2009 SOUTHEASTERN NATURALIST 8(3):411–426
Survival and Habitat Use of Feral Hogs in Mississippi
Robert Hayes1, Sam Riffell1,*, Richard Minnis1, and Brad Holder2
Abstract - Sus scrofa (Feral Hog) can cause extensive damage to agricultural
crops and native vegetation, is a potential disease vector, and competes with other
wildlife for food resources. Without site-specific information about survival and
habitat use, habitat management and control efforts may not be effective. We
examined home-range size, habitat use, and survival of 29 Feral Hogs in central
Mississippi using radio telemetry. Dry-and wet-season survival rates were 80.8%
and 41.4%, respectively. Hunting (primarily during the wet season) was the major
cause of mortality. Dry-season home ranges were larger (6.4 km2) than wet-season
home ranges (3.0 km2). During the dry-season, Feral Hog home ranges (2nd-order
selection) were associated with dense vegetation types (seasonally flooded old
fields, old fields, and managed openings). During the wet season, old fields and
agricultural fields were selected, but seasonally flooded old fields and managed
openings were not. Within home ranges (3rd-order selection), hogs selected old
fields and managed openings during the dry season. All habitats were used randomly
within home ranges during the wet season. Flooding of preferred habitats,
changes in food availability, and hunting pressure likely caused these changes in
habitat use and home-range placement.
Introduction
Since first being introduced to North America in the 1500s, Sus scrofa
L. (Feral Hog) has steadily spread over much of the United States (Gipson
et al. 1998). Feral Hogs disturb the soil by rooting in search of tubers and
larval and adult insects. They can cause extensive damage to native plant
communities (Engeman et al. 2007), destroy agricultural crops (Beach
1993), and increase the potential for soil erosion, which can, in turn, impact
water quality and aquatic fauna (Kaller and Kelso 2006). Foraging
behavior of Feral Hogs may impact native wildlife by reducing nesting
cover and available forage. In times of low mast availability, competition
for resources may limit both Odocoileus virginianus Zimmerman (Whitetailed
Deer) and Feral Hog populations (Yarrow 1987). Feral Hogs may
prey on eggs of ground-nesting birds (Henry 1969, Matschke 1965), reptiles
and amphibians (Ditchkoff and West 2007), and White-tailed Deer
fawns (Springer 1977).
Although concern about effects of Feral Hogs is growing (Ditchkoff
and West 2007), basic life-history information is lacking for many parts of
1Department of Wildlife and Fisheries, Box 9690, Mississippi State, MS 39762. 2Mississippi
Department of Wildlife, Fisheries and Parks, 1505 Eastover Drive, Jackson,
MS 39211. *Corresponding author - sriffell@cfr.msstate.edu.
412 Southeastern Naturalist Vol. 8, No. 3
the southeastern United States. Our objectives were to gather life-history
data to better inform population management in Mississippi and to better
understand how Feral Hogs select and use habitats. We gathered basic information
about survival, home-range size and habitat use of Feral Hogs in
an agriculture-dominated region of central Mississippi. Understanding basic
life-history of Feral Hog populations is critical for controlling Feral Hogs,
reducing negative impacts on native wildlife and agriculture, and minimizing
human-wildlife confl icts.
Field-site Description
We worked in the southwestern corner of Grenada County, MS on the
3837-ha Malmaison Wildlife Management Area (MWMA) and surrounding
lands located within the alluvial fl oodplain of the Yalobusha River. Available
habitats at MWMA included agricultural fields, bottomland hardwoods,
old fields, managed wildlife openings, seasonally fl ooded bottomlands, and
seasonally fl ooded old fields.
Bottomland hardwood habitat was dominated by Platanus occidentalis
L. (Sycamore), Ulmus spp. (elms), Celtis laevigata Willd. (Sugarberry),
Liquidambar styraciflua L. (Sweetgum), Quercus nigra L. (Water Oak),
Q. phellos L. (Willow Oak), Q. lyrata Walt. (Overcup Oak), Q. michauxii
Nutt. (Swamp Chestnut Oak), Q. pagoda Raf. (Cherrybark Oak), and Q.
buckleyi Nixon and Dorr (Nuttall Oak). Seasonally flooded bottomlands
were similar to bottomland with the addition of some Taxodium spp. (cypress)
and Nyssa spp. (black gum). Old field was characterized by very
thick vegetation comprised mainly of Rubus spp. (blackberry), Lonicera
japonica Thung. (Honeysuckle), and a variety of grasses, forbs, and woody
species. Vegetation in old-field habitat was often dense enough to reduce
vision to ≤1 m. Areas that were characterized by dense ground cover, including
hardwood regeneration areas, clear-cuts, and fallow fields 2–8
years old, were classified as old field. Bottomland hardwoods and old
fields that were flooded ≥3 months of the year were considered seasonally
flooded. Seasonally flooded old-field habitat was similar to old field in
vegetative structure, except for the presence of some aquatic plants such as
Polygonum sp. (smartweed) and Lotus spp. (water lilies). Ground cover in
seasonally flooded bottomlands was sparse.
Managed openings included small openings (<3 ha) that were planted
and managed specifically as wildlife food plots. These openings were planted
with Zea mays L. (Corn), Echenechloa spp. (millet), Sorghum spp. (sorghum),
or Trifolium spp. (clover) to benefit waterfowl, Meleagris gallopavo
L. (Wild Turkey), and White-tailed Deer. Agricultural fields were planted in
Corn, Gossypium spp. (cotton), and Glycine spp. (soybean). More detailed
description of these habitats is in Hayes (2007).
