Spatially Explicit Capture-recapture Research Articles

Abundance of the northernmost jaguar Panthera onca breeding population

Jaguars are the apex predators of tropical and subtropical ecosystems in the Americas but their conservation is threatened due to habitat loss, poaching, and conflicts over predation. The Northern Jaguar Reserve (NJR) was created by Naturalia A.C. as a project to protect the northernmost jaguar population within an area of 59.3 thousand acres (24.4 ha) of private property. The property also includes other protected species. The present study was conducted to verify the status of this jaguar population in order to evaluate the effectiveness of the NJR, and to contribute to knowledge regarding its ecology. We estimated density using mark-resight and spatially explicit capture-recapture (SERC) models, and evaluated home range through a survey with 56 camera-trap stations over 12 bimonthly periods. In total, we obtained a sampling effort of 40,880 camera-days, registering 2.32 records from 1,000 camera-days. Jaguar density, calculated using the capture-recapture models, ranged from 0.3 to 1.44 indiv./100 km2, while the SERC model estimated a density of between 0.21 and 3.04 indiv./100 km2. The home range was 78.9 km2 for males and 45.1 km2 for females. Long-term population monitoring together with the establishment of Areas Voluntarily Assigned to Conservation, an initiative recently created by the Mexican government, such as the NJR, could be a valid strategy to strengthen jaguar conservation and thereby maintain key ecological processes for other species and their habitat in northwestern Mexico.

Density estimation of the globally threatened fishing cat Prionailurus viverrinus through a participatory science approach in the Chilika lagoon, eastern India

The fishing cat Prionailurus viverrinus is an Indo-Malayan wetland-dependent felid which is listed as Vulnerable on the IUCN Red List. Its main prey, fish, has low energetic demands and is abundant in productive aquatic habitats. This facilitates high piscivore densities and potentially alters expected scaling patterns that link population density to the body mass of small cats. With local community participation, we estimated the density of the fishing cat in Chilika, Asia’s largest brackish water lagoon, located in the state of Odisha, eastern India, with community participation. The study was carried out in 2 phases in different habitats during 2021 and 2022: a homogeneous marshy habitat and its buffer (Northern Block), and a heterogeneous matrix of different land-use types (Southern Block). We deployed a total of 144 camera traps across 4380 trap nights. Using spatially explicit capture-recapture (SECR), we estimated mean (±SE) fishing cat density to be 0.69 ± 0.1 ind. km-2 in the Northern Block and 0.67 ± 0.33 ind. km-2 in the Southern Block. The population abundance estimates for the Northern and the Southern Blocks were 159 ± 23 and 185 ± 91 respectively. In the former, SECR modelling indicated an effect of anthropogenic habitat modification upon the species’ home range extent. Our density estimates are amongst the highest reported for the species outside protected areas. The results imply that Chilika holds an abundant population of the fishing cat, the continued persistence of which requires mitigation of local and external threats to fish populations. Furthermore, our study, with its inclusive approach, sets a precedent for the use of camera trapping for obtaining robust density estimates of species with uniquely marked individuals in wetland habitats.

Where the small things are: Modelling edge effects on mouse lemur population density and distribution in northwestern Madagascar.

Edge effects result from the penetration to varying depths and intensities, of abiotic and biotic conditions from the surrounding non-forest matrix into the forest interior. Although 70% of the world's forests are within 1 km of a forest edge, making edge effects a dominant feature of most forest habitats, there are few empirical data on inter-site differences in edge responses in primates. We used spatially explicit capture-recapture (SECR) models to determine spatial patterns of density for two species of mouse lemurs (Microcebus murinus and Microcebus ravelobensis) in two forest landscapes in northwestern Madagascar. The goal of our study was to determine if mouse lemurs displayed spatially variable responses to edge effects. We trapped animals using Sherman live traps in the Mariarano Classified Forest (MCF) and in the Ambanjabe Forest Fragment Site (AFFS) site within Ankarafantsika National Park. We trapped 126 M. murinus and 79 M. ravelobensis at MCF and 78 M. murinus and 308 M. ravelobensis at AFFS. For M. murinus, our top model predicted a positive edge response, where density increased towards edge habitats. In M. ravelobensis, our top model predicted a negative edge response, where density was lower near the forest edges and increased towards the forest interior. At regional and landscape-specific scales, SECR models estimated different density patterns between M. murinus and M. ravelobensis as a result of variation in edge distance. The spatial variability of our results using SECR models indicate the importance of studying the population ecology of primates at varying scales that are appropriate to the processes of interest. Our results lend further support to the theory that some lemurs exhibit a form of ecological flexibility in their responses to forest loss, forest fragmentation, and associated edge effects.

