A machine-learning model can detect the sound of plants, and it could be used to monitor drought, drought and water quality in farming and horticulture

There are sounds of 20 to 100 kilohertz made by thirsty plants, which are the reason you don’t hear them. That means they are so high-pitched that very few humans could hear them. Some animals, however, probably can. Bats, mice and moths could potentially live in a world filled with the sounds of plants, and previous work by the same team has found that plants respond to sounds made by animals, too.

The team created a machine-learned model to determine whether a plant had been cut or was stressed by water with about 70% accuracy. This result suggests a possible role for the audio monitoring of plants in farming and horticulture.

The tomatoes sound high in pitch, and mice and moths can hear it. Since tomato plants have some insects on them, they could listen in to find out which plant gives their young the best chance of survival. Other kinds of plants have also been found to react to sound and might start to prepare themselves for drought, for instance.

He doesn’t think animals listen to plants’ moans. “It is unlikely that these animals are really able to hear the sound at such distances,” he says. He doesn’t think that the sounds would be very strong. There needs to be further research on the matter. But Pyke says he’s perfectly willing to accept that plants ‘squeal’ when stressed.

“Even in a quiet field, there are actually sounds that we don’t hear, and those sounds carry information,” Lilach Hadany, senior author of the paper and an evolutionary biologist and theoretician at Tel Aviv University, says in a press release.

Hadany and her colleagues wrote about how tomato and tobacco plants can make clicking noises when they are dehydrated or cut.

Researchers uploaded a recording of plant sounds to the internet. They lowered the frequency so that humans can hear it. The recording does kind of sound like Bubble Wrap popping.

Farmers could set up listening sensors to gauge whether their crops are healthy or need extra attention. And since they could potentially hear how dry the plants are — whether they’re just starting to get dehydrated or nearing the point of no return — this kind of application could help farmers water crops more efficiently.