Camera trapping is a highly useful wildlife monitoring technique that uses non-intrusive motion and thermal remote sensing devices that capture georeferenced photographic evidence of species at a particular time and location, showcasing species presence, behaviour, patterns, and visual traits. Photographs from camera traps enable the identification of individual animals and estimation of population sizes using robust methods, such as capture-mark-recapture modelling (Wearn & Glover-Kapfer 2017). They play an important role in studying elusive and uncommon species (Whitworth et al. 2016), estimating species occupancy and population density (O’Connell & Bailey 2011), and exploring species–habitat relationships and preferences (Rovero et al. 2013). Such information is useful for identifying areas in need of conservation interventions (Cordier et al. 2022). Camera traps can be deployed in the field for extended periods of several weeks to months to monitor medium- to large-sized terrestrial mammals across vast spatial and temporal scales (Kays et al. 2020), and they have recently been shown to be effective in studying small mammals and canopy-associated animals (Bowler et al. 2016).

In South Africa, camera traps are used by national and provincial conservation authorities, non-government organisations (NGOs), private nature reserves and private individuals as a research tool to monitor wildlife in South Africa. National and provincial conservation authorities collaborate with NGOs, such as Panthera, the Cape Leopard Trust (CLT), Endangered Wildlife Trust (EWT), and Snapshot Safari (Pardo et al. 2021), in addition to conducting their own camera trap monitoring. The Snapshot Safari camera trap network is one of the largest globally. It monitors population trends of southern and east African mammals in many private reserves and national parks in South Africa (Pardo et al. 2021). Panthera and Wildlife ACT monitor leopards in provincial reserves in KwaZulu-Natal (KZN) Province (Hudson 2018), and the CLT monitors leopard distribution, density, and population trends in CapeNature reserves (CapeNature 2022) and private properties in the Western Cape Province. Cape Leopard Trust staff also gather baseline data on other mammal species in the same areas (CLT 2023). Panthera monitors leopard occurrence, distribution, individuals and population density (i.e. leopards/100 km²), using spatially explicit capture-recapture techniques to model and evaluate population trends in private reserves in South Africa, such as the uMkhuze Game Reserve, and Tembe Elephant Park both in KZN, and Pilanesberg National Park and Game Reserve in Northwest Province. The results and data from these surveys are also passed on to the South African National Biodiversity Institute (SANBI) for further independent analysis (G. Mann [Panthera] pers. comm., 22 November 2023).

CapeNature collaborates with additional organisations to conduct monitoring assisted by the use of camera traps: EWT monitors riverine rabbits using spatial distribution models, and SANBI conducts research on Cape mountain zebra (CapeNature 2022). In addition, the Wilderness Foundation deploys camera traps on private properties and conservation-worthy land in potential ecological corridor areas. Images from these cameras provide very useful insights for landowners on the wildlife present on their properties, leading to enhanced and collective conservation efforts (R. Brand [Wilderness Foundation Africa] pers. comm., 22 November 2023).

