How Do Trees Pull Water Up
100 Metres High?
However hard you suck on a pump, water will only rise 10 metres at most. That's not a limit of engineering — it's a limit set by physics. Yet the world's tallest tree stands 115 metres high, and water still reaches the leaves at the top. Trees do something a pump simply cannot.
When you sip a drink through a straw, you aren't "pulling" the liquid up. You're simply removing air from your mouth. The air pressing down on the surface of the drink then pushes the liquid up to fill the gap.
In other words, it isn't you lifting the water — it's atmospheric pressure. And there's a limit to how high the atmosphere can push. Even with a perfect pump, that limit is about 10 metres. No matter how complete the vacuum, it won't go any higher.
Now look up at a tree. A street tree might be 20 metres tall, a big cedar 50 metres, and the world's tallest tree 115 metres. Water reaches every leaf, even the ones at the very top.
Trees exceed the limit of atmospheric pressure tenfold. That's because they aren't sucking at all.
A pump pushes water up from underneath, so atmospheric pressure sets its ceiling. Trees work differently. They pull the column of water up from the leaves.
Water molecules cling tightly to one another. Inside a thin tube, a column of water behaves like an unbreakable rope, so it can be pulled from above.
And what generates that pulling force isn't a pump or a motor — it's water evaporating from the leaves. Let's look at each step.
Why a pump stops at 10 metres
All a pump does is remove air from inside a tube. The actual lifting is done by the atmosphere pressing on the water's surface.
The force the atmosphere can exert has a limit. The height of a water column that this force can support is about 10 metres. Even a perfect vacuum inside the tube won't push it any higher. "Sucking harder" simply isn't possible, as a matter of principle.
This is why a simple well pump doesn't work for a deep well — instead, deep wells sink the pump into the water itself and push from below.
What trees do instead ― pulling from above
Inside a tree runs a vast network of thin tubes connecting the roots to the leaves. Some are thinner than a human hair. These tubes are filled with water, forming one continuous column from root to leaf.
Water is constantly evaporating from the surface of the leaves. As water molecules leave a leaf, the tip of the column is pulled along with them. That pull then drags the entire column up from below.
Here's where water's special property comes in. Water molecules attract one another very strongly. So inside a thin tube, the column of water doesn't break apart — it behaves like a single rope. Pull the top, and the whole thing rises together.
This method has no ceiling like atmospheric pressure does. It can pull as far as the rope's strength allows.
A tree pulls from above.
Water being pulled is in the opposite state to water being pushed. In terms of pressure, it's below zero — that is, lower than a vacuum. That's an unusual state to imagine, but it's exactly what happens inside a thin tube.
Near the top of a tall tree, this pulling tension is estimated to reach 10 to 20 times atmospheric pressure. A tree may look like it's standing quietly, but enormous forces are at work inside it.
The volumes involved are large, too. A big tree is thought to draw up and return several hundred litres of water a day to the sky on a sunny day. This is part of why forests are called "green dams."
But this method has a weakness
Water being pulled like a rope is actually in a highly unstable state. If even a tiny bubble forms in water under such strong tension, that bubble can suddenly expand and snap the column.
Once that happens, that tube can no longer carry water. This is known to happen often during prolonged dry spells, or when water freezes and thaws in winter.
Trees guard against this in several ways.
- Having many tubes ― if one stops working, the rest keep going. A tree trunk is a bundle of countless thin tubes
- Fine membranes between tubes ― these stop a bubble from spreading into the next tube
- Growing new tubes every year ― tree rings are, in part, a record of this
Even so, once the limit is exceeded, a tree can no longer transport water. When a forest dies in a drought, it's often not just that there's "no water" — the water-transport system itself has broken down.
Is there a limit to how tall a tree can grow?
Yes. The taller a tree, the harder its topmost leaves must pull to get water. In response, leaves tend to close the pores on their surface to avoid losing water.
But those same pores are also the entry point for the carbon dioxide needed for photosynthesis. Close them, and the tree can't grow. Protecting water and growing come into direct conflict.
Balancing these two forces, the upper limit for tree height is estimated at roughly 120 to 130 metres. The tallest tree actually recorded is about 115 metres — a good match for this estimate. Trees, it seems, have come right up to the edge of what physics allows.
Something you can check in your kitchen
- Dissolve food colouring in a glass of water to make it a deep colour
- Cut the base off a pale celery stalk (with leaves) or a white flower and stand it in the water
- Leave it somewhere bright and well-ventilated for a few hours to half a day
- The leaf veins or flower petals will start to take on the colour
- Slice the stem crosswise and you'll see coloured dots arranged in a ring — these are the tubes the water travels through
There's no pump or motor at work, yet the water still rises. The only driving force is water leaving through the leaves. As a test, try covering one stalk's leaves with a plastic bag so they can't evaporate — the colour will rise more slowly. Comparing the two makes it clear that "evaporation is doing the pulling."
Summary
Trees can carry water up to 100 metres high because they don't push it from below like a pump — they pull it from above using the force of evaporation from their leaves. Pushing has a ceiling set by atmospheric pressure; pulling has none.
