Screaming into the void: What plants are trying to tell us

Imagine walking through a field on a hot, dry summer day. There is nobody around and the world is blissfully quiet – at least, you think it is. If only you could hear sounds at ultrasonic frequency, you would in fact hear loud “screaming” coming from all around you. You would hear the surrounding plants screaming out, “I’m thirsty!” 

Plant Noises

In a paper published in the journal Cell in March 2023 (Khait et al.), scientists at Tel Aviv University in Israel demonstrated that plants produce audible, airborne sounds. And not only do plants make sound, but the sounds they make are indicative of the state the plant is in. For example, by listening to what the plant is saying, you can tell if it is stressed, thirsty, or injured. 

Prior to this study, it had been shown that plants, like many other living things, respond to sound. Certain sound frequencies can drive plants to alter the specific genes they express (Jeong et al, Ghosh et al.) or the amount of sugar in their nectar (Veits et al.). Additionally, it has been demonstrated that plants can communicate with each other and other organisms, such as herbivores which may wish to eat them (Lev-Yadun). One way they do this is by emitting “volatile organic compounds”, chemical signals they release into the air. For example, plants will emit these molecules when exposed to drought or animals that try to eat them. These compounds can cause neighboring plants to respond by increasing their own resistance to the threat (Paré and Tumlinson, Holopainen and Gershenzon). Essentially, plants are helping each other out by warning each other about potential dangers. 

But who would have guessed plants are talking? It is hard to imagine how plants can even produce sound, with no muscles, vocal cords nor even brains.

The Experiment

To perform this study, the scientists used two different types of plant species: tomato and tobacco plants. First, they put the tomato and tobacco plants in an acoustic box, which blocks out all outside noise, which was equipped with highly sensitive microphones. They then cut some of the plants, restricted water to others, and treated some normally and they continued to record. By training machine learning algorithms with the resulting sounds emitted by the plants, they were able to eventually decipher what the plants were “saying”, as well as which plant was saying it. In other words, the specific acoustic output could tell them what kind of plant was talking, as well as what condition the plant was in: normal, injured, or dehydrated. The stressed plants (injured or dehydrated) had much more to say than their healthy counterparts. 

A) The experimental setup where the plants are located in an acoustic box with microphones. B) The number of sounds made per hour for the different groups, tomato or tobacco, normal, dehydrate, or injured. The dehydrated and injured plants made significantly more noise. C) Visual representation of the plant noises. The X axis shows time and the Y axis shows the amplitude of the noise. You can see here the plants make short bursts of sound. D) A visual representation of the spectrum of sound made in panel C.
A) The number of sounds a tomato plant made per day plotted over the course of nine days of not being watered. B) The proportion of sounds made per hour plotted over the course of a 24 hour day. The plants are far more “talkative” during the day than at night.

Next, the researchers wanted to see if they could identify what the plants were saying in a greenhouse, with all its background noises. After training the machine learning algorithm in the greenhouse, they were able to use it to correctly identify what kind of plant was talking as well as what it was saying (injured or thirsty) inside the greenhouse. They then measured a single tomato plant every day for 10 days without watering the plant to see how the tomato plant responded over time. They saw that the number of sounds the plant made rose daily for the first 4-5 days and then started to drop off after that. They also found that the plants didn’t make noise consistently over the course of the day, but rather talked more in the morning and evening and hardly spoke at night.

Examples of recorded sound made by wheat, corn, grapevine, cactus, deadnettle, and infected tomato.

Not an Isolated Incident

To ensure that it is not only tomato and tobacco plants that emit sounds in response to stress, the team tested several other plant species to see if they made sound and if those sounds changed when those plants were stressed. Of the six plants they tested, all emitted sound and the amount of noises went up significantly when the plants were stressed. So far, all plants tested have been “non-woody” plants (plants that do not have hard stems), so it is still yet to be determined if “woody” plants, such as almond trees, exhibit the same behavior. 

How Does It Work?

