Camera Traps Research Articles

INFERRING BIRD COMMUNITIES ON REMOTE FRESHWATER LAKES THROUGH TIME-LAPSE IMAGERY

Assessing bird diversity and associated ecological patterns in remote freshwater lakes presents challenges that require innovative approaches. Here we evaluated the utility of time-lapse images from camera traps for this purpose using two lakes in Haida Gwaii, British Columbia, Canada. We consider four key factors: 1) manual versus automated image processing, 2) data validation through in-person observations, 3) the ability of time-lapse data to capture known ecological patterns, and 4) variation in sampling effort. We find that (1) MegaDetector, a common AI approach, is not effective at detecting birds from time-lapse images – necessitating manual screening, (2) relative bird abundances were correlated between time-lapse and in-person observer data, (3) time-lapse data captures previously-documented ecological variation in space and time, and (4) sampling effort per camera trap can be, under certain scenarios, scaled down, but camera trap position and time-lapse frequency greatly influence bird detectability. Our research builds on the few previous studies that use time-lapse imagery to detect birds, and our work is the first to focus on detecting ecological patterns on freshwater lakes in remote landscapes. Camera trap technologies can shed light on avifauna in remote freshwater lakes, but additional developments are needed to maximize utility of such applications.

Camera trapping observations of the Kloof Frog, Natalobatrachus bonebergi, in South Africa, reveals effects of environmental conditions on breeding behaviour patterns

ABSTRACT The threatened Kloof Frog (Natalobatrachus bonebergi) is associated with coastal forested rocky streams in the KwaZulu-Natal and Eastern Cape provinces, South Africa. Their expanded toe tips allow them to navigate rocks, branches and leaves, where the species lays distinct egg clumps that overhang slow-flowing streams, into which hatching tadpoles drop to complete their development. A decade of implementation of a provincial monitoring and surveillance protocol led to the observation that there were regular egg clump laying sites, suggesting that this species was an ideal candidate to examine behaviour in relation to environmental factors using camera traps. The main objective of the study was to determine the efficacy of using camera trap imagery to understand this species’ behaviour to inform conservation management. From January to December 2021, a camera trap was installed at a site where egg clumps are regularly laid in Vernon Crookes Nature Reserve, KwaZulu-Natal, South Africa. Data on breeding and non-breeding behaviour were captured from camera trap images. Data on environmental variables, including temperature, rainfall and lunar albedo, were also collected. Results showed that the temperatures of habitats used by N. bonebergi were significantly cooler than the surrounding areas. Natalobatrachus bonebergi was active throughout the year and bred from September to April. Breeding activity duration decreased as lunar albedo increased while activity duration increased as temperature decreased. This is the first use of camera trapping to observe amphibian behaviour in South Africa and proved valuable for documenting ethological patterns that could be used to predict how changes in habitat and climate could impact the species.

Community Camera Trapping: A Novel Method for Encouraging Human–Big Cat Coexistence on Human‐Dominated Land

ABSTRACTHuman–wildlife conflict (HWC) poses a significant threat to both human livelihoods and conservation, particularly for large, potentially dangerous species such as big cats. Here, we provide an overview of community camera trapping (CCT) and the expanded community camera‐trapping plus (CCT+) programs, which are innovative, inclusive approaches that can help reduce conflict and improve coexistence. We illustrate how CCT, through a combination of participatory monitoring and tangible benefits directly linked to wildlife presence, helps create initial changes in attitude and tolerance towards big cats and other wildlife. CCT+ builds on this by introducing additional incentives and disincentives that promote conservation‐friendly behavior change, leading to tangible improvements in coexistence. We explain the approaches, highlight key aspects for practitioners to consider, and discuss the potential to scale these approaches to strengthen human–wildlife coexistence across diverse contexts.

Landscape-Scale Effects of Season and Predation Risk on the Terrestrial Behavior of Chacma Baboons (Papio ursinus).

