ArticlePDF Available

Snapshot Safari: A large-scale collaborative to monitor Africa’s remarkable biodiversity

Abstract and Figures

Nature is experiencing degradation and extinction rates never recorded before in the history of Earth.1,2 Consequently, continuous large-scale monitoring programmes are critical, not only to provide insights into population trends but also to aid in understanding factors associated with altering population dynamics at various temporal and spatial scales.3 Continuous monitoring is important not only for tracking rare or threatened species but also to detect the increase of potentially invasive species4, and the trends in the populations of common species, which in some regions are declining even more rapidly than are rare species2. The combination of citizen science and cutting-edge technologies has improved monitoring programmes.5 In this regard, camera traps have become a popular tool to engage with society while obtaining accurate scientific data.3 The importance of advances in technological monitoring has even been highlighted by the United Nations Environment Programme (UNEP) through the proposed ‘Digital Ecosystem framework’, a complex distributed network or interconnected socio-technological system.6 Monitoring species and ecosystems is critical to Africa – a highly biodiverse continent with numerous mammal species threatened by human activities such as poaching, overhunting, and climate and land-use change.7 Over half the terrestrial mammals in Africa have experienced range contractions of as much as 80% on average, including predator species such as lions (Panthera leo) and large ungulates.2 In sub-Saharan Africa, human impacts are projected to increase, and trigger an increased extinction risk.7 However, information on the conservation status of many species is limited, and many areas in Africa lack the baseline biodiversity data necessary to assess the outcomes of existing conservation programmes.5 Further, the lack of standardised methods to assess biodiversity patterns limits our ability to detect and respond to changes in mammal populations caused by environmental and anthropogenic factors.8 In attempting to address some of the above challenges, we have formed the Snapshot Safari Network (www. snapshotsafari.org) – a large-scale international camera trap network to study and monitor the diversity and ecological dynamics of southern and eastern African mammals. Snapshot Safari (hereafter Snapshot) is one of the largest camera trap networks in the world. It began in 2010 with a single camera trap grid in Serengeti National Park, Tanzania9, and the model and protocols have since been expanded in Tanzania as well as into five other countries: Botswana, Kenya, Mozambique, South Africa, and Zimbabwe (Figure 1). Participating locations represent a wide variety of habitats, wildlife communities, management types and protected area sizes. Here, we introduce this multidisciplinary initiative which combines citizen science and advanced machine learning techniques for the analysis of millions of animal photographs. We also introduce a set of high priority research questions emanating from expert consultation in 2019.
Content may be subject to copyright.
1Volume 117| Number 1/2
January/February 2021
Commentary
https://doi.org/10.17159/sajs.2021/8134
Snapshot Safari: A large-scale collaborative to
monitor Africa’s remarkable biodiversity
AUTHORS:
Lain E. Pardo1,6
Sara Bombaci2
Sarah E. Huebner3
Michael J. Somers4,5
Herve Fritz1,6
Colleen Downs7
Abby Guthmann3
Robyn S. Hetem8
Mark Keith4
Aliza le Roux9,10
Nokubonga Mgqatsa11
Craig Packer3
Meredith S. Palmer12
Daniel M. Parker9,13
Mike Peel7,14,15
Rob Slotow7
W. Maartin Strauss16
Lourens Swanepoel17,18
Craig Tambling19
Nairobi Tsie4
Mika Vermeulen1,6
Marco Willi20
David S. Jachowski7,21
Jan A. Venter1,6
AFFILIATIONS:
1School of Natural Resource
Management, Nelson Mandela
University, George, South Africa
2Department of Fish, Wildlife, and
Conservation Biology, Colorado
State University, Fort Collins,
Colorado, USA
3College of Biological Sciences,
University of Minnesota, St. Paul,
Minnesota, USA
4Eugène Marais Chair of Wildlife
Management, Mammal Research
Institute, Department of Zoology and
Entomology, University of Pretoria,
Pretoria, South Africa
5Centre for Invasion Biology,
University of Pretoria, Pretoria,
South Africa
6REHABS, International Research
Laboratory, French National Centre
for Scientific Research (CNRS)
/ University of Lyon 1 / Nelson
Mandela University, George,
South Africa
7School of Life Sciences, University
of KwaZulu-Natal, Durban,
South Africa
8School of Animal, Plant and
Environmental Sciences, University
of the Witwatersrand, Johannesburg,
South Africa
9Department of Zoology and
Entomology, University of the Free
State, Phuthaditjhaba, South Africa
10Afromontane Research Unit,
University of the Free State,
Phuthaditjhaba, South Africa
11Wildlife and Reserve Management
Research Group, Department of
Zoology and Entomology, Rhodes
University, Makhanda, South Africa
12Department of Ecology and
Evolutionary Biology, Princeton
University, Princeton, New
Jersey, USA
13School of Biology and
Environmental Sciences, University
of Mpumalanga, Mbombela,
South Africa
14Agricultural Research Council,
Animal Production Institute,
Rangeland Ecology, Pretoria,
South Africa
Nature is experiencing degradation and extinction rates never recorded before in the history of Ear th.1,2 Consequently,
continuous large-scale monitoring programmes are critical, not only to provide insights into population trends but
also to aid in understanding factors associated with altering population dynamics at various temporal and spatial
scales.3 Continuous monitoring is important not only for tracking rare or threatened species but also to detect the
increase of potentially invasive species4, and the trends in the populations of common species, which in some
regions are declining even more rapidly than are rare species2.