Every day, quietly, trees break the limit a pump can't.
The only power source is water leaving through the leaves.
The tubes that carry water actually change thickness with the seasons. That record, preserved in the trunk, is what forms tree rings. This is explained in Why Do Tree Rings Form One Per Year?
Then again, how does a root even know which way is "down" as it pushes into the soil? Why Do Plant Roots Grow Down and Shoots Grow Up? looks at how a heavy particle sinking inside a cell sets the direction.
There's another living thing that also "draws water up from the ground," but by a completely different method. Why Do Mushrooms Suddenly Appear After Rain? covers how they rise simply by filling their cells with water and swelling up.
The same surface tension that lifts water inside a tree also works to repel water elsewhere. Why Don't Waterbirds' Feathers Get Wet Even After a Full Day Swimming? explains how narrow gaps keep water out and hold it on the surface of their feathers.
Want to know more? ― terms, equations, and how this fits the textbooksFrom middle-school science to open research questions — each level is labelled
- MSCovered in middle-school science
- HSCovered in high-school "Basic Biology" / "Basic Physics"
- HS+Covered in high-school "Biology" / "Physics", or treated as advanced/sidebar material in textbooks
- UnivNot covered in high school — content from university-level courses (plant physiology, physical chemistry)
- ResearchNot settled even at university level — an active research question
MSTerms: a plant's water pathway
- Xylem: the tubes carrying water from root to leaf. This is the "thin tube" referred to in the article. Made of dead cells joined end to end into a tube.
- Phloem: a separate set of tubes that carries the nutrients made in the leaves. Its role differs from the water pathway.
- Transpiration: water leaving the leaf as water vapour. This is the article's "driving force."
- Stomata: tiny pores on the leaf surface. They open to take in carbon dioxide, and water escapes through them at the same time. Water vapour leaves through here, and carbon dioxide enters. Open them and you lose water; close them and photosynthesis stops.
- Cohesion: the mutual attraction between water molecules. This is why the water column stays unbroken.
HSChecking with equations: both obvious explanations vanish under calculation
Water reaches the top of a 100 m tree. How? The obvious guesses are "the roots push it up" or "it climbs through thin tubes." Neither holds up once you calculate. Let's check them in order.
h ＝ P ÷ (ρ × g)
| h height that can be lifted | in m |
| P atmospheric pressure | 101300 Pa |
| ρ density of water | 1000 kg/m³ |
| g gravity | 9.8 m/s² |
This is the same idea as drinking through a straw. However hard the sucking side tries, it's atmospheric pressure doing the lifting, so there's a hard ceiling.
| Compute the denominator | 1000 × 9.8 ＝ 9800 |
| Divide | 101300 ÷ 9800 ≒ 10.3 |
| Limit height | 10.3 m |
10.3 m is the ceiling. No matter how powerful the pump, this method cannot exceed it. For a 100 m tree, it covers only a tenth of the distance. Explanation 1 is ruled out.
This is capillary action — water rising by itself inside a narrow tube. The narrower the tube, the higher it climbs. And the tree's water pathways really are thin tubes.
h ＝ (2 × σ) ÷ (ρ × g × r)
| σ surface tension of water | 0.072 N/m |
| r radius of the tube | in m |
| Radius of a tree's water pathway | r ＝ 0.00002 m (20 micrometres) |
| Compute the numerator | 2 × 0.072 ＝ 0.144 |
| Compute the denominator | 1000 × 9.8 × 0.00002 ＝ 0.196 |
| Divide | 0.144 ÷ 0.196 ≒ 0.73 |
| Height reached | 0.73 m |
73 cm. That's even less than Explanation 1. "It climbs because the tube is thin" doesn't even get it up to knee height. Explanation 2 is out too.
※ An even narrower tube would rise higher, but narrower tubes also carry water more slowly — too slowly to supply what a tree actually needs.
Both explanations are gone. That means the water is neither being "pushed up" nor "climbing up" on its own. The remaining possibility is that it's being pulled from above.
When water evaporates from a leaf, that much water gets pulled along. Because water molecules attract each other so strongly, the column stays joined like a thread and is dragged upward from below. Picture it not as sucking through a straw, but as pulling on the water itself as though it were a rope.
The force required can also be calculated.
| Pressure needed to lift 100 m | 1000 × 9.8 × 100 ＝ 980000 Pa |
| As a multiple of atmospheric pressure | 980000 ÷ 101300 ≒ 9.7 times |
Add friction inside the tubes on top of that. The tension at work inside a tree exceeds ten times atmospheric pressure.
Here's the interesting part. This means the pressure is below zero — a regime where ordinary water would boil and turn to bubbles. Water inside a tree is being pulled continuously in a state where it "should" be turning into bubbles, yet doesn't.
Ruling out the two obvious explanations by calculation led us to a far stranger answer. Why this state can be maintained is covered in the next section.
HSWhere does "10 metres" come from?
