Content uploaded by Andrea B Webster
Author content
All content in this area was uploaded by Andrea B Webster on Dec 14, 2021
Content may be subject to copyright.
The determinants of mesocarnivore
activity patterns in highveld grassland
and riparian habitats
Andrea B. Webster1( )§, Mariëtte E. Pretorius1( ) & Michael J. Somers1,2*( )
1Mammal Research Institute, Department of Zoology and Entomology, University of Pretoria, Pretoria, South Africa
2Centre for Invasion Biology, Department of Zoology and Entomology, University of Pretoria, Pretoria, South Africa
Received 29 August 2021. To authors for revision 20 September 2021. Accepted 28 November 2021
Despite the diversity of mesocarnivores and the broad geographic ranges of these species,
our understanding of their behaviour and ecology at multi-species and community levels is
limited. Our study was conducted between April and mid-July 2015 and used data collected
over 105 days from 39 camera traps to quantify activity patterns of sympatric meso-
carnivores in riparian and grassland habitats of Telperion Nature Reserve, South Africa. A
total of 13 mesocarnivore species were detected within this relatively small (~7350 ha)
reserve. Sufficient records (≥10 records) were obtained for rusty-spotted genet (
Genetta
maculata
), black-backed jackal (
Canis mesomelas
), otter species (African clawless otter,
Aonyx capensis
, and spotted-necked otter,
Hydrictis maculicollis
), serval (
Leptailurus
serval
), slender mongoose (
Galerella sanguinea
), yellow mongoose (
Cynictis penicillata
) and
marsh mongoose (
Atilax paludinosus
). Generalized linear models were used to investigate
whether species ID, temperature, vegetation characteristics or moon phase best predicted
temporal activity.To assess which species had the highest potential for competitive interac-
tion, we also quantified the coefficient of activity overlap. Our results show that species ID
and temperature were the best predictors of mesocarnivore activity. Slender and yellow
mongooses displayed the highest coefficient of activity overlap (0.90), followed by marsh
mongoose and rusty-spotted genet (0.80), and serval and rusty-spotted genet (0.79). These
species are likely to have the highest potential for competitive interactions, but preferences
for different vegetation characteristics and variations in the estimated relative abundance
may point to coexistence through spatial and fine-scale temporal partitioning. The other spe-
cies exhibited lower coefficients of activity overlap with each other, suggesting they may
coexist through temporal partitioning of resources.
Keywords: carnivore guild, activity patterns, species coexistence, small carnivores, biodioversity.
INTRODUCTION
Understanding coexistence mechanisms between
ecologically similar species is necessary to under-
stand ecological communities and their continued
persistence (Violle
et al
., 2012). Interspecific
competition affects the ability of each species to
access limited resources and ultimately shape
ecological niches (Chesson, 2000). Species may
adapt to interspecific competition by partitioning
resources along three main, but not mutually
exclusive, dimensions. 1) Spatial partitioning,
exploiting the same or similar resources in differ-
ent areas, 2) temporal partitioning, exploiting the
same resources at different times or 3) trophic
niche partitioning, exploiting different resources
altogether. These strategies ensure that interac-
tions and, therefore, inter-species competition is
reduced (Donadio & Buskirk, 2006). Inter-species
differences in morphology, behaviour, and physi-
ology can also mediate interspecific competition
(Loveridge & Macdonald, 2003). In small nature
reserves or urban parks, interspecific competition
may intensify when available habitat is limited.
Subsequent interspecific contact rates, and the
possibility of one or more competitive species
impacting the occurrence and activity times of
another, may result in some species becoming
subordinate (Gehrt
et al
., 2013; Massara
et al
.,
2016).
The effects of interspecific interactions are most
prominent within the carnivore guild (Palomares &
Caro, 1999). Understanding how carnivore–carni-
*To whom correspondence should be addressed.
E-mail: michael.somers@up.ac.za
RESEARCH ARTICLE
African Journal of Wildlife Research 51: 178–192 (2021)
ISSN 2410-8200 [Online only] — DOI: https://doi.org/10.3957/056.051.0178
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
vore interactions affect patterns of co-occurrence
and the traits that influence interspecific interac-
tions are important components of understanding
niche dynamics, coexistence and mesocarnivore
release (Monterroso
et al
., 2020). Compared to
large predators, mesocarnivores are habitat and
resource generalists, are species-rich and gener-
ally more abundant (Roemer, Gompper & Van
Valkenburgh, 2009). Mesocarnivores are small to
medium sized carnivores weighing ≤15 kg, with a
diet of 50–70% meat that occupy a trophic position
below the large carnivores (weighing >20 kg)
(Ritchie & Johnson, 2009; Bird & Mateke, 2013). A
single ecosystem may support several mesocarni-
vore species; however, a mesocarnivore in one
ecosystem may fill the role of an apex predator in
another (Roemer, Gompper & Van Valkenburgh,
2009).
Transformed and urbanized environments favour
generalist species (Clavel, Julliard & Devictor,
2011). South Africa is a developing country with
one of the fastest urbanization rates worldwide
(Saghir & Santoro, 2018). It also hosts a large
number of protected areas and some of the richest
biodiversity in the world (Skowno
et al
., 2019),
including a variety of mammalian mesocarnivore
species (Skinner & Chimimba, 2005). Meso-
carnivore community composition within protec-
ted areas is influenced by various abiotic and
biotic factors, including temperature, habitat
requirements, interspecific relationships, human
pressures and protected area attributes (Tambling
et al
., 2018). In general, carnivore occurrence
within protected areas in southern Africa is influ-
enced by the location of permanent water sources,
with higher mesocarnivore occupancy closer to
water (Schuette
et al
., 2013; Rich
et al
., 2017).
Dense vegetation along riparian habitats also
attracts mesocarnivores, as it may provide
concealment during hunting and refuge from
interspecific predation (Boydston
et al
., 2003;
Santos
et al
., 2011). Increased habitat variability,
vegetation and terrain diversity may also support
more generalist carnivore species (Roemer,
Gompper & Van Valkenburgh, 2009).
Despite the variety of mesocarnivores in South
Africa, studies have primarily been on single
species, descriptive and focused mainly on diurnal
species (Do Linh San
et al
., 2013). Consequently,
we still know little about mesocarnivore behaviour
and many aspects related to their ecology
and intraguild interactions at multi-species and
community levels (González-Maya, Schipper &
Benitez, 2009; Do Linh San
et al
., in press). The
overall aim of this study was twofold: 1) to identify
which mesocarnivore species were present in a
South African nature reserve comprising grass-
land and riparian habitats, and 2) to determine the
activity patterns of each mesocarnivore species as
a measure of coexistence mechanisms (spatial or
temporal niche partitioning) within this guild. We
further evaluated environmental factors that could
influence daily activity periods for each meso-
carnivore species, predicting that 1) the vegeta-
tion characteristics of the landscape would most
affect mesocarnivore activity and 2) temperature
would affect mesocarnivore species activity times
differently.
METHODS
Study area
The study was conducted at the Telperion
Nature Reserve (between the latitudes of 25°38’
and 25°44’S and longitudes of 28°55’E and
29°02’E), from April to mid-July 2015. The
property (~7350 ha) lies in the ecotone between
the Rand Highveld Grassland biome of Gauteng
province near Bronkhorstspruit, and the Loskop
Mountain Bushveld Savanna Biome near
Emalahleni in Mpumalanga province, South Africa
(Mucina & Rutherford, 2006; Coetzee, 2012).