The combination of citizen science and cutting-edge technologies has improved monitoring programmes.5 In
this regard, camera traps have become a popular tool to engage with society while obtaining accurate scientific
data.3 The importance of advances in technological monitoring has even been highlighted by the United Nations
Environment Programme (UNEP) through the proposed ‘Digital Ecosystem framework’, a complex distributed
network or interconnected socio-technological system.6
Monitoring species and ecosystems is critical to Africa – a highly biodiverse continent with numerous mammal
species threatened by human activities such as poaching, overhunting, and climate and land-use change.7 Over half
the terrestrial mammals in Africa have experienced range contractions of as much as 80% on average, including
predator species such as lions (Panthera leo) and large ungulates.2 In sub-Saharan Africa, human impacts are
projected to increase, and trigger an increased extinction risk.7 However, information on the conservation status
of many species is limited, and many areas in Africa lack the baseline biodiversity data necessary to assess the
outcomes of existing conservation programmes.5 Further, the lack of standardised methods to assess biodiversity
patterns limits our ability to detect and respond to changes in mammal populations caused by environmental and
anthropogenic factors.8
In attempting to address some of the above challenges, we have formed the Snapshot Safari Network (www.
snapshotsafari.org) – a large-scale international camera trap network to study and monitor the diversity and
ecological dynamics of southern and eastern African mammals. Snapshot Safari (hereafter Snapshot) is one of the
largest camera trap networks in the world. It began in 2010 with a single camera trap grid in Serengeti National Park,
Tanzania9, and the model and protocols have since been expanded in Tanzania as well as into five other countries:
Botswana, Kenya, Mozambique, South Africa, and Zimbabwe (Figure 1). Participating locations represent a wide
variety of habitats, wildlife communities, management types and protected area sizes. Here, we introduce this
multidisciplinary initiative which combines citizen science and advanced machine learning techniques for the
analysis of millions of animal photographs. We also introduce a set of high priority research questions emanating
from expert consultation in 2019.
Figure 1: Current Snapshot Safari study locations in southern Africa.
2Volume 117| Number 1/2
January/February 2021
Commentary
https://doi.org/10.17159/sajs.2021/8134
Snapshot Safari: African monitoring programme
Page 2 of 4
15Applied Behavioural Ecology and
Ecosystems Research Unit, University
of South Africa, Johannesburg,
South Africa
16Department of Environmental
Sciences, University of South Africa,
Johannesburg, South Africa
17Department of Zoology, University
of Venda, Thohoyandou, South Africa
18African Institute for Conservation
Ecology, Makhado, South Africa
19Department of Zoology and
Entomology, University of Fort Hare,
Alice, South Africa
20School of Physics and Astronomy,
University of Minnesota, Minneapolis,
Minnesota, USA
21Department of Forestry and
Environmental Conservation,
Clemson University, Clemson, South
Carolina, USA
CORRESPONDENCE TO:
Sarah Huebner
EMAIL:
huebn090@umn.edu
HOW TO CITE:
Pardo LE, Bombaci S, Huebner SE,
Somers MJ, Fritz H, Guthmann A,
et al. Snapshot Safari: A large-scale
collaborative to monitor Africa’s
remarkable biodiversity. S Afr J Sci.
2021;117(1/2), Art. #8134. https://
doi.org/10.17159/sajs.2021/8134
ARTICLE INCLUDES:
Peer review
Supplementary material
KEYWORDS:
citizen science, camera trap,
conservation, machine learning,
mammals
PUBLISHED:
29 January 2021
Standardised methods
Our camera trap design consists of regular grids of 5 km2 per location (grid sites). Each grid has 8 to 245 camera
traps depending on the objectives and area to cover. Cameras are secured with steel cases and fixed at ~50 cm
height to detect medium- to large-bodied mammals. Each camera is programmed to take a series of three images
within 1–5 seconds of each other (a ‘capture event’) when passive infrared sensors are triggered by motion or heat
during the day and one image at night.9 Most grids have operated continuously since 2018and are intended to run
for a decade or longer. Data collected are forwarded to the University of Minnesota Lion Center for curation and
management of the citizen science component. Metadata on camera placements and habitat characteristics are
also collected in a standardised manner to facilitate cross-site comparisons.
Camera traps generate large volumes of photographs which makes the classification of species a time-consuming
task. To facilitate efficient image processing, we combine citizen science efforts with advanced machine learning
techniques.10 Snapshot partners with the citizen science platform www.zooniverse.org, on which volunteers
identify species and annotate other information such as the number of individuals or behaviour. They can also
interact directly with researchers on the talk boards or via social media. Each of our 24 current projects has its
own webpage within the Snapshot organisation. More than 150 000 volunteers worldwide have classified over
9 million photographs since the relaunch of Snapshot as a network in February 2018. These responses exhibit 97%
accuracy, confirming the reliability of citizen science in rapidly processing large volumes of data.11
Machine learning
The Snapshot network also incorporates machine learning algorithms prior to uploading data to Zooniverse
to decrease the number of citizen scientists required to view each capture event.10 Two image classifiers
(convolutional neural networks) are employed; one to identify empty images and one to predict the species, counts,
and behaviours in images containing wildlife. The algorithms’ predictions and confidence levels are uploaded to
Zooniverse with the image manifest and used to dynamically determine the level of agreement and number of
volunteers required to confirm the algorithms’ label (see Figure 2). The millions of images generated annually by the
Serengeti grid have been utilised as training data for many deep learning algorithms developed to identify African
mammal species automatically.10,12,13
Figure 2: Snapshot Safari data processing workflow. (1) Data from each location is checked for errors (bad
flash, date-time reset) and run through machine learning algorithms during the pre-processing phase.
(2) Data and machine learning predictions are uploaded to Zooniverse for classification. Every image is
classified by multiple volunteers, and their votes are aggregated to assign labels of species, count, and
behaviours. (3–5) Citizen scientists’ labelled annotations are used to iteratively train and refine the machine
learning algorithm. (6–7) Data are returned to researchers and reserve management for publications and
conservation assessments.
