Based on sounds: AI determines species diversity

According to a study, an automated analysis of animal sounds can provide a good indication of the development of species diversity in the area. The analysis was tested with the help of artificial intelligence (AI) on tropical reforestation areas.

a dense jungle

The sounds of birds, amphibians and mammals were taken into account, as a team led by forest scientist Jörg Müller from the University of Würzburg reported.

Studies had previously shown that the occurrence of animal species in the course of forest destruction or its restoration can be tracked using soundscapes. The research team now installed recording devices in the north of Ecuador in South America on pastures and cacao plantations that are no longer cultivated and where forest is gradually reestablishing itself. Subsequently, the recordings of the different animal sounds were analysed with AI models.

The sound of species communities

"The research results show that the sound data reflect quite excellently the return of biodiversity in the abandoned agricultural areas," Müller shared. "It is above all the sound of the species communities that depicts the recolonisation very well." These communities have a very characteristic composition in the forest and are clearly different from those on agricultural land that is still active.

From the researchers' point of view, the AI models used could provide a basis for studying biodiversity on other reforestation sites as well. In a next step, the scientists would like to improve and expand their models to be able to record more species.

Written by: sda

Photos: Wikimedia Commons

Read more