Camera-Trap Monitoring Report for Lux national monitoring (valerian) 2023-2025, Luxembourg and Germany
Ecological insights from camera-trap data for wildlife monitoring
Cyrille Charles, Alain Dohet, Martin Heyeres, Viola Huck, Lionel L, Youri Martin, Xavier Mestdagh, Sarah Vray, and Claudio Walzberg
Luxembourg Institute of Science and Technology
June 19, 2026
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 Lux national monitoring (valerian) 2023-2025, Luxembourg and Germany. The aim is to estimate and monitor wild mammals 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 Lux national monitoring (valerian) 2023-2025, located in Luxembourg and Germany. The site is geographically defined by the coordinates 49.500°–50.100°N and 5.800°–6.500°E and covers an area of approximately 2205.31 km². The site supports a diverse range of wildlife, with approximately 70 species recorded. The most frequently observed species include Procyon lotor, Meles meles, Capreolus capreolus, Felis silvestris, and Vulpes vulpes.
2.2 Sampling
The sampling design at this site is targeted. In a targeted design, cameras are placed non-randomly at locations expected to maximize detections of particular species, such as trails or water sources. Camera surveys at this site were conducted using the following camera model: Reconyx-HyperFire 2 Professional HP2W. A mixture of baited and unbaited camera deployments was used in this survey. Multiple camera heights were used at this site, ranging from 0.3 to 2.9 m above the ground. Media were captured using time-lapse and 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 |
|---|---|---|---|---|
| 2023 | 22 | 2023-01-24 - 2023-06-30 | 24 January - 23 February | June 30 - June 30 |
| 2024 | 27 | 2024-01-29 - 2024-07-18 | 29 January - 15 February | July 18 - July 18 |
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.
2023
2024
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:
- 2023: 93 cameras broke down during the study period, the cameras operated for an average of 131 days (range: 12 - 149 days), and the total sampling effort was 12724 days, equivalent to 34.86 years.
- 2024: 91 cameras broke down during the study period, the cameras operated for an average of 134 days (range: 113 - 167 days), and the total sampling effort was 12832 days, equivalent to 35.16 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 2023 and 2024, the survey recorded 43 wild species and 3 domestic species. Human presence was also detected on 315 occasions, most likely associated with field visits for camera installation or maintenance.
The report focuses on wild mammals 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.
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 wild mammals recorded in at least 25 capture events during the study period in Lux national monitoring (valerian) 2023-2025. 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 wild mammals 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 wild mammals group across survey years.
Number of Locations
Figure 5. Trends in the number of camera-trap locations where each species was recorded in the wild mammals 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).
Capreolus capreolus
Figure.6: Estimated daily activity pattern for Capreolus capreolus, 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.
Cervus elaphus
Figure.7: Estimated daily activity pattern for Cervus elaphus, 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.
Felis silvestris
Figure.8: Estimated daily activity pattern for Felis silvestris, 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.
Lepus europaeus
Figure.9: Estimated daily activity pattern for Lepus europaeus, 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.
Martes foina
Figure.10: Estimated daily activity pattern for Martes foina, 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.
Martes martes
Figure.11: Estimated daily activity pattern for Martes martes, 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.
Meles meles
Figure.12: Estimated daily activity pattern for Meles meles, 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.
Mustela putorius
Figure.13: Estimated daily activity pattern for Mustela putorius, 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.
Procyon lotor
Figure.14: Estimated daily activity pattern for Procyon lotor, 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.
Sciurus vulgaris
Figure.15: Estimated daily activity pattern for Sciurus vulgaris, 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.
Sus scrofa
Figure.16: Estimated daily activity pattern for Sus scrofa, 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.17: 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.
2023
Figure.18: Species richness based on the species focus group in this report, across camera-trap locations for survey year 2023.
2024
Figure.19: Species richness based on the species focus group in this report, across camera-trap locations for survey year 2024.
Total
Figure.20: Species richness based on the species focus group in this report, across camera-trap locations for the cumulative dataset across all years.