Camera traps have been deployed in 11 national parks (NPs; Figure 1, Table 1) by scientists, regional ecologists and biotechnicians. In the Garden Route National Park (GRNP) in the Western Cape Province, camera traps have been used to monitor the Knysna elephant and elusive and vulnerable species, such as blue duiker, and to investigate the human–wildlife interface. Camera traps are used to monitor rhino populations in some national parks and to investigate the pollination of Quiver trees in Augrabies Falls NP, Northern Cape Province. In Table Mountain NP, 88 cameras are systematically placed in a grid array to investigate species diversity, relative abundances, and human–wildlife interactions and monitor the presence of invasive mammal species (Table 1). In parks where camera traps are used to monitor target species, such as rhinos, bycatch photographs are often neglected and not analysed because of lack of capacity. Therefore, additional useful information is waiting to be extracted from these photographs. Camera traps are also used by park operations for management purposes. For example, in Mokala NP, camera traps are used to monitor animals that may need management interventions, and in Bontebok NP in the Western Cape Province, there are prospects of deploying camera traps to monitor zebra. Many of the cameras used by the Scientific Services division of SANParks have been funded externally. For example, 40 camera traps in Addo Elephant NP (AENP) have been funded by the International Union for Conservation of Nature-Save Our Species (IUCN-SOS) programme (Bissett et al. 2022), the Peace Parks Foundation funded camera traps for a white rhino project in Marakele NP in Limpopo Province, and the Department of Forestry, Fisheries, and the Environment (DFFE) has funded 88 camera traps in Table Mountain National Park (TMNP), Western Cape Province. Camera traps are stationed in the field continuously for weeks to months, generating large volumes of data. Several camera trap projects have been conducted in parks by researchers from the CLT, Snapshot Safari, South African Environmental Observation Network (SAEON) and universities (Figure 1, Table 1). Some examples of camera trap projects include investigating the impact of fence removal on medium- to large-sized mammals in Addo Elephant NP, leopard populations in Agulhas NP (Western Cape Province), species diversity and ecological dynamics of southern African mammals and herbivore density and distribution patterns in the Kruger NP.

In marine environments, BRUV stations are used to study marine life. A BRUV station comprises a baited camera system suspended in the water column by a weighted metal frame designed to attract fish and other marine organisms into the field of view (Langlois et al. 2020). Two types of BRUV stations exist: mono-BRUV stations have a single camera, and stereo-BRUV stations have two cameras mounted to a horizontal plane. These underwater video systems are a non-destructive yet highly efficient and repeatable means of recording the occurrence of species, relative abundances (MaxN), and fish sizes, and characterising habitat types and water conditions (Langlois et al. 2020). They are also useful for monitoring temporal changes in marine ecosystems and assessing the effectiveness of conservation measures (Espinoza et al. 2014). For example, it is expected that fish sizes will be larger in marine protected areas where fishing is prohibited.

The Kogelberg Small Scale Fishery Improvement Project (2014) in CapeNature reserves in the Western Cape Province in South Africa used BRUV stations for multi-species surveys of line fish and West Coast rock lobster in the Betty’s Bay Marine Protected Area (MPA) along the southern coast of the province (CapeNature 2020). Moreover, CapeNature, in partnership with the Dyer Island Conservation Trust, deploys BRUV stations to investigate species diversity and monitor the relative abundance of fish species in the waters surrounding Dyer Island, Gansbaai, Western Cape Province (CapeNature Report 2022:29). This project is of national importance and forms a priority monitoring site for the National BRUV Working Group led by the South African Institute for Aquatic Biodiversity (SAIAB). CapeNature also deploys BRUV stations in the Stillbaai, Goukamma, Robberg and De Hoop MPAs along the southern coast of the Western Cape Province (CapeNature 2022:29). Their assessments of fish diversity and abundance both inside and outside MPAs inform management decisions and actions, mainly through the monitoring of reef fish species assemblages, abundances, and distributions using BRUV stations (CapeNature 2022:29).

The SAEON oversees long-term monitoring projects, such as the Benthic Ecosystem Long-Term Ecological Research stations, using BRUV stations and jump cameras to monitor benthic ecosystem changes over extended periods in Algoa Bay, Agulhas ecoregion in South Africa (T. Parker-Nance [SAEON] pers. comm., 17 October 2023). Similarly, the South African DFFE uses BRUV stations in fishing hotspots and MPAs to determine species compositions, abundances, and the size frequencies of reef-associated line fish (DFFE 2023). Ezemvelo KZN Wildlife leads the expansion planning of MPAs across the province, working closely in collaboration with national initiatives. In addition, they lead the African Coelacanth Ecosystem Programme Surrogacy and Spatial Solutions projects in South Africa, Mozambique, Tanzania, Kenya, the Comoros, the Seychelles and Madagascar. They have deployed BRUV stations within and outside MPAs to monitor changes in the diversity of fish species over time and to study patterns in and relationships between fish assemblages (Harris 2018).

