Study
Sounds emitted by plants under stress are airborne and informative
Khait, Itzhak; Lewin-Epstein, Ohad; Sharon, Raz; Saban, Kfir; Perelman, Revital; Boonman, Arjan; Yovel, Yossi; Hadany, Lilach
Cell, 2023
Ultrasonic microphones recorded tomato and tobacco plants in an acoustically isolated box and later in a greenhouse. Plants were drought-stressed, cut, or left untouched, and a machine-learning classifier was trained to distinguish the recordings by treatment.
- Studied in
- Solanum lycopersicum, Nicotiana tabacum
- Sample size
- recordings from drought-stressed, cut and control plants
Stressed plants emit airborne ultrasonic clicks in the 20–100 kilohertz range, at rates of dozens per hour, against near-silence from unstressed plants. A classifier could distinguish drought-stressed from cut from control plants above chance, and could do so in greenhouse conditions. The sounds are detectable at a distance of several metres for an animal with suitable hearing.
What it means
The authors' reading, and ours. Where they differ, that difference is the point.
How the authors put it
Plants emit informative airborne sounds under stress, which organisms with ultrasonic hearing could in principle use.
How NatureHQ reads it
This is the paper behind every "plants scream" headline, and the gap between it and the headline is the point. What was measured is clicking, in a range no human can hear, at a rate of a few dozen an hour — closer to a dripping tap than to a scream, and produced by cavitation, water columns snapping under tension. There is no evidence it is a signal: nothing indicates the plant produces it for a receiver, and the paper does not say so. "Informative" is doing precise work in that title — it means the sound correlates with the plant's state well enough that a listener *could* learn something, which is the definition of a cue rather than a signal. Whether anything actually listens is the open question, and the honest answer today is that it is being investigated.
- Two crop species under controlled conditions; the greenhouse extension is a smaller dataset.
- Classifier performance above chance is not the same as a receiver being able to use the information in the field.
- No receiver behaviour was tested in this study, so no communicative function is established.
- Mechanism is inferred as cavitation and not directly imaged here.
Listening to plants at frequencies nobody can hear
Do stressed plants make sounds, and does the sound say anything about the stress?
What this study is used for on NatureHQ
One study can inform several subjects. Here is everywhere this one is cited.
Published by: Cell PressPublished in a Cell Press journal.
doi.org/10.1016/j.cell.2023.03.009
This reference was resolved automatically against Crossref on 2026-08-11.
NatureHQ summarises research in its own words and does not reproduce published text. Reviewed 2026-08-11.