"Terrestrial" primates are not common nor well defined across the order. In those species that do use the ground, terrestriality is rarely documented outside daylight hours. Predation risk is thought to have shaped conserved behaviors like primates' selection of arboreal sleep sites, but it is less clear-particularly at the landscape scale-how predation risk interacts with other ecological and seasonal variables to drive terrestriality. This camera trapping study investigates patterns in terrestrial behavior both spatially and temporally across neighboring populations of chacma baboons. We use camera trap data from two terrestrial grids, one established within and one outside the boundaries of Gorongosa National Park, Mozambique. We model how baboon terrestrial activity varies with woody cover, proximity to water, season, anthropogenic variables, as well as predation risk. We also model how terrestrial activity varies across the diel cycle and use overlap analyses to explore differences in the baboon populations' activity patterns. We find no significant predictors of geospatial variation in the terrestrial activity of baboons across each grid but do find evidence of higher terrestrial activity in the late dry season. We also find significantly different diel patterns of baboon activity detected across each grid. Baboons likely use the ground more in the dry season for accessing water and resources when arboreal foods are less abundant. Diel variation between the two populations suggests that baboons might utilize the ground more during "riskier" crepuscular and nocturnal hours where leopards are not present.

Optimization techniques applied on image segmentation process by prediction of data using data mining techniques

The research work presents an enhanced method that combines rule-based color image segmentation with fuzzy density-based spatial clustering of applications with noise (FDBSCAN). This technique enhances super-pixel robustness and improves overall image quality, offering a more effective solution for image segmentation. The study is specifically applied to the challenging and novel task of predicting the age of tigers from camera trap images, a critical issue in the emerging field of wildlife research. The task is fraught with challenges, particularly due to variations in image scale and thickness. Proposed methods demonstrate that significant improvements over existing techniques through the broader set of parameters of min and max to achieve superior segmentation results. The proposed approach optimizes segmentation by integrating fuzzy clustering with rule-based techniques, leading to improved accuracy and efficiency in processing color images. This innovation could greatly benefit further research and applications in real-world scenarios. Additionally, the scale and thickness variations of the present barracuda panorama knowledge base offer many advantages over other enhancement strategies that have been proposed for the use of these techniques. The experiments show that the proposed algorithm can utilize a wider range of parameters to achieve better segmentation results.

Indigenous-Led Analysis of Important Subsistence Species Response to Resource Extraction.

Subsistence hunting, or "country food," on traditional territories is essential for numerous Indigenous Peoples who face food insecurity. For many First Nations of Canada, subsistence hunting is also inextricably linked to traditional conservation practices, as hunting is an important way of engaging with nature. In Canada's boreal forest, large game such as moose (Alces alces) is a primary source of protein. However, resource extraction-including forestry and oil and gas-has shifted large game distributions and affected the availability and abundance of food resources. Here, the Indigenous authors designed the study and processed remote camera trap data, then sought out Western scientists to generate generalized linear models to evaluate moose habitat use and spatial-numerical responses to possible stressors in north-central Alberta, including fire, harvest, oil and gas extraction, and other disturbances. Together, through the coproduction of knowledge, we examined the effects of human-caused stressors on moose habitat use by sex and age class. The proportion of various land cover types and human land use for resource extraction was important in moose habitat use. Notably, male, female, and young moose all used habitat differently and at different spatial scales. However, young moose (with their mothers) strongly selected natural forest disturbances such as burned areas but avoided human-created disturbances such as petroleum exploration "seismic" lines. Female moose with young attempts to maximize forage opportunities do not use human-disturbed forests in the same ways they use naturally disturbed areas. Our findings, in the context of Indigenous interpretation from remote cameras and community insights, have linked human disturbance to declines in moose densities and displacement from traditional hunting grounds. Evaluating and predicting shifts in large game distributions is critical to supporting Indigenous food security and sovereignty and identifying where industries operating on First Nations lands can better engage responsibly with First Nations.

Long-Term Changes in Survival of Eurasian Lynx in Three Reintroduced Populations in Switzerland.