Using photo by-catch data to reliably estimate spotted hyaena densities over time

Abstract Protected areas are becoming increasingly isolated refugia for large carnivores but remain critical for their survival. Spotted hyaenas (Crocuta crocuta) are important members of the African large carnivore guild but, like other members of the guild, routinely come into conflict with people because of their large home ranges that are not always confined to protected areas. To effectively conserve spotted hyaena populations, it is paramount to monitor their abundance through reliable and cost-effective techniques. We estimated the density of spotted hyaenas in Hluhluwe–iMfolozi Park (HiP), South Africa using camera trap images and a spatially explicit capture-recapture (SECR) framework between 2013 and 2018. We estimated an average of 18.29 ± 3.27 spotted hyaenas per 100 km2 between 2013 and 2018, with an annual estimated high of 20.83/100 km2 in 2014 and a low of 11.98/100 km2 in 2015. Our results demonstrate that camera trap by-catch data can be used for estimating spotted hyaena densities over time. We believe that given the widespread use and deployment of camera traps across Africa, collaborative efforts to use existing data to improve regional and continental estimates and population trends for spotted hyaenas should be a priority.

Spotted hyaena population density across habitat and land use types in southern Tanzania

AbstractAlthough the spotted hyaena (Crocuta crocuta) has been widely considered to be resilient to human disturbance, the species is now thought to be undergoing widespread population declines. Nevertheless, only a handful of population density estimates are available for the species, despite the importance of this information for informing conservation management. This is a consequence of both a lack of surveys and logistical challenges associated with processing spotted hyaena data. In this study, we collaborated with a cohort of students to process camera trap data from the Ruaha‐Rungwa landscape initially collected to estimate lion (Panthera leo) and leopard (Panthera pardus) population density. By doing so, we provide the first spatially explicit population density estimates for spotted hyaena in Tanzania, via spatially explicit capture‐recapture (SECR) modelling. We also examine the relationship between population densities of spotted hyaena, lion and leopard at each site. Spotted hyaena densities varied from 3.55 ± 0.72 adults and sub‐adults per 100 km2 in a miombo (Brachystegia‐Julbernardia) woodland area of Ruaha National Park, to 10.80 ± 1.08 per 100 km2 in a prey‐rich open woodland savannah habitat in Ruaha National Park, with intermediate densities recorded in Rungwa Game Reserve and MBOMIPA Wildlife Management Area. Our results suggest that spotted hyaena density is influenced by prey availability and protection, and the species may be less resilient to human pressures than widely thought. Spotted hyaena densities were generally positively correlated with densities of lion and leopard, suggesting that prey availability and anthropogenic disturbance had a greater impact than interspecific effects in shaping large carnivore densities in this system. Overall, our study provides some of the first insights into an under‐studied species in an under‐researched part of its range, while shedding light into the impact of anthropogenic versus interspecific effects in shaping population status of spotted hyaena in human‐impacted African systems.

Accounting for heterogeneous density and detectability in spatially explicit capture–recapture studies of carnivores

AbstractReliable estimates of population density are fundamental for managing and conserving wildlife. Spatially explicit capture–recapture (SECR) models in combination with information‐theoretic model selection criteria are frequently used to estimate population density. Variation in density and detectability is inevitable and, when unmodeled, can lead to erroneous estimates. Despite this knowledge, the performance of SECR models and information‐theoretic criteria remain relatively untested for populations with realistic levels of variation in density and detectability. We addressed this issue using simulations of American black bear (Ursus americanus) populations with variable density and detectability between sexes and across study areas. We first assessed the reliability of Akaike information criterion adjusted for small sample sizes (AICc) to correctly identify the true data generating model or a good approximating model. We then assessed the bias, accuracy, and precision of density estimates when such a model was selected or not. We demonstrated that unmodeled heterogeneity in detection and, more importantly, density can lead to pronounced bias. However, when a good approximating model is included in the candidate set, models with lower AICc included important forms of variation and yielded accurate estimates. We encourage researchers and practitioners to consider the impact of unmodeled variation in SECR models when making inferences and to strive to include covariates likely to be the most influential based on the species biology and ecology in candidate model sets. Doing so can improve the robustness of wildlife density estimation methods that can be leveraged to make more sound conservation and management decisions.

Estimating wolf (Canis lupus) densities using video camera traps and spatial capture–recapture analysis

AbstractEstimating population density is critical for effective species conservation, wildlife management planning, and long‐term monitoring. Obtaining accurate estimates is especially important for the wolf (Canis lupus), a widely distributed northern hemisphere apex predator whose management and conservation are highly controversial in most of its range, and whose presence usually generates high‐profile media coverage. The peculiarities of wolf social spatial organization and behavior can violate the assumptions of capture–recapture models (uniformity and independence, respectively) to a greater or lesser extent and make it difficult to obtain precise and reliable density estimates. This paper presents a case study, which estimated the population density of the Iberian wolf in the Dorsal Gallega mountain ridge (Galicia, NW Spain) based on the identification of individual wolves from their traits and behavior using video camera traps and spatially explicit capture–recapture (SCR) analyses. The study followed three phases. Firstly, field data were collected by installing camera traps and changing their location until the entire area was sampled. Second, a complete morphological and behavioral study of the wolves recorded was performed to facilitate individual recognition. Third, overdispersion due to gregariousness and other sources of heterogeneity was modeled in the SCR analyses comparing Poisson and negative binomial observation models with different random effects on the baseline detection probability. We estimated a density of 2.88 (SD: 0.37) wolves/100 km2 in the study area. We concluded that estimating wolf population size using camera trap videos, individual identification, and SCR provides a feasible method and can be used for estimating the density in similar species.