2009 R. Hayes, S. Riffell, R. Minnis, and B. Holder 413
Methods
Trapping techniques
We began trapping animals throughout the study area on 1 April 2005
and ended with the onset of hunting season in early October 2005. Once
a suitable area with abundant hog sign was discovered, we baited the area
with corn. If the hogs returned to the area, a trap was placed at the site
and baited again. We set the traps after the hogs became accustomed to
the trap (usually <2 nights). We used box-style panel live traps with roottype
gates (Littauer 1993).
We anesthetized hogs with a mixture of 3.2 mg/kg Telazol® and 1.6
mg/kg Xylezine (Sweitzer et al. 1997). We weighed each individual, determined
sex, and marked it with a unique eartag. We fitted animals >23 kg in
mass with collar transmitters from Lotec® Lotec Wireless (Newmarket, ON).
Hogs were processed and released on site as quickly as possible to minimize
handling stress. Handling and marking procedures were approved by the
Institutional Animal Care and Use Committee (IACUC), Mississippi State
University (IACUC Protocol # 04-090).
Telemetry procedures
We located collared hogs using a 3-element Yagi antenna and a multifrequency
receiver (Advanced Telemetry Systems, Isanti, MN) using a
rotational schedule to evenly distribute sampling over a complete 24-hr
period (night and day). Each sampling period was divided into 6-hr sampling
shifts (0600–1200 CST, 1200–1800 CST, etc.). On each successive collection
period, we progressed to the next sampling shift. We attempted to locate
each radio-collared hog on ≥3 days each week from the time of capture until
the animal’s death or the end of the study period (31 March 2006).
To triangulate hog locations, we took 3–4 bearings from
spatially referenced telemetry stations in the direction of the strongest
signal. Occasionally, hogs were visually located in open habitats. In these
cases, we recorded UTM coordinates of observer location, one bearing
toward the animal, and a distance estimate. Because hogs did not stop
foraging or move in response to observation in these instances, collection
of occasional locations by visual observation had minimal effects on
hog locations. Bearing data were entered into LOAS software (Ecological
Software Solutions, Urnäsch, Switzerland) to generate UTM coordinates
(MLE estimator) of hog locations.
We estimated telemetry error for both winter (leaf-off) and summer
(leaf-on) conditions encountered during our study period using 2 methods.
First, a technician held dummy transmitters in a habitat similar to,
but separate from, our study area while 3 bearings toward the strongest
signal were collected. Second, we used radio-collared hogs sending mortality
signals as test transmitters (Withey et al. 2001). Mean distance error
was 50 m (SE = 6, n = 30). This is a reasonable error considering the
414 Southeastern Naturalist Vol. 8, No. 3
dense vegetation where hogs were often located. To minimize telemetry
error due to hog movements, we kept the time between the first and last
bearing under 12 minutes (Holder 2006). Any locations with an error ellipse
that was greater than 3 times the interquartile range above the 75th
percentile was rejected (Phillips et al. 2004).
Estimating available habitat
We used a combination of aerial photo interpretation and ground-truthing
to delineate habitat types and create a digital habitat coverage map in ArcGIS
(ESRI 2004). We delineated the following habitat types: agricultural fields
(AG), bottomland hardwood forests (BLHD), seasonally fl ooded bottomland
hardwoods (FLHD), old fields (OF), seasonally fl ooded old fields (FLOF),
and managed openings (MGDO).
We conducted analyses for two separate time periods because seasonal
flooding and hunting season co-occurred and created a distinct change in
habitats. We defined the dry season as 1 April 2005–31 October 2005 and
the wet season as 1 November 2005–31 March 2006. Most of the seasonal
flooding occurred—either naturally or mechanically to provide winter
waterfowl habitat (M. Cooper, Backwater Brake Hunting Club, MS, pers.
comm.)—after 1 November. Additionally, general gun hunting season
began on 18 November, and agricultural crops such as corn were maturing
in November. These events had the potential to profoundly influence
habitat use and/or Feral Hog distributions within a few days of 1 November.
The beginning date (April 1) for the dry season was when most of the
water had receded from the flooded areas, and general gun hunting season
had closed.
Survival
We used the nest-survival model within program MARK (Dinsmore
et al. 2002, White and Burnham 1999) to determine effects of a priori
selected factors on daily survival rates (DSR) of Feral Hogs. We modeled
effects of season, sex, and initial weight on DSR. We selected the model
with the lowest adjusted Akaike’s information criterion (AICc: Burnham
and Anderson 2002) as the best approximating model (White and
Burnham 1999) to estimate survival rates. To calculate seasonal survival
estimates, we raised the DSR to the power of the days included in the season
(DSR# days in season).
Calculating home ranges
Seasonal home ranges for each hog were generated using a 95% adaptive
kernel estimator within the Animal Movement Extension (Hooge and
Eichenlaub 1997) in ArcView 3.2 (ESRI 2002). Because hogs formed groups
during the course of our study, some home ranges were not truly independent.