Rhyolite, sandstone and minor shale support
predominantly dry and wet degraded grassland
(Coetzee, 2012) and the foothills of rocky ridges
support a general woodland community, which
may extend into grassland. In the lower-lying
riparian areas, dense vegetation gives way to
rocky ledges and sandy patches along the river
(Helm, 2006). Several small tributaries feed into
the Wilge River, a perennial water source that
flows south to north through the reserve for
19.66 km to form wetlands and reed beds, enhanc-
ing water availability in the dry season (Helm,
2006; Coetzee, 2012). Mean temperatures range
from 14°C to 27°C in the wet season from Septem-
ber to March, and mean annual rainfall is ~650 mm
(Bronkhorstspruit Weather Station). In the dry
season, from April to August, mean temperatures
range between 4°C and 18°C with frost in the early
mornings (Witbank Weather Station).
The property is privately owned, and except
for the active management of fire, is managed
using a non-intervention approach. Rehabilitation
from historical crop and livestock farming for
wildlife conservation, eco-tourism, education and
Webster
et al.
: Mesocarnivore activity patterns 179
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
research is ongoing. The reserve supports a diver-
sity of herbivores, some of which historically
occurred in the province, while others have been
introduced (Helm, 2006; Coetzee, 2012).
Study design
We used a random stratified sampling method to
deploy 40 Bushnell HD Trophy Cam camera traps
(Model 119537, 8 Megapixel sensor with 32
Hyper-night vision LED flash) throughout grass-
land (
n
= 29) and riparian (
n
= 11) areas of the
reserve (Fig. 1). However, camera number 7 in the
riparian zone went missing within the first week of
data collection. Different characteristics within
riparian and grassland landscapes were deter-
mined visually (Table 1), and active game paths
were identified through track and sign interpreta-
tion (Liebenberg, 1990). A handheld Global
Positioning System (GPS) device (Garmin eTrex
20) was used to record each trap site’s longitude
and latitude coordinates. Trap sites held one
camera, elevated between 50 cm and 1 m above
ground depending on natural attachment sites
on or near a game path or game path junction to
maximize visibility and capture potential for
mesocarnivore species (O’Connell, Nicols &
Karanth, 2011; Hamel
et al
., 2013). The standard-
ized minimum distance (Karanth & Nichols, 2000;
Jackson
et al
., 2005; Kelly, 2008) was maintained
between traps at ≥800 m in grassland but were
reduced to ≥400 m along only 8.37 km of accessi-
ble riverfront in the riparian habitat. All cameras
were set to normal sensitivity, synchronized for
time and pre-programmed to record the date,
temperature, and phase of the moon. The temper-
atures recorded on the camera traps were consid-
180 African Journal of Wildlife Research Vol. 51, 2021
Fig. 1. Random stratified placement of 11 riparian (light grey marker) and 29 grassland (black marker) camera trap
stations at Telperion Nature Reserve, on the border between the Gauteng and Mpumalanga province, South Africa,
between 1 April and 17 July 2015.
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
ered to be correlated with ambient air temperature
(Hofmann
et al
., 2016). A 30-second time delay
was used between capture events, and a single
photograph was produced in response to motion
detection (Karanth & Nicols, 2000; Kelly, 2008;
O’Connell
et al
., 2011). Camera trap data were
collected over six weeks, with cameras recording
data continuously over 24 hours for a total of 105
days. Cameras, batteries and secure digital (SD)
card capacity were checked every two weeks from
1 April to 17 July 2015. Data from SD cards were
downloaded to a laptop computer in the field.
SD cards were re-formatted, and camera set-
tings checked and corrected after each survey.
Batteries were replaced as necessary.
Detection rates and mesocarnivore activity
We calculated trap detection rates by summing
the total number of detections that yielded
mesocarnivore data for each trap within each
habitat type (grassland or riparian) and divided this
by the number of traps deployed in each area.
Using photo-detections from all 13 mesocarnivore
species detected (Table 2), we determined
species composition on the property. The propor-
tion (%) of time each mesocarnivore species was
detected in each landscape was determined by
dividing the number of detections per species per
landscape type (riparian or grassland), and the
times at which each mesocarnivore species was
active were recorded and plotted. Detections of
the same species recorded at 60-second intervals
or less were removed to minimize autocorrelation
(Sollmann, 2018; Havmøller
et al
., 2021; Kays
et al
., 2021). In all cases, this removed successive
detections of the same individual at a specific trap
site. Rather than identifying individuals, our focus
was on identifying the mesocarnivore species
present and comparing detection rates between
species and areas. Therefore, we did not expect
the time between independent photographs to
introduce a bias toward either one of these factors
(Jenks
et al
., 2011). Two large carnivore species,
leopards (
Panthera pardus
) and brown hyaenas
(
Parahyaena brunnea
), were also detected from
photographs but were not included in the analy-
Webster
et al.
: Mesocarnivore activity patterns 181
Tab le 1. Classifications assigned to the vegetation characteristics, timing of activity, moon phase and other environ-
mental variables used for classifying explanitory variables in the statistical models of mesocarnivore activity in the
Telperion Nature Reserve, South Africa. Numbers in brackets in the vegetation category indicate number of camera
traps deployed.
Category Rationale
Habitat Grassland Dry Open (10) Relatively homogenous, dry open grassland, various elevations
Rocky Outcrops (11) Rocky outcrops in dry grassland, little vegetation cover, various
elevations
Wet near tributaries (5) Vleis, marsh, or in close proximity to tributary in open
grassland – cannot be characterized as true riparian
Wooded areas (3) Concentrated clusters of trees surrounded by open dry
grassland
Riparian Dense vegetation (4) Little sky visible – dense vegetation along game paths
Rocky ledges adjacent to river (2) Rocky ledges adjacent to the river
Drainage line pathway (2) Open areas along game paths in Riparian vegetation – patches
of little cover
Sandy patches adjacent to the river (3) Sand predominates, minimal grass and little tall vegetation
Time of activity Midnight to dawn 00:01 to 06:00
Dawn to midday 06:01 to 12:00
Midday to dusk 12:01 to 18:00
Dusk to midnight 18:01 to 24:00
Moon phase New moon and waning crescent 0–25% illumination
First quarter and waxing gibbous 50–75% illumination
Full moon and waning gibbous 100–75% illumination
Last quarter and waning crescent 50–25% illumination
Temperature °C
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
182 African Journal of Wildlife Research Vol. 51, 2021
Tab le 2. Complete list of carnivores and mesocarnivores identified at the study site.The number of detections per species (
n
), proportion of detections in each habitat type
for seven species of mesocarnivores in the grassland and riparian landscapes and the overall relative abundance index (RAI) and the characteristics of detection locations
in the Telperion Nature Reserve from April–July 2015.
