Plants have been screaming all around us, and we've never heard a thing.
For as long as humans have tended fields and walked through forests, plants have been speaking in a language beyond our hearing. Researchers at Tel Aviv University have now confirmed that stressed plants — drought-stricken or physically injured — emit ultrasonic popping sounds between 20 and 100 kilohertz, detectable up to five metres away by organisms attuned to those frequencies. Published in the journal Cell, the findings suggest that the botanical world has long possessed an acoustic dimension operating quietly alongside us, and that learning to listen may reshape how we grow food, understand ecosystems, and conceive of plant life itself.
- Plants have been broadcasting distress signals in ultrasonic frequencies all along — we simply lacked the ears to hear them.
- The discovery creates an urgent rethinking of what 'communication' means in the natural world, unsettling long-held assumptions about plant passivity.
- Machine learning can already decode these sounds to identify plant species and the specific type of stress they are under, turning noise into diagnosis.
- Agriculture stands at the threshold of a precision revolution — microphone arrays in fields could one day alert farmers to drought stress before crops visibly suffer.
- A deeper, more speculative question is now open: if plants emit these signals and neighbouring plants can respond to sound, an acoustic warning network may already exist in nature.
Plants have been emitting distress calls all around us, and we've never heard a thing — until now. Researchers at Tel Aviv University recorded ultrasonic sounds produced by tomatoes, tobacco, grapevines, maize, wheat, and other species when subjected to drought or physical cutting. These signals, ranging from 20 to 100 kilohertz and inaudible to humans, travel through the air up to five metres — well within range of many other creatures. The study, published in Cell, reveals an acoustic dimension to plant life that has been operating invisibly alongside us.
The experimental setup was precise: microphones placed 10 centimetres from plant stems captured sounds under drought conditions and after cutting, then compared them to recordings from healthy plants and empty pots. Stressed plants emitted far more acoustic signals. When slowed into the human audible range, the recordings produced a distinctive popping — each pop a moment of physiological distress. As drought deepened, the pops grew more frequent, tapering only as the plant neared complete desiccation. The likely mechanism is cavitation: air bubbles forming and bursting inside the xylem, the plant's water-conducting tissue, as internal water dynamics shift under stress.
What elevates this beyond curiosity is that the sounds carry structured information. Using machine learning, researchers trained systems to identify not only which species had produced a sound, but what kind of stress it was experiencing — a drought-stressed tomato sounds different from a cut one. Many organisms, including moths, operate in the ultrasonic range and may already be 'reading' these signals in real time.
The agricultural implications are immediate: microphone systems in fields could enable precision irrigation, delivering water exactly when and where plants signal need rather than on fixed schedules. For home gardeners, the vision is simpler still — a phone notification that your houseplant is thirsty. Further out lies a more open question: if plants emit these signals and other plants respond to sound vibrations, could a form of acoustic plant-to-plant communication already be shaping how ecosystems respond to drought or stress? One limitation has emerged — woody stems in trees appear largely silent in this acoustic register — but the broader revelation stands. We have been moving through a world of botanical sound we were never equipped to hear, and now that we know how to listen, the question is what we will do with what we learn.
Plants have been screaming all around us, and we've never heard a thing. Researchers at Tel Aviv University have now recorded the ultrasonic distress calls that tomato, tobacco, grapevines, maize, wheat, and several other species emit when stressed by drought or physical injury. These sounds exist in frequencies between 20 and 100 kilohertz—well above the range of human hearing—but they travel through the air up to five metres away, loud enough for other creatures to detect. The findings, published in the journal Cell, suggest that plants possess an acoustic dimension to their existence that has been operating invisibly alongside us, a hidden conversation we're only now learning to listen to.
To conduct the study, researchers placed microphones just 10 centimetres from plant stems under two conditions: severe drought (soil moisture below 5 per cent) and physical cutting near the soil line. When they compared these recordings to sounds from healthy plants and empty pots, the pattern was unmistakable. Stressed plants emitted significantly more acoustic signals than their unstressed counterparts. When the team downsampled the recordings into the human audible range and sped them up, the result was a distinctive popping sound—each pop representing a moment of distress. As drought intensified, the frequency of these pops increased, though the pattern eventually declined as the plant approached complete desiccation.
