One Hundred Meters, No Pump: How Trees Lift Water

Bölüm metni

A suction pump can draw water ten meters up at most; that is a hard limit set not by the quality of the pump but by atmospheric pressure. And yet in California the topmost leaf of a redwood is drinking water one hundred sixteen meters above the ground, inside a living thing with no heart and no motor. The tree does not break the limit — it does not push the water, it pulls it, and stretches it like a wire about to snap.

`podcast-science.json` bu kanalı `"language": "en"`, `kaynak_kanal: "bilim"` olarak tanımlıyor; bugünün `bilim.txt` bölümü de tam bu konu. Sert kısıt "çeviri kaynağa sadık kalır" dediği için metni Türkçe yeniden yazmadım (o zaten `bilim.txt`) — kaynağın İngilizce çevirisini verdim; jingle ve kapanış satırlarını çıkardım.

A suction pump can raise water about 10 meters and no higher. That ceiling is set not by the quality of the pump but by the weight of the atmosphere. And yet in California the topmost needle of a redwood is drinking water 116 meters above the ground, inside a living thing with no heart and no motor anywhere in it. The tree does not break the limit. It does not push the water upward — it pulls, and it holds that water stretched like a wire about to snap.

In the Florence of the 1630s, in the gardens of the Grand Duke of Tuscany, there was a problem nobody could solve. The court's well-diggers had fitted a suction pump to a deep well, drawn the piston, watched the water climb the pipe, and watched it stop at a certain point for no reason anyone could give. They replaced the seals; nothing changed. They widened the pipe; nothing changed. They swapped the man on the handle for a stronger one; still nothing. The water would come up to roughly 18 braccia, about 10 meters, and then flatly refuse to rise further. Empty pipe above it, plenty of water below it, and nothing at all in between.

They took the problem to the most celebrated natural philosopher of the age, the aging Galileo. His answer was wrong, but the way in which it was wrong was oddly ahead of its time. As he saw it, the rising column of water was like a rope hanging from a ceiling. The void had the power to pull matter, but that power was not unlimited; once the column of water passed a certain length it snapped under its own weight, exactly as a rope would. In other words, Galileo thought water could be pulled, but that there was a breaking point to the pulling. Hold on to that idea, because three hundred years later it comes back from an entirely different direction.

The real answer arrived a few years afterward, from Galileo's student Torricelli. In Rome, a man named Gasparo Berti had rigged an experiment against the wall of his house with an enormous glass tube more than 10 meters tall, and had shown that the water always came to rest at the same height. In 1643 Torricelli ran the same experiment in miniature: he used mercury, 13.6 times denser than water, and found the mercury column standing at 76 centimeters. The ratio between the two numbers was precisely the ratio of the two densities. The meaning was plain, and it inverted the intuition of the age. The pump was not pulling the water up. The air itself, with the weight of the kilometers-thick column of atmosphere overhead, was pressing down on the surface of the water in the well and pushing it into the emptied pipe. At sea level that push amounted to a force of roughly 1 kilogram on every square centimeter, which is exactly enough to hold up a column of water 10.33 meters tall.

This is not a problem engineering can solve. It is a piece of arithmetic. Build the most perfect pump on earth, draw a flawless vacuum inside the pipe, and at sea level you still cannot suck water higher than 10.33 meters, because the only force working in your favor is the weight of the atmosphere and that force has a hard ceiling. That is the physical limit of pulling from above. If you want water to go higher, you have to push it from below.

Now go to the coast redwood forests of northern California. There, in a valley whose location is kept secret, stands a tree called Hyperion. When it was found in 2006 it was a little over 115 meters; the measurements show it is still growing, and today it has passed 116 meters. Eleven times the wall that stopped the pump. Inside this tree there is no piston, no valve, no motor, not one contracting muscle fiber. Wood is not a material that moves. And yet on a hot summer day a large tree takes hundreds of liters of water out of the soil and carries it all the way to its crown, continuously enough to keep the needles up there wet.

Let us clear the easy answers out of the way. Roots really can generate pressure. That is why water weeps from a maple or birch branch cut in spring, and why droplets gather on the tips of grass blades at dawn. But root pressure tops out around 0.1 megapascals, which at best buys you a 10-meter column. It largely vanishes the moment the leaves begin to transpire, and in most conifers it is barely present at all. Pushing from below is not enough.