For conservation or management programs, basic data on vital rates are important but often hard to acquire for long-lived and elusive wildlife species such as large carnivores. In this study, we analyzed long-term changes in survival rates for different sexes and age classes (juvenile, subadult, adult) in three reintroduced Swiss lynx populations (Alps, Jura, Northeastern Switzerland). A novel modeling approach allowed us to combine picture data from camera trapping and lynx pictures resulting from chance observations, telemetry data, and dead recoveries over a monitoring period of 25 years (1997-2022). Mean annual survival of adult lynx varied between 0.71 and 0.81 for males and between 0.70 and 0.85 for females. Mean survival of subadults ranged between 0.59 and 0.89 among populations. Juvenile survival was highly variable and low on average (< 0.4). Our findings highlight that unknown sources of mortality exist in some populations and that future studies on mortality causes and potential effects of inbreeding on survival are needed to ensure long-term conservation of the lynx in Switzerland. Our study can serve as a basis for future studies on population viability and conservation threats to the species in human-dominated landscapes and demonstrates the complexity and high variation of survival between different age and sex classes in space and time, potentially leading to source-sink dynamics.

Mammalian diversity and conservation in the mixed-evergreen forest of Baraiyadhala national park, Chattogram, Bangladesh

The study was carried out to assess the mammalian diversity and conservation in Baraiyadhala National Park between October 2021 and June 2023 spanning a total of 78 field days. Field data were collected following a combination of different methods including transect walk, night survey, mist netting and camera trapping. Ten automated, infrared sensor camera traps were set up at 48 camera trap station for 14 months covering a total of 2793 camera days. Fifty-five species of mammals were recorded of which 41.82% was rare, followed by 29.09% common, 16.36% uncommon and 12.73% very common. Mammalian community of this national park were mostly nocturnal (32.73%), followed by metaturnal (23.64%), crepuscular (21.82%) and diurnal (21.82%). Overall mammalian diversity of this national park was very high (D = 0.06, H’ = 3.35). Among the recorded mammals, 17 species are threatened in different categories. Hunting and poaching with intentional forest fire were the major threats to the mammals of this national park. Effective management plan involving local community, habitat restoration and introducing Spatial Monitoring and Reporting Tool (SMART) patrolling could minimize the threats to mammals in this national park. Bangladesh J. Zool. 52(3): 269-280, 2024

Local knowledge of Komodo dragons (Varanus komodoensis) in north-central Flores Island (Indonesia) and implications for conservation of the species

Context Apart from Komodo and other small islands to the east, Varanus komodoensis is found near the western extremity of Flores Island and along the island’s north coast. How far east along this coast the species extends remains a question. Aims Ethnographic research undertaken in 2008 resulted in reports of Komodo dragons in the region of Nggolo Nio in north-central Flores Island. In 2018 and 2024, I returned to this region to collect further information on local knowledge and survival of the lizards. Methods Open-ended and directed questioning of informants was conducted in the national language, Bahasa Indonesia. Key results Eleven men and three women had observed V. komodoensis locally and provided descriptions of the lizards. The most recent sightings occurred during the last 5–10 years. Repeated sightings of a specimen that, unusually, had entered a settlement in the western part of the Nggolo Nio region occurred in 2017. This was observed by eight interviewees, including the three women. The species was described with reference to size, colour, vocalisation, diet, behaviour, habitat and relations with humans. All reports focused partly on differences between V. komodoensis and the smaller species Varanus salvator. Conclusions All accounts indicated that the same species was being described and that V. komodoensis survives in Nggolo Nio territory, though in small numbers that villagers stated had declined over the past two to three decades. Though ambiguous, the findings of camera trapping by herpetological researchers in 2014–2019 lend some support to these local claims. Implications The study shows how local knowledge of and attitudes toward rare animals can complement zoological study and contribute to conservation efforts.

The discovery of a male Malay Crestless Fireback &lt;Lophura erythrophthalma&gt; (Raffles, 1822) (Aves: Galliformes: Phasianidae) at Ulu Sat Forest Reserve, Machang, Kelantan, Peninsular Malaysia

A study was conducted in the Ulu Sat Forest Reserve, Machang to record the presence of terrestrial vertebrate animals by using the camera trapping method from April 2023 until May 2023, and seven camera trap units (Reconyx Hyperfire) were used. During the study, a male Malay Crestless Fireback Lophura erythrophthalma was recorded. This discovery is considered something important and needs to be documented because this species has been classified as ‘Critically Endangered’ by the International Union for Conservation of Nature (IUCN) Red List of Threatened Species. We believe that the record of this species can increase efforts in the management and conservation of the area.