Detections (%) in each Relative abundance
Body mass habitat type index
Family Scientific name Common name (kg) (n) Riparian Grassland (RAI) Vegetation characteristic
1 Felidae Felis nigripes Black-footed cat 1.6–2.45 1 – 100% Dry open grassland
2Leptailurus serval Serval 9–18 15 1% 99% 0.53 Wet grassland, Dry open grassland, Riparian sandy
patches
3Caracal caracal Caracal 7–19 6 – 100% Dry open grassland, Rocky outcrop, Wet grassland
4 Canidae Canis mesomelas Black-backed jackal 6–13 505 9% 91% 12.47 All characteristics of grassland, all characteristics of
Riparian except rocky ledges
5 Hyaenidae Proteles cristatus Aardwolf 10–14 6 50% 50% Dry open grassland, wet grassland, riparian rocky
ledges
6 Viveridae Genetta maculata Rusty spotted genet 1.3–3 136 77% 23% 3.32 Woodland and rocky outcrops in grassland, All
characteristics of Riparian
7 Herpestidae Galerella sanguinea Slender mongoose 0.46–0.57 118 56% 44% 2.88 Rocky outcrops, wet grassland, All characteristics of
Riparian
8Cynictis penicillata Yellow mongoose 0.6–3 85 – 100% 2.08 Dry open grassland
9Atilax paludinosus Marsh mongoose 2.5–2.9 82 44% 56% 2.00 Wet grassland, All characteristics of Riparian
10 Suricata suricatta Meerkat 0.62–0.97 5 – 100% Dry open grassland
11 Mustelidae Aonyx capensis African clawless otter 3–6.5 41 12% 88% 1.22 Wet grassland (fast-flowing shallow tributaries),
Lutra maculicollis Spotted-necked otter Riparian dense vegetation
12 Mellivora capensis Honey badger 9–16 1 – 100% Grassland rocky outcrops
13 Ictonyx striatus Striped polecat 0.6–1.3 8 – 100% Dry open grassland, Rocky outcrops
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
ses. Insufficient data (<10 detections) were recor-
ded for caracal (
Caracal caracal
), honey badger
(
Mellivora capensis
), black-footed cat (
Felis
nigripes
), striped polecat (
Ictonyx striatus
),
aardwolf (
Proteles cristatus
), and meerkat
(
Suricata suricatta
) and were removed from the
dataset before statistical modelling. Data were
pooled for African clawless otter (
Aonyx capensis)
and spotted-necked otter (
Hydrictis maculicollis
),
as it was not always possible to identify individuals
to species level. In addition to pooled data from
‘otter sp.’, rusty-spotted genet (
Genetta macu-
lata)
, black-backed jackal (
Canis mesomelas
),
serval (
Leptailurus serval
), slender mongoose
(
Galerella sanguinea
), marsh mongoose (
Atilax
paludinosus
) and yellow mongoose (
Cynictis
penicillata
) were detected and used for statistical
analyses and mesocarnivore activity plots. We
calculated the relative abundance index (RAI) for
each of the seven species to evaluate differences
in the detection rates by taking the sum of all
detections for each species multiplied by 100 and
divided by the total sampling effort (39 × 105)
(Karanth & Nichols, 1998; Jenks
et al
., 2011).
We acknowledge that care must be taken when
interpreting RAIs, since they are influenced by
the behaviour and movement patterns of study
species, camera trap setup and the size of the
study area (Sollmann
et al
., 2013).
Statistical analysis and predictors of
mesocarnivore activity
To test whether the predictor variables; species
ID (categorical), temperature (°C, numeric), vege-
tation characteristic (Rocky Grassland, Wet
Grassland, Wooded Grassland, Dense Riparian,
Open Riparian, Rocky Riparian, and Sandy Ripar-
ian or moon phase (Full, Last Quarter, New Moon,
Waning Crescent, Waning Gibbous, Waxing Cres-
cent and Waxing Gibbous) predicted meso-
carnivore time of detection (numeric response
variable), generalized linear models (GLMs) with
Gaussian error distributions were constructed in
the software R (version 1.1.463: 2013) (The R
Foundation for Statistical Computing 2013). The
time of detection was converted from hh:mm:ss to
decimal hours (hh:mm:ss × 24) and used for the
GLM modelling. We tested for multi-colinearity
using Variance Inflation Factors (VIF) using the R
car
package (Fox & Weisberg, 2011).All VIFs were
<5, indicating low co-linearity between covariates
(Sheather, 2009) and all covariates were subse-
quently retained in the global model.
Thirteen separate stepwise GLM models were
then constructed. To select the most suitable GLM
model, we performed a multi-model selection
using the Akaike Information Criterion (AIC), the
difference between the best model in each set
(lowest AIC value) and all other models (Δ
i
) and
Akaike weights (
W
i
) (Barto½, 2019). Models where
Δ
i
< 3 were deemed the most informative (Burn-
ham, Anderson & Huyvaert, 2011). To assess the
goodness-of-fit of the various models, we plotted
the sample and theoretical quantiles of the model
residuals (Kery & Royle 2015).The relative impor-
tance of each predictor variable (
x
i
) was then
calculated as the sum of the AIC weights of each
informative model that included the predictor
(Burnham & Anderson, 2002). Because analyses
yielded multiple parsimonious models, we used
model averaging for all models where Δ
i
<3
(Burnham & Anderson, 2002). We calculated the
model-averaged parameter estimates (average
model coefficients), adjusted standard errors
(S.E.) and associated
z
-values for variables for
the top-ranked generalized linear models and
compared the direction and magnitude of effects
of the various levels within factors (
e.g.
species ID
within the species factor). To investigate the over-
lap of the activity patterns between the seven
mesocarnivore species, the time of detection was
converted to radians following the requirements of
the
overlap
package (Ridout & Linkie, 2009). The
coefficient of overlapping (Δ1for small sample
sizes
n
< 50) and 95% confidence interval was
calculated, generating 1000 bootstrap estimates
per species (Ridout & Linkie, 2009). The overlap
coefficient ranges from 0, indicating no overlap, to
1, indicating complete overlap. All graphing was
conducted using the package
ggplot2
(Wickham,
2009).
RESULTS
Detection rates
The total sampling effort using 39 traps over
105 days yielded 420 783 total detections, of
these only 802 contained identifiable mesocarni-
vores. Camera trap detections of mesocarnivores
differed between the landscapes, with riparian
traps having 150% higher detection rate (mean ±
standard deviation = 38 ± 27.86 detections per
trap) than grassland traps (16 ± 21.31 detections).
Community composition differed between the
landscapes; six (46.15%) mesocarnivore species
(yellow mongoose, black-footed cat, caracal,
Webster
et al.
: Mesocarnivore activity patterns 183
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
meerkat, honey badger and striped polecat) were
detected only in grassland, while seven (53.84%)
mesocarnivore species (otter spp., serval,
black-backed jackal, aardwolf, rusty-spotted
genet, slender mongoose and marsh mongoose)
were detected in both landscape types (Table 2).
None of the species were exclusively detected in
riparian habitat. Black-backed jackals were the
most abundant species detected in the study area
(RAI = 12.47), with detections of servals being the
least abundant (RAI = 0.53). Black-backed jackals
also had the highest overall detection rate, with the
most detections recorded in the grassland land-
scape (10.65), whilst servals had the lowest detec-
tion rate (0.48) (Fig. 2). The rusty-spotted genet
was the most frequently detected species in ripar-
ian areas (8.54). The different mesocarnivores
were also detected in different vegetation types
throughout the study site (Table 2). Although
slender mongooses were detected in both riparian
and grassland landscapes, they were not detected
in dry open grassland; areas occupied predomi-
nantly by yellow mongooses. Instead, when
utilizing grassland, slender mongooses confined
their activities to the rocky outcrops in grassland
where the other two mongoose species were not
detected. When using the riparian areas, slender
mongooses were detected mostly in densely
vegetated areas.
Mesocarnivore activity patterns
Black-backed jackals, rusty-spotted genets and
marsh mongooses showed crepuscular peaks of
activity, with low to zero activity detected during the
day (Fig. 3A,B,D).Servals also showed crepuscu-
lar peaks in activity but were most active two
hours after dusk (Fig. 3E). Slender and yellow
mongooses showed diurnal peaks in activity
(Fig. 3F,G). The otter species were most active in
the four hours following dawn, with low activity
around dawn and dusk (Fig. 3C).