Image of an air bubble in a long tube of a plant (called a xylem tube)

You may be wondering how exactly the plants are able to make noise. While it has not been specifically tested in these plants, one potential way the plants are producing sounds is called cavitation. This is a phenomenon that can occur when small cavities form inside a liquid. When these cavities, called “bubbles” or “voids”, collapse, they can generate shock waves which result in sound production (Cavitation, The Daily Garden). This phenomenon has been shown to occur in the long, fluid-filled chambers inside of plants. The resulting sounds from cavitation would also change in response to changes in water levels or damage to the plant, which further validates the hypothesis that this may be how the plants are talking. 

As for why plants are making noise in response to stress: it is possible that plant screams are just a byproduct of their physiology and that the phenomenon didn’t necessarily evolve for any particular reason. That being said, the authors propose the idea that plants might be using sound to communicate with other organisms, such as other plants and animal species. Remember, plants have been shown to respond to sound. For example, plants might be warning their neighbors that they themselves are dehydrated or injured, causing their neighbors to increase their own defenses. Given that plants warn each other using airborne molecules, these plant conversations may not be as far-fetched of an idea as they seem. The sounds emitted could be detected as far as 5 meters (16 feet) away and within the hearing range of creatures such as mice and moths, so it is also conceivable that other animals are listening to what plants have to say and making decisions based on this information. 

Future Implications 

Whether stressed plants are intentionally screaming out or not, this newfound information will likely lead to the development of sound-based plant monitoring systems in agriculture, ecology, and studies of plant disease. The study concludes with the suggestion that airborne sound monitoring may one day be used for precision agriculture that could save up to 50% of water used in agriculture as well as significantly increase crop yield. With the increasingly obvious effects of climate change, we need to start utilizing new and creative methods of sustaining life and minimize our destruction of the planet. Perhaps if humans could hear in the sound range of plants screaming out in agony, we would have been more thoughtful about how we’ve treated the planet. 

Works Cited

“Cavitation.” The Daily Garden, 18 July 2019, www.thedailygarden.us/2/post/2019/07/cavitation.html.

Ghosh, Ritesh, et al. “Exposure to Sound Vibrations Lead to Transcriptomic, Proteomic and Hormonal Changes in Arabidopsis.” Scientific Reports, vol. 6, no. 1, Springer Science and Business Media LLC, Sept. 2016. Crossref, https://doi.org/10.1038/srep33370.

Holopainen, Jarmo K., and Jonathan Gershenzon. “Multiple Stress Factors and the Emission of Plant VOCs.” Trends in Plant Science, vol. 15, no. 3, Elsevier BV, Mar. 2010, pp. 176–84. Crossref, https://doi.org/10.1016/j.tplants.2010.01.006.

Jeong, Mi-Jeong, et al. “Plant Gene Responses to Frequency-specific Sound Signals.” Molecular Breeding, vol. 21, no. 2, Springer Science and Business Media LLC, July 2007, pp. 217–26. Crossref, https://doi.org/10.1007/s11032-007-9122-x.

Khait, Itzhak, et al. “Sounds Emitted by Plants Under Stress Are Airborne and Informative.” Cell, vol. 186, no. 7, Elsevier BV, Mar. 2023, pp. 1328-1336.e10. Crossref, https://doi.org/10.1016/j.cell.2023.03.009.

Lev-Yadun, Simcha. Defensive (anti-Herbivory) Coloration in Land Plants. 2016. Bowker, https://doi.org/10.1007/978-3-319-42096-7.

Paré, Paul W., and James H. Tumlinson. “Plant Volatiles as a Defense Against Insect Herbivores.” Plant Physiology, vol. 121, no. 2, Oxford UP (OUP), Oct. 1999, pp. 325–32. Crossref, https://doi.org/10.1104/pp.121.2.325.

Veits, Marine, et al. “Flowers Respond to Pollinator Sound Within Minutes by Increasing Nectar Sugar Concentration.” Ecology Letters, edited by Christoph Scherber, vol. 22, no. 9, Wiley, July 2019, pp. 1483–92. Crossref, https://doi.org/10.1111/ele.13331.