To avoid pseudoreplication, we dropped all but one animal in each
group of closely associated animals, retaining the animal with the largest
2009 R. Hayes, S. Riffell, R. Minnis, and B. Holder 415
number of locations. We estimated annual home ranges (i.e., across both
wet and dry seasons) for hogs monitored ≥8 months (n = 7). We estimated
seasonal home ranges for hogs with ≥30 locations (Seaman et al. 1999) in
either season (dry n = 10, wet n = 4). To allow comparison with other studies
(Saunders and McLeod 1999), we also calculated seasonal home ranges
using minimum convex polygons.
We used unequal variance t-tests to test the hypotheses that mean
home-range size did not differ between sexes and seasons. We used linear
regression to determine the effects of mid-point weight (estimated wt. at
mid-point of the observation period for each individual) on home-range
size. We were able to obtain multiple weights for 12 hogs (5 boars and
7 sows) through recapture or hunter cooperation after harvest. Including
quadratic and cubic terms did not improve model fit over the simple
linear model (ΔAICc > 16.0), so we used the simple linear model to calculate
a simple estimate of weight gain: 0.115 kg/day for boars (n = 5) and
0.049 for sows (n = 7). We used these rates of gain to estimate weight at
the mid-point of the observation period for each animal. Although this assumes
that weight gain was linear during the observation period, it does
represent a better estimate of the size of the animal across the observation
period than would initial weight.
Habitat use
We restricted our analysis to independent animals with ≥30 locations
during either season (dry n = 10; wet n = 4). To estimate 2nd-order habitat
use (comparing habitat within the home range to habitat available in
the study area; Johnson 1980), we generated 10,000 random points in the
study area and calculated distances between random points and the nearest
polygon representative of each habitat type using ArcView Calculate
Distance command (ESRI 2002). We calculated distances from animal
locations to each habitat type (Conner and Plowman 2001). We compared
the mean of these distances for each animal to the mean distances of random
points to each habitat type. All habitats within the study area were
considered available. To define the extent of the study area, we created
a 100% minimum convex polygon of all hog locations and then created a
500-m buffer around that polygon (Harveson et al. 2004). We used the
same distance methods to describe 3rd-order habitat use (use of habitat
types within a home range; Johnson 1980), except that we generated 1000
random points within each animal’s home range rather than for the entire
study area.
Typically, MANOVA is used to control Type I error rate before using
univariate t-tests to test the hypotheses that use of each specific habitat type
did not differ from random (Conner and Plowman 2001:278). However,
we modified this approach because small sample sizes precluded using
MANOVA for the wet season. Instead of using MANOVA to control Type
416 Southeastern Naturalist Vol. 8, No. 3
I error rate, we used a sequential Bonferroni correction (which provides
stricter control than a MANOVA) on the t-tests for the individual habitat
types (Holm 1979, Westfall et al. 1999). To minimize high rates of Type II
error (low power) associated with small sample sizes, we used α = 0.10 for
all analyses.
Results
Survival
We observed 29 hogs for periods ranging from 15–346 days. Twelve
of the 15 mortalities (6 boars and 6 sows) were from hunting or trapping.
The other mortalities appeared to be related to reproductive stress, collar
mortality, and unknown factors, but we could not rule out disease, hunting,
or aggressive encounters as potential causes. A reduced 2-season
model was the best approximating model (ΔAICc = 0.00, wi = 0.92;
Table 1), indicating that hogs had lower daily survival in the wet season
(0.9946, 95% CI = 0.9906–0.9970) than in the dry season (0.9990, 95%
CI = 0.9970–0.9997; Fig. 1). This approximation translated into seasonal
survival rates of 80.8% over the 204-day dry season and 41.4% over the
164-day wet season. There was little evidence to suggest that sex or initial
weight had any effect on DSR.
Home range
Annual home ranges (Table 2) averaged 10.0 (± 2.6) km2 for all hogs, 8.1
(± 3.2) km2 for 5 boars;, and 15.0 (± 1.9) km2 for 2 sows (mean # locations/
hog = 83). Home-range sizes were larger in the dry season (6.4 ± 1.3 km2)
than in the wet season (3.0 ± 0.7 km2; t11.8 = 2.23, P = 0.046). We did not
observe any significant effect of sex (t12 = 0.80, P = 0.437) or weight (t12 =
-0.66, P = 0.522) on home-range sizes. Home-range estimates are listed with
other published estimates in Table 2.
2nd-order habitat use
Available habitat in the study area was 24% agricultural fields (AG),
40% bottomland hardwoods (BLHD), 10% seasonally fl ooded hardwoods
(FLHD), 4% seasonally fl ooded old fields (FLOF), 2% managed openings
(MGDO), and 20% old field (OF).
In the dry season, hogs were found closer to seasonally fl ooded old fields,
old fields, and managed openings than at random (Table 3). There were no
Table 1. Model selection for daily survival of Feral Hogs at Malmaison WMA, MS and
surrounding lands, 1 April 2005–31 March 2006.
Model AICc ΔAICc AICc wi k Deviance
Seasonal survival 194.27 0.00 0.92 2 190.27
Constant survival 200.56 6.29 0.04 1 198.56
Initial mass 200.56 6.29 0.02 2 198.56
Sex 201.97 7.70 0.02 2 197.97
2009 R. Hayes, S. Riffell, R. Minnis, and B. Holder 417
Table 2. Feral Hog body mass estimates (kg) and home-range estimates (km2) in the southeastern
US from published sources.
