Presets
Playback
Parameters
Camera
Layers
Quality
Model check
The science
What is modelled
Two plants release the same number of seeds into the same wind. One cohort keeps its feathery parachute (the pappus), the other loses it. Every seed is a point mass with quadratic air drag that falls through a turbulent, height-dependent wind, so you can watch how a tiny parachute turns a drop of one metre into tens of metres of travel.
Equations
- Seed:
m dv/dt = m g + ½ ρ (CdA) |u − v| (u − v), with air density ρ = 1.2 kg/m³. - Terminal speed in still air:
vt = sqrt(2 m g / (ρ Cd A)). The pappus is modelled as an effective disc of diameter d with Cd = 1.2; the naked achene as a 3 mm x 0.8 mm body with Cd = 1.0 and mass 0.5 mg. - Reynolds number
Re = vt d / νwith ν = 1.5e-5 m²/s; for the default seed it is about 200, the regime where a separated vortex ring forms behind a porous disc (Cummins et al., Nature Communications 2018). - Mean wind, neutral surface layer:
U(z) = U10 ln(z/z0) / ln(10/z0)with roughness length z0 = 0.03 m (short grass). - Turbulence: an Ornstein-Uhlenbeck process per seed,
du' = −u'/τ dt + σ sqrt(2/τ) dW, with σ = I U(z) horizontally, half of that vertically, and τ = z / max(U, 0.5) limited to 0.3..4 s. A thermal adds a steady updraft W. - Expected drift (no turbulence):
X = ∫0h U(z) dz / (vt − W). This is the yellow line on the histogram. If W exceeds vt the seed never comes down.
Checks run in the page
- Time-stepped terminal speed against the closed form
sqrt(2mg/(ρ Cd A))within 1 %. - Drift in a uniform 2 m/s wind against
U h / vtwithin 6 %. - Sample variance of 1.5 million Ornstein-Uhlenbeck steps against the target σ² within 6 %.
Honest simplifications
Seeds are points: no rotation dynamics, no lift or flutter, a constant drag coefficient instead of one that depends on Reynolds number, and independent turbulence for every seed (real gusts are correlated across the plant). The seeds are drawn at the chosen magnification (default 10 times real size); the real pappus is about 1 cm across. The wind blows in one fixed direction over flat grass, the ground is a smooth hillock, and seeds that pass 110 m are counted as escaped. Numbers such as 0.6 mg and 9 mm effective diameter are round, plausible values, not a measurement of one species.
What to try
- Choose Still air: both cohorts land within a few centimetres of their plants; the parachute only lengthens the time aloft (compare the tracked seeds).
- Set 6 m/s: parachute seeds fly several times farther than naked ones, and the histogram spreads out.
- Add a thermal of 0.35 to 0.7 m/s: updraft cancels part of the fall speed and a share of seeds soars beyond 100 m.
- Shrink the pappus to 4 mm and watch the terminal-speed curve (graph 3) climb towards the naked seed.
- Use the Seed close-up camera to look at the 130 bristles of the Blender seed model.