Camera trap and questionnaire dataset on
ecosystem services provided by small carnivores
in agro-ecosystems in South Africa
Samual T. Williams
, Naudene Maree
, Peter Taylor
Steven R. Belmain
, Mark Keith
, Lourens H. Swanepoel
Department of Zoology, School of Mathematical & Natural Sciences, University of Venda, Private bag X5050,
Thohoyandou 0950, South Africa
Department of Anthropology, Durham University, Durham DH1 3LE, United Kingdom
South African Research Chair on Biodiversity Value & Change, University of Venda, Private bag X5050,
Thohoyandou 0950, South Africa
School of Life Sciences, University of KwaZulu-Natal, Private Bag X54001, Durban 4000, South Africa
Natural Resources Institute, University of Greenwich, Chatham Maritime, Kent, United Kingdom
Eugène Marais Chair of Wildlife Management, Mammal Research Institute, University of Pretoria, 0002,
Received 19 December 2017
Received in revised form
3 March 2018
Accepted 12 March 2018
Available online 22 March 2018
This dataset includes data derived from camera trap surveys and
questionnaire surveys relating to small carnivores in agro-ecosys-
tems in the Vhembe Biosphere Reserve, South Africa. The data
were collected as part of the study “Predation by small mammalian
carnivores in rural agro-ecosystems: An undervalued ecosystem
service?”(Williams et al., 2017a) . Camera trap locations were
stratiﬁed by land use: settlement, crops, and grazing areas. The
camera trap data provide an insight into the ecology of the nine
species of small carnivores that were recorded: striped polecat
(Ictonyx striatus), honey badger (Mellivora capensis), large-spotted
genet (Genetta maculata), African civet (Civettictis civetta), slender
mongoose (Galerella sanguinea), Meller's mongoose (Rhynchogale
melleri), Selous' mongoose (Paracynictis selousi), white tailed
mongoose (Ichneumia albicauda), and dwarf mongoose (Helogale
parvula). We also recorded domesticated animals such as domestic
cats (Felis catus), domestic dogs (Canis lupus familiaris), and cattle
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/dib
Data in Brief
2352-3409/&2018 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license
DOI of original article: https://doi.org/10.1016/j.ecoser.2017.12.006
Corresponding author at: Department of Zoology, School of Mathematical & Natural Sciences, University of Venda, Private
bag X5050, Thohoyandou 0950, South Africa.
E-mail address: firstname.lastname@example.org (S.T. Williams).
Data in Brief 18 (2018) 753–759
(Bos taurus) on the camera traps. The questionnaire data are
comprised of responses of stakeholders to questions regarding the
impacts of these species on rural farming communities. In the
accompanying data repository hosted on Figshare (doi 10.6084/
m9.ﬁgshare.4750807, (Williams et al., 2017b) ) we provide raw
data, along with processed data and R code used to analyse these
data to determine the impact of land use and domestic animals on
the species richness and occupancy of small carnivores in rural
agro-ecosystems (Williams et al., 2017a) .
&2018 The Authors. Published by Elsevier Inc. This is an open
access article under the CC BY license
Subject area Biology
More speciﬁc subject area Conservation biology, ecology
Type of data Text ﬁle, shapeﬁle, R code
How data was acquired Camera traps, questionnaires
Data format Raw, processed
Experimental factors Camera trap data were stratiﬁed by land use
Experimental features Camera trap surveys, questionnaire surveys
Data source location Ka-Ndengeza (S23.310028, E30.409812) and Vyeboom (S23.151735,
E30.392782) villages, Limpopo Province, South Africa
Data accessibility Data are available from Figshare (doi:10.6084/m9.ﬁgshare.4750807;
Related research article This is a companion article to 
Value of the data
The raw camera trap data could be useful for studying the biodiversity and distribution of small
carnivores in agro-ecosystems.
The processed camera trap data may be useful for the study of small carnivore occupancy.
The questionnaire data may be of interest to researchers studying the opinions of people towards
These data could be compared with other data collected in protected areas, other geographic
locations or among other groups of stakeholders, and could contribute to spatial data sets (e.g.
Global Biodiversity Information Facility).