Other
Body mass Time/ MCPB
methods
StateA Boar Sow Season Boar (n) Sow (n) MethodC Boar Sow Source
MS 67 50 ≈6 mo/wet 2.1 (3) 3.8 (1) 95% AK 2.9 3.2 This study
MS 62 53 ≈6 mo/dry 7.6 (6) 4.1 (4) 95% AK 7.5 4.6 This study
MS 59 46 >8 mo 8.1 (5) 15.1 (2) 95% AK 8.1 15.0 This study
SC 58 57 3–10 mo 5.3 (4) 4.4 (1) MMA 10.2 8.1 Kurz and
Marchinton 1972
SC - - <1 yr 1.1 (6) 1.2 (3) - - - Wood and
Brenneman 1980
SC - - >1 yr 2.3 (3) 1.8 (3) - - - Wood and
Brenneman 1980
TN - - Summer 3.8 (9) 3.5 (4) - - - Singer et al. 1981
TN - - Winter 3.9 (9) 2.7 (4) - - - Singer et al. 1981
TX - - ≤24 mo - (18) - (27) MOU 19.6 16.2 Yarrow 1987
TX - - Annual 15.8 (6) 6.5 (10) - - - Mersinger and
Silvy 2007
TX - - 10 mo 35.0 (7) 28.3 (6) 95% AK 58.7 43.4 Adkins and
Harveson 2007
TX - - 2 yrs - - - 5.95 (18) Gabor et al. 1999
AMS = Mississippi, SC = South Carolina, TN = Tennessee, and TX = Texas.
BMinimum convex polygon technique.
CAbbreviations for methods as follows: AK = adaptive kernel, HM = harmonic mean, MMA =
modified minimum area, MOU = multivariate Ornstein-Uhlenbeck.
Figure 1. Daily survival rates and 95% CI for Feral Hogs at Malmaison WMA, and
surrounding lands, 1 April–31 March 2006.
418 Southeastern Naturalist Vol. 8, No. 3
differences in distances to hog locations compared to distances to random
points for bottomland hardwood, agricultural fields, and seasonally fl ooded
hardwood habitats (Table 3). Thus, 2nd-order habitat use of Feral Hogs during
the dry season can be ranked as follows: the greatest preference was for
seasonally fl ooded old field; followed by old field and managed openings.
No preference for bottomland hardwood, agricultural fields, or seasonally
fl ooded hardwoods was detected.
During the wet season, hog locations were closer to old-field habitat at
random. Agricultural fields, managed openings, seasonally fl ooded old field,
seasonally fl ooded hardwoods, and bottomland hardwoods were used no differently
than expected by chance (Table 3). Wet season 2nd-order habitat- use
analysis indicated that the greatest preference was for old field followed by
agricultural fields, managed openings, seasonally fl ooded hardwoods, seasonally
fl ooded old fields, and bottomland hardwoods.
3rd-order habitat use
The mean composition for hog home ranges for the dry season was
29% agricultural fields, 32% bottomland hardwood, 5% seasonally fl ooded
hardwoods, 3% seasonally fl ooded old fields, 6% managed openings, and
25% old fields. Mean composition for wet-season home ranges were 28%
agricultural fields, 34% bottomland hardwood, 4% seasonally fl ooded hardwoods,
2% seasonally fl ooded old fields, 6% managed openings, and 26%
old fields.
During the dry season, hog locations were closer to old fields and managed
openings compared to random points within each home range. Hog
Table 3. Ranking and associated P and t values for 2nd- and 3rd-order habitat selection by Feral
Hogs at Malmaison WMA, MS and surrounding lands, 1 April 2005–31 March 2006.
Dry Wet
Habitat ρ T P Habitat ρ T P
2nd-order
FLOF 0.402 -4.81 <0.001* OF 0.380 -5.54 0.012*
OF 0.623 -6.58 <0.001* AG 0.422 -3.84 0.031
MGDO 0.663 -3.33 0.007* MGDO 0.906 -0.33 0.763
BLHD 0.932 -0.52 0.612 FLOF 1.223 0.36 0.740
AG 1.067 0.49 0.637 FLHD 1.042 0.11 0.919
FLHD 1.321 1.99 0.075 BLHD 1.813 3.67 0.035
3rd-order
OF 0.807 -4.56 0.001* OF 0.691 -1.17 0.327
FLOF 0.850 -2.38 0.039 MGDO 0.818 -2.18 0.118
MGDO 0.884 -2.82 0.018* FLHD 0.871 -2.06 0.131
BLHD 0.918 -1.83 0.097 FLOF 0.985 -0.35 0.751
AG 0.976 -0.59 0.567 AG 1.097 0.73 0.520
FLHD 1.065 1.36 0.204 BLHD 1.114 0.73 0.520
*Significant using a sequential Bonferroni correction (experiment-wise α = 0.10).
2009 R. Hayes, S. Riffell, R. Minnis, and B. Holder 419
locations were also closer to seasonally fl ooded old fields, but not significantly after a Bonferroni sequential adjustment (P = 0.039). Bottomland
hardwoods, agricultural fields, and seasonally fl ooded hardwoods were used
randomly (Table 3). The ranking of dry-season habitat use was: old fields
were most preferred, followed by seasonally fl ooded old fields and managed
openings, with no preference for bottomland hardwoods, agricultural fields,
or seasonally fl ooded hardwoods. None of the habitats were used disproportionately
to their availability within the home ranges during the wet season
(Table 3).
