Jorg Muller, an ecologist at the University of Würzburg, with his colleagues, proposes a way to measure the biodiversity—listen to the jungle by AI, in a paper published in Nature Communications.
The rainforests are very important and always alive with the sounds of animals, which is useful to ecologists. When it comes to measuring the biodiversity of a piece of land, listening out for animal caller is more effective than uncovering the bushes looking for tracks and paw prints. The latter analysis method is time-consuming, and it requires an expert pair of ears. Muller's idea was to apply the principle of smartphone apps which can identify the sounds of birds, bats and mammals to conservation work.
The researchers took recordings from across 43 sites in the rainforests. Some sites were relatively primitive, old-growth forests. Others were areas that had just been cleared for pasture(牧场) recently. And some other original forests had been cleared but then abandoned, allowing themselves to regrow. The various calls were identified by an expert,and then used to construct a list of the species present. As expected, the longer the land had been free from agricultural activity, the greater biodiversity it hosted. Then it was the computer's turn. "We found that the AI tools could identify the sounds as well as the experts," says Dr. Muller.
Of course, not everything in a rainforest makes a noise. Dr. Muller and his colleagues used light traps to catch night-flying insects, and DNA analysis to identify them. They found that the diversity of noisy animals was a reliable representative for the diversity of the quieter ones, too.
Besides measuring the biodiversity, the results are also expected to be applied to outside ecology departments. Under pressure from their customers, firms like L'Oreal, a make-up company, and Shell, an oil firm, have been spending money on forest restoration projects around the world. Dr. Muller hopes that an automated approach to checking on the results could help monitor such efforts, and give a standard way to measure whether they are working as well as their sponsors say.