Scat DNA as a non-invasive method for estimating the abundance of the vulnerable mala (Lagorchestes hirsutus)

Context Population-monitoring programs often use direct (e.g. live capture or spotlighting) or indirect (e.g. scats sightings) observations to estimate population abundance. Such methods, however, are often inadequate for rare, elusive, or cryptic species due to the difficulty in achieving sufficient encounters or detection rates. The mala (Lagorchestes hirsutus), a small native Australian macropod, listed as Vulnerable by the IUCN, is difficult to capture, susceptible to capture myopathy, and not easily sighted in their dense habitat; consequently, the population size cannot always be estimated. The use of molecular markers to identify individual genotypes from non-invasively collected samples is increasingly being used in wildlife conservation and may be an alternative approach for mala. Aim The aim of this study was to evaluate the efficacy of non-invasive scat DNA sampling to estimate the population abundance of mala. Methods A panel of microsatellite markers was developed for the identification of individual mala via profiling of their scats. Scats were systematically collected from a wild mala population located in an 1100-ha fenced reserve in Western Australia. Individual genotypes were determined using the microsatellite markers, and the abundance of mala was estimated using the genotypes with spatially explicit capture–recapture (SECR) and mark–resight analyses. Key results The genetic markers proved variable and with sufficient exclusionary power to confidently identify unique individuals (mean locus genotyping error rate: 3.1%). Individual genetic identification from scat sampling, when used with traditional mark–recapture/resight analytical models, provides feasible estimates of population abundance. This is the first reliable abundance estimate of this mala population, suggesting a >70% increase in population size since the initial reintroduction of 64 individuals in 2011–13. Conclusions Given the inherent difficulties in surveying mala, this approach would be valuable to ensure effective monitoring of the few remaining fenced and island mala populations to prevent further decline of this vulnerable species. Implications This is the first study to identify species-specific microsatellite markers for mala and use genetic-capture sampling with scat DNA to estimate the abundance of a mala population. The study provides an evaluation of a valuable species monitoring technique that can be applied to other rare, elusive, or cryptic threatened species.

Application of genetic and Spatially Explicit Capture-Recapture analyses to design adaptive feral cat control in a large inhabited island

Faunas of oceanic islands have a high proportion of endemic species which contribute to the uniqueness of island communities. Island species are particularly naïve and vulnerable to alien predators, such as cats (Felis catus). On large, inhabited islands, where the complete eradication of feral cat populations is not considered feasible, control represents the best management option to lower their detrimental effects on native fauna. The first objective of our study was to investigate population genetics of feral cats of Réunion Island. The second objective was to understand the space use of feral cats established near the breeding colonies of the two endemic and endangered seabirds of Réunion Island, the Barau’s Petrel (Pterodroma baraui) and the Mascarene Petrel (Pseudobulweria aterrima). We evaluated genetic diversity, population structure and gene flow amongst six groups of feral cats located at a maximum of 10 km from known petrel colonies. We also analysed the behaviour and space use of one of these feral cat groups using camera-trap data and Spatially Explicit Capture-Recapture (SECR) models. Genetic analyses revealed that feral cats were structured in three genetic clusters explained mostly by the island topography. Two clusters were observed at five sampled sites, suggesting high connectivity amongst these sites. The last cluster was found in only one site, suggesting high isolation. This site was a remote mountain area located in the vicinity of one of the main Barau’s Petrel colonies. The behavioural study was conducted on this isolated feral cat population. Mark recapture analysis suggested that feral cats were present at low density and had large home ranges, which is probably explained by reduced food availability. Finally, we make several recommendations for refining feral cat management programmes on inhabited islands.