Discussion
Survival
Although others have documented greater Feral Hog mortality during
dry weather (Massei et al. 1997, Woodall 1983), we observed the opposite
pattern in our study area. Feral Hogs often experience food shortages during
periods of low rainfall because of hardening ground which makes rooting
more difficult (Massei et al. 1997, Woodall 1983). But even the dry season
in Mississippi can have substantial rainfall, and coupled with the abundant
wetland habitats in our study area, conditions at our study area likely did not
mirror the arid environments in the studies cited above.
In contrast, we observed lower survival during the wet season, and we
are confident this was not related to food availability for two reasons. First,
natural food resources (i.e., hard mast), maturing agricultural crops, and
wildlife food plots were plentiful at this time (R.C. Hayes, pers. observ.).
Second, general gun hunting season was open during most of the wet season,
and hunting was the major cause of mortality among the animals that died
during the study.
Neither initial weight nor sex affected hog survival in our study. Weight
can be an indicator of overall health. Conversely, larger animals are often
targeted by hunters (thus, larger hogs could have had a greater likelihood of
being shot). However, hogs in our study area often traveled in mixed groups
of males and females of varying mass, and hunters tended to shoot the
first hog that presented itself (J.R. Lee, Delta Wings Hunt Club, MS, pers.
comm.). Jezierski (1977) and Massei et al. (1997) found that for European
Wild Boar, juvenile males tend to experience higher mortality than juvenile
females. However, adult females had higher mortality than adult males
(Jezierski 1977, Massei et al. 1997), a result which was attributed to costs of
reproduction and caring of young incurred by females during summer (Massei
et al. 1997).
Feral Hogs in our study area suffered very little mortality from causes
other than hunting. This result suggests that hunting may be one way to
control Feral Hog populations. Evidence supports the potential of hunting as
a measure for reducing Feral Hog populations in some areas, thus reducing
420 Southeastern Naturalist Vol. 8, No. 3
damage and confl ict (Waithman et al. 1999). In California, hog densities were
lower in areas that received high hunting pressure compared to areas where
hunting was restricted (Pine and Gerdes 1973, Sweitzer et al. 2000). Damage
to sensitive wetlands in hunted areas can be less than half that of unhunted
areas, and supplemental removal of hogs in those unhunted areas reduced
damage substantially (Engeman et al. 2007). An important caveat, however,
is that a large portion of the total population may have to be removed each
year to maintain stable hog numbers (Caley 1993, Dzieciolowski et al. 1992,
Sweitzer et al. 2000).
Home range
Home-range sizes of Feral Hogs in our study area were generally comparable
to elsewhere in the southeastern US (i.e., South Carolina [Kurz and
Marchinton 1972, Wood and Brenneman 1980] and Tennessee [Singer et al.
1981]) except markedly smaller than in Texas (Adkins and Harveson 2007,
Gabor et al. 1999, Mersinger and Silvy 2007, Yarrow 1987). Home-range size
may be inversely related to precipitation (and hence food resources). Adkins
and Harveson (2007) described a similar relationship between precipitation
and Feral Hog densities across an east–west gradient within Texas.
In South Carolina, hog home ranges were larger during the winter–spring
season (which overlapped our wet season) than either summer or fall (Kurz
and Marchinton 1972), but Feral Hogs in our study area occupied smaller
home ranges during winter. Several factors may have contributed to this
difference. First, food sources were more available during winter (our wet
season) compared to summer (our dry season), the opposite of of Kurz and
Marchinton (1972). Early in the wet season, waste grain was abundant in
agricultural fields, acorns were available in bottomlands, and crops planted
in managed openings were maturing. Increased food resources may allow
animals to meet their energetic requirements in a smaller area (Saunders
1966), resulting in smaller home ranges. This increased food availability may
partially account for the smaller home ranges observed in our wet season.
Second, many areas were fl ooded during the wet season, both naturally
and mechanically (to create winter waterfowl habitat). Thus, home ranges
may have been restricted during the wet season simply because hogs were
excluded from part of their home range (i.e., these habitats may have been
unavailable). However, the water levels (and hence, true availability) varied
over both time and space, and hogs could have potentially accessed (waded
in or through) much of the seasonally fl ooded habitat. Additionally, fl ooded
habitats represented a maximum of 13% and 8% of the total habitat for 2nd
and 3rd-order selection, respectively. Thus, fl ooding of habitats cannot entirely
account for the large difference in home-range size between the wet
and dry seasons.
Third, hunting activity (either direct hunting of hogs or activities related
to waterfowl hunting) may have caused hogs to avoid portions of
2009 R. Hayes, S. Riffell, R. Minnis, and B. Holder 421
their dry-season home range during this period as well. Beyers and Labisky
(2005) documented a similar phenomenon with White-tailed Deer.
Although several studies have suggested that boars tend to have larger
home ranges than sows (Baber and Coblentz 1986; Saunders and Kay 1991,
1996), Saunders and McLeod (1999) pooled data from several home-range
studies and found that there were no differences in home-range size after
accounting for body mass. We found no evidence of effects of either body
mass or sex on home-range size. This finding suggests that home-range size
in the southeastern US may be driven by availability of suitable habitat and/
or food resources.
