Camera-Trap Monitoring Report for Data from three camera trapping pilots in the Amsterdam Water Supply Dunes of the Netherlands, Netherlands

Ecological insights from camera-trap data for wildlife monitoring

1 Introduction

Terrestrial wildlife populations play a fundamental role in ecosystem structure and function. Across taxonomic groups, wildlife contributes to key ecological processes such as nutrient cycling, energy flow, seed dispersal, trophic interactions, and habitat modification, thereby supporting biodiversity and strengthening ecosystem resilience (Dirzo et al., 2014). However, these populations are increasingly exposed to anthropogenic pressures, especially habitat alteration and land-use change, which may affect their distribution and abundance across space and time (Newbold et al., 2015). Understanding and monitoring these responses is therefore crucial for conservation planning, effective protected-area management, and evidence-based decision-making (Coad et al., 2015).

To better understand and monitor these responses, camera trapping has become a cost-effective, reliable, and non-invasive method for monitoring wildlife populations. By automatically capturing images or videos when animals move through the detection zone of infrared sensors, camera traps generate robust information on species occurrence, habitat use, activity patterns, and relative abundance while minimizing observer-related disturbance. Consequently, camera-trap data have become an important source of biodiversity information and are widely applied in ecological research, biodiversity monitoring, and adaptive conservation management (Kays et al., 2009). These data are valuable to both scientists and practitioners, as they support ecological analyses, advance understanding of species ecology, and provide critical insights into population status and trends (O’Connell et al., 2011).

This report presents the results of camera-trap surveys conducted in Data from three camera trapping pilots in the Amsterdam Water Supply Dunes of the Netherlands, Netherlands. The aim is to estimate and monitor red fox and european rabbit over time and across deployments, and to provide insight into wildlife status at this site, including behaviour and changes in spatial and temporal patterns.

2 Methods

2.1 Study Area

The study was conducted in Data from three camera trapping pilots in the Amsterdam Water Supply Dunes of the Netherlands, located in Netherlands. The site is geographically defined by the coordinates 52.307°–52.323°N and 4.489°–4.503°E and covers an area of approximately 0.583 km². The site supports a diverse range of wildlife, with approximately 16 species recorded. The most frequently observed species include Oryctolagus cuniculus, Vulpes vulpes, Corvus corone, Erinaceus europaeus, and Pica pica.

2.2 Sampling

The sampling design at this site is opportunistic. In an opportunistic design, cameras are deployed in an ad hoc way, often without a predefined sampling frame, which can yield useful records but is generally not suitable for rigorous population- or community-level inference. Camera surveys at this site were conducted using the following camera model: SnyperCommander4GWireless-SY4.0CG-Rwildlifecamera. No bait was used during camera deployment. Cameras were mounted at approximately 0.3 m above the ground. Media were captured using activity detection. This project was not specifically designed to identify individual animals, but rather to support broader wildlife monitoring. More detailed information on camera-trap deployments by survey year is provided in the summarized information table below.

📷 Camera Deployment Summary
Table 1: Details of camera deployments per year
Year	Camera Traps	Deployment Period	Setup Period	Pickup Period
2021	3	2021-08-13 - 2022-12-31	13 August - 14 August	December 31 - December 31

2.3 Camera Locations

The maps below display the locations of camera traps deployed during different years. Use the tabs above to explore data for each sampling year. The last tab shows the study area with all camera locations in this site. Locations are color-coded by habitat type. Click on the points for additional deployment information.

2021

Total

Figure.1: Interactive map of camera trap locations. Click on a marker to view the corresponding location details and additional metadata.

2.4 Sampling Efforts

Sampling effort refers to the total number of camera trap operational days, accounting for both the number of active cameras and their duration of deployment. It provides critical context for interpreting species detection rates and comparing data across years. A summary of yearly sampling effort is provided below, including camera breakdowns, average operational days, and total effort per year:

• 2021: No cameras broke down during the study period, the cameras operated for an average of 505 days (range: 504 - 505 days), and the total sampling effort was 1515 days, equivalent to 4.15 years.

The following figure provides detailed information on sampling effort over time, and the slider can be used to focus on specific time periods.

Figure.2: Number of active camera traps per survey year. Adjust the slider to focus on specific time periods.