First records of the Amazonian brown brocket deer Passalites nemorivagus (Cuvier, 1817) (Artiodactyla: Cervidae) from Isla de Barú, Departamento de Bolívar, Colombia.

In this report, we present the first documented records of Passalites nemorivagus on Isla de Barú, Departamento de Bolívar, Colombia. These records were obtained through direct evidence (camera trapping) and indirect evidence (tracks, scats, and latrines). Our findings indicate that this species inhabits the dry forests of this island.

Thirteen years of Andean cat research in Chile: occurrence records and environmental patterns

ABSTRACT The Andean cat is the only felid listed as Endangered in the Americas according to IUCN. We present an update on the records of this species from 2010 to 2023. We used 350 camera trap units along 1,500 km in Chile, sightings, and records of skins in landowners’ houses. We detected Andean cat presence 114 times in 58 sites from northernmost Putre to Farellones and Alfalfal in central Chile, recording 13 sightings and one skin. From the records obtained, we developed an analysis based on the review of the distribution patterns of the species in the environmental space, as well as its potential distribution, intending to evaluate its environmental niche and determine priority sites for conducting new surveys in search of the species. This research aims to compile numerous unpublished Andean cat records across its distribution range in Chile, to create a valuable resource for species conservation and future research efforts.

Camera traps offer reliable estimates compared to ground surveys for monitoring duck pairs and broods

AbstractMonitoring breeding waterfowl populations with ground‐based pair and brood surveys informs management and conservation decisions. However, surveys are often limited temporally and may miss individuals that are not present or available for detection at the time of the survey. Alternative methods to monitor waterfowl such as camera traps may be more appropriate to measure relative abundance, but it is unknown how camera trap surveys compare to ground‐based surveys. We conducted concurrent walk‐up pair and brood surveys on 20 wetlands in Manitoba, Canada and deployed cameras set to take pictures at 10‐min intervals during daylight hours. We compared indices of relative abundance and species richness of ducks and ducklings on small prairie wetlands (&lt;3.8 ha) detected with ground‐based and camera trap surveys and make recommendations regarding the time of day and duration of camera surveys. As predicted, camera surveys detected more ducks, ducklings, and duck species than ground surveys counted. Importantly, camera surveys detected ducks and ducklings at wetlands that ground surveys did not. Both duck and duckling observations were positively associated between survey methods (ducks: R2 = 0.22, F1,17 = 4.83, P = 0.04, ducklings: R2 = 0.49, F1,15 = 14.16, P = 0.002). We found that cameras are a useful tool to survey relative duck abundance, and the extended temporal surveillance of cameras reduces false negatives.

Seasonal use of different tree strip variants within a modern silvoarable agroforestry system by large and medium-sized mammals

Silvoarable agroforestry systems (AFS) with short rotation coppice strips (CS) combine fast-growing trees with croplands, offering potential biodiversity benefits in agricultural landscapes. However, their actual impact on biodiversity remains underexplored. This study investigated the use of CS within a modern silvoarable AFS in northern Germany by large and medium-sized mammals in different seasons (winter, summer) and various tree strip variants (Conventional, Aspen, Ecological). A total of ten species were detected during camera trapping surveys, including farmland (European hare) and forest species (e.g. pine marten, wildcat) along with generalists (e.g. beech marten, raccoon). The European hare showed the highest activity (detected on nearly 90% of days in winter and on all days in summer), followed by roe deer (almost 40% of days in winter and 80% in summer) and red fox (about 60% resp. 40% of days in winter/summer), while most other species were detected sporadically or only in one season. Mammal activity was higher in summer than in winter, with minor differences between the tree strip variants in both seasons. The results indicate that CS in modern AFS are used by a wide range of mammal species and could especially benefit the threatened European hare, while potentially supporting forest species like the strictly protected wildcat with regard to woodland connectivity. However, this study does not provide information on the habitat function and quality of CS compared to other habitats like hedgerows. Therefore, and due to the spatial and temporal limitations of the study, there is a great need for further research on the use of CS by mammals.

Assessing the effects of landscape attributes in medium and large terrestrial mammal richness inside a tropical rainforest biosphere reserve.

