Particle.news
Download on the App Store

Scientists Identify Second Lion Roar, Boosting AI-Based Monitoring

Machine-learning analysis points to passive acoustic surveys as a promising tool for counting vulnerable lion populations.

Overview

  • A peer-reviewed study in Ecology and Evolution reports two distinct lion roars: the classic full‑throated call and a newly described intermediary roar that is shorter and lower-pitched.
  • Algorithms trained on tens of thousands of hours of audio from Tanzania and Zimbabwe classified call types and identified individual lions with reported accuracy as high as 95.4%.
  • Automated identification outperformed human experts when isolating full‑throated roars, indicating potential for scalable, less biased population monitoring.
  • Researchers say the communicative role of the intermediary roar remains unknown because recordings lacked direct behavioral context.
  • The University of Exeter–led collaboration with Oxford’s Wildlife Conservation Unit and African partners presents passive acoustics as a complement to camera traps for tracking declining populations estimated at 20,000–25,000.