A willow by still water, from a seed to one last leaf, with its music.
An aesthetic simulation. It is a piece of music with a picture, and it is not an experiment. A willow grows from a seed, meets the wind, loses its leaves in a storm, grows them again, and is sung to by birds. It lasts a little over six minutes. Sound begins only when you press Play.
The score is the one written for the first version of this piece: the same notes in D minor at 66 beats a minute, and the same moments for every gust, creak, click and drip. The instruments are the same too. What has changed is the engineering. Notes now start on the audio clock, not on the next frame of the picture; the sound is in stereo; clicks at the starts of sounds are gone; and a limiter keeps it from clipping.
The robins take the rise and fall of their phrases from three songs Doris Day sang. The parrot’s song, “What a Lovely Day”, was written for the piece. It is sung twice. The second time a quiet second voice hums a third below, which is new, and the piece now ends when the song does.
A weeping willow’s young shoots are long, thin and limp, so they hang. A hanging shoot swings as a pendulum does, more slowly the longer it is. The force of the wind on a leaf grows with the square of the wind’s speed, and leaves and shoots give way and stream with it, which lowers the force on the tree (Vogel, 1989; de Langre, 2008).
Leonardo noticed that where a branch divides, the cross-sections of the parts add up to that of the whole. Eloy (2011) showed that a tree built to resist wind evenly would come out that way.
Water is drawn up a tree under tension (Dixon and Joly, 1894). When the tension breaks a thread of water, the vessel fills with vapour. Milburn and Johnson (1966) heard these breaks as clicks; most of the sound is too high for us to hear (Tyree and Dixon, 1983; Tyree and Sperry, 1989).
A drop that takes a bubble of air under water rings at a pitch set by the bubble’s size (Minnaert, 1933).
The picture follows the score. The wind you see is built from the wind you hear; a creak bends a limb; a click of cavitation shows as a point of light in the wood; each drip is a drop that falls and rings the water. The shoots are chains hung from the branches and pushed by that wind. The branch radii are set by Leonardo’s rule.
One thing is not scored: the turning of the leaves. Each leaf adds up the wind’s push on it, turns its pale side out when the sum reaches a fixed amount, and starts again from nothing. The timing rule here is a count, a threshold and a restart, which is the grammar of the CRR programme (coherence, rupture, regeneration). That programme is under investigation. With annotations on, the page shows the spread of the times between turns as measured in the run you are watching, beside the programme’s candidate value of 0.16. Nothing has been adjusted to bring them together.
Almost everything about a real tree. This one is flat, with three layers standing in for depth, and it grows in two minutes. Its wind is a few travelling waves and has no eddies. Its leaves do not shade or touch one another. The storm takes every leaf and they are back within a minute, which no willow manages. The clicks fall where the music wants them. Robins do not sing Doris Day, and parrots do not live in English willows.
The new score was compared with the original’s, event by event, on the same random numbers: 1,085 events and no differences. The original page was then run on a simulated clock. Its 1,044 sounding events came between 0 and 17 thousandths of a second later than the score asks, because each waited for the next frame of the picture; here they are on time. The song has the same 26 syllables at the same pitches and lengths. A render of the whole piece peaks at 5 decibels below full scale, and at 1.3 below with every mixer slider at its maximum. Over the whole piece the wind heard and the movement of the shoots seen correlate at 0.82. In a test run the spread of the leaf-turn times came out between 0.21 and 0.45, depending on the movement; the candidate value is 0.16. One check failed: on a busy test machine with software graphics, the script for one frame took 6 thousandths of a second at the median against a bet of 4.
Sources, cited from memory; please check before relying on them. Dixon, H. H. and Joly, J. (1894), On the ascent of sap, Phil. Trans. R. Soc. B. Minnaert, M. (1933), On musical air-bubbles and the sounds of running water, Philosophical Magazine. Milburn, J. A. and Johnson, R. P. C. (1966), The conduction of sap II: detection of vibrations produced by sap cavitation in Ricinus xylem, Planta. Tyree, M. T. and Dixon, M. A. (1983), Cavitation events in Thuja occidentalis L.? Ultrasonic acoustic emissions from the sapwood can be measured, Plant Physiology. Tyree, M. T. and Sperry, J. S. (1989), Vulnerability of xylem to cavitation and embolism, Annual Review of Plant Physiology and Plant Molecular Biology. Vogel, S. (1989), Drag and reconfiguration of broad leaves in high winds, Journal of Experimental Botany. de Langre, E. (2008), Effects of wind on plants, Annual Review of Fluid Mechanics. Eloy, C. (2011), Leonardo’s rule, self-similarity and wind-induced stresses in trees, Physical Review Letters.