Field notes on things that run themselves
The Air Moves Because the Day Does
In a shaded forest near Bangalore stands a mound of clay a metre or two high, ridged with vertical buttresses called flutes. Underneath, out of sight, is the colony: termites a few millimetres long, and the fungus they farm, which consumes more oxygen than they do. The mound above is nearly empty. No fan, no pump, not one muscle assigned to moving air. The air moves anyway, and it took a century to work out what pushes it.
Something has to. Carbon dioxide takes about four days to diffuse two metres through still air, and a colony breathing hard underneath cannot wait four days. Ventilation here has to mean bulk flow — actual air travelling — and every account of what drove it rested on inference.
There were two candidates. In 1961 Martin Lüscher proposed a steady thermosiphon: colony heat warms the core, warm air climbs the chimney, cool air enters below. Elegant — and the version that reached architecture. Four decades later J. Scott Turner, running tracer gases through Namibian mounds, argued the density differences were far too small for that, and that the mound regulated gas, not temperature. He proposed wind instead: gusts sloshing air through the walls, more breathing than chimney.
Nobody had measured the air inside a mound directly, because it is hard. The flows run at centimetres per second, too slow for commercial instruments; a sensor that stays warm generates its own convection and measures itself; and termites cement over any foreign object within about ten minutes.
In 2015 Hunter King, Samuel Ocko and L. Mahadevan built a probe for it: three thermistor beads, each three-tenths of a millimetre across. A pulse through the middle bead releases a bolus of warm air; the outer two watch which way it drifts. They drilled into twenty-five live mounds, worked a few minutes per reading to stay ahead of the repair crews, and never measured a flute twice.
A schedule appeared, across seventy-eight flutes. Through the day, air drifts gently up the outer flutes. At night it pours down them, hard enough to run off the end of the calibration. Inside the nest, carbon dioxide climbs through the daylight hours to nearly six percent, while the chimney above holds below one. Around dusk the profile inverts, the circuit turns over, and the mound flushes.
Both old answers failed on the same trip. An abandoned mound kept the same daily gradients and flows as the living ones: no metabolism required. Then they measured a cone cut from a mound wall — thirty-seven to forty-seven percent air by volume, pores about five microns wide. A breathable windbreaker. Gas crosses it freely by diffusion; pressure cannot push air through it. Wind could force at most a hundredth of a millimetre per second across it; a powerful fan aimed at a mound changed nothing inside.
What is left is the shape of the thing sitting in the weather. The flutes are thin and follow the outside temperature closely; the core is massive and lags behind. So the flutes run warmer than the centre in the afternoon and colder before dawn; air rises on whichever side is warm, sinks on the other, and the loop reverses twice a day.
Notice what the mound is not living on. Over a full day the average difference between flute and core is close to nothing; the mound ends where it started. It has no gradient to spend. What it has is a difference that keeps changing sign, and a geometry that turns the changing into transport. Issue No. 40 described a room selecting one frequency out of broadband noise; this is a structure taking one frequency and rectifying it. Nearly everything covered here runs on a difference. This runs on a difference that will not sit still.
The physics is still rough, and the authors say so. Model the mound as a simple convective loop and you predict about thirty-five centimetres per second, ten times what anyone measures; the real interior is a tangle of bottlenecks. The mechanism is established, the arithmetic is not. And the much-repeated claim that Harare’s Eastgate Centre copies a termite mound rests on Lüscher’s version — the one the measurements displaced.
Two years later the same group went to Namibia, to Turner’s species and Turner’s field site, and did it again on Macrotermes michaelseni — which builds no flutes. The mechanism held. The sun added a second rhythm: east flank warm in the morning, west in the afternoon, the south face cool all day. Turner and Rupert Soar are co-authors on that paper. Wind was not absent, only demoted.
Which leaves the mound where the best of these systems end up. Give it a day with no day in it — bring the mound and the air around it into equilibrium — and the ventilation stops. It is not built to survive the fluctuation outside. It is built out of it.
One loop I’m watching
Next: thirty kilometres up over the equator, the wind reverses direction, east to west and back, about every twenty-eight months. Nothing outside sets that period — not the year, not the sun, not the tides. Waves rising out of the weather below deposit their momentum wherever the shear lets them break, which drags the shear zone downward, which changes where the next waves break. A pattern that regenerates itself at the top and erases itself at the bottom, keeping a clock nobody wound.
Tip: the ← and → arrow keys move between issues.
Prefer feeds? Follow by RSS or JSON Feed.
Want something playable between issues? Thought Toys has its own email list and RSS feed.
New to The Standing Wave? Start here →