This could contribute towards scientiﬁc research or policy documents, for example the Interna-
tional Union for Conservation of Nature and Natural Resources (IUCN) Red List of Threatened
Species, The Red List of Mammals of South Africa, Swaziland and Lesotho, South Africa's National
Biodiversity Strategy and Action Plan, or local wildlife management plans.
We provide data collected using camera trap and questionnaire surveys conducted in agro-eco-
systems (Fig. 1). Data collection focussed on small carnivores, deﬁned as members of the order Car-
nivora with a body mass under 15 kg . A total of nine species of wild small carnivores, and two
species of domesticated carnivores, were detected (Table 1), along with domestic cattle (Fig. 2). The
S.T. Williams et al. / Data in Brief 18 (2018) 753–759754
Fig. 1. Location of the study sites and the camera traps, showing the settlement, crops and grazing land use types.
Summary of photographs of carnivore species collected during the camera trap study. The table is ordered according to family
level (all capitals).
Number of independent detections per 1000 camera trap
Common name Scientiﬁc name Settlement Crops Grazing Settlement Crops Grazing IUCN Red List 
Domestic dog Canis lupus familiaris 9324 1270 308 5160 201 37
Striped polecat Ictonyx striatus 0 0 5 0 8 0 Least concern
Honey badger Mellivora capensis 0 0 0 0 0 6 Least concern
Domestic cat Felis catus 324 0 10 720 0 6
Large-spotted genet Genetta maculata 0 643 217 22 173 228 Least concern
African civet Civettictis civetta 0 0 0 0 8 0 Least concern
Slender mongoose Galerella sanguinea 0 254 25 0 148 86 Least concern
Meller's mongoose Rhynchogale melleri 0 48 0 0 0 0 Least concern
Selous' mongoose Paracynictis selousi 0 71 0 0 33 0 Least concern
White tailed mongoose Ichneumia albicauda 0 151 0 27 8 19 Least concern
Dwarf mongoose Helogale parvula 0 32 0 4 4 31 Least concern
S.T. Williams et al. / Data in Brief 18 (2018) 753–759 755
data were collected in two villages in Limpopo province, South Africa, and camera trap locations were
stratiﬁed by land use (settlement, crops, and grazing areas).
2. Experimental design, materials, and methods
2.1. Study area
We conducted the study at two villages in the Vhembe Biosphere Reserve, South Africa: Ka-
Ndengeza (S23.310028°E30.409812°), and Vyeboom (S23.151735°E30.392782°)(Fig. 1). Both sites
receive an annual rainfall of 700–800 mm per year, with a hot wet season from October to March and
a cool dry season from May to August . Natural vegetation is classiﬁed as Granite Lowveld and
Fig. 2. Example photographs collected during the camera trap surveys. These images show a) striped polecat, b) honey badger,
c) large-spotted genet, d) African civet, e) slender mongoose, f) Meller's mongoose, g) Selous' mongoose, h) white tailed
mongoose, i) dwarf mongoose, j) domestic cat, k) domestic dog, and l) domestic cow.
S.T. Williams et al. / Data in Brief 18 (2018) 753–759756
Gravelotte rocky bushveld . Vegetation is characterised by tall shrubs with few trees to moderately
dense low woodland on the deep sandy uplands dominated by Combretum zeyheri and Combretum
apiculatum. Low lying areas are characterised by dense thicket to open Savanna with Senegalia
(Acacia) nigrescens, Dichrostachys cinerea, and Grewia bicolor dominating the woody layer, particularly
the Granite Lowveld .