The first stage of classification occurs on the Zooniverse in a binary workflow, in which volunteers evaluate whether
animals can be seen in the images. If two people agree with the algorithm that an image is empty, it is removed from
the data set. If one person disagrees with the computer, the image remains in circulation to accumulate three more
human votes, at which point majority consensus is accepted. Images marked as containing wildlife are moved
to the second stage of classification, in which the species, count, and behaviours are annotated. At this stage,
dynamic rules within the Zooniverse infrastructure are used to retire captures based on the species, agreement of
the volunteers with the computer and one another, and the algorithm’s confidence in its own prediction.
As a result of quickly removing blank images and those of humans and common species, citizen scientists are
presented with more images of rare and cryptic species, improving the volunteer experience14 and providing
valuable data to refine the machine learning algorithm’s capabilities. This approach has improved efficiency by
43% thus far on the Serengeti project.10 Citizen science and machine learning stages are applied iteratively to
constantly improve performance and efficiency. The labelled images from all projects retrain the existing model to
© 2021. The Author(s). Published
under a Creative Commons
Attribution Licence.
3Volume 117| Number 1/2
January/February 2021
Commentary
https://doi.org/10.17159/sajs.2021/8134
Snapshot Safari: African monitoring programme
Page 3 of 4
improve the machine’s predictive capability in a variety of habitats, and
the refined model is run on new data sets as they are collected from the
field (Figure 2).
Snapshot also provides useful data for computer scientists who wish
to build artificial intelligence algorithms to identify wildlife species
automatically without human assistance. Data from six South African
sites and the Serengeti are publicly available for download on the
Microsoft-hosted site Labelled Image Library of Alexandria – Biology &
Conservation (LILA-BC) at www.lila.science.com. More data are added
to the labelled image repository as datasets are finalised. In all cases,
images of humans and rhinoceros are withheld from the repository
for privacy and poaching concerns, respectively. Metadata on habitat
and environmental characteristics are available to researchers outside
of the network upon request. This ensures that local researchers and
reserve management are aware of how the data are being used and can
contribute and collaborate.
Outputs and general results
Snapshot is documenting the presence, distribution, diversity, and
ecology of at least 85 medium- to large-sized African mammals. The
variety of Snapshot locations provides numerous opportunities to
answer questions in wildlife ecology and conservation, test ecological
hypotheses and analytical methods, and measure the impacts of
anthropogenic disturbances across multiple spatiotemporal scales. For
example, the data produced by the camera traps running continuously
in the Serengeti provided the basis for papers on spatiotemporal
partitioning15-17, behavioural interactions18-20, and advancing modelling
techniques used with the camera trap data11,19,21.
Scientists visit sites regularly to maintain and manage the camera
trap grids and also to improve community participation. For example,
in Ruaha National Park (Tanzania), researchers trained community
members to deploy and maintain the grid22, thus creating jobs and
infrastructure. In South Africa, we are building strong relationships
with local communities, such as the Khomani San in the Kalahari and
managers of private reserves, who will use the camera trap information
to improve their wildlife management practices.
Snapshot Safari – South Africa
Like other African countries, South Africa lacks adequate conservation
data for many species, and about 17% of mammal species are threatened
with extinction.8 However, South Africa has a long history of conservation
intervention and is, therefore, a testing ground for different hypotheses.
In South Africa, we have surveyed 31 locations (Figure 1), of which 21
are permanent grids for long-term monitoring purposes. These represent
a total of 1 408 cameras deployed for grids on permanent locations and
873 installed in the roaming locations. More than 43 000 volunteers have
annotated approximately 18 months of data for 19 of these sites and
classified more than 2 million photos to date.
Our grids and collaborative research allow us to combine different
questions and leverage the full potential of camera traps. For example,
in the Kruger National Park we have set up the cameras along with
studies of the phenology of vegetation cover (e.g. tree and herbaceous
composition, cover and structure), which has created an opportunity to
monitor the vegetation dynamic while accounting for mammal presence.
Further, by-catch data (not targeted species), such as birds or human
activity, will represent an interesting opportunity to investigate other
species in the future. The potential role of this by-catch data of camera
trap studies to conservation efforts or ecological studies has recently
been highlighted.23
Our grids have documented sightings of some threatened and
elusive species outside known distribution ranges as well as unusual
behavioural interactions. In 2019, for example, we detected a leopard
(Panthera pardus) near Karoo National Park. Similarly, we photographed
a brown hyaena (Parahyaena brunnea) in Camdeboo National Park.
None of these records was expected as no official or anecdotal records
have existed since colonial times and therefore these sightings represent
significant recent geographic range shifts for these species. In June
2019 in the Karoo National Park, one of the Snapshot cameras took
21 pictures of three to five meerkats (Suricata suricatta) and a yellow
mongoose (Cynictis penicillata) foraging together. An analysis of the
photographic sequence suggested they were sharing vigilance and
staying near each other. These observations constitute new evidence to
support the only previous observation of this cooperative behaviour in
the Andries Vosloo Kudu Nature Reserve in the Eastern Cape Province.24
We aim to be as collaborative as possible and share data to facilitate
other conservation and research projects. For example, we have shared
~5500 records with the MammalMAP initiative, which provides a
platform for citizen scientists and researchers to contribute biodiversity
information for South Africa (http://vmus.adu.org.za/). We expect
to make data available on other platforms such as the Foundational
Biodiversity Information Programme in South Africa (http://fbip.
co.za/) and the Global Biodiversity Information Facility to help evaluate
progress toward governmental agreements such as the UN Sustainable
Development Goals. Evaluating trends at the broad geographic scale of
the Snapshot network can inform IUCN Red List entries on species for
which we currently lack accurate estimates of extant population sizes.