Count me in: Leopard population density in an area of mixed land‐use, Eastern Cape, South Africa

Although the leopard (Panthera pardus) has the widest range of any felid in the world is designated as a vulnerable species, mainly because of human-induced conflict (Jacobson et al., 2016). Our study focuses on a population of leopards on privately owned, mixed-use farmland (Baviaanskloof Hartland – BH hereafter) which is adjacent to the Baviaanskloof Mega Reserve (BMR) in the Baviaanskloof UNESCO World Heritage Site of the Eastern Cape, South Africa. Given the unique make-up of the region, with sometimes conflicting management objectives, the status of leopards in the broader Baviaanskloof is of particular interest to a range of stakeholders. However, despite the need for management decisions to be based on reliable and regular population monitoring efforts (Elliot et al., 2020), the last formal assessment of the leopard population in the Baviaanskloof was performed in 2011/2012 but published 9 years later (Devens et al., 2018). The only other assessment of the status of leopards in the region was an unpublished Master's project (McManus, 2009). Here, we use photographic captures of leopards and a Spatially Explicit Capture Recapture (SECR) analytical framework in the mixed-use BH region of the Baviaanskloof to generate an up-to-date leopard population density estimate that can inform conservation management of the species in this important World Heritage Site. The Baviaanskloof World Heritage Site falls within the Cape Floristic Region, recognised for its extraordinary biodiversity (Myles, 2018). The landscape consists of a central valley and its tributary valleys and floodplains, dominated by steep, mountainous terrain. The area hosts an array of unique vegetation biomes including Albany Thicket, Fynbos, Succulent Karoo, Nama-Karoo, Grassland, Savannah and Forest (Skowno et al., 2021). The BMR is a roughly 500,000 ha area of protected land consisting of the 200,000 ha Baviaanskloof Nature Reserve (BNR), under the management of the Eastern Cape Parks and Tourism Agency (ECPTA), and several state-owned protected areas and privately owned lands under conservation stewardships (Rapeneau, 2015). The BH falls within the greater BMR and is situated west of the BNR (Figure 1). The BH is a cooperative of privately owned agricultural lands (Rapeneau, 2015). Specific camera locations with a minimum convex polygon area of 75 km2 were placed in the BH (Figure 1) in four of the Bavianskloof's major vegetation biomes: Fynbos, Albany Thicket, Savannah and Forest (Appendix 1). We deployed 28 un-baited motion-activated camera traps across the study area, from 28 January 2020 to 5 February 2020 and monitored them for 7 months. We placed one Cuddeback X-Change™ Color Model 1279 camera per site (Figure 2). Cameras were placed to be representative of the variations in vegetation type in the study area and spaced close enough together such that no individual leopard present could be missed (Figure 2). This camera spacing was aligned with known leopard home range sizes for the area (females = 106–130 km2; males = 85–498 km2) (McManus, 2009). Cameras were placed in areas where there were two or more signs of animal activity (i.e., tracks, scat or signs of foraging), and were positioned in such a way as to minimise false triggers, over-exposure, and interference from sun and vegetation. Cameras were visited every 30–45 days to download photos, replace batteries, ensure continued camera functionality and remove any obstructions that may have appeared, such as vegetation that had grown in front of the camera and obscured its view. Some traps were affected by camera or battery malfunctions, as well as interference from animals. These issues were corrected and recorded during camera checks. During checks, photos were downloaded directly from the cameras' SD cards, and then uploaded to PhotoGoFer (Rapid Imaging, Inc.), a photo data capture program that securely stores image files and associated data in a secure database. In PhotoGoFer, relevant data for each photo, such as species, sex, age, unique attributes and individual ID, were associated with the image file in a process called tagging. Individual leopards were identified by matching their unique spot patterns between photos. Photos were manipulated in Google Slides to align photos of spot patterns directly on top of one another and to confirm identifications. All leopard identifications were made by SB and cross-checked by TWP and MB. We generated leopard population density estimates independently for left-side and right-side captures in R (R Core Team, 2021) using the ‘secr’ package (Efford, 2021). We used the first 60 days of our sampling period for our estimates to address the assumption of a closed population (Karanth & Nichols, 1998). We defined our camera traps as ‘count’ detectors to account for their ability to capture an individual several times at the same occasion. We selected a buffer of 20 km and used a half-normal detection function following Devens et al. (2018). The addition of the 20 km buffer to our study area made our total effective area sampled 2643 km2. Of 13,819 photographic capture events over a period of 7 months and 5012 camera trap days, we captured 89 photos of leopards. We identified a total of six adult leopards from the left side and five adult leopards from the right side. During the initial 60-day period with 4740 camera trap days, four adult leopards were identified from the left side and the right-side photos respectively. From these data, we generated population density estimates of 0.74 leopards/100 km2 for left-side photos and 1.4 leopards/100 km2 for right-side, resulting in a mean estimate of 1.07 leopards/100 km2 (Table 1). Additionally, we captured photos of two reproductive females. The first was sighted with a single cub on 5 February 2020 and 21 February 2020. We positively identified the cub on 21 February 2020 sighting to be one of two cubs we frequently sighted together; therefore, we reasonably assume this leopard to be the mother of two cubs. The other reproductive female was sighted with a single cub on 5 March 2020. Our estimate of 1.07 leopards/100 km2 is comparable to the first population estimate generated in the region (McManus, 2009), but higher than the later assessment (Devens et al., 2018). McManus (2009) produced a population density estimate of 1.1 leopards/100 km2 and 1.3 leopards/100 km2 using camera traps and GPS data, respectively, for the entire Baviaanskloof. Devens et al. (2018), using the SPACECAP package, and an SECR framework, reported a much lower estimate of 0.24 leopards/100 km2. Differences in methodology between our study and Devens et al. (2018) may have resulted in this disparity. For example, Devens et al. (2018) generated their estimates using a 90-day sample period (Karanth & Nichols, 1998) and Bayesian SECR modelling. Fine scale differences in photo-capture success across different camera sites between the two studies is also possible. The disparity in estimates may, of course, also be because of changes in the leopard population in the intervening decade between the two studies. Indeed, we know leopard population densities and their estimates can fluctuate dramatically over time and/or depending on the field techniques and analytical framework used (Amin et al., 2022). Nevertheless, our estimate of 1.07 leopards/100 km2 likely reflects our study area being situated on non-protected farmland and its proximity to a large, protected area with abundant natural prey and one that provides a mosaic of suitable habitat types for leopards (cf. Amin et al., 2022). The density of leopards on farmlands has been found to be much lower than adjacent protected areas, and human–farmer conflict resulting in edge effects that may seriously weaken the effectiveness of reserves to conserve species like leopards (Balme et al., 2010; Marker & Dickman, 2005). Even in areas like the Phinda-Mkhuze Complex in KwaZulu-Natal, South Africa, which has shown consistently high leopard population densities averaging around 7 leopards/100 km2, leopards show much greater mortality rates near the borders of the protected area compared to the central areas (Balme et al., 2010). Although the BNR prohibits hunting or capturing of any animals on the reserve, leopards can move freely in and out of the reserve, increasing their potential to come into conflict with farmers (Balme et al., 2010). The status of leopards in the Baviaanskloof is of great economic and socio-political relevance to the farmers of the BH, who are under significant stress due to the reduced profitability of agriculture in an unfavourable economic environment and on lands degraded due to past unsustainable agricultural practices (Kerley et al., 1995). The majority of agricultural production in the Baviaanskloof is small livestock, for which farmers suffer losses due to depredation (Rapeneau, 2015). Actual or perceived threats of loss to carnivores has been shown to prompt farmers to resort to lethal protection of their livestock using traps, poison, hunting dogs, and often direct hunting and killing of perceived or potential offenders (Minnie et al., 2015). In addition, 67% of Baviaanskloof farmers harboured negative attitudes towards leopards, despite leopards only being responsible for 0.7% of livestock losses per year (Minnie et al., 2015). By killing leopards on their land, farmers not only threaten the leopard population but also their own livelihoods in an already unfavourable economic environment. Balme et al. (2010) recommend mitigating conflicts along reserve borders by implementing strategies on both sides of the borders, including systems for the regulation of problem animal control and community interventions to reduce human–leopard conflicts. We believe that such interventions should at least be attempted in the Baviaanskloof to reduce human–leopard conflict. Our leopard population estimate provides up-to-date insight on the status of leopards in an important mixed land-use part of the Baviaanskloof World Heritage Site. However, to be relevant and useful for all stakeholders in the region, it is vital that up-to-date leopard population monitoring be conducted, and suitable human–leopard conflict mitigation measures put in place. Funding for this project was provided by Furman University's Office of Undergraduate Research. Our research was performed in association with Furman Cougar Project and the Baviaanskloof Wildlife Project. We thank the Eastern Cape Parks and Tourism Agency, Rhodes University, the National Research Foundation, Living Lands, and the landowners of the Baviaanskloof Hartland for their support of this project. The authors confirm that they have no interests to declare. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Occupancy data improves parameter precision in spatial capture-recapture models.

