as an explanation for the increase. However, there was a marked shift in den selection,
toward unprotected dens and away from multiple use, that is not likely attributable to either release from
exploitative competition or to island recovery. Thus, our results suggest that, although both of these processes
have likely played a role, release from interference competition also may be a factor behind the dramatic increase
in number of skunks. Skunks on Santa Cruz Island may be nearing or even exceeding carrying capacity, and the
future of the population and its effects on the endangered island fox remain unclear.
Key words: Channel Islands, competitive release, interference competition, intraguild predation, island fox, island recovery,
radiotelemetry, Spilogale gracilis, Urocyon littoralis
Islands are of particular conservation value because they
tend to support disproportionately high levels of endemism
(Ceballos and Brown 1995; Courchamp et al. 2003). Further,
islands are of great conservation concern because of heightened
vulnerability at both the species and community levels. Insular
endemic species are extinction-prone because of their highly
limited geographic ranges and increased susceptibility to exotic
species (Purvis et al. 2001). In addition, islands typically are
depauperate in their biota because of isolation, resulting in re-
duced community resilience to novel disturbances (Courchamp
et al. 2003; Lomolino et al. 2005; Simberloff 1986). Although
populations in isolated insular communities tend to be particu-
larly stable (Adler and Levins 1994), their sensitivity to novel
perturbations can lead to dramatic population fluctuations
across the community (e.g., North et al. 1994).
Islands are often lacking in mammals, especially carnivores
(Alcover and McMinn 1994; Lomolino et al. 2005); hence,
Santa Cruz Island, off the coast of California, is unusual in
supporting 2 insular endemic carnivores, the island spotted
skunk (Spilogale gracilis amphiala) and the island fox
(Urocyon littoralis santacruzae). The island spotted skunk is
endemic at the subspecific level and exists exclusively on 2 of
the 8 Channel Islands, Santa Cruz and Santa Rosa. Previous
accounts indicate that the skunk was rare on Santa Cruz Island
throughout much of the 1900s (Crooks 1994a; Laughrin 1977;
Sheldon 1990); however, between 1992 and 1999 capture
success of skunks, assumed to be a good index of population
size (Roemer et al. 2002), increased dramatically and un-
expectedly (Crooks and Van Vuren 2000; Roemer et al. 2002).
Two explanations have been proposed for this striking in-
crease: competitive release due to a decline in island foxes
(Crooks and Van Vuren 2000; Roemer et al. 2002) and island
* Correspondent: kljones@ucdavis.edu
 2008 American Society of Mammalogists
www.mammalogy.org
Journal of Mammalogy, 89(1):75–86, 2008
75

vegetative recovery due to removal of feral livestock (Crooks
and Van Vuren 1994, 1995, 2000).
Concurrent with the increase in the population of skunks, the
population of island foxes declined from approximately 1,300
individuals in 1993 to an estimated 133 in 1998 (Roemer et al.
2001); the decline has been attributed to hyperpredation by
golden eagles (Aquila chrysaetos) that colonized during the
early 1990s (Roemer et al. 2001). Research in 1992 indicated
that there was significant niche overlap on multiple axes
between the foxes and skunks, with the fox likely the dominant
competitor (Crooks and Van Vuren 1995). Exploitative
competition, where 1 competitor is more efficient at utilizing
limited resources than another, is common between carnivores
as between many other taxa. However, interference competi-
tion, in which 1 species limits or prevents access of the other to
shared resources, also is known to be particularly fierce among
carnivores (Creel et al. 2001; Dayan and Simberloff 1998).
Thus, the drastic decline of the island fox may have precipi-
tated the concomitant population growth of the island spotted
skunk via competitive release.