Systematic long-term monitoring is crucial to understand the trends
in populations and species through time and to facilitate informed
management decisions. However, data collection via camera traps has
increased at a much faster rate than have our technical capabilities
to analyse such large data sets.25 Another limitation is the access to
research groups and computational centres with sufficient processing
infrastructure to train and run machine learning algorithms. In our
project, for example, the information is analysed by the Lion Center at
the University of Minnesota. All data are sent to that lab group for the
management of the online classification process and data curation. This
can create a bottleneck and slow the data pipeline. One potential solution
is to set up multiple hubs within a network, which we plan to enact by
copying existing infrastructure onto South Africa’s supercomputer at
the Centre for High Performance Computing. This continues the trend
of distributed management and provides additional opportunities for
training and professional development.
Future directions and conclusions
The Snapshot collaborators in South Africa meet annually to review
current projects, prioritise future research, and plan funding activities. At
the 2019 meeting, we defined five research themes:
• Theme 1: The role of anthropogenic landscapes in shaping
biodiversity, distributions, populations, and communities.
• Theme 2: Investigating ecological interactions and food webs
through space and time.
• Theme 3: Understanding and predicting the consequences of
climate change for mammal behaviour, distribution, adaptation,
and community composition.
• Theme 4: Assessment of conservation priorities and protected
area effectiveness.
• Theme 5: Merging camera trap data with other data to address
more specific questions and improve monitoring.
Snapshot’s integration of scientists, citizens, and technology can
provide meaningful assessments of the status of southern African
mammals at fine or broad scales. Our collaboration also holds the
potential to contribute to advancing statistical and technological capacity
in South Africa as well as to support environmental education for the
general public. With the current advances in data processing, we are
hopeful that Snapshot will provide timely recommendations and relevant
ecological information to support southern Africa’s national and
international commitments to conserving one of the most biodiversity
rich regions in the world. However, only political will and multi-sectoral
commitments will make it possible to leverage the full potential of
technology to produce practical effects.
4Volume 117| Number 1/2
January/February 2021
Commentary
https://doi.org/10.17159/sajs.2021/8134
Snapshot Safari: African monitoring programme
Page 4 of 4
Reserves and researchers who wish to join the Snapshot network
should contact Sarah Huebner at huebn090@umn.edu or Jan Venter at
Jan.Venter@mandela.ac.za
Acknowledgements
We thank all the Zooniverse volunteers who contribute classifications
to Snapshot Safari, and the moderators who donate their time and
expertise to our projects. We also thank sponsors of our work, including
the South African National Biodiversity Institute (SANBI), Foundational
Biodiversity Information Programme (FBIP), South African National
Research Foundation (NRF), Fynbos Trust, Nelson Mandela University,
Fairfields, Detroit Zoological Society, Zoo Miami, Cincinnati Zoo
Angel Fund, Seneca Park Zoo, and The Living Desert. We thank the
Minnesota Supercomputing Institute for providing resources for data
storage and processing, among others. Finally, we thank all the people
and institutions participating in Snapshot South Africa, including the
reserve managers and owners for providing access and supporting this
programme, National Parks Institutions of every country, students and
volunteer groups helping to maintain the grids, and the governmental
institutions overseeing these parks and reserves.
References
1. Díaz S, Settele J, Brondízio ES, Ngo HT, Agard J, Arneth A, et al. Pervasive
human-driven decline of life on Earth points to the need for transformative
change. Science. 2019;366(6471), eaax3100. https://doi.org/10.1126/
science.aax3100
2. Ceballos G, Ehrlich PR, Dirzo R. Biological annihilation via the ongoing sixth
mass extinction signaled by vertebrate population losses and declines. Proc
Natl Acad Sci USA. 2017;114(30):E6089–E6096. https://doi.org/10.1073/
pnas.1704949114
3. Ahumada JA, Hurtado J, Lizcano D. Monitoring the status and trends of
tropical forest terrestrial vertebrate communities from camera trap data: A tool
for conservation. PLoS ONE. 2013;8(9), e73707. https://doi.org/10.1371/
journal.pone.0073707
4. Donlan CJ, Tershy BR, Campbell K, Cruz F. Research for requiems: The need for
more collaborative action in invasive species management and conservation.
Conserv Biol. 2003;17(6):1850–1851. https://doi.org/10.1111/j.1523-
1739.2003.00012.x
5. Barnard P, Altwegg R, Ebrahim I, Underhill LG. Early warning systems for
biodiversity in southern Africa – How much can citizen science mitigate
imperfect data? Biol Conserv. 2017;208:183–188. https://doi.org/10.1016/j.
biocon.2016.09.011
6. Jensen D, Campbell J. Discussion paper: The case for a digital ecosystem
for the environment: Bringing together data, algorithms and insights for
sustainable development. Science Policy Business Forum. Nairobi: UN
Environment; 2019.
7. Tilman D, Clark M, Williams DR, Kimmel K, Polasky S, Packer C. Future threats
to biodiversity and pathways to their prevention. Nature. 2017;546(7656):73–
81. https://doi.org/10.1038/nature22900
8. Child M, Roxburgh L, Do Linh San E, Raimondo D, Davies-Moster H. The
2016 red list of mammals of South Africa, Swaziland and Lesotho. South
Africa: South African National Biodiversity Institute, Endangered Wildlife Trust;
2016.