Predictors of mesocarnivore activity
Species ID and temperature (relative impor-
tance
x
1and
x
2= 100% across two informative
models) were predictors of variation in the activity
time between the seven mesocarnivore species
used for statistical modelling (Table 3). Vegetation
characteristics (relative importance
x
3= 20% for
one model) and moon phase (relative importance
x
4= 14% for one model) predicted variation in
times of activity for mesocarnivores to a lesser
extent.
Parameter estimates for the parsimonious
model differed in direction (+/–) and magnitude
across the four predictor variables used to explain
the variation observed in mesocarnivore activity
(Fig. 5). Serval and marsh mongoose activity times
were not significantly different when compared to
184 African Journal of Wildlife Research Vol. 51, 2021
Fig. 2. Detection rate (number of detections/total number of camera traps) for the grassland (light grey,
n
= 29 traps)
and riparian (dark grey,
n
= 11 traps) for rusty-spotted genet (Genet,
Genetta maculata
), black-backed jackal (Jackal,
Canis mesomelas
), otter (
Lutra maculicollis and Aonyx capensis
), marsh mongoose (
Atilax paludinosus
), serval
(
Leptailurus serval
), slender mongoose (
Galerella sanguinea
) and yellow mongoose (
Cynictis penicillata
)inthe
Telperion Nature Reserve, South Africa, from April–July 2015.
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
rusty-spotted genet activity times (Table 4). Tem-
perature had a significant positive effect on activity
times of modelled mesocarnivore species, with
overall activity increasing as ambient tempera-
tures increased (Table 4). Rusty-spotted genets
and marsh mongooses were active above 5°C,
while black-backed jackals were active over the
full range of ambient temperatures recorded on
traps from –10°C to 40°C. Servals were active
within a narrow range of temperatures between
Webster
et al.
: Mesocarnivore activity patterns 185
Fig. 3. Detection counts across 24-hour periods (0:00 to 23:00) for (A) Black-backed jackal (
Canis mesomelas
),
(B) marsh mongoose (
Atilax paludinosus
), (C) otter (
Lutra maculicollis and Aonyx capensis
), (D) rusty-spotted genet
(
Genetta maculata
), (E) serval (
Leptailurus serval
), (F) slender mongoose (
Galerella sanguinea
) and (G) yellow
mongoose (
Cynictis penicillata
) in riparian (dashed black line) and grassland (solid black line) landscapes in the
Telperion Nature Reserve, South Africa, from April–July 2015. Solid grey vertical lines denote dawn (6:00) and dusk
(18:00). Note differences in
y
-axis scales.
Tab le 3. Ranked Akaike Information Criterion (AIC), difference between the top-ranked model and the
i
th model (Δ
i
)
with AIC weight (
W
i
) from generalized linear models investigating if species, temperature (Temp), moon phase
(Moon) or vegetation characteristics (Vegetation) explain the variation in the activity time of sevenmesocar nivorespe-
cies in the Telperion Nature Reserve.
Model d.f. AIC Δ
iWi
Species + Temp 9 4953.7 0 0.611
Species + Temp + Vegetation 18 4954.8 1.07 0.209
Species + Temp + Moon 19 4955.9 2.96 0.139
Species + Temp + Moon + Vegetation 28 4990.6 5.40 0.041
Temp 3 4991.6 36.90 0
Temp + Moon 13 4994.4 40.69 0
Species + Vegetation 15 4994.4 40.68 0
Temp + Vegetation 12 4995.3 41.61 0
Species 8 4997.8 44.11 0
Species + Moon 15 4998.3 44.62 0
1 (null) 2 5011.7 57.99 0
Moon 9 5012.7 59.01 0
Vegetation 9 5017.9 64.24 0
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
5°C and 15°C, while slender and yellow mon-
gooses were active during the hottest part of the
day (Fig. 4). Activity times of modelled meso-
carnivore species were influenced positively or
negatively by moon phase and vegetation charac-
teristic. The magnitude of these variables could
also be seen across species; however, neither of
these effects were statistically significant.
Slender and yellow mongooses had the highest
coefficient of activity overlap (Δ1= 0.905), followed
by marsh mongooses and rusty-spotted genets
(Δ1= 0.804) and servals and rusty-spotted genets
(Δ1= 0.798; Table 5). The lowest coefficient of
activity overlap was observed between yellow and
slender mongooses with servals (Δ1= 0.071).
DISCUSSION
Our study aimed to identify mesocarnivore
species within the Telperion Nature Reserve,
South Africa, and investigate their activity patterns.
A diverse group of 13 mesocarnivore species was
identified within this relatively small area. All
species broadly followed similar activity patterns
to those found previously throughout Africa
(reviewed in Skinner & Chimimba (2005) and
Kingdon & Hoffman (2013)) except for servals,
which in our study were detected only at night. The
type of mesocarnivore species and ambient
temperature best explained overall variation in
times of activity for the seven modelled species,
namely the otter sp., rusty-spotted genet, black-
backed jackal, serval, slender mongoose, marsh
mongoose and yellow mongoose. The various
species detected differed in their estimated
relative abundances, with black-backed jackals
detected most often, and servals detected
the least, although these results may relate to
behaviour and movement patterns of the different
study species (Sollmann
et al
., 2013). We also
acknowledge that the relatively short sampling
period may have affected the rate at which species
were detected on the camera traps and that more
longitudinal monitoring would be needed to inves-
tigate activity patterns across multiple seasons.
Black-backed jackals are highly mobile and
wide-ranging when foraging for a wide variety of
food (Kaunda, 2001), which may contribute to
higher detection rates. In comparison, servals are
more cryptic with a preference for rodent prey and
dense grassland habitats (Ramesh
et al
., 2016),
which may lead to lower detection rates. The high-
est coefficient of activity overlap was observed
between slender and yellow mongooses, followed
by marsh mongooses and rusty-spotted genets,
186 African Journal of Wildlife Research Vol. 51, 2021
Fig. 4. Boxplot showing medians and interquartile ranges (Tukey-style whiskers extend to 1.5 × IQR) of the tempera-
ture ranges during which rusty-spotted genet (
Genetta
.
maculata
), black-backed jackal (Jackal,
Canis mesomelas
),
otter (
Lutra maculicollis
and
Aonyx capensis
), marsh mongoose (
Atilax paludinosus
), serval (
Leptailurus serval
),
slender mongoose (
Galerella sanguinea
) and yellow mongoose (
Cynictis penicillata
) were active at the Telperion
Nature Reserve, South Africa, from April to July 2015.
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
and servals and rusty-spotted genets, indicating
that these species are likely to have the highest
potential for competitive interaction in this study
area.
Slender and yellow mongooses are diurnal and
have comparable omnivorous diets (Wilson &
Reeder, 2005).However, yellow mongooses were
exclusively detected in the grassland landscape,
whilst slender mongooses were detected more
often in the riparian landscape. Similar to our
findings, other studies have demonstrated differ-
ences in fine-scale habitat selection; yellow
mongooses select open habitats with short grass-
land vegetation, whereas slender mongooses
prefer covered areas with rocky outcrops (Cronk &
Pillay, 2020). Additionally, whilst both yellow and
slender mongoose species are diurnal, they have
several peaks in their activity throughout the day.
Our study was conducted during autumn–early
winter, and temporal overlap between yellow and
slender mongooses may be greater during colder
months when day length is shorter and resources
are scarcer than in warmer months (Cronk & Pillay,
2020). This finding suggests that yellow and
slender mongoose species use a combination of
spatial and fine-scale temporal partitioning as
coexistence mechanisms (Donadio & Buskirk,
2006).