Population size is one of the basic demographic parameters for species management and conservation. Among different estimation methods, spatially explicit capture–recapture (SCR) models allow the estimation of population density in a framework that has been greatly developed in recent years. The use of automated detection devices, such as camera traps, has impressively extended SCR studies for individually identifiable species. However, its application to unmarked/partially marked species remains challenging, and no specific method has been widely used. We fitted an SCR‐integrated model (SCR‐IM) to stone marten Martes foina data, a species for which only some individuals are individually recognizable by natural marks, and estimate population size based on integration of three submodels: (1) individual capture histories from live capture and transponder tagging; (2) detection/nondetection or “occupancy” data using camera traps in a bigger area to extend the geographic scope of capture–recapture data; and (3) telemetry data from a set of tagged individuals. We estimated a stone marten density of 0.352 (SD: 0.081) individuals/km2. We simulated four dilution scenarios of occupancy data to study the variation in the coefficient of variation in population size estimates. We also used simulations with similar characteristics as the stone marten case study, comparing the accuracy and precision obtained from SCR‐IM and SCR, to understand how submodels' integration affects the posterior distributions of estimated parameters. Based on our simulations, we found that population size estimates using SCR‐IM are more accurate and precise. In our stone marten case study, the SCR‐IM density estimation increased the precision by 37% when compared to the standard SCR model as regards to the coefficient of variation. This model has high potential to be used for species in which individual recognition by natural markings is not possible, therefore limiting the need to rely on invasive sampling procedures.