9. Swanson A, Kosmala M, Lintott C, Simpson R, Smith A, Packer C. Snapshot
Serengeti, high-frequency annotated camera trap images of 40 mammalian
species in an African savanna. Sci Data. 2015;2(1):150026. https://doi.
org/10.1038/sdata.2015.26
10. Willi M, Pitman RT, Cardoso AW, Locke C, Swanson A, Boyer A, et al.
Identifying animal species in camera trap images using deep learning
and citizen science. Methods Ecol Evol. 2019;10(1):80–91. https://doi.
org/10.1111/2041-210X.13099
11. Swanson A, Kosmala M, Lintott C, Packer C. A generalized approach for
producing, quantifying, and validating citizen science data from wildlife
images. Conserv Biol. 2016;30(3):520–531. https://doi.org/https://doi.
org/10.1111/cobi.12695
12. Norouzzadeh MS, Nguyen A, Kosmala M, Swanson A, Palmer MS, Packer
C, et al. Automatically identifying, counting, and describing wild animals
in camera-trap images with deep learning. Proc Natl Acad Sci USA.
2018;115(25):E5716–E5725. https://doi.org/10.1073/pnas.1719367115
13. Gomez A, Salazar A, Vargas F. Towards automatic wild animal monitoring:
Identification of animal species in camera-trap images using very deep
convolutional neural networks. Ecol Inform. 2016;41:24–32. https://doi.
org/10.1016/j.ecoinf.2017.07.004
14. Spiers H, Swanson A, Fortson L, Simmons BD, Trouille L, Blickhan S, et
al. Everyone counts? Design considerations in online citizen science. J Sci
Commun. 2019;18(1):A04. https://doi.org/10.22323/2.18010204
15. Anderson TM, White S, Davis B, Erhardt R, Palmer M, Swanson A, et al. The
spatial distribution of African savannah herbivores: Species associations and
habitat occupancy in a landscape context. Philos Trans R Soc B Biol Sci.
2016;371(1703), 20150314. https://doi.org/10.1098/rstb.2015.0314
16. Hepler SA, Erhardt R, Anderson TM. Identifying drivers of spatial
variation in occupancy with limited replication camera trap data. Ecology.
2018;99(10):2152–2158. https://doi.org/10.1002/ecy.2396
17. Palmer MS, Fieberg J, Swanson A, Kosmala M, Packer C. A ‘dynamic’
landscape of fear: Prey responses to spatiotemporal variations in predation
risk across the lunar cycle. Ecol Lett. 2017;20(11):1364–1373. https://doi.
org/10.1111/ele.12832
18. Allen ML, Peterson B, Krofel M. No respect for apex carnivores: Distribution
and activity patterns of honey badgers in the Serengeti. Mamm Biol.
2018;89:90–94. https://doi.org/10.1016/j.mambio.2018.01.001
19. Palmer MS, Packer C. Giraffe bed and breakfast: Camera traps reveal
Tanzanian yellow-billed oxpeckers roosting on their large mammalian hosts.
Afr J Ecol. 2018;56(4):882–884. https://doi.org/10.1111/aje.12505
20. Swanson A, Arnold T, Kosmala M, Forester J, Packer C. In the absence of
a “landscape of fear”: How lions, hyenas, and cheetahs coexist. Ecol Evol.
2016;6(23):8534–8545. https://doi.org/10.1002/ece3.2569
21. Cusack JJ, Swanson A, Coulson T, Packer C, Carbone C, Dickman AJ, et al.
Applying a random encounter model to estimate lion density from camera
traps in Serengeti National Park, Tanzania. J Wildl Manage. 2015;79(6):1014–
1021. https://doi.org/10.1002/jwmg.902
22. Dickman AJ, Hazzah L, Carbone C, Durant SM. Carnivores, culture and
“contagious conflict”: Multiple factors influence perceived problems with
carnivores in Tanzania’s Ruaha landscape. Biol Conserv. 2014;178:19–27.
https://doi.org/10.1016/j.biocon.2014.07.011
23. Hofmeester TR, Young S, Juthberg S, Singh NJ, Widemo F, Andrén H, et al.
Using by-catch data from wildlife surveys to quantify climatic parameters and
the timing of phenology for plants and animals using camera traps. Remote
Sens Ecol Conserv. 2019;6(2):129-140. https://doi.org/10.1002/rse2.136
24. Do Linh S, Somers MJ. Mongooses on the move : An apparent case of
interspecific cooperative vigilance between carnivores. South Afr J Wildl Res.
2006;36(2):201–203.
25. Ahumada JA, Fegraus E, Birch T, Flores N, Kays R, O’Brien TG, et al. Wildlife
insights: A platform to maximize the potential of camera trap and other
passive sensor wildlife data for the planet. Environ Conserv. 2019;47(1):1–6.
https://doi.org/10.1017/s0376892919000298
... This research was part of an ongoing camera trap monitoring project called Snapshot Safari South Africa (hereafter Snapshot) which contains multiple surveys across several PAs (Pardo et al. 2021). We filtered the available data to those surveys conducted between 2018 and 2019 considering only a single survey of approximately 3 months per reserve to approximate the closure population assumption of occupancy models (MacKenzie et al. 2002 see below). ...
... Images were classified with the help of citizen scientists through the Zooniverse platform (Zooniverse.org, Pardo et al. 2021;Willi et al. 2019), except for three sites (Goegap Nature Reserve, Augrabies Falls National Park, and Khamab Kalahari Game Reserve). These images were tagged manually by a lab technician using Digikam (www. ...
Article
Black-backed jackals (Canis mesomelas) are opportunistic mesopredators occupying a variety of ecosystems across South Africa (SA). They can move between protected areas (PAs) and surrounding human-dominated landscapes where they are prone to conflict with wildlife and livestock farmers and subsequently face high persecution rates. However, it remains unclear to what extent the anthropogenic landscape matrix in which PAs are embedded affects black-backed jackal occupancy within PAs at large spatial scales. Therefore, in this study, we explore how different sources of environmental variation inside and outside PAs influence the patterns of jackal’s occupancy within PAs. We used 309 camera traps across 15 PAs in SA to respond to the following questions: (i) How does the landscape matrix surrounding PAs affect black-backed jackal occupancy compared to the landscape characteristics inside PAs? (ii) How does the presence of large carnivores affect black-backed jackal occupancy under these varying conditions of anthropogenic and PA landscape characteristics? When contrasting the effect of landscape variables and large predators at different spatial scales (fine-scale at site vs coarse-scale at reserve level), we found overwhelming support for tree cover at the camera site level as the main factor driving jackal occupancy with a higher occupancy in open areas. Our results suggest that neither anthropogenic context around PAs nor large predators influence the geographic variation in jackal’s occupancy at large scales and that fine-scale habitat attributes are more important. Our study sheds light on the role of bottom-up over top-down mechanisms in driving jackals’ distribution, confirming the ecological plasticity of this species to occupy different environments and suggesting that management of this species must be planned at local scales.