The landscape's structure significantly impacts how communities assemble due to the environment filtering and the limitation of dispersal processes. Human activities can enhance or alter these factors, resulting in changing environments and isolated animal populations. In studying this, we used 39 camera trap stations during 102 continuous days in preserved and disturbed areas to assess medium and large terrestrial mammals in the Selva El Ocote Biosphere Reserve in Chiapas, Mexico (REBISO). We identified various groups of mammals and the factors influencing their presence and distribution through VIF, Clustering, RDA, NDMS, ANOSIM multivariate, and niche decomposition (OMI) analyses. The redundancy analysis (RDA) indicated that the most significant variables were altitude, distance to main roads and settlements, and forest cover. The optimal multivariate indicator (OMI) analysis accounted for 88.75% of the variability in niche structure. It revealed that Puma concolor exhibited the highest level of specialization (marginality = 2.96), while Nasua narica displayed the most generalist behavior (marginality = 0.26). Natural elements and human impact played a crucial role in the species' distribution, resulting in patterns in two distinct conditions: one characterized by preserved natural environments and the other affected by significant human impact. Notably, 63% of the species were common in both regions. For instance, Cuniculus paca positively correlated with distance to main roads and altitude, whereas Leopardus pardalis negatively responded to proximity to settlements. This study emphasizes the importance of maintaining habitat connectivity to preserve terrestrial mammal species.

Restoration of giant panda habitat requires balancing single‐ and multi‐species benefits

Efforts to restore habitat for wildlife often target single species, with limited consideration of the potential benefits provided to sympatric species. On the basis of range‐wide data from the Fourth National Giant Panda Survey and infrared camera trapping, we used species distribution models to project the outcomes of five habitat restoration scenarios—designed to benefit giant pandas (Ailuropoda melanoleuca)—for giant pandas as well as for sympatric birds and mammals. Scenario outcomes, particularly those involving the conversion of plantation forests and shrublands into suitable forests, demonstrated a significant enhancement in giant panda habitat suitability, but with contrasting effects for sympatric species. Moreover, while restoration of giant panda habitat may enhance species richness and functional diversity, especially when shrublands are converted into forests, such action could also reduce phylogenetic diversity. Our findings suggest that single‐species habitat restoration may have negative outcomes for sympatric species, highlighting the need to consider trade‐offs between focal and non‐focal taxa.

Exploring the use of REM in compact hunting grounds comparing site specific to average parameters

Wildlife management decisions require accurate knowledge of population parameters such as density. The random encounter model (REM) is effective for estimating densities of unmarked animals but has not been tested in areas smaller than 10 km2. This study evaluates REM's precision in small areas, considering study area size, shape, and species aggregation behaviours. It also examines whether REM can be simplified by using averaged REM parameters (AVE) instead of site-specific parameters (SSP). Camera trap data from 19 alpine and continental areas in Austria, covering 28 ungulate populations of roe deer Capreolus capreolus, red deer Cervus elaphus, chamois Rupicapra rupicapra, and wild boar Sus scrofa, were analysed. The mean coefficient of variation (CV) was 0.45 ± 0.03, comparable to studies in larger areas (46.6 km2—501 km2). Study area size and shape did not significantly affect CV. The ratio of cameras with contact negatively impacted CV, indicating higher precision for evenly distributed species like roe deer (overdispersion factor k = 1.0 ± 0.11) compared to aggregated species like red deer (k = 0.5 ± 0.12). SSP and AVE densities were not statistically different but could not be explained by the same parameters, such as sea level, ecoregion, accessibility, road length, and feeding. Thus, a simplified approach using averaged REM parameters does not yield sufficiently comparable results. For wildlife management in small areas, REM can be used without restrictions related to the size and shape of the areas sampled.