Habitat use
During the dry season, ranking of habitats based on animal locations
were very similar for both 2nd- and 3rd-order selection. Animal locations were
closer than expected to seasonally fl ooded old fields (2nd-order only), old
fields, and managed openings (Table 3). Old-field habitats in our study area
were comprised of dense vegetation that hogs prefer (Baber and Coblentz
1986, Barrett 1982, Mersinger and Silvy 2007). Additionally, the seasonally
fl ooded old fields may have provided cool areas for wallowing and available
food sources during the dry season (Wood and Brenneman 1980). Hogs were
observed feeding on smartweed (a common moist soil plant in these areas)
in seasonally fl ooded old fields after the water had receded early in the dry
season (R.C. Hayes, pers. observ.).
Hogs also were associated with managed openings (at both the 2nd
and 3rd order) during the dry season. These wildlife food plots may have
provided another important food source for hogs. Baber and Coblentz
(1986) reported that cultivated areas, similar to managed openings, composed
only 1% of their study area, yet were heavily exploited. Our results
suggest that hogs preferentially selected habitats that permitted access
to these potential food sources. Land managers around MWMA often
reported hog damage to these openings (D. Adams, Malmaison Wildlife
Management Area, MS, pers. comm.). Considering the time and money
invested in creating managed openings by landowners, this points to a potential
conflict with Feral Hog populations.
The remaining 3 habitats—bottomland hardwoods, agricultural fields,
and seasonally fl ooded hardwoods—were not used differently from random.
It was surprising that bottomland hardwoods did not infl uence home-range
placement because bottomland is traditionally considered a key habitat for
Feral Hogs in the southeastern US (Kurz and Marchinton 1972). One possible
explanation is that bottomland hardwood habitat may not have been
limiting in our study area. Hogs had access to ample bottomland hardwood
habitat wherever their home range was positioned.
It was not surprising, however, that seasonally flooded hardwoods
were not preferred. Flooded hardwood stands may produce fewer acorns
422 Southeastern Naturalist Vol. 8, No. 3
than adjacent, unflooded hardwoods (Francis 1983). Our observations
suggest this was the case in our study area (R.C. Hayes, pers. observ.).
Aside from poor mast production, these areas offered little groundcover
vegetation. Neither was it surprising that agricultural fields were not
preferred by hogs during the dry season because there was little food or
cover available in agricultural habitat until crops matured towards the end
of the dry season.
We observed changes in both 2nd- and 3rd-order habitat selection during
the wet season. Although home-range placement continued to be associated
with old fields, they were no longer associated with fl ooded old fields or
managed openings (Table 4). These changes may be explained by the onset
of hunting season and changes in management. Specifically, fl ooding of the
seasonally fl ooded bottomland and old-field habitats may have excluded the
hogs from using these habitats during this part of the year.
The dense cover provided by old-field habitat may have been more important
when hunting pressure began. Old-field habitat was rarely disturbed
by humans (R.C. Hayes, pers. observ.) and may have provided safe loafing
and feeding areas when hogs were being hunted and excluded from seasonally
fl ooded habitats. Similarly, Swenson (1982) found that Odocoileus
hermionus Rafinesque (Mule Deer) selected habitats with more cover during
hunting season, presumably in response to hunting pressure (see also
Kilgo et al. 1998). Old-field habitat was frequently adjacent to agricultural
habitat where crops such as corn and cotton were maturing. Thus, old-field
habitat may have provided excellent escape cover immediately adjacent to a
viable food source during a time of high hunting pressure.
A second important change in 3rd-order selection was that all habitat types
appeared to be used randomly within the home range during the wet season.
Flooding of previously used habitats and hunting pressure may have caused
hogs to shift their habitat-use patterns to find necessary cover and alternate
food sources. Such shifts could cause habitat selection to appear random.
Furthermore, our small sample size for wet season may have resulted in low
power to detect non-random habitat use.
Our results suggest that hunting may infl uence small-scale patterns of
habitat use. Thus, hunting might be effective at reducing hog damage to localized
resources such as wildlife food plots, even though it may not reduce
hog numbers over large areas. Hunting, trapping, and general harassment
of hogs near sensitive areas may cause hogs to relocate to areas where they
may not cause as much damage (Barrett and Birmingham 1995, Maillard and
Fournier 1995). Engeman et al. (2007) found that although sport hunting removed
only a small number of animals, hog damage in areas open to hunting
was less than half that in unhunted areas. Hunting in areas and during times
when specific resources are most susceptible to damage may help reduce
damage and confl icts at a local scale.
2009 R. Hayes, S. Riffell, R. Minnis, and B. Holder 423
Acknowledgments
D. Adams, M. Cooper, and J. Lee provided logistical support in the field. S.
Demarais provided input at several stages of this project. Anonymous reviewers
provided helpful comments on an earlier version of this manuscript. We thank the
Mississippi Department of Wildlife, Fisheries, and Parks (Study #63, Project #
W–84–53) and the Mississippi Agricultural and Forestry Experiment Station for
financial support. This manuscript is publication #WF260 of the Forest and Wildlife
Research Center, Mississippi State University.
Literature Cited
Adkins, R.H., and L.A. Harveson. 2007. Demographic and spatial characteristics
of Feral Hogs in the Chihuahuan Desert, Texas. Human-Wildlife Confl icts
1:152–160.
Baber, D.W., and B.E. Coblentz. 1986. Density, home range, habitat use, and reproduction
in Feral Hogs on Santa Catalina Island. Journal of Mammalogy
67:512–525.