... Motion-activated remote cameras (henceforth 'camera traps') have emerged as a popular non-invasive tool for monitoring terrestrial vertebrate communities [13][14][15] . Their decreasing cost and greater reliability have recently led to the application of camera traps for long-term, continuous deployment aiming to monitor entire wildlife communities across multiple seasons and years 1,[16][17][18] . Compared with one-time or annual surveys, continuous monitoring reveals new insights into wildlife responses to local, regional and global environmental changes, as well as to conservation interventions. ...
... This need is likely to grow exponentially over the coming decades as more monitoring sites are set up. Although only one or two experts are needed to validate each wildlife image, it is common practice that multiple (5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) volunteers or citizen scientists look at each image to produce a high-accuracy 'consensus' classification (~97% accurate compared to expert identifications 16 ). This duplication of effort needed to generate accurate results using volunteers further perpetuates the classification bottleneck. ...
Article
Full-text available
Camera trapping is increasingly being used to monitor wildlife, but this technology typically requires extensive data annotation. Recently, deep learning has substantially advanced automatic wildlife recognition. However, current methods are hampered by a dependence on large static datasets, whereas wildlife data are intrinsically dynamic and involve long-tailed distributions. These drawbacks can be overcome through a hybrid combination of machine learning and humans in the loop. Our proposed iterative human and automated identification approach is capable of learning from wildlife imagery data with a long-tailed distribution. Additionally, it includes self-updating learning, which facilitates capturing the community dynamics of rapidly changing natural systems. Extensive experiments show that our approach can achieve an ~90% accuracy employing only ~20% of the human annotations of existing approaches. Our synergistic collaboration of humans and machines transforms deep learning from a relatively inefficient post-annotation tool to a collaborative ongoing annotation tool that vastly reduces the burden of human annotation and enables efficient and constant model updates.
... Camera traps are time efficient in field as the maintenance hours of camera traps are low, but high personnel hours are needed for data processing (McCleery et al. 2014;Torrents-ticó et al. 2017). However, there is current research into lessening the data processing time, like using citizen science and artificial intelligence (Green et al. 2020;Swanson et al. 2016;Pardo et al. 2021). Camera traps are commonly used to monitor medium to large-sized mammals and have been found to be effective at detecting nocturnal, threatened, or cryptic species (Bowler et al. 2017;De Bondi et al. 2010). ...
... Our assessment overlapped with camera traps deployed for the Snapshot Safari project (Pardo et al. 2021), using Cuddeback (Cuddeback Digital, Green Bay, WI, USA) Black Flash (model# 1231) and IR Plus White Flash (model# 1309) camera traps. The camera traps were positioned on a tree or pole within a 250-m radius of a predetermined GPS point at the center of a 5-km 2 grid cell. ...
Article
Full-text available
Management and conservation actions are only as effective as our ability to monitor and assess biodiversity trends. We therefore compared the cost efficiency and effectiveness of several standard methods to assess mammal diversity using camera traps, live traps, track plates, mist nets for bats, acoustic bat surveys, spotlight surveys, and block transects recording individual animals, scat, and tracks. We also assessed local knowledge through interviews. We surveyed on two contrasting arid ecosystems in South Africa. Our data indicated that block transects were the most cost-efficient and effective method at ascertaining terrestrial mammal species richness. Depending on the goal of the study and the area, a combination of block transects with camera traps or spotlight surveys is a viable option. However, our study indicated the best combination to detect species across different taxonomic groups was block transects and live traps. Local knowledge interviews can be a good addition to a survey as it assesses mammal diversity for longer time period and not just the survey season and it provides knowledge on species that are difficult to detect.
... Animal-Centric Datasets. Existing computer vision datasets in animal ecology primarily target the tasks of species classification [9,12,96,120,126,127,128], detection [2,6,9,10,21,32,36,66,99,106,110,117,120,121,122,141,150], and individual identification [58, 80,100]. These datasets consist predominantly of RGB imagery where visual features are key signals for recognition; in contrast, our dataset consists of single-channel sonar video, in which the animals of interest are difficult to distinguish from background, debris, and each other. ...
Preprint
We present the Caltech Fish Counting Dataset (CFC), a large-scale dataset for detecting, tracking, and counting fish in sonar videos. We identify sonar videos as a rich source of data for advancing low signal-to-noise computer vision applications and tackling domain generalization in multiple-object tracking (MOT) and counting. In comparison to existing MOT and counting datasets, which are largely restricted to videos of people and vehicles in cities, CFC is sourced from a natural-world domain where targets are not easily resolvable and appearance features cannot be easily leveraged for target re-identification. With over half a million annotations in over 1,500 videos sourced from seven different sonar cameras, CFC allows researchers to train MOT and counting algorithms and evaluate generalization performance at unseen test locations. We perform extensive baseline experiments and identify key challenges and opportunities for advancing the state of the art in generalization in MOT and counting.
