THe is the rainforest Living in the sounds of animals. Apart from the joy of religion, it is also useful for ecologists. If you want to measure the biodiversity of a piece of land, look no further than listening to animal calls or rummaging through the undergrowth in search of tracks or spurs. But such “bioacoustic analysis” is still time-consuming, and requires an expert pair of ears.
In a paper published on October 17 Nature communication, a group of researchers led by Jörg Müller, an ecologist at the University of Würzburg, described a better way: let a computer do the work. Smartphone applications already exist that will detect birds, bats or mammals simply by listening to their sounds. Their idea was to apply principles to conservation work.
The researchers took recordings from 43 sites in the Ecuadorian rainforest. Some sites were relatively primitive, old-growth forests. Others are areas that have recently been cleared for pasture or cocoa planting. And some were cleared but then abandoned, allowing the forest to grow back.
Sound recordings were taken four times every hour for two weeks. Different calls were manually identified by an expert and then used to create a list of the species present. As expected, the longer the land was free from agricultural activity, the greater the biodiversity.
Then it’s the computer’s turn. The researchers fed their recordings to artificial-intelligence models that were trained, using sound samples from elsewhere in Ecuador, to identify 75 bird species from their calls. “We saw that A.I The tools can identify experts as well as words,” says Dr. Müller.
Of course, not everything in the rainforest makes noise. Dr. Müller and his colleagues used light traps to capture night-flying insects and DNA Analysis to identify them. Reassuringly, they found that the diversity of noisy animals is also a reliable proxy for the diversity of quiet animals.
The findings may be relevant beyond the ecology department. Under pressure from their customers, L’Oréal, a make-up company, and Shell, an oil company, are spending money on forest restoration projects around the world. Dr. Muller hopes that an automated method for checking results can help monitor such efforts and provide a standardized way to measure whether they are working as well as their sponsors.