Chapter 09 · Sustainability & AISustainability & AI

Bioacoustic monitoring

Meaning statusEstablishedSource recordDirect document linkedWhy these are different

Definition

Using environmental sound recordings and AI analysis to detect and identify species — and human disturbances — across landscapes continuously.

References

Ecological Informatics 2021Kahl et al. — BirdNET: deep learning for avian acoustic monitoring

This source provides part of the technical or institutional basis for the definition.

tinyML FoundationtinyML community and foundation

This source supports the explanation of how the term is applied, measured or governed in practice.

Overview

What it means

Passive acoustic sensors record soundscapes for months at negligible cost; machine-learning classifiers such as BirdNET (Kahl et al. 2021) then identify calls, turning audio into species occurrence data. The approach covers birds, bats, amphibians and insects, and can flag chainsaws or gunshots in protected areas.

How it is used

Researchers and conservation bodies deploy sensor networks for population trends, restoration monitoring and anti-poaching response. Data volumes are enormous, making automated classification — with human validation of uncertain calls — the only viable workflow.

Why it matters

Sound reveals what cameras miss: nocturnal, canopy and underwater life. Bioacoustics plus AI gives conservation a persistent, scalable sensory system — one whose classifications, like any model output, require calibration against expert ground truth.