What about capillarity? It is true that the water-carrying pipes inside wood are extremely fine, and that in a narrow tube water will climb on its own by clinging to the walls. But how far it climbs is set by the diameter of the tube, and the arithmetic is merciless. In a typical tree conduit 20 micrometers wide, capillary rise comes to about 1.5 meters. If you wanted capillarity alone to lift water 100 meters, the conduits would have to be 0.15 micrometers across, and in practice no water would flow through a pipe like that at all; friction would lock everything solid. So the tree is neither pushing from below nor climbing. It is doing something else.

To find the answer you have to go to the top and look inside a leaf.

Cut a leaf across and you see that its interior is made of loosely packed cells with air spaces between them. The outer walls of those cells are a spongy, permanently damp mesh woven out of cellulose. On the surface of those wet walls, water meets air. When the pores on the underside of the leaf, the stomata, open, the outside air is drier than the air within, and water begins to evaporate. Every molecule that leaves drags the water behind it deeper into the pores of the cellulose mesh. As the water retreats, the surface at the mouth of each pore dishes inward like a tiny bowl.

The entire mechanism is hidden in that curvature. The surface tension of water tries to flatten a curved surface, and in trying, it hauls the water underneath upward. The strength of that pull depends on how sharply the surface is bent: the narrower the pore, the stronger the pull. The pores in the wall of a leaf cell are on the scale of nanometers. In a pore with a radius of 10 nanometers the pull can reach 15 megapascals. For comparison, atmospheric pressure is 0.1 megapascals. The damp wall of a leaf in contact with dry air can therefore generate a suction a hundred times stronger than the finest vacuum pump. And it needs no mechanism and no expenditure of energy to do it; all it needs is dry air and a wet surface.

What carries that pull all the way down to the roots is a peculiarity of water itself. Water molecules grip one another through hydrogen bonds, and that grip is astonishingly strong. In the laboratory, pure water with no air bubbles in it can withstand tens of megapascals of tension before it parts. Under the right conditions, in other words, water behaves less like a liquid than like a wire. And that is exactly what the tree builds: a single unbroken thread of water running from leaf to root with no gap anywhere along it. A droplet evaporating at the top twitches the thread upward, and the thread, at its lowest end, draws a fresh droplet out of the soil through the root hairs. Galileo's rope hanging from the ceiling was right, once you turn it upside down.

The pipes the water runs through are made of dead cells inside the wood. The tree builds the cells, stiffens their walls with lignin, then empties out their contents, leaving hollow cylinders stacked end to end with their ends opened. In conifers like pine and redwood these are narrow cells a few millimeters long and 10 to 40 micrometers wide, and water seeps from one into the next through thin membraned pits in their walls. In flowering trees the cell ends have dissolved away completely, producing true pipes meters in length. That structural difference shows up directly in the speed of flow: in a pine, water advances 1 to 2 meters an hour, while in a wide-vesseled oak it can reach 40 meters an hour.

This explanation was put forward in the 1890s independently by Dixon, Joly and Askenasy, and for a long time the idea was thought unbelievable. The reason was this: if the water is being pulled, the pressure inside the pipe has to be below zero. A pressure lower than a vacuum. Measuring that took half a century, and it became possible in the 1960s with the pressure chamber technique Scholander developed. A leaf is cut and sealed in a chamber with its stalk protruding, and the pressure inside is raised until water appears at the cut end. The pressure required is the mirror image of the tension that was inside that leaf. The measurements confirmed the theory. At midday the tension at the crown of a tall tree runs between minus 1 and minus 2 megapascals, and in desert plants far more extreme values are recorded.

That closes the energy account as well. The gravitational cost of carrying water 100 meters up is about 0.01 megapascals per meter, so 1 megapascal in total, with friction losses in the pipes on top of it. The Sun pays every bit of that bill, because the thing doing the work is evaporation itself. The tree does not spend a single molecule of sugar lifting water. Its cost lies elsewhere: to take in carbon dioxide it has to open its stomata, and through those open pores it loses between 200 and 400 molecules of water for every molecule of carbon dioxide it gains. That is why the tree moves water at all. The rising column is the invoice photosynthesis pays.

But water under tension is in a dangerous state.