... Similarly, Snapshot Safari 33 has found success with cyclical development in collaboration with multiple stakeholders and technologists. The project provides a platform for camera trap data processing and has slowly expanded to include more study sites, expanded functionality to allow tagging by citizen scientists, and added machine learning to pre-process images 33 . Additionally, our limited data from technologists suggest a different set of feature priorities for conservation technologies and highlights the importance of involving end-users from the beginning to ensure that tool specifications meet conservation needs 30 . ...
Article
Full-text available
Amid accelerating threats to species and ecosystems, technology advancements to monitor, protect, and conserve biodiversity have taken on increased importance. While most innovations stem from adaptation of off-the-shelf devices, these tools can fail to meet the specialized needs of conservation and research or lack the support to scale beyond a single site. Despite calls from the conservation community for its importance, a shift to bottom-up innovation driven by conservation professionals remains limited. We surveyed practitioners, academic researchers, and technologists to understand the factors contributing to or inhibiting engagement in the collaborative process of technology development and adoption for field use and identify emerging technology needs. High cost was the main barrier to technology use across occupations, while development of new technologies faced barriers of cost and partner communication. Automated processing of data streams was the largest emerging need, and respondents focused mainly on applications for individual-level monitoring and automated image processing. Cross-discipline collaborations and expanded funding networks that encourage cyclical development and continued technical support are needed to address current limitations and meet the growing need for conservation technologies.
Article
Proper management of the earth's natural resources is imperative to combat further degradation of the natural environment. However, the environmental datasets necessary for informed resource planning and conservation can be costly to collect and annotate. Consequently, there is a lack of publicly available datasets, particularly annotated image datasets relevant for environmental conservation, that can be used for the evaluation of machine learning algorithms to determine their applicability in real-world scenarios. To address this, the Time-evolving Data Science and Artificial Intelligence for Advanced Open Environmental Science (TAIAO) project in New Zealand aims to provide a collection of datasets and accompanying example notebooks for their analysis. This paper showcases three New Zealand-based annotated image datasets that form part of the collection. The first dataset contains annotated images of various predator species, mainly small invasive mammals, taken using low-light camera traps predominantly at night. The second provides aerial photography of the Waikato region in New Zealand, in which stands of Kahikatea (a native New Zealand tree) have been marked up using manual segmentation. The third is a dataset containing orthorectified high-resolution aerial photography, paired with satellite imagery taken by Sentinel-2. Additionally, the TAIAO web platform also contains a collated list of other datasets provided and licensed by our data partners that may be of interest to other researchers.
Article
Full-text available
Research to inform a sustainable future for southern African mountains as socialecological systems requires major investment. This is needed to strengthen existing relationships, build new relationships among academia, policy, and practice, and drive a robust research capacity program. This is particularly important in disciplines where there is currently limited capacity for mountain-related research in the region. For many pertinent issues in southern African mountains, the urgent need for foundational research is a reality; without this, it is impossible to build toward multidisciplinary outcomes and to drive transdisciplinary efforts. Keys to strengthening solution-oriented research are improved coordination between actors in similar disciplines (eg water security), strong relationships to achieve maximum synergy instead of competition, and major investment in emerging young researchers. The Afromontane Research Unit is leading the way for southern African mountains.
Article
Full-text available
The time is now For decades, scientists have been raising calls for societal changes that will reduce our impacts on nature. Though much conservation has occurred, our natural environment continues to decline under the weight of our consumption. Humanity depends directly on the output of nature; thus, this decline will affect us, just as it does the other species with which we share this world. Díaz et al. review the findings of the largest assessment of the state of nature conducted as of yet. They report that the state of nature, and the state of the equitable distribution of nature's support, is in serious decline. Only immediate transformation of global business-as-usual economies and operations will sustain nature as we know it, and us, into the future. Science , this issue p. eaax3100
Article
Full-text available
Abstract Gaining a better understanding of global environmental change is an important challenge for conserving biodiversity. Shifts in phenology are an important consequence of environmental change. Measuring phenology of different taxa simultaneously at the same spatial and temporal scale is necessary to study the effects of changes in phenology on ecosystems. Camera traps that take both time‐lapse as well as motion‐triggered images are increasingly used to study wildlife populations. The by‐catch data of these networks of camera traps provide a potential alternative for measuring several climatic and phenological variables. Here, we tested this ability of camera traps, and quantified climatic variables as well as the timing of changes in plant and animal phenology. We obtained data from 193 camera‐unit deployments during a year of camera trapping on a peninsula in northern Sweden aimed at studying wildlife. We estimated daily temperature at noon and snow cover using recordings provided by cameras. Estimates of snow cover were accurate, but temperature estimates were higher compared with a local weather station. Furthermore, we were able to identify the timing of leaf emergence and senescence for birches (Betula sp.) and the presence of bilberry berries (Vaccinium myrtillus), as important food sources for herbivores. These were linked to the timing of the growth of antlers and the presence of new‐born young for three ungulate species as well as the presence of migratory Eurasian cranes (Grus grus). We also identified the timing of spring and autumn moulting of mountain hares (Lepus timidus) in relation to snow cover. In this novel study, we show the potential of (by‐catch) data from camera traps to study phenology across a broad range of taxa, suggesting that a global network of camera traps has great potential to simultaneously track wildlife populations and the phenology of interactions between animals and plants.
Article
Full-text available
Wildlife is an essential component of all ecosystems. Most places in the globe do not have local, timely information on which species are present or how their populations are changing. With the arrival of new technologies, camera traps have become a popular way to collect wildlife data. However, data collection has increased at a much faster rate than the development of tools to manage, process and analyse these data. Without these tools, wildlife managers and other stakeholders have little information to effectively manage, understand and monitor wildlife populations. We identify four barriers that are hindering the widespread use of camera trap data for conservation. We propose specific solutions to remove these barriers integrated in a modern technology platform called Wildlife Insights. We present an architecture for this platform and describe its main components. We recognize and discuss the potential risks of publishing shared biodiversity data and a framework to mitigate those risks. Finally, we discuss a strategy to ensure platforms like Wildlife Insights are sustainable and have an enduring impact on the conservation of wildlife.
