Model. Two chemicals u (food) and v (catalyst) obey the Gray-Scott equations ut = Du∇²u − uv² + F(1−u), vt = Dv∇²v + uv² − (F+k)v. The step U + 2V → 3V is autocatalytic: v makes more v from u. F feeds fresh u and washes both out, k removes v.
Why patterns appear. Alan Turing (1952) showed that two chemicals that diffuse at different speeds (here Du = 2 Dv) can turn a flat mixture into spots or stripes. In Gray-Scott most patterns grow from a finite-size seed that invades the empty state u = 1, v = 0, which is stable to small noise, so the linear Turing band is only part of the story. The F-k plane shows where non-trivial uniform states exist at all.
Numerics. Explicit Euler, 5-point Laplacian, periodic square box, grid units dx = dt = 1, Du = 0.2, Dv = 0.1 (Pearson 1993 uses 2e-5 and 1e-5 m²/s on a 0.01 m grid). Stability needs 4 D dt/dx² ≤ 1 (here 0.8).
Honest simplifications. The dish is a round window onto a periodic square, not a no-flux wall; real Gray-Scott chemistry is run in gel reactors (Lee, McCormick, Ouyang, Swinney 1993) and the pattern names here describe what appears in this build. The parameter pairs are typical Pearson-map values; the appearance is sensitive to them, to the seed and to the numerics. The agar picture is decoration (Blender render), not data. Colours: amber agar is v = 0, blue-violet dye is high v.
Independent checks (run on load, the same core drives the 3D page):
| running… |