Background

Loss of terrestrial vertebrates — defaunation — is occurring at an unprecedented rate. But the decline of animal species is not uniform. Variation in species sensitivity translates directly to their probability of decline and endangerment. Wildlife management requires efficient, easy-to-use methods to assess species occupancy and abundance over space and time with high accuracy. 

Technologies like camera traps are more effective than traditional methods because they are noninvasive, record data over long periods and capture a wide range of species. Similarly, acoustic recorders can record cryptic and arboreal species that camera traps fail to capture. Although rarely combined, pairing both methods alongside other new technologies like terrestrial LiDAR provides a more comprehensive biodiversity picture than any one method alone.

Project Description

The scientific goal of this project is to measure wildlife habitat and occupancy/abundance. In the process, the team will test, for the first time, the ability of combining data from camera traps, acoustic recorders and terrestrial LiDAR to improve estimates of species occupancy. The team will also assess species’ habitat needs and threats by combining terrestrial LiDAR and animal data. The two areas surveyed will be Duke Forest, North Carolina, and the tropical forest in Ivindo National Park, Gabon. This will allowing team members to compare the efficacy of the technologies, apprehend the ecological complexity of temperate and tropical systems, and recognize the different ethical considerations that arise when performing research in different locations.

The team will begin by developing a survey design to monitor species in three divisions of Duke Forest and employ terrestrial LiDAR scans and camera trap animal observations. The same methods will be used in Ivindo National Park by subteams for each technology — camera traps, acoustic recorders and terrestrial LiDAR. 

After the data gathering is complete, the subteams will use various computer software to process their findings. Finally, team members will analyze the data, estimating species occupancies using Bayesian models.

Anticipated Outputs

Long-term study design and database; data on species occupancy; minimum of two peer-reviewed papers; curriculum for Wildlife Camera Trap Analysis course

Student Opportunities

Ideally, this project team will include 6 graduate students and 3 undergraduate students with diverse perspectives and majors. Students with interests in environmental science, ecology, computer science, engineering and statistics would all benefit from and contribute to the project.

Students will have opportunities to collect data with technologies such as camera traps, acoustic monitors and terrestrial LiDAR and improve cross-cultural communication skills. They will interact with Gabonese graduate students and participate in workshops on decolonizing ecology, inclusive mentoring, sexual harassment in field science and intercultural mindfulness. Three students will also have the opportunity to travel internationally to Gabon and practice French. Team members will conduct research, contributing to all steps from study design to data analysis to writing of peer-reviewed publications. Graduate students will gain and practice project management skills. 

The project includes two types of optional summer work in 2023. Three students will travel to Gabon for 10 weeks to collect data in Ivindo National Park. These students will work full-time on data collection. Other students will have the option of starting to process the data collected in Duke Forest during the spring. 

Caroline Rowley will serve as project manager.

See the related Data+ project for Summer 2023; there is a separate application process for students who are interested in this optional component.

Timing

Summer 2023 – Spring 2024

• Summer 2023 (optional): Collect data in Ivindo National Park using same methods working with Gabonese student collaborators; or start to process Duke Forest data
• Fall 2023: Engage in inclusive mentoring/teaching and Duke Forest history; do potential second round of data collection in Duke Forest; process data; identify animals from camera traps, acoustic recorders; process LiDAR data
• Spring 2024: Analyze data; conduct modeling on forest characteristics; draft paper(s); present poster or talk at Bass Connections Showcase

Crediting

Academic credit available for fall and spring semesters; summer funding available

This Team in the News

Two Graduating Nicholas School Master's Students Share Virlis Fischer Award

See related Data+ project, Innovations in Research Technology to Assess Forest Wildlife (2023).

Image: Biology professor Ron Grunwald and students in his herpetology class identify reptiles and amphibians they find in Duke Forest, a teaching and research laboratory established in 1931, by Jared Lazarus/Duke University