Rusty-spotted genets, servals and marsh
mongooses all showed crepuscular peaks of activ-
ity. Rusty-spotted genets and servals favour
mammalian (mainly rodent) prey, which can
comprise more than 80% of the diet of both
species (Ramesh & Downs, 2015; Zemouche,
2018). Considering their similar activity patterns, it
seems probable that there may be competition
between rusty-spotted genets and servals for
similar food resources in the reserve. The low
detection rates of servals and the predominant use
of grassland habitat by this species compared to
rusty-spotted genets detected mainly in riparian
habitat suggest that encounter rates between
these two species may be low. This result may
indicate spatial partitioning and aid in predation
risk avoidance (Ramesh & Downs, 2015) or is
simply due to generally low serval densities in
nature reserves (Taylor, 2020). Insects (28%) and
crabs (26%) form substantial components of
Webster
et al.
: Mesocarnivore activity patterns 187
Tab le 4. Model-averaged parameter estimates (average model coefficients), adjusted standard error (S.E.) and
associated
z
-values for variables for the top-ranked generalized linear models testing the effects of species, tempera-
ture (Temp), moon phase or vegetation characteristics on the variation of seven mesocarnivore species’ activity time.
For each factor, the category used as the intercept is indicated in brackets.Parameter estimates not overlapping zero
(showing statically significant effects) are highlighted in bold.
Factor Estimate ± S.E.
z
-value Pr (>|
t
|)
(Intercept) 12.74 ± 1.46 8.704 <0.001
Species (Genet) Jackal –3.52 ± 1.17 3.00 <0.01
Otter –5.00 ± 1.87 2.66 <0.01
Serval 2.03 ± 2.09 0.97 0.33
Slender mongoose –3.94 ± 1.04 3.78 <0.001
Marsh mongoose 0.94 ± 1.12 0.84 0.40
Yellow mongoose –5.72 ± 1.45 3.92 <0.001
Temperature Temp 0.20 ± 0.03 6.60 <0.001
Moon phase (First quarter) Full –0.84 ± 1.16 0.73 0.46
Last quarter 1.34 ±1.40 0.96 0.33
New moon –0.24 ± 0.87 0.28 0.77
Waning crescent 0.23 ± 0.86 0.27 0.78
Waning gibbous –0.75 ± 1.09 0.69 0.49
Waxing crescent –0.49 ± 1.04 0.48 0.63
Waxing gibbous 0.54 ± 1.07 0.51 0.61
Vegetation (Grassland:Open) Grassland: Rocky –0.94 ± 1.25 0.75 0.45
Grassland: Wet 0.03 ± 0.56 0.06 0.95
Grassland: Wooded –1.22 ± 1.57 0.77 0.43
Riparian: Dense –0.84 ± 1.10 0.77 0.44
Riparian: Open –087 ± 1.28 0.68 0.49
Riparian: Rocky –1.77 ± 1.66 1.06 0.28
Riparian: Sandy 1.01 ± 1.26 0.80 0.42
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
marsh mongoose diet (Somers & Purves, 1996).
Although marsh mongoose activity overlap with
rusty-spotted genets was high, coexistence
between these species is likely facilitated by
resource partitioning (
e.g.
food) and different
foraging strategies (Mills
et al
., 2019).
Temperature was the other important factor
affecting the behaviour of mesocarnivores in the
Telperion Nature Reserve. Behavioural strategies
are affected by various intrinsic and external
factors, including ambient temperature (Caraco
et al
., 1990). Given that midday ambient tempera-
tures for this region can peak above 40°C during
summer, the risk of thermoregulatory stress may
inhibit the activity of most mesocarnivores around
midday (Monterroso, Alves & Ferreras, 2014). Of
the seven species compared, only slender and
yellow mongooses were active during the hottest
parts of the day. This behaviour suggests that
these species may take advantage of food
resources not utilized by other species because of
a narrow thermal tolerance range, although further
studies would be required to investigate diurnal
activity for all seasons for these two species
Vegetation characteristics predicted variation in
times of activity to a lesser extent when compared
to temperature. Vegetation characteristics within
the landscape may affect mesocarnivore activity
by influencing the distribution and availability of
prey species (Schuette
et al
., 2013). Rodents, for
example, may be associated with denser vegeta-
tion and taller grass (Thompson & Gese, 2007),
which provides suitable habitat for hiding or resting
(Krofel, 2008; Pretorius, 2019). The selection of
different microhabitats within a landscape causes
differences in species distributions, allowing for
the coexistence of species with seemingly similar
activities and diets (Noor
et al
., 2017). Animals
modify their spatial territory when faced with
increased interspecific competition, resulting in
successful coexistence (Yang
et al
., 2018; Zhao
et al
., 2020). Habitat heterogeneity, particularly in
small nature reserves, may be crucial to support-
ing the coexistence of a diverse set of meso-
carnivores (Moreira-Arce
et al
., 2016; Carricondo-
Sanchez
et al
., 2019), like those that inhabit the
Telperion Nature Reserve.
Moon phase somewhat predicted variation in
times of activity of the mesocarnivores, with no
clear pattern of effect for our analysis on times
of activity between the modelled species. This
result may be influenced by the fact that three of
the modelled species were exclusively diurnal
188 African Journal of Wildlife Research Vol. 51, 2021
Tab le 5. Activity overlap coefficients (Δ1, in bold) and 95% confidence intervals (in brackets) for each species pair among seven mesocarnivores at the Telperion Nature
Reserve, South Africa, from April to July 2015.
Genet Jackal Otter Serval Slender mongoose Marsh mongoose Yellow mongoose
Genet 1
Jackal 0.605 (0.54–0.66) 1
Otter 0.194 (0.07–0.33) 0.442 (0.29–0.58) 1
Serval 0.798 (0.61–0.94) 0.516 (0.34–0.67) 0.197 (0.03–0.36) 1
Slender mongoose 0.051 (0.01–0.09) 0.383 (0.32–0.44) 0.325 (0.18–0.46) 0.071 (–0.02–0.17) 1
Marsh mongoose 0.804 (0.70–0.88) 0.711 (0.61–0.80) 0.257 (0.11–0.40) 0.711 (0.52–0.88) 0.214 (0.12–0.32) 1
Yellow mongoose 0.055 (0.01–0.10) 0.382 (0.31–0.44) 0.366 (0.22–0.50) 0.071 (–0.01–0.17) 0.905 (0.80–0.98) 0.216 (0.12–0.31) 1
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
(otter, slender mongoose and yellow mongoose),
three species exhibited crepuscular activity
(black-backed jackal, marsh mongoose and
rusty-spotted genet), whilst only servals exhibited
a nocturnal peak in activity two hours after dusk.
Visually orientated hunters experience increased
detectability under moonlit conditions, which
increases predation risk for nocturnal mammals
and may result in suppressed activity (Prugh
& Golden, 2014). In large African carnivores,
African wild dogs (
Lycaon pictus
) and cheetahs
(
Acinonyx jubatus
) show heightened nocturnal
activity and better hunting opportunities during
moonlit nights despite the increased mortality
risk from lions (
Panthera leo
) and spotted
hyaenas (
Crocuta crocuta
) (Cozzi
et al
., 2012). To
our knowledge, the effects of moon phase on
nocturnal South African mesocarnivores have not
yet been examined and would warrant further
investigation.
Mesocarnivores are important components
of healthy ecosystems (Roemer
et al
., 2009).