Technical Report
Full-text available
This discussion paper explores the global political economy of environmental data and how public and private sector actors can jointly generate global public goods while avoiding data and technology monopolies and governance processes which lack transparency, inclusion and accountability. It also touches upon the question of social media and consumer behavior. If power and consumption patterns are increasingly based in data and digital social networks, what is the science-society strategy to support those networks in driving transformations to deliver global sustainability?
Article
Full-text available
Effective classification of large datasets is a ubiquitous challenge across multiple knowledge domains. One solution gaining in popularity is to perform distributed data analysis via online citizen science platforms, such as the Zooniverse. The resulting growth in project numbers is increasing the need to improve understanding of the volunteer experience; as the sustainability of citizen science is dependent on our ability to design for engagement and usability. Here, we examine volunteer interaction with 63 projects, representing the most comprehensive collection of online citizen science project data gathered to date. Together, this analysis demonstrates how subtle project design changes can influence many facets of volunteer interaction, including when and how much volunteers interact, and, importantly, who participates. Our findings highlight the tension between designing for social good and broad community engagement, versus optimizing for scientific and analytical efficiency.
Article
Full-text available
1.Ecologists often study wildlife populations by deploying camera traps. Large datasets are generated using this approach which can be difficult for research teams to manually evaluate. Researchers increasingly enlist volunteers from the general public as citizen scientists to help classify images. The growing number of camera trap studies, however, makes it ever more challenging to find enough volunteers to process all projects in a timely manner. Advances in machine learning, especially deep learning, allow for accurate automatic image classification. By training models using existing datasets of images classified by citizen scientists and subsequent application of such models on new studies, human effort may be reduced substantially. The goals of this study were to (i) assess the accuracy of deep learning in classifying camera trap data, (ii) investigate how to process datasets with only a few classified images that are generally difficult to model, and (iii) apply a trained model on a live online citizen science project. 2.Convolutional neural networks (CNNs) were used to differentiate among images of different animal species, images of humans or vehicles, and empty images (no animals, vehicles or humans). We used four different camera trap datasets featuring a wide variety of species, different habitats, and a varying number of images. All datasets were labeled by citizen scientists on Zooniverse. 3.Accuracies for identifying empty images across projects ranged between 91.2% and 98.0%, while accuracies for identifying specific species were between 88.7% and 92.7%. Transferring information from CNNs trained on large datasets (‘transfer‐learning’)was increasingly beneficial as the size of the training dataset decreased and raised accuracy by up to 10.3%. Removing low‐confidence predictions increased model accuracies to the level of citizen scientists. By combining a trained model with classifications from citizen scientists, human effort was reduced by 43% while maintaining overall accuracy for a live experiment running on Zooniverse. 4.Ecology researchers can significantly reduce image classification time and manual effort by combining citizen scientists and CNNs, enabling faster processing of data from large camera trap studies. This article is protected by copyright. All rights reserved.
Article
Full-text available
Occupancy models are widely used in camera trap studies to analyze species presence, abundance, and geographic distribution, among other important ecological quantities. These models account for imperfect detection using a latent variable to distinguish between true presence/absence and observed detection of a species. Under certain experimental setups, parameter estimation in a latent variable framework can be challenging. Several studies have issued guidelines on the number of independent replicated observations (surveys) needed for each unchanging occupancy field (season) to ensure reliable estimation. In this paper we present a spatio‐temporal occupancy model, and show through a simulation study that it can be fit to data obtained from a single survey per season, so long as the number of seasons is sufficiently large. We include an application using camera‐trap data on the Thomson's gazelle in the Serengeti in Tanzania This article is protected by copyright. All rights reserved.
Article
Full-text available
Honey badgers are cryptic carnivores that occur at low densities and range across large areas. The processes behind site-level honey badger abundance and detection rates are poorly understood, and there are conflicting results about their avoidance of larger carnivores from different regions. We used data from 224 camera traps set up in the Serengeti National Park, Tanzania to evaluate patterns in detection rates, spatial distribution, and activity patterns of honey badgers. Our top models showed that the relative abundance of larger carnivores (e.g., African lions, Panthera leo, and spotted hyenas, Crocuta crocuta) was important, but surprisingly was positively related to honey badger distribution. These results suggest that honey badgers were not avoiding larger carnivores, but were instead potentially seeking out similar habitats and niches. We also found no temporal avoidance of larger carnivores. Honey badgers exhibited seasonal variation in activity patterns, being active at all times during the wet season with peaks during crepuscular hours, but having a strong nocturnal peak during the dry season. Our detection rates of honey badgers at individual camera traps were low (3402 trap nights/detection), but our study shows that with adequate effort camera traps can be used successfully as a research tool for this elusive mustelid.
Article
Ambiguous empirical support for ‘landscapes of fear’ in natural systems may stem from failure to consider dynamic temporal changes in predation risk. The lunar cycle dramatically alters night-time visibility, with low luminosity increasing hunting success of African lions. We used camera-trap data from Serengeti National Park to examine nocturnal anti-predator behaviours of four herbivore species. Interactions between predictable fluctuations in night-time luminosity and the underlying risk-resource landscape shaped herbivore distribution, herding propensity and the incidence of ‘relaxed’ behaviours. Buffalo responded least to temporal risk cues and minimised risk primarily through spatial redistribution. Gazelle and zebra made decisions based on current light levels and lunar phase, and wildebeest responded to lunar phase alone. These three species avoided areas where likelihood of encountering lions was high and changed their behaviours in risky areas to minimise predation threat. These patterns support the hypothesis that fear landscapes vary heterogeneously in both space and time.