Using Camera Traps for Wildlife Population Monitoring

Project Overview

The use of camera traps has emerged as a powerful tool for estimating moose and other wildlife populations in Sweden. This project, conducted in the Ödmården Moose Management Area and surrounding regions, aims to refine and validate camera-based methods for wildlife monitoring. The project is a collaboration between the University of Gävle, the Swedish Forest Agency, the County Administrative Board, the Federation of Swedish Farmers (LRF), and the Swedish Hunters’ Association. It is funded by the European Agricultural Fund, with the University of Gävle leading the research.

Since its inception in 2020, approximately 30 cameras have been deployed across a 20,000-hectare area near Ockelbo, positioned at mineral licks in cooperation with hunting teams and landowners. The collected images allow researchers to analyse population dynamics, track seasonal variations, and assess the feasibility of using AI-driven image recognition for improved accuracy. The project has been extended until the end of 2024 to further develop data analysis techniques and enhance automation capabilities.

Key Objectives:

  • Develop a standardized method for estimating moose population size using camera traps.
  • Compare data collected from cameras with traditional methods such as the Moose Observation Index and pellet group counts.
  • Improve understanding of local variations in wildlife populations.
  • Explore the potential for monitoring additional ungulate species.
  • Address technical challenges, including optimal camera placement, data retrieval logistics, and minimizing disturbances.
  • Increase collaboration with landowners and hunting teams to enhance acceptance and practical implementation of the method.

The project has already demonstrated that camera traps provide a more continuous and objective monitoring approach compared to conventional methods. Data collected directly after calving offers a clearer picture of moose reproduction rates, allowing for better-informed management decisions regarding hunting quotas. By standardizing camera placement across different areas, the method enables reliable comparisons over time and between regions, contributing to long-term wildlife population assessments.

Looking ahead, the integration of AI tools will further enhance the ability to process large datasets efficiently. While individual recognition of moose remains a technological challenge, advancements in automated image classification and species identification continue to refine population estimates. This research will serve as a foundation for future applications in wildlife monitoring and sustainable management across Sweden.

For further information, please contact:

  • University of Gävle: Lars Hillström, +46 72-168 23 98
  • Swedish Forest Agency: Marcus Larsson, +46 70-511 49 10
Logotype: Europeiska jordbruksfonden för landsbygdsutveckling