However, major overlaps in diet and activity times,
in addition to limited space within small nature
reserves, may heighten competition between
members of this guild (Massara
et al
., 2016),
particularly in South Africa with its rich mesocarni-
vore diversity (Skinner & Chimimba, 2005). Our
study detected several mesocarnivore species
that showed comparable patterns of temporal
activity and probably shared similar diets, indicat-
ing possible competition for resources (food and
space) between these species. Our results also
provide a useful first description of mesocarnivore
diversity and activity patterns for this small reserve
for future studies to build on.
Variable detection rates for various species in
the riparian and grassland landscapes suggest
that most mesocarnivore species in the Telperion
Nature Reserve employ spatial partitioning rather
than temporal or trophic niche partitioning as a
coexistence mechanism (Donadio & Buskirk,
2006). With its mixture of riparian and grassland
vegetation types, the heterogeneity of the nature
reserve, and the related diversity of available prey
species (Afonso
et al
., 2021), was likely the main
contributor to the rate of species detected, despite
its relatively small size. Landscape heterogeneity
has been shown to facilitate niche partitioning and
enable coexistence (Fisher
et al
., 2013) and is
particularly important in increasingly human-
dominated landscapes (Manlick
et al
., 2020).
Therefore, maintaining heterogeneity in enclosed
nature reserves in South Africa is one important
consideration to promote mesocarnivore bio-
diversity, as demonstrated in other parts of the
world (Moreira-Arce
et al
., 2016; Curveira-Santos
et al
., 2017).
CONCLUSION
Interspecific competition may intensify when avail-
able habitat is limited, such as in small nature
reserves like Telperion Nature Reserve. Our
results show that Telperion hosts a variety of
mesocarnivore species that follow diverse activity
patterns across grassland and riparian land-
scapes. Species ID and temperature were shown
to be the best predictors of activity patterns, likely
related to the different behavioural strategies and
different tolerances to thermal stress. Several
species-pairs showed high degrees of activity
overlap, which included slender and yellow
mongooses, followed by marsh mongooses and
rusty-spotted genets, and servals and rusty-
spotted genets. These species pairs likely utilized
different spatial, temporal and resource (in the
case of marsh mongoose and rusty-spotted
genet) partitioning strategies to facilitate coexis-
tence in this relatively small nature reserve. This
shows that preserving habitat heterogeneity in
small nature reserves is likely essential to the
continued persistence of the diverse mesocarni-
vore guild observed in this study.
ACKNOWLEDGEMENTS
E. Oppenheimer and Son, and its manager for
research and conservation, Duncan MacFadyen,
are acknowledged for granting permission to
conduct the study. Samantha and Brendon
Schimmel are thanked for their assistance with
data collection. We thank the Telperion ecologist,
Elsabe Bosch, students and staff for providing
background information on the study site. The
University of South Africa is thanked for facilitating
accommodation during fieldwork. Lourens
Swanepoel is acknowledged for initial discussions
on sampling design. Emmanual do Lihn San is
thanked for comments on an earlier draft of the
manuscript and for helping with some of the
species identifications.
§ORCID iDs
A.B. Webster: orcid.org/0000-0002-7136-4421
M.E. Pretorius: orcid.org/0000-0002-4821-1013
M.J. Somers: orcid.org/0000-0002-5836-8823
Webster
et al.
: Mesocarnivore activity patterns 189
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
190 African Journal of Wildlife Research Vol. 51, 2021
REFERENCES
Afonso, B.C., Swanepoel, L.H., Rosa, B.P., Marques,
T.A., Rosalino, L.M., Santos-Reis, M. & Curveira-
Santos, G. (2021). Patterns and drivers of rodent
abundance across a South African multi-use land-
scape.
Animals
, 11, 2618.
https://doi.org/10.3390/ani11092618
Barto½, K. (2019). MuMIn:multi-model inference. R pack-
age version 1.43.6.
https://CRAN.R-project.org/package=MuMIn
(accessed 20 October 2020).
Boydston, E.E., Kapheim, K.M., Watts, H.E., Szykman,
M. & Holekamp, K.E. (2003). Altered behaviour in
spotted hyenas associated with increased human
activity.
Animal Conservation
, 6, 207–219.
https://doi.org/10.1017/S1367943003003263
Burnham, K.P. & Anderson, D.R. (2002).
Model selection
and multimodel inference a practical informa-
tion-theoretic approach
. New York, U.S.A.: Springer.
Burnham, K.P., Anderson, D.R. & Huyvaert, K.P. (2011).
AIC model selection and multimodel inference in
behavioral ecology: some background, observations,
and comparisons.
Behavioral Ecology and Socio-
biology
, 65, 23–35.
https://doi.org/10.1007/s00265-010-1029-6
Caraco, T., Blanckenhorn, W.U., Gregory, G.M.,
Newman, J.A., Recer, G.M. & Zwicker, S.M. (1990).
Risk-sensitivity: ambient temperature affects forag-
ing choice.
Animal Behaviour
, 39, 338–345
https://doi.org/10.1016/S0003-3472(05)80879-6
Carricondo-Sanchez, D., Odden, M., Kulkarni, A. &
Vanak, A.T. (2019). Scale-dependent strategies for
coexistence of mesocarnivores in human-dominated
landscapes.
Biotropica
, 51, 781–791.
https://doi.org/10.1111/btp.12705
Chesson, P. (2000). Mechanisms of maintenance of
species diversity.
Annual Review of Ecology and Sys-
tematics
, 31, 343–366.
https://doi.org/10.1146/annurev.ecolsys.31.1.343
Clavel, J., Julliard, R. & Devictor, V. (2011). Worldwide
decline of specialist species: toward a global func-
tional homogenization?
Frontiers in Ecology and the
Environment
, 9, 222–228.
https://doi.org/10.1890/080216
Coetzee, C. (2012).
The effect of vegetation on the
behaviour and movements of Burchell’s zebra
, Equus
burchelli
(Gray 1824) in the Telperion Nature Reserve
,
Mpumalanga
,
South Africa
. (Unpublished M.Sc.
thesis). Pretoria, South Africa:University of Pretoria.
Cozzi, G., Broekhuis, F., McNutt, J.W., Turnbull, L.A.,
Macdonald, D.W. & Schmid, B. (2012). Fear of the
dark or dinner by moonlight? Reduced temporal
partitioning among Africa’s large carnivores.
Ecology
,
93, 2590–2599.
https://doi.org/10.1890/12-0017.1
Cronk, N.E. & Pillay, N. (2020). Spatiotemporal co-
occurrence and overlap of two sympatric mongoose
species in an urban environment.
Journal of Urban
Ecology
, 6, juaa013.
https://doi.org/10.1093/jue/juaa013
Curveira-Santos, G., Marques, T.A., Björklund, M. &
Santos-Reis, M. (2017). Mediterranean mesocarni-
vores in spatially structured managed landscapes:
community organisation in time and space.
Agricul-
ture,
Ecosystems & Environment
, 237, 280–289.
https://doi.org/10.1016/j.agee.2016.12.037
Do Linh San, E., Ferguson, A.W., Belant, J.L., Schipper,
J., Hoffmann, M., Gaubert, P., Angelici, F.M. &
Somers, M.J. (2013). Conservation status, distribu-
tion and species richness of small carnivores in
Africa.
Small Carnivore Conservation
, 48, 4–18.
Do Linh San, E., Sato, J.J., Belant, J.L.& Somers, M.J. (In
press). The world’s small carnivores: definitions, rich-
ness, distribution, conservation status, ecological
roles, and research efforts. In Do Linh San, E., Sato,
J.J., Belant, J.L. & Somers, M.J. (Eds),
Small carni-
vores: evolution, ecology, behaviour and conserva-
tion
. London, U.K.: Wiley.