In SANParks’ MPAs, BRUV stations are used to investigate the effectiveness of MPA zonation and the effects of coastal fishing. Marine biologists at the Cape Research Centre have deployed mono-BRUV stations since 2018 and stereo-BRUV stations since 2022 in different zones of MPAs: the no take, controlled, and sanctuary zones. Information derived from the analysed BRUV data is used to assess the effectiveness of the West Coast, Robben Island and Table Mountain National Park MPAs in the Western Cape Province. Stereo-BRUV stations are preferred by SANParks researchers because they can be used to measure fish sizes, and the size frequency distribution of fish populations is often a better indicator of MPA effectiveness than relative abundance alone. In Garden Route NP, mono-BRUV stations were initially deployed for monitoring purposes from 2014 to 2017, and later stereo-BRUV stations were deployed from 2018 to 2021. These BRUV stations were deployed in marine and estuarine ecosystems to investigate the impact of the Tsitsikamma MPA’s (Eastern Cape Province) re-zonation on coastal fishing and local fishing communities. Furthermore, a trial study in the Knysna Estuary (Western Cape Province) investigated fish-habitat associations in the marine bay regions of the estuary, and monitoring of fish-habitat associations in the Swartvlei Estuary (Western Cape Province) between open and closed phases of the river mouth. A once-off study also used BRUV stations to investigate the estuarine fish community in the Touws River Estuary, Western Cape Province. The SAEON and SAIAB also have long-term monitoring projects with sites in the Tsitsikamma MPA, monitoring sub-tidal reef fish in the Garden Route region.

Baited remote underwater video surveys are sea and weather-dependent and have occurred annually in the GRNP and bi-annually in the Cape Region. On average, sampling is conducted for 2–3 days, and the number of deployments varies between 6 and 18 per survey. This generates large amounts of 1-h BRUV data.

Managing camera trap data effectively maximises its value in biodiversity research and conservation efforts. Best practices include establishing a systematic data management system that organises images, metadata, and associated information in a standardised format. This system should include data collection, storage, and analysis protocols to ensure consistency and reproducibility. Appropriate metadata standards and annotation tools, such as TrapTagger, can facilitate data interpretation and sharing among researchers. Other fundamental best practice recommendations include ensuring that datasets conform to standardised metadata fields and attributes, such as using standardised coordinate and date formats, as well as regularly checking and servicing cameras in the field. In addition, regular data quality checks, backups, and documentation of processing steps are essential to maintain data integrity and reliability, and increase dataset longevity (Figure 2).

Standardised procedures are also needed to manage BRUV data and extract its maximum value efficiently. This entails following best practices during fieldwork, securely transferring and backing up footage to storage systems, and organising it into standardised folders with consistent naming conventions. Subsequently, data processing should be conducted using robust tools, such as EventMeasure, followed by quality assurance checks with tools such as CheckEM, and standardised, repeatable statistical analysis in platforms such as R. Once scientific or internal reports on the data have been generated, the data and associated metadata should be archived in global repositories, such as the Ocean Biodiversity Information System (2024) and GlobalArchive (2024), to enable data sharing and synthesis across localities (Figure 2). Importantly, findings from these data also need to be shared with local fishing communities at stakeholder engagement events.

To maintain the vision of maximising the value of SANParks’ visual data, SANParks must stay aligned with global best practices and use available resources to ensure SANParks’ data management follows international standards. By doing so, data handling and processes in SANParks will be streamlined and ensure that SANParks can effectively manage the ever-growing volumes of data that play a crucial role in conservation efforts. Best practice data governance and stewardship will enable SANParks to know what datasets are available, where they reside, who is authorised to use them, and how to maximise SANParks’ data value.