At pressures below zero, liquid water is in what physicists call a metastable condition: on paper it should already have boiled away, and the only thing keeping it liquid is the refusal of its molecules to let go of one another. Let the smallest air bubble into a column like that and the bubble does not close up. The opposite happens: because the pressure outside is higher than the pressure within, the bubble expands explosively, fills the whole conduit in an instant, and puts that conduit permanently out of service. The process is called cavitation, and the blocked pipe is called an embolism.

The air usually gets in from an empty neighboring conduit, sucked through the pores of the pit membrane between them. As drought deepens the tension rises, and as tension rises air can pass through ever smaller pores, so the blockage spreads from one pipe to its neighbors. This is an audible event. In the middle of the 1960s, researchers pressed sensitive detectors against drying trunks and managed to count the ultrasonic clicks given off as the columns snapped. A tree running out of water crackles, in the literal sense of the word.

Trees have defenses against this. The first is sheer number: millions of conduits run in parallel in the trunk, and the loss of a few hundred does not stop the flow. The second is design. In conifers the pit membrane between cells carries a thick plug at its center; when the neighboring cell fills with air, the pressure difference shoves that plug against the mouth of the pit and stoppers the hole, sealing the blockage off before it can spread. The third is repair. Gas bubbles released from water that froze over winter leave much of the tree blocked come spring, and this is where root pressure genuinely earns its keep: it is used to refill the columns. That spring flow, the source of maple syrup, is in fact a repair operation.

For all these defenses, a tree still has to place a bet on how close to its breaking point it is willing to run. Close its stomata early and it stays safe but goes hungry; close them late and it grows but blocks its pipes. A wide-ranging review published in 2012 pooled data on 226 forest species from 81 different regions of the world, and the result was striking: 70 percent of the species were operating with a safety margin narrower than 1 megapascal from their blockage threshold. What is more, this did not distinguish wet tropical forest from arid scrub. Forests, wherever they grow, appear to have chosen to live right at the edge of their own limits.

Which brings us to the ceiling on height. In a study led by Koch and colleagues in 2004, researchers climbed California's tallest redwoods on rope and harness, among them the tallest known at the time, a tree of 112.7 meters, a few meters shorter than today's record holder. What they measured at the top was this: even with the soil soaking wet and the roots under no strain at all, gravity and pipe friction alone left the tension inside the topmost needles about 1 megapascal greater than in the needles below. And the marks of that difference were written into the leaves themselves. The higher you went, the shorter and thicker the needles became, the heavier per unit area, the more tightly they pressed against the stem. The reason is simple: a leaf grows by its cells taking up water and swelling. The tension at the crown is so high that the cells cannot swell enough, and the leaf cannot expand. And when the leaf cannot expand, photosynthesis falls. When the researchers extended these height-dependent curves, they found a crossing point: the height at which leaves could no longer expand at all, and therefore at which the tree could not invest another centimeter upward, came out between 122 and 130 meters.

So the tallest trees alive today have struck the ceiling. And what sets that ceiling is not the strength of the wood, not wind, not soil, not nutrients. It is whether a handful of needles at the very top can pull enough water from a stretched thread of water to swell.

That limit carries a new meaning now. As the air warms, its drying power rises even at the same humidity, which means the pull bearing down on the leaf, and so the tension in the column, increases. For trees whose safety margins were already narrow, that amounts to a systematic drift toward the breaking threshold. It is what happened in the great Amazon droughts of 2005 and 2010: across wide areas, trees died not of thirst but because their conducting systems filled with air. Even a tree with its roots in wet soil becomes unable to reach water once the column in its pipes has parted.

Engineering's attempts to imitate this plumbing are still at the crawling stage. In 2008 researchers managed to hold water at a stable tension of around minus 1 megapascal inside a tiny structure built from a gel with nanometer-scale pores, and to make it flow by evaporation: a working artificial tree in the laboratory. But that device was a few centimeters across, and it lasted hours rather than weeks. Nobody has yet built a version 100 meters tall that repairs itself and keeps running for 2,000 years. The forest does it every summer, quietly, without a single moving part, on nothing but the drying power of the Sun and the stubbornness of water molecules that will not let go of one another.

Nöron Science'de bugünkü bölümün sonuna geldik. Yeni bir hikâyede yeniden buluşmak üzere.