Along with the decline in island foxes, Santa Cruz Island
simultaneously experienced widespread vegetative recovery
following removal of exotic livestock. Since the 1860s, the
island supported feral sheep at exceptionally high densities
(Van Vuren and Coblentz 1989), resulting in the large-
scale destruction of tree- and shrub-dominated communities
(Brumbaugh 1980; Van Vuren and Coblentz 1989; Wehtje
1994). During the 1980s, 37,000 sheep were removed, leaving
90% of the island sheep-free (Schuyler 1993), and all domestic
cattle (approximately 1,500) were removed as well (Brenton
and Klinger 1994). A prolonged drought from 1987 through
1991 delayed vegetative recovery (Brenton and Klinger 1994;
Klinger et al. 1994), but regrowth of woody vegetation accel-
erated thereafter. During the 1990s, communities such as
coastal sage scrub and pine (Pinus muricata) forest expanded
into areas formerly dominated by annual grasslands, and there
was an obvious increase in both low-growing woody vege-
tation and herbaceous biomass across habitats (D. H. Van
Vuren and P. T. Schuyler, in litt.). Spotted skunks are often as-
sociated with habitats dominated by woody vegetation (Kinlaw
1995; Verts et al. 2001) and areas that provide cover (Carroll
2000; Kinlaw 1995; Rosatte and Lariviere 2003). In addition,
island spotted skunks are considered carnivorous, consuming
primarily deer mice (Peromyscus maniculatus) and insects
(Crooks and Van Vuren 1995); deer mice on Santa Cruz Island
are abundant only in habitats dominated by woody vegetation
(Mayfield et al. 2000). Thus, regrowth of trees and shrubs may
have facilitated the population increase of skunks by improving
habitat quantity and quality and prey availability.
Our general objective was to evaluate the causes and con-
sequences of the dramatic increase in spotted skunks on Santa
Cruz Island by investigating their ecology during abundance in
2003–2004 and comparing it with that during rarity in 1992
(Crooks 1994a, 1994b; Crooks and Van Vuren 1995). We as-
sessed abundance, body size, home-range size, spatial resource
use (den sites and habitat use), temporal resource use, and diet.
Both proposed causes for the population increase are based on
increased resource availability, so we expected an improvement
in body condition, as reflected by increased body mass. How-
ever, if density had reached the point of intense intraspecific
competition, we would expect no change in body mass. With
increased resource availability, we also expected home ranges
of skunks to decrease in size, as is common for many ver-
tebrates (Boutin 1990; Mares and Lacher 1987). Finally, we
expected that, if competitive release was the cause of the in-
crease, skunks would either shift or broaden their niches along
1 or more axes for which there is overlap with the island fox
(Crooks and Van Vuren 1995). In contrast, because island
recovery has produced a general increase in habitat quantity
and quality, we expected no major shift in resource utilization
if this was the cause of the increase.
MATERIALS AND METHODS
Study area.—Santa Cruz Island (34809N, 1198459W) is
located 40 km off the mainland coast south of Santa Barbara,
California. The island is approximately 39 km long and varies
from 3 to 11 km wide, with a total area of roughly 250 km2.
The climate is maritime Mediterranean, with pronounced dry
(May–November) and wet (December–April) seasons. Re-
search was conducted from August 2003 through September
2004 at 2 sites on the island: Valley Anchorage (VA) and
Willows Drainage (WD). The VA site (3.7 km2) lay at the
eastern end of the island’s Central Valley and was character-
ized by chaparral, coastal sage scrub, and open grassland. The
WD site (4.1 km2) was on the south side of the island in
Willows drainage and had steeper topography with more barren
slopes, significant chaparral and open grassland, and very little
coastal sage scrub. These sites differed from those used by
Crooks and Van Vuren (1995) because of logistical constraints.
Trapping and radiotelemetry.—Skunks were live-trapped in
Tomahawk single-door box traps (Tomahawk Live Traps Co.,
Tomahawk, Wisconsin) baited with commercial cat food and
On Target fruit paste lures (On Target, Cortland, Illinois) during
August 2003–February 2004 and August–September 2004.
Trapped individuals were ear-tagged and weighed, sex was
determined, and age was estimated based on tooth wear as
adult, juvenile (,1 year), or questionable. Forty animals were
anesthetized with ketamine hydrochloride (20 mg/kg) and fitted
with Holohil radiocollars (12 g for females; 15 g for males;
Holohil, Carp, Ontario, Canada); collars never exceeded 5%
of body weight. Captured animals were released immediately
after processing at the point of capture, except for anesthetized
individuals, which were returned to their trap until fully re-
covered before release. All fieldwork was carried out humanely
in accordance with guidelines of the American Society of
Mammalogists (Gannon et al. 2007) and those of the Institutional
Animal Care and Use Committee at the University of California.
The study was conducted under a Memorandum of Understand-
ing between the California Department of Fish and Game and the
University of California, Davis (expiration 31 December 2007).
Because skunks are nocturnal (Crooks and Van Vuren 1995),
telemetry for location triangulations was conducted primarily in
the hours between dusk and dawn, with each study site visited
76 JOURNAL OF MAMMALOGY Vol. 89, No. 1