Efficient Discretization of Movement Kernels for Spatiotemporal Capture–Recapture

Spatially explicit capture–recapture (SECR) models treat detection probability as a function of the distance between each animal and its notional activity centre. Open-population variants of these models (open SECR) are increasingly used to estimate the vital rates (survival and recruitment) of spatial populations subject to turnover between sampling times. If activity centres also move between sampling times then modelling the movement can reduce bias in estimates of vital rates. The usual movement model in open SECR is a random walk with step length governed by a probability kernel. Space is discretized in open SECR for computational convenience, and in some implementations this includes truncation of the probability kernel. Computations for the movement submodel are nevertheless very time-consuming owing to the repeated convolution steps and the need to manage boundary effects. A novel ‘sparse’ discretized kernel is proposed that greatly reduces fitting time. The sparse kernel was tested by simulation and applied to two datasets. Differences between models fitted using the sparse and full kernels were minor and unlikely to matter in practice. The sparse kernel extends the practical limits of the movement modelling in open SECR to greater dispersal distances and greater spatial resolution. Supplementary materials accompanying this paper appear online.

Assessing small-mammal trapping design using spatially explicit capture recapture (SECR) modeling on long-term monitoring data.

Few studies have evaluated the optimal sampling design for tracking small mammal population trends, especially for rare or difficult to detect species. Spatially explicit capture-recapture (SECR) models present an advancement over non-spatial models by accounting for individual movement when estimating density. The salt marsh harvest mouse (SMHM; Reithrodontomys raviventris) is a federal and California state listed endangered species endemic to the San Francisco Bay-Delta estuary, California, USA; where a population in a subembayment has been continually monitored over an 18-year period using mark-recapture methods. We analyzed capture data within a SECR modeling framework that allowed us to account for differences in detection and movement between sexes. We compared the full dataset to subsampling scenarios to evaluate how the grid size (area) of the trap design, trap density (spacing), and number of consecutive trapping occasions (duration) influenced density estimates. To validate the subsampling methods, we ran Monte Carlo simulations based on the true parameter estimates for each specific year. We found that reducing the area of the trapping design by more than 36% resulted in the inability of the SECR model to replicate density estimates within the SE of the original density estimates. However, when trapping occasions were reduced from 4 to 3-nights the density estimates were indistinguishable from the full dataset. Furthermore, reducing trap density by 50% also resulted in density estimates comparable to the full dataset and was a substantially better model than reducing the trap area by 50%. Overall, our results indicated that moderate reductions in the number of trapping occasions or trap density could yield similar density estimates when using a SECR approach. This approach allows the optimization of field trapping efforts and designs by reducing field efforts while maintaining the same population estimate compared to the full dataset. Using a SECR approach may help other wildlife programs identify sampling efficiencies without sacrificing data integrity for long term monitoring of population densities.

The population density and trap-revealed home range of short-eared possums (

Investigating how the population density of a species changes over time is an integral step in determining whether that species is stable or needs assistance from conservation managers. The short-eared possum (Trichosurus caninus) is a species that has been poorly studied with only one previous population density estimate. Short-eared possums were live-trapped between August 2020 and January 2021 in the Northern Tablelands, New South Wales, to estimate their current density using a spatially explicit capture–recapture (SECR) model. The average density of short-eared possums was 0.46 possums/ha (95% CI: 0.32–0.66) in temperate rainforest and 0.13 possums/ha (95% CI: 0.06–0.28) in wet sclerophyll forest. No individuals were caught in dry sclerophyll forest. Trap-based home ranges were estimated to be 12.5 ha (95% CI: 8–19) for males and 5.5 ha (95% CI: 3–11) for females. This study provides a reference for determining trends in short-eared possum population density within the Northern Tablelands in the future. Adequate conservation of temperate rainforest and wet sclerophyll forest habitat is important to the conservation of the species.

