LONDON: A groundbreaking artificial intelligence tool has demonstrated the ability to assess the health of forests by simply listening to the sounds they produce, offering a cost-effective solution for monitoring forest restoration and conservation initiatives.
Healthy forests, teeming with diverse wildlife, emit a symphony of squawks, squeals, trills, and yelps that change when trees are cut down. Deciphering these sounds accurately has been a challenge for human researchers, but this new AI technology might hold the key.
Researchers have reported in the journal Nature that the AI algorithm effectively matched expert assessments of biodiversity across both recently degraded forests and old, healthy ones. This breakthrough could revolutionize how conservationists monitor the success of forest restoration projects, a task that has been challenging and costly until now.
With governments increasingly focusing on conserving and restoring forests to combat climate change and preserve biodiversity, large-scale restoration projects have become common.
Effectiveness of AI Tool for Forests’ Conservation
However, ensuring the success of these projects is far from guaranteed and often controversial. Traditional tree-planting initiatives, while popular, often lack a comprehensive approach to biodiversity, focusing on monoculture – planting a single tree species – rather than creating diverse and sustainable ecosystems.
The new AI tool analyzes the soundscape of forests, offering a cost-effective and efficient means to assess the effectiveness of conservation efforts. Scientists can interpret sound data in various ways, examining factors such as sound frequency and complexity. The researchers developed an AI that combines multiple methods of analyzing sound data to monitor biodiversity across different forest environments, even identifying species that produce minimal noise.
In testing, the AI was applied to recordings from 43 different plots in Ecuador, representing various stages of forest regrowth, from active cacao plantations and pastures to old-growth forests. The AI’s predictions of biodiversity levels at each plot were then compared to assessments made by biologists in the field.
