Drones in Wildlife Monitoring

Purpose and Context of Wildlife Monitoring

• Wildlife monitoring tracks population abundance, behavioral patterns, and ecosystem interactions while detecting early risks like disease or human pressure.

• Traditional methods include on-foot surveys, thermal cameras, tagging, and satellite imagery.

• Drones, also known as Remotely Piloted Aircraft Systems ($RPAS$), Unmanned Aerial Systems ($UAS$), or Unmanned Aerial Vehicles ($UAV$), offer faster, cheaper, and more accurate data collection.

• High-quality data is essential for conservation policies and global biodiversity metrics, such as the $IUCN$ Red List of Threatened Species.

Technical Capabilities and Sampling Methods

• Drones carry diverse payloads: cameras, microphones, speakers, and sensors ($LiDAR$, multispectral, hyperspectral, and infrared thermal imaging).

• $LiDAR$ (Light Detection and Ranging) is used to create $3D$ point clouds of vegetation structures.

• Emerging techniques include capturing whale vocalizations or using speakers to simulate predator sounds.

• Standard sampling methods:

  • Line transect: Single or series of straight lines.

  • Grid-based: A "lawnmower" pattern over a specific area.

  • Hover and capture: Stationary imaging.

  • Waypoint: Following a preprogrammed sequence of $GPS$ coordinates.

Case Study: Monitoring the Hainan Gibbon

• The Hainan gibbon is the rarest primate in the world, with approximately $27$ individuals remaining in China.

• The study used drones with thermal infrared sensors to detect the gibbons under dense forest canopies, which previous ground surveys could not achieve.

• Research parameters:

  • Focused on "Group C," consisting of $8$ adults and $1$ infant.

  • Drones flew at approximately $50\,m$ above the canopy to avoid disturbance.

  • Conducted grid-based sampling at an elevation of $800\,m$.

• Results: Successfully identified $8$ temperature hotspots and documented nocturnal sleeping sites in Chinese sweetgum trees for the first time.

Limitations and Future Directions

• Knowledge gaps remain regarding the cumulative stress and disturbance caused by drone frequency and altitude.

• Ethical concerns involve the deployment of drones near nesting and breeding sites.

• Operational challenges include limited battery life, weather restrictions, and evolving regulatory/licensing requirements.

• Data analysis is shifting toward automated methods such as Machine Learning and Computer Vision to handle large datasets effectively.