Comparison of methods for estimating density and population trends for low-density Asian bears

Populations of bears in Asia are vulnerable to extinction and effective monitoring is critical to measure and direct conservation efforts. Population abundance (local density) or growth (λ) are the most sensitive metrics to change. We discuss the value in implementing spatially explicit capture-recapture (SCR), the current gold standard for density estimation, and open population SCR (OPSCR) to monitor changes in density over time. We provide guidance for designing studies to provide estimates with sufficient power to detect changes. Because of the wide availability of camera traps and interest in their use, we consider six density estimation methods and their extensions developed for use with camera traps, with specific consideration of assumptions and applications for monitoring Asian bears. We conducted a power analysis to calculate the precision in estimates needed to detect changes in populations with reference to IUCN Red List criteria. We performed a systematic review of empirical studies implementing camera trap abundance estimation methods and considered sample sizes, effort, and model assumptions required to achieve adequate precision for population monitoring. We found SCR and OPSCR, reliant on “marked” individuals, are currently the only methods with enough power to reliably detect even moderate to major (20–80%) declines. Camera trap methods with unmarked individuals rarely achieved precision sufficient to detect even large declines (80–90%), although with some exceptions (e.g., situations with moderate population densities, large number of sampling sites, or inclusion of ancillary local telemetry data. We describe additional estimation options including line transects, direct observations, monitoring age-specific survival and reproductive rates, and hybrid/integrated methodologies that may have potential to work for some Asian bear populations. We conclude monitoring changes in abundance or density is possible for most Asian bear populations but will require collaboration among researchers over broad spatial extents and extensive financial investment to overcome biological and logistical constraints. We strongly encourage practitioners to consider study design and sampling effort required to meet objectives by conducting simulations, power analyses, and assumption checks prior to implementing monitoring efforts, and reporting standardized dispersion measures such as coefficients of variation to allow for assessment of precision. Our guidance is relevant to other low-density and wide-ranging species.

Monitoring sitatunga (Tragelaphus spekii) populations using camera traps

AbstractPopulation densities are an important consideration for wildlife management. For hunted species, estimates of population size are essential for establishing quotas. There are limited data regarding sitatunga (Tragelaphus spekii) abundance, and few studies have used camera traps for estimating population densities. We estimated sitatunga density in the Mayanja River area of central Uganda using the time in front of the camera (TIFC) method, which yields population density estimates using camera‐trap data without having to identify each individual. Density estimates from the TIFC method averaged 11.1 sitatunga/km2, similar to estimates from a spatially explicit capture–recapture (SECR) study, which requires perfect identification of individuals. These results suggest that TIFC methods accurately estimate population densities and might be an alternative for species where individual identity cannot be determined, or in habitats where typically identifiable species might not be accurately identified to meet the requirements of SECR. In addition, the resulting density estimates can influence management decisions and quotas for hunting.

A noninvasive and integrative approach for improving density and abundance estimates of moose

AbstractAcquiring demographic data for moose (Alces alces) can be difficult because they are solitary in nature, they prefer densely vegetated and mountainous habitats, and they often occur at low density. Such data, however, are essential for long‐term population monitoring, evaluating management practices, and effective conservation. Winter aerial surveys are the standard method for estimating moose population parameters, but they can be logistically challenging, expensive, and subject to sightability correction, which necessitates the capture of study animals for initial model development. Herein, we demonstrate a noninvasive alternative approach for estimating population parameters of moose in northern Yellowstone National Park, where aerial surveys were attempted but proved ineffective. We determined individual moose genotype and sex using microsatellite polymerase chain reaction amplification of DNA extracted from fecal pellets, integrated ancillary pellet sample data (i.e., metadata) in genotype analysis to aid in the identification of matching genotypes, and used spatially explicit capture‐recapture (SECR) modeling to estimate sex‐specific density and abundance. We collected 616 samples over 3 consecutive winters (Dec 2013–Apr 2016) and within 2 sampling occasions each winter. We recorded 514 captures of 142 individual moose (69 males, 73 females). Overall density ranged between 0.062 moose/km2 and 0.076 moose/km2 and averaged 0.034/km2 for females and 0.033/km2 for males. Abundance estimates were 150 moose in 2013 (female = 76, 95% CI = 55–105; male = 74, 95% CI = 54–103), 186 in 2014 (female = 95, 95% CI = 63–142; male = 91, 95% CI = 60–138), and 160 in 2015 (female = 79, 95% CI = 58–108; male = 81, 95% CI = 59–110). Average population sex ratio was 0.99 males/female. We demonstrate that SECR analysis of fecal DNA genotypes, using metadata in genotype analysis to help identify matching moose genotypes, is a promising alternative method for estimating sex‐specific density and abundance of a low‐density moose population in a mountainous and forested landscape.