Donadio, E. & Buskirk, S.W. (2006). Diet, morphology,
and interspecific killing in Carnivora.
The American
Naturalist
, 167, 524–536.
https://doi.org/10.1086/501033
Fisher, J.T., Anholt, B., Bradbury, S., Wheatley, M. &
Volpe, J.P. (2013). Spatial segregation of sympatric
marten and fishers: the influence of landscapes and
species-scapes.
Ecography
, 36, 240–248.
https://doi.org/10.1111/j.1600-0587.2012.07556.x
Fox, J.& Weisberg, S. (2011
)
.
An R companion to applied
regression
.
Second edition
. Thousand Oaks CA:Sage.
http://socserv.socsci.mcmaster.ca/jfox/Books/Com-
panion
Accessed 22 October 2020.
Gehrt, S.D., Wilson, E.C., Brown, J.L. & Anchor, C.
(2013). Population ecology of free-roaming cats and
interference competition by coyotes in urban parks.
PLOS ONE
, 8, e75718.
https://doi.org/10.1371/journal.pone.0075718
González-Maya, J.F., Schipper, G.J.I. & Benítez, A.
(2009). Activity patterns and community ecology of
small carnivores in the Talamanca region, Costa
Rica.
Small Carnivore Conservation
, 41, 9–14.
Havmøller, L.W., Loftus, J.C., Havmøller, R.W., Alavi,
S.E., Caillaud, D., Grote, M.N., Hirsch, B.T.,
Tórrez-Herrera, L.L., Kays, R.& Crofoot, M.C. (2021).
Arboreal monkeys facilitate foraging of terrestrial
frugivores.
Biotropica
, 53, 1685–1697.
Helm, C.V. (2006).
Ecological separation of the black and
blue wildebeest on Ezemvelo Nature Reserve in the
highveld grasslands of South Africa
. (Unpublished
M.Sc. thesis). Pretoria, South Africa: University of
Pretoria.
Jenks, K. E., Chanteap, P., Kanda, D., Peter, C., Cutter,
P., Redford, T. & Leimgruber, P. (2011).Using relative
abundance indices from camera-trapping to test wild-
life conservation hypotheses—an example from
Khao Yai National Park, Thailand.
Tropical Conser va-
tion Science
, 4, 113–131.
https://doi.org/10.1177/194008291100400203
Karanth, K. U. & Nichols, J. D. (1998). Estimation of tiger
densities in India using photographic captures and
recaptures.
Ecology
, 79, 2852–2862.
Karanth, K.U. & Nichols, J.D. (Eds) 2000 Monittoring
tigers and their prey: a mannual for researchers,
managers,and conservationists in tropical Asia.
Banglore: Centre for wildlife studies.
Kaunda, S.K.K. (2001). Spatial utilization by black-
backed jackals in southeastern Botswana.
African
Zoology
, 36(2), 143–152.
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
Webster
et al.
: Mesocarnivore activity patterns 191
Kays, R., Hody, A., Jachowski, D.S. & Parsons, A.W.
(2021). Empirical evaluation of the spatial scale and
detection process of camera trap surveys.
Movement
Ecology
, 9(41), 41
https://doi.org/10.1186/s40462-021-00277-3
Kery, M. & Royle, J.A.(2015).
Applied hierarchical model-
ing in ecology: Volume 1: Prelude and static models
.
Elsevier Science.
Kingdon, J. & Hoffmann, M. (eds). 2013.
Mammals of
Africa. Volume V: Carnivores, pangolins, equids and
rhinoceroses
. London, U.K.: Bloomsbury Publishing.
ISBN-978-1-4081-2255-6 (print); ISBN-978-1-4081-
8994-8 (epdf).
Krofel, M. (2008). Opportunistic hunting behaviour of
black-backed jackals in Namibia.
African Journal of
Ecology
, 46, 220.
https://doi.org/10.1111/j.1365-2028.2007.00809.x
Loveridge, A. & Macdonald, D. (2003). Niche separation
in sympatric jackals (
Canis mesomelas
and
Canis
adustus
).
Journal of Zoology
, 259, 143–153.
https://doi.org/10.1017/S0952836902003114
Manlick, P. J., Windels, S. K., Woodford, J.E. & Pauli, J.N.
(2020). Can landscape heterogeneity promote carni-
vore coexistence in human-dominated landscapes?.
Landscape Ecology
, 35, 2013–2027.
https://doi.org/10.1007/s10980-020-01077-7
Massara, R.L., Paschoal, A.M.O., Bailey, L.L., Doherty,
P.F. & Chiarello, A.G. (2016). Ecological interactions
between ocelots and sympatric mesocarnivores in
protected areas of the Atlantic Forest, southeastern
Brazil.
Journal of Mammalogy
, 97, 1634–1644.
https://doi.org/10.1093/jmammal/gyw129
Mills, D.R., Do Linh San, E., Robinson, H., Isoke, S.,
Slotow, R. & Hunter, L. (2019). Competition and
specialization in an African forest carnivore commu-
nity.
Ecology and Evolution
, 9, 10092-10108.
https://doi.org/ 10.1002/ece3.5391
Monterroso, P., Alves, P.C.& Ferreras, P. (2014).Plastic-
ity in circadian activity patterns of mesocarnivores in
southwestern Europe: implications for species coex-
istence.
Behavioral Ecology and Sociobiology
, 68,
1403–1417.
https://doi.org/10.1007/s00265-014-1748-1
Monterroso, P., Diaz-Ruiz, F., Lukacs, P.M., Alves, P.C. &
Ferreras, P. (2020). Ecological traits and the spatial
structure of competitive coexistence among carni-
vores.
Ecology
, 101, e03059.
https://doi.org/10.1002/ecy.3059
Moreira-Arce, D., Vergara, P.M., Boutin, S., Carrasco, G.,
Briones, R., Soto, G.E. & Jimenez, J.E. (2016).
Mesocarnivores respond to fine-grain habitat struc-
ture in a mosaic landscape comprised by commercial
forest plantations in southern Chile.
Forest Ecology
and Management
, 369, 135–143.
https://doi.org/10.1016/j.foreco.2016.03.024
Mucina, L. & Rutherford, M.C. (2006).
The vegetation of
South Africa
,
Lesotho and Swaziland
. Pretoria, South
Africa: South African National Biodiversity Institute.
Noor, A., Mir, Z.R., Veeraswami, G.G.& Habib, B.(2017).
Activity patterns and spatial co-occurrence of
sympatric mammals in the moist temperate forest of
the Kashmir Himalaya, India.
Journal of Vertebrate
Biology
, 66, 231–241.
https://doi.org/10.25225/fozo.v66.i4.a4.2017
Palomares, F. & Caro, T.M. (1999). Interspecific killing
among mammalian carnivores.
The American Natu-
ralist
, 153, 492–508.
https://doi.org/10.1086/303189
Pretorius, M.E. (2019).
Mesocarnivores in protected
areas: ecological and anthropogenic determinants of
habitat use in northern KwaZulu
-
Natal
,
South Africa
.
(Unpublished M.Sc. thesis). Cape Town, South
Africa: University of Cape Town.
Prugh, L.R. & Golden, C.D. (2014). Does moonlight
increase predation risk? Meta-analysis reveals diver-
gent responses of nocturnal mammals to lunar
cycles.
Journal of Animal Ecology
, 83, 504–514.
https://doi.org/10.1111/1365-2656.12148
Ramesh T., Downs, C.T., Power, R.J., Laurence, S.,
Matthews, W. & Child, M.F. (2016). A conservation
assessment of
Leptailurus serval
. In F. Child, L.