The height a water column can be pushed up is set by the balance atmospheric pressure ＝ density of water × gravitational acceleration × height.
| Atmospheric pressure | about 101,300 Pa |
| Density of water × gravity | 1,000 × 9.8 ＝ 9,800 |
| Height ＝ 101,300 ÷ 9,800 | ≒ 10.3 m |
※ Mercury is 13.6 times denser, so for the same atmospheric pressure it rises only about 76 cm — this is the principle behind the mercury barometer.
This figure has nothing to do with pump performance. Once you choose "sucking" as your method, the ceiling is already fixed.
HSHS+Capillary action alone isn't enough
Some readers may have thought of capillary action — water rising on its own inside a narrow tube. But once you run the numbers, it falls well short on its own.
The height reached is h ＝ 2γ cosθ /(ρ g r). With a xylem radius of 20 micrometres, it rises at most 1 to 2 metres — nowhere near 100 metres.
But the gaps in the cell walls of leaves are far, far narrower. At a radius of a few nanometres, that same equation produces an enormous pulling force. Capillary action isn't the force that lifts the water — it's the "tension-generating device" at the leaf tip. That distinction is the key point here.
UnivThe cohesion-tension theory and water's metastable state
This mechanism is called the cohesion-tension theory, proposed in the late 19th century. Its framework rests on three points: ① evaporation at the leaf generates negative pressure (tension), ② the cohesion of water maintains the continuity of the water column, and ③ adhesion to the xylem walls provides support.
The core idea is that water can exist in a pulled (negative-pressure) state. Thermodynamically, this is a metastable state. It should, in principle, vaporise and form a cavity (cavitation), but without a bubble nucleus to start from, water can withstand considerable tension. Experiments have reported water inside tiny inclusions withstanding tension exceeding −100 MPa — roughly 1,000 times atmospheric pressure.
Trees typically operate at a water potential of around −1 to −3 MPa. The absence of gas nuclei inside the tubes, combined with the extreme narrowness of the xylem, is what allows this metastable state to be maintained.
To prepare for cavities that do form (embolism), xylem walls contain a fine pit membrane that prevents bubbles from spreading into neighbouring tubes. This membrane involves a trade-off between water conductivity and safety, however — it's known that species more resistant to drought tend to conduct water less freely.
ResearchWhat's still unresolved
- How a blocked tube recovers is not understood. A xylem tube once blocked by a bubble has been observed to refill with water again. But how a bubble can be pushed out while the surrounding water is under tension (negative pressure) is thermodynamically hard to explain. The leading theory is that living cells release sugars to locally draw water in, but results differ depending on the observation method, and the question remains unresolved.
- Directly measuring tension inside a tree remains difficult. The widely used measurement methods are indirect, and there has long been debate over discrepancies with direct measurements using a fine probe inserted into the tissue. The field faces a distinctive difficulty: trying to measure the system tends to disturb it.
- The process by which trees die in drought also isn't fully modelled. Whether a tree dies because "its water pathways break down" or because "it keeps its stomata shut until it runs out of nutrients" varies by species and conditions. Because this connects directly to predicting how forests will respond under climate change, it's one of the most actively studied areas right now.
- Even the height limit isn't a fixed number. Besides the constraint from water transport, mechanical buckling, wind damage, and supply from the roots are all factors. There are records suggesting taller trees existed in the past, but verifying them is difficult.
Right now, in an ordinary street tree near you, something is happening at the very edge of what physics allows.
How this maps to the textbooks (by level)
| Level | Subject / unit | Where in this article |
|---|---|---|
| MS | Science ― plant structure and function (xylem, transpiration) / atmospheric pressure | The water pathway, transpiration as the driving force, the coloured-water observation |
| HS | Basic Physics ― pressure / Basic Biology ― water uptake in plants | The 10.3 m calculation, stomata opening and closing |
| HS+ | Physics ― surface tension and capillary action / Biology ― water potential | h ＝ 2γcosθ/(ρgr), how the cell-wall gaps work |
| Univ | Plant physiology / physical chemistry | Cohesion-tension theory, metastable state, cavitation, pit membrane |
| Research | Plant physiology / ecology (unresolved) | Embolism repair, direct measurement of tension, drought-driven death, height limit |
- Dixon, H. H. & Joly, J., On the ascent of sap, Philosophical Transactions of the Royal Society B 186, 563–576, 1895 (proposed the cohesion-tension theory).
- Koch, G. W. et al., The limits to tree height, Nature 428, 851–854, 2004 (estimate of the upper limit of tree height).
- Tyree, M. T. & Zimmermann, M. H., Xylem Structure and the Ascent of Sap (the standard textbook in this field).
- Wheeler, T. D. & Stroock, A. D., The transpiration of water at negative pressures in a synthetic tree, Nature 455, 208–212, 2008 (an experiment reproducing negative-pressure transport in an artificial "tree").
- Choat, B. et al., Triggers of tree mortality under drought, Nature 558, 531–539, 2018 (discussion of drought-driven mortality).
※ Figures for tension and height vary by species and conditions. This article presents commonly cited approximate values.
※ This article is a general-audience science explainer. The figures given are approximations meant to aid understanding and may vary with conditions. For any hands-on observation, please use purchased or home-grown plants rather than taking them from parks or other people's property.