Using camera traps to estimate density of snowshoe hare (Lepus americanus): a keystone boreal forest herbivore

AbstractBoreal ecosystems are experiencing extensive changes because of anthropogenic stressors such as climate change. Information on density of species at multiple sites is vital to understand and manage the impact of these changing conditions on boreal forest communities. Yet, for most boreal forest species, including the vast majority of mammals, obtaining reliable estimates of density is exceedingly difficult. Recently developed methods for the estimation of densities of unmarked animals from camera-trapping data could help to overcome this hurdle, but have not yet been empirically validated in many ecosystems. Here, we assess the ability of camera traps to estimate density of snowshoe hare (Lepus americanus) using three different models: the random encounter model (REM), the random encounter and staying time (REST) model, and the time-to-event (TTE) model. We additionally evaluate the relationship between hare density and two simple indices based on camera detection rate and pellet counts. Across 13 sites in North Central Washington, United States, we compared live-trapping spatially explicit capture–recapture (SECR) estimates of density to the three camera-based density models and the two indices. We found that the camera-based models, in particular the REM and REST models, performed well in estimating densities consistent with the live-trapping data, with an average difference in density from SECR-based estimates of only 0.12 and 0.13 hares/ha, respectively. Both indices also had strong predictive relationships with hare density. Our results show that, owing to their noninvasive nature and relative ease of application, camera-based methods could be used to obtain hare density estimates at much larger spatiotemporal scales than have been applied to date. Given the keystone role of hare in boreal ecosystems, and emerging evidence of hare range retraction, the ability to estimate densities across many sites is a key tool for hare conservation and management. Moreover, our results are highly encouraging for the application of camera-based methods to obtain density estimates on a wide variety of boreal forest species, though additional validation will be necessary.

Population densities and home range of the vulnerable Pyrenean brook newt in its core aquatic habitat

Abstract Accurate estimations of population size and space-use are critical issues, for assessing population trends and extinction risk. In this study, we applied spatially explicit capture-recapture (SECR) modelling customized for linear habitats to a three-year Capture-Mark-Recapture (CMR) monitoring of three populations of stream-dwelling Pyrenean brook newt (Calotriton asper), an endemic species of the Pyrenean mountain range. Our aim was to explore (1) spatial and temporal variation in population densities, home range size and individual detection probability during breeding season between populations and years, and (2) the influence of water temperature and flow on activity of newts. Estimates of population density range from 3044 to 4641 individuals/km of stream across the three years of the study. Linear home range along the stream was estimated to be 13.31 m from the home range centre on average over the three years. All detection probability estimates were similar, both between years and between sites, with a mean probability of 0.09, except in Fougax in 2020 where detection probability was 0.03. Activity of the Pyrenean brook newt was influenced by water temperature and flow, but these factors acted at different time scales. Overall, population densities are locally high in both populations but home range of the Pyrenean brook newt during breeding season is very small indicating a strong site attachment. The importance of water temperature and flow on activity emphasizes the sensitivity of the species to these factors that are forecasted to change in coming decades.

Leopard and spotted hyena densities in the Lake Mburo National Park, southwestern Uganda.

Robust measures of animal densities are necessary for effective wildlife management. Leopards (Panthera pardus) and spotted hyenas (Crocuta Crocuta) are higher order predators that are data deficient across much of their East African range and in Uganda, excepting for one peer-reviewed study on hyenas, there are presently no credible population estimates for these species. A lack of information on the population status and even baseline densities of these species has ramifications as leopards are drawcards for the photo-tourism industry, and along with hyenas are often responsible for livestock depredations from pastoralist communities. Leopards are also sometimes hunted for sport. Establishing baseline density estimates for these species is urgently needed not only for population monitoring purposes, but in the design of sustainable management offtakes, and in assessing certain conservation interventions like financial compensation for livestock depredation. Accordingly, we ran a single-season survey of these carnivores in the Lake Mburo National Park of south-western Uganda using 60 remote camera traps distributed in a paired format at 30 locations. We analysed hyena and leopard detections under a Bayesian spatially explicit capture-recapture (SECR) modelling framework to estimate their densities. This small national park (370 km2) is surrounded by Bahima pastoralist communities with high densities of cattle on the park edge (with regular park incursions). Leopard densities were estimated at 6.31 individuals/100 km2 (posterior SD = 1.47, 95% CI [3.75–9.20]), and spotted hyena densities were 10.99 individuals/100 km2, but with wide confidence intervals (posterior SD = 3.35, 95% CI [5.63–17.37]). Leopard and spotted hyena abundance within the boundaries of the national park were 24.87 (posterior SD 7.78) and 39.07 individuals (posterior = SD 13.51) respectively. Leopard densities were on the middle end of SECR studies published in the peer-reviewed literature over the last 5 years while spotted hyena densities were some of the first reported in the literature using SECR, and similar to a study in Botswana which reported 11.80 spotted hyenas/100 km2. Densities were not noticeably lower at the park edge, and in the southwest of our study site, despite repeated cattle incursions into these areas. We postulate that the relatively high densities of both species in the region could be owed to impala Aepyceros melampus densities ranging from 16.6–25.6 impala/km2. Another, potential explanatory variable (albeit a speculative one) is the absence of interspecific competition from African lions (Panthera leo), which became functionally extinct (there is only one male lion present) in the park nearly two decades ago. This study provides the first robust population estimate of these species anywhere in Uganda and suggests leopards and spotted hyenas continue to persist in the highly modified landscape of Lake Mburo National Park.