Roxburgh, E. Do Linh San, D. Raimondo, H.T.
Davies-Mostert (Eds),
The Red List of mammals of
South Africa
,
Swaziland and Lesotho
. South Africa:
South African National Biodiversity Institute and
Endangered Wildlife Trust.
Ramesh, T. & Downs, C.T. (2015). Diet of serval (
Leptail-
urus serval
) on farmlands in the Drakensberg Mid-
lands, South Africa.
Mammalia
, 79, 399–407.
https://doi.org/10.1515/mammalia-2014-0053
Rich, L., Miller, D., Robinson, H., McNutt, J. & Kelly, M.
(2017). Carnivore distributions in Botswana are
shaped by resource availability and intraguild
species.
Journal of Zoology
, 303, 90–98.
https://doi.org/10.1111/jzo.12470
Ridout, M.S. & Linkie, M. (2009). Estimating overlap of
daily activity patterns from camera trap data.
Journal
of Agricultural
,
Biological
,
and Environmental Statis-
tics
, 14, 322–337.
https://doi.org/10.1198/jabes.2009.08038
Ritchie, E.G. & Johnson, C.N. (2009). Predator interac-
tions, mesopredator release and biodiversity conser-
vation.
Ecology Letters
, 12, 982–998.
https://doi.org/10.1111/j.1461-0248.2009.01347.x
Roemer, G.W., Gompper, M.E. & Van Valkenburgh, B.
(2009). The ecological role of the mammalian
mesocarnivore.
BioScience
, 59, 165–173.
https://doi.org/10.1525/bio.2009.59.2.9
Saghir, J. & Santoro, J. (2018).
Urbanization in sub-
Saharan Africa
. Washington, DC, U.S.A.: Centre for
Strategic and International Studies Report.
https://csis-prod.s3.amazonaws.com/s3fs-public/pu
blication/180411_Saghir_UrbanizationAfrica_Web.
pdf?o02HMOfqh99KtXG6ObTacIKKmRvk0Owd
Santos, M.J., Matos, H.M., Palomares, F. & Santos-Reis,
M. (2011). Factors affecting mammalian carnivore
use of riparian ecosystems in Mediterranean cli-
mates.
Journal of Mammalogy
, 92, 1060–1069.
https://doi.org/10.1644/10-MAMM-A-009.1
Schuette, P., Wagner, A.P., Wagner, M.E. & Creel, S.
(2013). Occupancy patterns and niche partitioning
within a diverse carnivore community exposed to
anthropogenic pressures.
Biological Conservation
,
158, 301–312.
https://doi.org/10.1016/j.biocon.2012.08.008
Sheather, S.J. (2009).
A modern approach to regression
with R
. New York.: Springer.
https://doi.org/10.1007/978-0-387-09608-7
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
Skinner, J.D. & Chimimba, C.T. (2005).
The mammals of
the southern African sub-region
. Cambridge, U.K.:
Cambridge University Press.
https://doi.org/10.1017/CBO9781107340992
Skowno, A.L., Poole, C.J., Raimondo, D.C., Sink, K.J.,
Van Deventer, H., Van Niekerk, L., Harris, L.R.,
Smith-Adao, L.B., Tolley, K.A., Zengeya, T.A., Foden,
W.B., Midgley, G.F. & Driver, A. (2019).
The National
Biodiversity Assessment 2018: the status of South
Africa’s ecosystems and biodiversity
.
Synthesis
Report
. Pretoria, South Africa.: South African
National Biodiversity Institute, an entity of the Depart-
ment of Environment, Forestry and Fisheries.
http://hdl.handle.net/20.500.12143/6362
Sollmann, R. (2018).A gentle introduction to camera-trap
data analysis.
African Journal of Ecology
, 56, 740–749.
https://0-doi-org.innopac.wits.ac.za/10.1111/aje.125
57
Sollmann, R., Mohamed, A., Samejima, H. & Wilting, A.
(2013). Risky business or simple solution—Relative
abundance indices from camera-trapping.
Biological
Conservation
, 159, 405–412.
https://doi.org/10.1016/j.biocon.2012.12.025
Somers, M. & Purves, M. (1996). Trophic overlap
between three syntopic semi-aquatic carnivores:
Cape clawless otter, spotted-necked otter and water
mongoose.
African Journal of Ecology
, 34, 158–166.
https://doi.org/10.1111/j.1365-2028.1996.tb00609.x
Tambling, C., Avenant, N., Drouilly, M. & Melville, H.
(2018). The role of mesopredators in ecosystems:
potential effects of managing their populations on
ecosystem processes and biodiversity. In G.I.H.
Kerley, S.L. Wilson & D.Balfour (Eds),
Livestock pre-
dation and its management in South Africa: a scien-
tific assessment
(pp.205–227). Port Elizabeth, South
Africa: Centre for African Conservation Ecology,
Nelson Mandela University.
Taylor, J. (2020).
From big spots to little spots: influence
of camera trap deployment on spatial capture–recap-
ture estimates of servals
(Leptailurus serval)
in Ithala
Game Reserve
. (M.Sc. thesis), Cape Town, South
Africa: University of Cape Town.
Thompson, C.M. & Gese, E.M. (2007). Food webs and
intraguild predation: community interactions of a
native mesocarnivore.
Ecology
, 88, 334–346.
https://doi.org/10.1890/0012-9658(2007)88[334:FW
AIPC]2.0.CO;2
Violle, C., Enquist, B.J., McGill, B.J., Jiang, L., Albert,
C.H., Hulshof, C., Jung, V. & Messier, J. (2012). The
return of the variance: intraspecific variability in com-
munity ecology.
Trends in Ecology and Evolution
, 27,
244–252.
https://doi.org/10.1016/j.tree.2011.11.014
Wickham, H. (2009).
ggplot2: elegant graphics for data
analysis
. New York, U.S.A.: Springer-Verlag.
http://ggplot2.org (accessed 22 October 2020).
https://doi.org/10.1007/978-0-387-98141-3
Wilson, D.E. & Reeder, D.M. (2005).
Mammal species of
the world: a taxonomic and geographic reference
.
Bultimore, U.S.A.: Johns Hopkins University Press.
Yang, H., Zhao, X., Han, B., Wang, T., Mou, P., Ge, J. &
Feng, L. (2018). Spatiotemporal patterns of Amur
leopards in northeast China: influence of tigers, prey,
and humans.
Mammalian Biology
, 92, 120–128.
https://doi.org/10.1016/j.mambio.2018.03.009
Zemouche, J. (2018).
Trophic ecology of rusty-spotted
genet
Genetta maculata
and slender mongoose
Herpestes sanguineus
in Telperion Nature Reserve
,
with a focus on dietary segregation as a possible
mechanism of coexistence
. (M.Sc. thesis). Johan-
nesburg, South Africa: University of the Witwaters-
rand.
Zhao, G., Yang, H., Xie, B., Gong, Y., Ge, J. & Feng, L.
(2020). Spatio-temporal coexistence of sympatric
mesocarnivores with a single apex carnivore in a
fine-scale landscape.
Global Ecology and Conserva-
tion
, 21, e00897.
https://doi.org/10.1016/j.gecco.2019.e00897
Responsible Editor: B. Allen
192 African Journal of Wildlife Research Vol. 51, 2021
Downloaded From: https://bioone.org/journals/African-Journal-of-Wildlife-Research on 13 Dec 2021
Terms of Use: https://bioone.org/terms-of-useAccess provided by Southern African Wildlife Management Association