The mechanism behind these sounds likely involves cavitation, a process in which air bubbles form and burst inside the xylem—the water-conducting tissue that runs through plant stems. When a plant experiences drought or is cut, the water dynamics within its tissues shift dramatically, creating the conditions for these bubble formations. The physical rupture of these bubbles produces the acoustic signal. It's a byproduct of the plant's own physiology under stress, not necessarily an intentional communication strategy, though the distinction may matter less than the fact that the signal exists and carries information.
What makes this discovery particularly striking is that the sounds are not random noise. Using machine learning algorithms, researchers trained computer systems to listen to these recordings and identify not only which plant species had produced the sound, but also what type of stress it was experiencing. A tomato plant in drought sounds different from a cut tomato plant. The acoustic signature carries diagnostic information about the plant's condition. This suggests that if other organisms in the ecosystem can hear these frequencies—and many can, including moths that operate in the ultrasonic range—they may be receiving real-time updates about plant health status across their environment.
The implications ripple outward in several directions. In agriculture, the most immediate application is precision irrigation. Imagine a farmer with a microphone system monitoring a field, receiving alerts when specific plants or sections begin emitting distress signals. Water could be delivered exactly when and where it's needed, rather than on a fixed schedule or across an entire field. This kind of targeted approach could help address the growing pressure to feed a rising global population while using water more sustainably. For home gardeners, the vision is even simpler: a notification on your phone telling you that your neglected houseplant is thirsty.
Beyond agriculture lies a more speculative but equally intriguing possibility. If plants can emit these signals and other plants can hear them—research has already shown that plants respond to sound vibrations in their environment—then a form of plant-to-plant communication may be operating at scales we're only beginning to understand. Could stressed plants warn their neighbours of drought or pest pressure? Could this acoustic network help plant communities adapt more quickly to environmental change? These questions remain open, though the research has revealed one limitation: woody plant stems don't appear to produce detectable sounds, which means many tree species may be silent in this acoustic world, at least in their main structural tissues.
The discovery sits at the intersection of plant physiology and ecology, revealing that the botanical world has been far more complex and communicative than our sensory limitations allowed us to perceive. We've been walking through forests and fields, tending gardens and crops, all while missing an entire dimension of plant life playing out in frequencies our ears cannot reach. Now that we know how to listen, the question becomes what we'll do with this knowledge—and whether understanding plants as acoustic beings will change how we grow food, manage ecosystems, and think about the living world around us.
Notable Quotes
Plants can listen, but now they can talk— Research team at Tel Aviv University (paraphrased from source)
The Hearth Conversation Another angle on the story
So plants are making sounds when they're stressed. But are they actually trying to communicate, or is this just a side effect of what's happening inside them?
That's the crucial question, and honestly, we don't know yet. The popping sound comes from air bubbles bursting in the plant's water-conducting tissue—it's a physical consequence of drought or injury. Whether the plant "intends" to communicate through this, or whether other organisms have simply learned to interpret these sounds as signals, is still open.
But you said machine learning can tell the difference between a stressed plant and an unstressed one, and even what type of stress it's under. Doesn't that mean the signal carries real information?
It does carry information—that's undeniable. A moth or another organism listening in the ultrasonic range could pick up on these signals and learn something true about the plant's condition. But that's different from the plant deliberately encoding a message. It's more like reading a fever in someone's face—the information is there, but the person isn't necessarily trying to tell you they're sick.
What about the possibility of plant-to-plant communication? Could one plant hear another plant's distress call?
That's where it gets genuinely mysterious. We know plants respond to sound vibrations in their environment. We know they can detect chemical signals from neighbouring plants. But whether they're actually listening to these ultrasonic pops and adjusting their own behaviour in response—that hasn't been demonstrated yet. It's plausible, but it's speculation.
If a farmer could use this to know exactly when to water, would that actually change how agriculture works?
It could, significantly. Right now, irrigation is often done on a schedule or based on broad soil moisture readings. If you could listen to individual plants or sections of a field and water only when they're signalling stress, you'd use less water and potentially get better yields. That matters enormously as population grows and water becomes scarcer.
One last thing—why can't we hear these sounds ourselves? What's the evolutionary reason plants are screaming in frequencies we can't detect?
Plants didn't evolve these sounds with us in mind. The frequencies they use are probably determined by their physiology—the size and structure of their tissues, the speed at which bubbles form and burst. That those frequencies happen to be ultrasonic is just happenstance. Or perhaps it's useful to them precisely because we can't hear it, and neither can many herbivores. It's a private conversation.