Three major land-use types were identiﬁed in each of the villages. First, the settlement areas were
used for residential purposes (hereafter settlements) . The majority of households had large gar-
dens (50–80 m ×40–80 m) which were used to grow crops (maize (Zea mays), peanuts, beans
(Phaseolus vulgaris), ground nuts (Arachis hypogaea), avocados mangoes, bananas, litchis, and oran-
ges), and to overnight livestock (cattle, donkeys, sheep, goats, and poultry). The second land-use type
identiﬁed was cropping areas (hereafter crops). Residents of both villages practiced either rotational
cropping (maize, ground nuts, and beans) or intercropping (maize, beans, and pumpkins (Cucurbita
spp.)). Land preparation was usually by manual labour, and preparation typically began in October or
November, while planting commenced in early December. Harvesting of crops occurs in February
until late April (crop dependant). Farmers reported yields varying between 5–20 bags (each bag
weighing 50 kg) of maize and 3–10 bags of ground nuts (Swanepoel, unpublished data). Crop residues
were typically used for livestock fodder. The third land-use type was the grazing areas (hereafter
grazing), which comprised of short grass, shrubs and tall trees. In addition to communal grazing of
livestock, these grazing areas also served for areas where ﬁrewood were collected and informal
hunting took place. Due to poor land-management practices, however, the grazing areas were typi-
cally overgrazed, with woody plants (D. cinerea) replacing shrubs and grass, typically in low-lying
areas and drainage lines.
2.2. Camera trapping
We divided each study area into a settlement area, cropping area and grazing area, based on recent
satellite imagery , which was then overlaid with a regular spaced grid with a cell size of 300 ×
300 m (9 ha). The size choice of the grid cells was guided by the median home range size of small
carnivores expected to inhabit the study areas , to adhere to the independent assumptions of
occupancy models . We deployed one camera trap in each grid, which resulted in an average
spacing between camera traps of 193 m. Camera traps were set to record 24 h per day, with a 30 s
delay between detections, continuously for 10–12 days. We deployed camera traps at roads, drainage
lines, and well established animal paths. We placed cameras around 30 cm above the ground, and
cleared vegetation in front of camera traps to reduce the number of false triggers.
In the settlement grid cells we deployed 27–30 infra-red ﬂash cameras (Cuddeback Ambush 1194),
as these were less disruptive to the inhabitants of villages than cameras using a visible light ﬂash,
while in the crops and grazing areas we deployed 55–60 xenon ﬂash cameras (Cuddeback Ambush
1170). Camera traps were deployed between 2–26 June 2014 at Ka-Ndengeza and 17 June–27 July
2014 at Vyeboom. This resulted in a camera trapping effort of 810 trap days in Ka-Ndengeza and 738
trap days in Vyeboom. To classify land use we ﬁrst digitized the different land-use types using
satellite imagery from Google Maps , which we later ground-truthed. We classiﬁed crops as either
active ﬁelds, i.e. still showing agricultural activity, or as abandoned ﬁelds. For each camera trap we
calculated the percentage of crops, grazing and settlement that comprised the camera trapping grid
cell in which each camera trap was located. Camera trap images were catalogued using Camera Base
version 1.7 .
We assessed the opinions of community members towards small carnivores using a structured
questionnaire (based on the questionnaire used by Holmern and Røskaft ), completed by a total of
127 respondents (n¼58 in Ka-Ndengeza and n¼69 in Vyeboom). For each camera trap the
inhabitants of the nearest household were sampled, but when this was not possible another nearby
house was selected. If several households were equidistant to a camera trap, we sampled one of these
households at random. Photographs of small carnivore species were provided to ensure that the
S.T. Williams et al. / Data in Brief 18 (2018) 753–759 757
species were correctly identiﬁed. We asked interviewees whether they had seen each species of
carnivore, if they were good for the community, if they kill rodents, if they had impacted the
respondents negatively, and if they were aware if any small carnivore species that are killed by
people. The reasons for any positive and negative impacts of the species were also recorded. We also
asked whether interviewees consider poultry to be an important source of protein.
2.4. Data availability
Raw camera trap and questionnaire data are available in the online repository . We processed
these data to allow us to model the inﬂuence of land use on carnivore species richness, and model the
effect of the relative abundance index of domestic animals on carnivore occupancy (see  for details
of data processing and analysis). The processed data and R code are also available in .
2.5. Ethical approval
Ethical approval for the study was provided by the Ethics Committee of the University of Venda
(approval number SMNS/14/ZOO/03/2803). We also obtained consent to interview community
members of Ka-Ndengeza and Vyeboom from each community Chief in addition to community
members. We informed each respondent that anonymity would be maintained, and obtained written
consent from interviewees.