2.5 Image Processing

To support standardized, reproducible, and interoperable workflows, camera-trap data are increasingly managed using dedicated platforms such as Agouti (Casaer et al., 2019) and structured according to community standards such as CamtrapDP (Bubnicki et al., 2024). Agouti supports sequence-based annotation, AI-assisted classification, secure archiving, and standardized data export. In this workflow, images recorded within 120 seconds of one another were grouped into sequences representing putative single events and annotated using a combination of AI-assisted classification and manual review; a subset was subsequently validated by experienced observers. Where relevant for density estimation, calibration images and species-specific movement paths were also annotated and analysed using the photogrammetric tools in the R package camtrapDensity (Rowcliffe, 2014).

The images used in this report were processed and archived in Agouti , and the data were structured according to the CamtrapDP standard.

2.6 Data Processing

The input data were provided in Camtrap-DP standard format and pre-processed in camtrapReport, including cleaning, harmonisation, organisation, summarisation, and extraction of key attributes. Records were checked for completeness and internal consistency before being summarised into the descriptive statistics and ecological metrics presented in this report.

In the processing stage, ecological analysis modules operated on these standardised data to generate harmonised outputs, including tables, figures, text, and summary statistics. The report includes a broad range of ecological analyses, such as sampling effort, species richness, activity patterns, and species co-occurrence. These analyses draw on established ecological methods implemented in dedicated R packages, including activity and camtrapDensity.

In the post-processing stage, outputs generated by the selected modules were assembled into structured report sections and rendered into a coherent, article-style ecological report. Because sections can be customised by the user, the report structure remains flexible and can be adapted to different study objectives. This workflow provides a reproducible and extensible framework for transforming raw camera-trap data into standardised ecological insights.

3 Results

3.1 Captures

This section presents an overview of the camera-trap capture data collected during the study period. Across 2021, the survey recorded 1 wild species and 1 domestic species.The report focuses on red fox and european rabbit recorded in at least 25 capture events during the study period. The following table summarizes these frequently detected species and provides the main capture-based metrics for each species, including capture events, Relative Abundance Index (RAI; captures per 100 trap-days), number of capture locations, and total photos.

🐾 Table 2. Summary of Frequently Detected red_fox_and_european_rabbit
Species with ≥ 25 capture events
Common Name	Scientific Name	Captures	RAI	Capture Locations	Total Photos
European rabbit	Oryctolagus cuniculus	1893	124.99	3	11435
red fox	Vulpes vulpes	182	12.02	3	945

3.2 Capture-Based and Abundance Trends

Camera-trap capture data do not directly measure true population size, but they can provide useful indicators of temporal variation in relative abundance and species detectability, provided that survey effort and detection conditions remain reasonably comparable among years.

The figures below summarize capture-based trends for frequently detected red fox and european rabbit recorded in at least 25 capture events during the study period in Data from three camera trapping pilots in the Amsterdam Water Supply Dunes of the Netherlands. The tabs present complementary metrics, including the number of captures, capture rate (captures per 100 trap-days), number of capture locations, and, where available, REM-based density estimates. Together, these metrics support comparison of temporal patterns in species records across survey years.

Number of Captures

Figure 3. Trends in camera-trap detections over the survey years for each study species in the red fox and european rabbit group, based on the number of captures.

Capture Rate

Figure 4. Capture rates per 100 trap-days are shown on a log10 scale for each study species in the red fox and european rabbit group across survey years.

Number of Locations

Figure 5. Trends in the number of camera-trap locations where each species was recorded in the red fox and european rabbit group across survey years.

REM-Based Density

REM-based density estimates are not available for the selected species.

3.3 Activity Patterns

Activity patterns describe the temporal distribution of animal activity across the 24-hour day and are typically inferred from the timing of camera-trap detections. These patterns help identify whether species are primarily diurnal, nocturnal, or crepuscular, and can provide insight into daily behavior, ecological roles, and potential responses to environmental factors or human disturbance. The figures below present estimated daily activity curves for each study species, based on camera-trap detections collected across the monitoring period and derived using functionality from the activity R package (Rowcliffe, 2016).

Oryctolagus cuniculus

Figure.6: Estimated daily activity pattern for Oryctolagus cuniculus, aggregated across all sampling rounds. Dashed lines indicate the average sunrise (yellow) and sunset (gray) times across survey years. The solid line shows the fitted activity model, and the shaded region shows the 95% CI.