3.5 Species Co-occurrence
Species co-occurrence analysis was used to assess pairwise associations among species recorded across camera-trap locations. Relationships were summarized in a correlation matrix, where positive values indicate species pairs that tended to co-occur more frequently across sites and negative values indicate lower co-occurrence. Such patterns can provide insight into shared space use or spatial segregation among species, although they should be interpreted cautiously because co-occurrence alone does not imply direct interaction. (Gotelli, 2000).
2023
Figure.21: Species co-occurrence patterns for the selected species focus group in this report, shown for survey year 2023.
2024
Figure.22: Species co-occurrence patterns for the selected species focus group in this report, shown for survey year 2024.
Total
Figure.23: Species co-occurrence patterns for the selected species focus group in this report, shown for the cumulative dataset across all years.
3.6 Spatial Density
Spatial density maps derived from point pattern analysis illustrate the spatial distribution of species detections across the study area. These visualizations highlight variation in detection intensity and help identify areas with relatively high or low concentrations of records. The following maps are presented for each species, either by survey year or across the full monitoring period. Warmer colors represent higher estimated detection density, whereas cooler colors indicate lower density. Together, these maps support comparison of spatial detection patterns among species and across survey years.
2023
Capreolus capreolus
Figure.24: Spatial point pattern density for Capreolus capreolus based on the species focus group in this report, shown for survey year 2023.
Cervus elaphus
Figure.25: Spatial point pattern density for Cervus elaphus based on the species focus group in this report, shown for survey year 2023.
Felis silvestris
Figure.26: Spatial point pattern density for Felis silvestris based on the species focus group in this report, shown for survey year 2023.
Lepus europaeus
Figure.27: Spatial point pattern density for Lepus europaeus based on the species focus group in this report, shown for survey year 2023.
Martes foina
Figure.28: Spatial point pattern density for Martes foina based on the species focus group in this report, shown for survey year 2023.
Martes martes
Figure.29: Spatial point pattern density for Martes martes based on the species focus group in this report, shown for survey year 2023.
Meles meles
Figure.30: Spatial point pattern density for Meles meles based on the species focus group in this report, shown for survey year 2023.
Mustela putorius
Figure.31: Spatial point pattern density for Mustela putorius based on the species focus group in this report, shown for survey year 2023.
Procyon lotor
Figure.32: Spatial point pattern density for Procyon lotor based on the species focus group in this report, shown for survey year 2023.
Sciurus vulgaris
Figure.33: Spatial point pattern density for Sciurus vulgaris based on the species focus group in this report, shown for survey year 2023.
Sus scrofa
Figure.34: Spatial point pattern density for Sus scrofa based on the species focus group in this report, shown for survey year 2023.
Vulpes vulpes
Figure.35: Spatial point pattern density for Vulpes vulpes based on the species focus group in this report, shown for survey year 2023.
2024
Capreolus capreolus
Figure.36: Spatial point pattern density for Capreolus capreolus based on the species focus group in this report, shown for survey year 2024.
Cervus elaphus
Figure.37: Spatial point pattern density for Cervus elaphus based on the species focus group in this report, shown for survey year 2024.
Felis silvestris
Figure.38: Spatial point pattern density for Felis silvestris based on the species focus group in this report, shown for survey year 2024.
Lepus europaeus
Figure.39: Spatial point pattern density for Lepus europaeus based on the species focus group in this report, shown for survey year 2024.
Martes foina
Figure.40: Spatial point pattern density for Martes foina based on the species focus group in this report, shown for survey year 2024.
Martes martes
Figure.41: Spatial point pattern density for Martes martes based on the species focus group in this report, shown for survey year 2024.
Meles meles
Figure.42: Spatial point pattern density for Meles meles based on the species focus group in this report, shown for survey year 2024.
Mustela putorius
Figure.43: Spatial point pattern density for Mustela putorius based on the species focus group in this report, shown for survey year 2024.
Procyon lotor
Figure.44: Spatial point pattern density for Procyon lotor based on the species focus group in this report, shown for survey year 2024.