Barrett, R.H. 1982. Habitat preference of Feral Hogs, deer, and cattle on a Sierra
foothill range. Journal of Range Management 35:342–346.
Barrett, R.H., and G.H. Birmingham. 1995. Wild pigs. Pp. 65–70, In S.E. Hygnstrom,
R.M. Timm, and G.E. Larson (Eds.). Prevention and Control of Wildlife Damage.
Institute of Agricultural and Natural Resources, University of Nebraska,
Lincoln, NE.
Beach, R. 1993. Depredation problems involving Feral Hogs. Pp. 67–75, In C.W.
Hanselka and J.F. Cadenhead (Eds.). Feral Swine: A Compendium for Resource
Managers. Texas Agricultural Extension Service, College Station, TX.
Beyers, K.M., and R.F. Labisky. 2005. Infl uence of fl ood waters on survival, reproduction,
and habitat use of White-tailed Deer in the Florida everglades. Wetlands
25:659–666.
Burnham, K.P., and D.R. Anderson. 2002. Model Selection and Multi-model Inference:
A Practical Information-theoretic Approach. Springer-Verlag, New
York, NY.
Caley, P. 1993. Population dynamics of Feral Pigs (Sus scrofa) in a tropical riverine
habitat complex. Wildlife Research 20:625–636
Conner, L.M., and B.W. Plowman. 2001. Using Euclidian distances to asses nonrandom
habitat use. Pp. 275–290, In J.J. Millspaugh and J.M. Marzluff (Eds.).
Radio-tracking and Animal Populations. Academic Press, San Diego, CA.
Dinsmore, S.J., G.C. White, and F.L. Knopf. 2002. Advanced techniques for modeling
avian nest survival. Ecology 83:3476–3488.
Dtichkoff, S.S., and B.C. West. 2007. Ecology and management of Feral Hogs.
Human-wildlife Confl icts 1:149–151.
Dzieciolowski, R.M., C.M.H. Clark, and C.M. Frampton. 1992. Reproductive characteristics
of Feral Pigs in New Zealand. Acta Theriologica 37:259–270.
Engeman, R.M., A. Stevens, J. Allen, J. Dunlap, M. Danial, D. Teague, and B.
Constantin. 2007. Feral swine management for conservation of an imperiled
wetland habitat: Florida’s vanishing seepage slopes. Biological Conservation
134:440–446.
424 Southeastern Naturalist Vol. 8, No. 3
ESRI. 2002. ArcView, version 3.2. Environmental Systems Research Institute, Redlands,
CA.
ESRI. 2004. ArcMap, version 9.0. Environmental Systems Research Institute, Redlands,
CA.
Francis, J.K. 1983. Acorn production and tree growth of Nuttall Oak in a green-tree
reservoir. US Forest Service Publication SO-209, Southern Forest Experiment
Station, New Orleans, LA.
Gabor, T.M., E.C. Hellgren, R.A. Van Den Bussche, and N.J.Silvy. 1999. Demography,
sociospatial behavior and genetics of Feral Pigs (Sus scrofa) in a semi-arid
environment. Journal of Zoology 247:311–322.
Gipson, P.S., B. Hlavachick, and T. Berger. 1998. Range expansion by Wild Hogs
across the central United States. Wildlife Society Bulletin 26:279–286.
Harveson, P.M., M.E. Tewes, G.L. Anderson, and L.L. Laack. 2004. Habitat use by
ocelots in south Texas: Implications for restoration. Wildlife Society Bulletin
32:948–954.
Hayes, R.C. 2007. Feral Hogs in central Mississippi: Home range, habitat use, and
survival. M.Sc. Thesis. Mississippi State University, Starkville, MS.
Henry, V.G. 1969. Predation on dummy nests of ground-nesting birds in the southern
Appalachians. Journal of Wildlife Management 33:169–172.
Holder, B.D. 2006. Survival, habitat use, and nest-site characteristics of Wild
Turkeys in central Mississippi. M.Sc. Thesis. Mississippi State University,
Starkville, MS.
Holm, S. 1979. A sequentially rejective multiple test procedure. Scandinavian Journal
of Statistics 6:65–70.
Hooge, P.N., and B. Eichenlaub. 1997. Animal movement extension in ArcView,
version 1.1 and later. Alaska Biological Science Center, US Geological Survey,
Anchorage, AK.
Jezierski, W. 1977. Longevity and mortality rates in a population of Wild Boar. Acta
Theriologica 22:337–348.
Johnson, D.H. 1980. The comparison of usage and availability measurements for
evaluating resource preference. Ecology 61:65–71.
Kaller, M.D., and W.E. Kelso. 2006. Swine activity alters invertebrate and microbial
communities in a coastal watershed. American Midland Naturalist 156:165–179.
Kilgo, J.C., R.F. Labisky, and D.E. Fritzen. 1998. Infl uences of hunting on the behavior
of White-tailed Deer: Infl uences for conservation of the Florida Panther.
Conservation Biology 12:1359.
Kurz, J.C., and R.L. Marchinton. 1972. Radio telemetry studies of Feral Hogs in
South Carolina. Journal of Wildlife Management 36:1240–1248.
Littauer, G.A. 1993. Control techniques for Feral Hogs. Pp. 139–148, In C.W.