Vulpes vulpes

Figure.7: Estimated daily activity pattern for Vulpes vulpes, aggregated across all sampling rounds. Dashed lines indicate the average sunrise (yellow) and sunset (gray) times across survey years. The solid line shows the fitted activity model, and the shaded region shows the 95% CI.

3.4 Richness

Species richness refers to the number of unique species recorded at each camera-trap location. It provides a useful indicator of local biodiversity and helps assess how effectively the survey captured variation in the wildlife community across the study area. The figure below shows the spatial distribution of species richness, where each point represents a camera-trap location. Circle size reflects the number of unique species detected at that site, and color indicates richness intensity, with warmer colors representing higher richness values. By selecting a point, users can view additional information on species composition, richness count, and habitat type, if available. These features help identify biodiversity hotspots and potential detection gaps across the study area.

2021

Figure.8: Species richness based on the species focus group in this report, across camera-trap locations for survey year 2021.

Total

Figure.9: Species richness based on the species focus group in this report, across camera-trap locations for the cumulative dataset across all years.

3.5 Habitat Preferences

Habitat preferences were assessed by comparing proportional capture rates across habitat types for each species. Capture rate, expressed as detections relative to sampling effort, serves here as an indicator of habitat use and relative habitat association rather than a direct estimate of habitat selection. Differences in the distribution of capture rates among habitat types therefore highlight habitats in which species were more or less frequently recorded. Together, these patterns provide insight into species-specific habitat use within the study area and help identify broad differences in habitat association among the focal species.

Figure.10: Capture rate distribution across habitat types based on the species focus group. Each bar represents a species, with sections indicating the proportion of each habitat type.

4 Acknowledgements

This report was generated using the camtrapReport R package, developed by Elham Ebrahimi at Wageningen University & Research and Utrecht University, the Netherlands. The development of camtrapReport was supported by Biodiversa+ through the Big Picture project. We also gratefully acknowledge the European Observatory of Wildlife network for its contribution to package testing. Users are kindly requested to cite the package when using camtrapReport or publishing results derived from it.

Appendix

The following images showcase a selection of 2 species: Oryctolagus cuniculus, Vulpes vulpes, captured at different camera locations within this study site.

References

• Bubnicki, J. W. et al. (2024). Camtrap DP: an open standard for the FAIR exchange and archiving of camera trap data. Remote Sensing in Ecology and Conservation, 10, 283–295. https://doi.org/10.1002/rse2.374
• Casaer, J. et al. (2019). Agouti: A platform for processing and archiving of camera trap images. Biodiversity Information Science and Standards, 3, e46690. https://doi.org/10.3897/biss.3.46690
• Coad, L. et al. (2015). Measuring impact of protected area management interventions: current and future use of the Global Database of Protected Area Management Effectiveness. Philosophical Transactions of the Royal Society B: Biological Sciences, 370. https://doi.org/10.1098/rstb.2014.0281
• Dirzo, R. et al. (2014). Defaunation in the Anthropocene. Science, 345, 401–406. https://doi.org/10.1126/science.1251817
• Gotelli, N. J. (2000). Null model analysis of species co-occurrence patterns. Ecology, 81, 2606–2621. https://doi.org/10.1890/0012-9658(2000)081[2606:NMAOSC]2.0.CO;2
• Kays, R. et al. (2009). Camera traps as sensor networks for monitoring animal communities. In 2009 IEEE 34th Conference on Local Computer Networks, 811–818. https://doi.org/10.1109/LCN.2009.5355046
• Newbold, T. et al. (2015). Global effects of land use on local terrestrial biodiversity. Nature, 520, 45–50. https://doi.org/10.1038/nature14324
• O’Connell, A. F. et al. (2011). Camera traps in animal ecology: methods and analyses. Springer. https://doi.org/10.1007/978-4-431-99495-4
• Palencia, P. et al. (2022). Random encounter model is a reliable method for estimating population density of multiple species using camera traps. Remote Sensing in Ecology and Conservation, 8, 670–682. https://doi.org/10.1002/rse2.269
• Rowcliffe, J. M. et al. (2008). Estimating animal density using camera traps without the need for individual recognition. Journal of Applied Ecology, 1228–1236. https://doi.org/10.1111/2041-210X.12278
• Rowcliffe, M. & Rowcliffe, M. M. (2016). Package ‘activity’. Animal activity statistics R package version, 1(10.32614). https://doi.org/10.32614/CRAN.package.activity