We are grateful to the Chiefs and residents of Ka-Ndengeza and Vyeboom for granting permission
to collect these data, and for their support and hospitality. We are indebted to the participants that
completed the questionnaires, and the residents who allowed us to place camera traps on their land.
Data collection was funded by the Sasol Agricultural Trust (South Africa), Univen Niche Fund (SMNS/
17/Zoo/01), International Foundation for Science (D/4984-2), and the European Union through its ACP
S & T Programme (StopRats; FED2013-330223; http://www.acp-hestr.eu/). Further funding was
provided by the German Federal Ministry of Education and Science (“BMBF”) (01LL1304A) through
the project “Limpopo Living Landscapes - Understanding the Dynamics of Ecological and Cultural
Landscapes, in the Face of Global Change, in the Northern Limpopo Region of South Africa”of the
SPACES (Science Partnerships for the Assessment of Complex Earth Systems) consortium (https://
www.fona.de/mediathek/pdf/SPACES_Broschuere_Englisch.pdf). STW acknowledges funding from the
University of Venda postdoctoral grant. PJT acknowledges the support of the University of Venda, the
National Research Foundation, (107099), and the Department of Science and Technology under the
South African Research Chairs Initiative (SARChI) (87311) on Biodiversity Value and Change within the
Vhembe Biosphere Reserve, hosted at University of Venda and co-hosted by the Centre for Invasion
Biology at University of Stellenbosch. We are also grateful to the anonymous reviewer that helped to
improve this manuscript.
Transparency document. Supplementary material
Transparency document associated with this article can be found in the online version at http://dx.
Appendix A. Supplementary material
Supplementary data associated with this article can be found in the online version at http://dx.doi.
org/10.1016/j.dib.2018.03.071. These data include Google maps of the most important areas described
in this article.
S.T. Williams et al. / Data in Brief 18 (2018) 753–759758
 S.T. Williams, N. Maree, P. Taylor, S.R. Belmain, M. Keith, L.H. Swanepoel, Predation by small mammalian carnivores in rural
agro-ecosystems: an undervalued ecosystem service? Ecosyst. Serv. (2017), http://dx.doi.org/10.1016/j.ecoser.2017.12.006.
 S. Williams, N. Maree, P. Taylor, S. Belmain, M. Keith, L. Swanepoel Small carnivore ecosystem services data. Figshare [online
database], 2017b. https://doi.org/10.6084/m9.ﬁgshare.4750807. Available from: 〈https://ﬁgshare.com/articles/Small_carni
vore_ecosystem_services_data/4750807〉(Accessed 04 December 2017).
 R.J. Hijmans, S.E. Cameron, J.L. Parra, P.G. Jones, A. Jarvis, Very high resolution interpolated climate surfaces for global land
areas, Int. J. Climatol. 25 (15) (2005) 1965–1978. http://dx.doi.org/10.1002/joc.1276.
 L. Mucina, M.C. Rutherford, The Vegetation of South Africa, Lesotho and Swaziland, South African National Biodiversity
Institute, Pretoria, 2006.
 J.J.O. Odhiambo, V.N. Magandini, An assessment of the use of mineral and organic fertilizers by smallholder farmers in
Vhembe district, Limpopo province, South Africa, Afr. J. Agric. Res. 53 (5) (2008) 357–362.
 Google. Satellite imagery. Sources: Landsat/Copernicus. Image date 31 December 2013. Google Maps. 2014. Available from:
 D.I. Mackenzie, J.A. Royle, Designing occupancy studies: general advice and allocating survey effort, J. Appl. Ecol. 42 (6)
(2005) 1105–1114 . http://dx.doi.org/10.1111/j.1365-2664.2005.01098.x.
 M. Tobler, Camera base version 1.7 [computer program], 2015. Available from 〈http://www.atrium-biodiversity.org/tools/
 T. Holmern, E. Røskaft, The poultry thief: subsistence farmers' perceptions of depredation outside the Serengeti National
Park, Tanzania, Afr. J. Ecol. 41 (2014) 81–87.
 IUCN. The IUCN Red List of Threatened Species. Version 2016-3. 2016 (accessed 04.12.16). Available from: 〈http://www.
S.T. Williams et al. / Data in Brief 18 (2018) 753–759 759