Monitoring and habitat inferences change with population metric: a case study with mesocarnivores

Context Accurate and precise estimates of wildlife abundance and distribution are critical for robust ecological inference and effective management. However, obtaining this information for mesocarnivores is challenging because they are elusive and highly mobile. Aims To compare four common population metrics (occupancy, local abundance, relative abundance, and density) for monitoring unmarked populations and the influence of three habitat covariates on these population metrics. Methods For five mesocarnivores species we used data collected at 74 camera traps deployed in the northeastern USA in summer 2021 to fit (1) models that estimated probabilistic occupancy, (2) Royle–Nichols models that estimated local abundance, (3) Poisson distributed general linear models that estimated relative abundance, and (4) random encounter and staying time (REST) models that estimated density. We also quantified habitat relationships across these four different models and compared the resultant inferences. Key results Density and relative abundance had the highest correlation (Pearson correlation (r) = 0.91), whereas occupancy and density had the lowest correlation (r = 0.19). Density estimates for all species were consistent with expectations and similar to those reported in previous studies. The effects of habitat covariates changed across metrics, such that a significant effect of a covariate on one metric was not indicative of a significant influence on the other metrics. There were only two instances of a significant effect of a covariate on all metrics, and two instances where the influence of a covariate had opposite, albeit insignificant, effects on two metrics. Conclusions Estimates of occupancy and local abundance for mesocarnivores derived from camera traps may not be reliable proxies for density. However, relative abundance, as derived from detection rates, could be a promising means of monitoring density with less intensive data processing. Mesocarnivore habitat relationships changed across these metrics. Implications When designing monitoring or research programs, practitioners should be cautious about assuming that inferences derived from camera trap estimates of these four population metrics are interchangeable. Further, we highlight how the REST model offers a promising new means for monitoring multiple mesocarnivores simultaneously, and likely other unmarked species, via density estimates.

Comparing the efficacy of two camera trapping techniques for assessing the occupancy, detection and activity patterns of small Mustelids in Britain

Increasing concern over the unfavourable population trends and data gaps for many mustelids highlight the need to improve existing monitoring methods, to enable more accurate population assessments. Here, we investigated the efficacy of two camera trapping techniques for the detection of small mustelids, specifically least weasels Mustela nivalis, and stoats M. erminea in England. We deployed two concurrent camera trap techniques - a Mostela (a camera trap enclosed within a box) and an external camera trap at 20 sampling sites during summer 2021. We measured the number of daily detections over time and calculated the probability of detection and occupancy for each species. Our results highlighted the efficacy of the Mostela at detecting weasels whilst significantly reducing the amount of video footage to review. Stoats, however, were rarely detected with either technique, although the external camera appeared to be more reliable than the Mostela. European polecats Mustela putorius were regularly detected inside the Mostela and were detected at a similar rate as the external cameras after an initial period of avoidance. The data collected from the Mostela also highlighted the nocturnal activity of polecats and diurnal activity of weasels. Here, the Mostela provided an effective method to detect and estimate the occupancy of weasels and this may be just as efficient for polecats if deployed for long enough.

Social behaviour of the crested porcupine in urban and rural areas

Abstract The social behaviour of rodents has captivated scientists for decades, offering valuable insights into the evolution of mammalian sociality. One of the main critical gaps in understanding rodent sociality is the role of subadult individuals prolonging their stay with parents and younger siblings, particularly in nocturnal and elusive species. Moreover, urban environments are known to elicit changes in the social behaviour of mammals, leading to altered patterns of sociality. Yet, little is known on how rodent societies may change in altered environments. Thus, we aimed at investigating the social behaviour of the crested porcupine Hystrix cristata in central Italy, with a particular focus on group composition and on the prolonged occurrence of subadult individuals within family groups. We also tested the potential effects of natural and urban environments. We combined marking of free-ranging individuals with camera trapping campaigns across three ecologically contrasting areas, and showed a pair-living social system for adult crested porcupines. Most videos captured either single young individuals or pairs of adults. Subadult marked females were observed together with adults during the birth season in the year following their birth, suggesting a potential helping role. Conversely, subadult marked males were rarely observed, presumably dispersing. Groups exceeding two individuals ranged between three and five, and were typically observed following reproductive events, with porcupines commuting in line, parents positioned at head and tail of the row. Urban environments showed a slightly different pattern, with occasional aggregations of subadults. We highlight the importance of understanding subadult behaviour in crested porcupines, as well as of testing the effects of urban ecosystems upon mammalian social systems. These findings contribute to a broader understanding of social behaviour and evolutionary adaptations in rodents and, more in general, of mammals.