Hanselka and J.F. Cadenhead (Eds.). Feral Swine: A Compendium for Resource
Managers. Texas Agricultural Extension Service, College Station, TX.
Maillard, D., and P. Fournier. 1995. Effects of shooting with hounds on size of resting
range of Wild Boar (Sus scrofa) groups in Mediterranean habitat. Journal of
Mountain Ecology 3:102–107.
Massei, G., P.V. Genov, B.W. Staines, and M.L. Gorman. 1997. Mortality of Wild
Boar, Sus scrofa, in a Mediterranean area in relation to sex and age. Journal of
Zoology 242:400–404.
2009 R. Hayes, S. Riffell, R. Minnis, and B. Holder 425
Matschke, G.H. 1965. Predation by European Wild Hogs on dummy nests of groundnesting
birds. Pp. 154–156, In Proceedings of the 18th Annual Conference of
Southeastern Association of Game and Fish Commissioners. Tulsa, OK.
Mersinger, R.C., and N.J. Silvy. 2007. Range size, habitat use, and dial activity of
Feral Hogs on reclaimed surface-mined lands in east Texas. Human-wildlife
Confl icts 1:161–167.
Phillips, M.L., W.R. Clark, S.M. Nusser, M.A. Sovada, and R.J. Greenwood. 2004.
Analysis of predator movement in prairie landscapes with contrasting grassland
competition. Journal of Mammalogy 85:187–195.
Pine, D.S., and G.L. Gerdes. 1973. Wild Pigs in Monterey County, California. California
Fish and Game Quarterly 59:126–137.
Saunders, G.C. 1966. The study of mammal movements: A review. Journal of Wildlife
Management 30:215–235.
Saunders, G., and B. Kay. 1991. Movements of Feral Pigs (Sus scrofa) at Sunny
Corner, New South Wales. Wildlife Research 18:49–61.
Saunders, G., and B. Kay. 1996. Movements and home ranges of Feral Pigs in Kosciusko
National Park, New South Wales. Wildlife Research 23:711–719.
Saunders, G., and S. Mcleod. 1999. Predicting home-range size from the body mass
or population densities of Feral Pigs, Sus scrofa. Australian Journal of Ecology
24:538–543.
Seaman, D.E., J.J. Millspaugh, B.J. Kernohan, G.C. Brundige, K.J. Raedeke, and
R.A. Gitzen. 1999. Effects of sample size on kernel home-range estimates. Journal
of Wildlife Management 63:739–747.
Singer, F.J., D.K. Otto, A.R. Tipton, and C.P. Hable. 1981. Home ranges, movements,
and habitat use of European Wild Boar in Tennessee. Journal of Wildlife
Management 45:343–353.
Springer, M.D. 1977. Ecological and economic aspects of Wild Hogs in Texas. Pp.
37–64, In G.W. Wood (Ed.). Research and Management of Wild Hog Populations:
Proceedings of a Symposium. Belle W. Baruch Forest Science Institute,
Clemson University, Georgetown, SC.
Sweitzer, R.A., B.J. Gonzales, I.A. Gardner, D. Van Vuren, J.D. Waithman, and W.M.
Boyce. 1997. A modified panel trap and immobilization techniques for capturing
multiple Wild Pigs. Wildlife Society Bulletin 25:699–705.
Sweitzer, R.A., D. Van Vuren, I.A. Gardener, W.M. Boyce, and J.D. Waithman. 2000.
Estimating sizes of Wild Pig populations in the north and central coast regions of
California. Journal of Wildlife Management 64:531–543.
Swenson, J.E. 1982. Effects of hunting on habitat use by Mule Deer on mixed-grass
prairie in Montana. Wildlife Society Bulletin 10:115–120.
Waithman, J.D., R.A. Sweitzer, D. Van Vuran, J.D. Drew, A.J. Brinkhaus, I.A. Gardner,
and W.M. Boyce. 1999. Range expansion, population sizes, and management
of Wild Pigs in California. Journal of Wildlife Management 63:298–308.
Westfall, P.H., R.D. Tobias, D. Rom, R.D. Wolfinger, and Y. Hochberg. 1999. Multiple
Comparisons and Multiple Tests Using the SAS System. SAS Institute, Inc.,
Cary, NC.
White, G.C., and K.P. Burnham. 1999. Program MARK: Survival estimation from
populations of marked animals. Bird Study 46:120–139.
Withey, J.C., T.D. Bloxton, and J M. Marzluff. 2001. Effects of tagging and location
error in wildlife radiotelemetry studies. Pp. 43–75, In J.J. Millspaugh and J.M.
Marzluff (Eds.). Radio Tracking and Animal Populations. Academic Press, San
Diego, CA.
426 Southeastern Naturalist Vol. 8, No. 3
Woodall, P.F. 1983. Distribution and population of Dingoes (Canis familiaris) and
Feral Pigs (Sus scrofa) in Queensland, 1945–1976. Journal of Applied Ecology
20:85–95.
Wood, G.W., and R.E. Brenneman. 1980. Feral Hog movements and habitat use
coastal South Carolina. Journal of Wildlife Management 44:420–427.
Yarrow, G.W. 1987. The potential for interspecific resource competition between
White-tailed Deer and Feral Hogs in the Post Oak savannah region of Texas.
Ph.D. Dissertation. Steven F. Austin State University, Nacogdoches, TX.