Column height is not decorative — it follows the empirical Mastin et al. (2009) plume-height law used in real ash-transport forecasting, driven only by the mass eruption rate (MER) the "Ash output" slider maps to on a log scale:
H (km) = 2.00 · MER(kg/s)^0.241
umbrella (neutral-buoyancy) height ≈ 0.76 · H
Below the umbrella altitude, ash rides the buoyant column and drifts with the wind only a little; above it, ash detaches into the spreading umbrella cloud and is carried fully by the wind while gravity settles it. Fall speed is not a fixed number either — it is solved from a drag-based terminal-velocity balance (grain diameter from the "Grain size" slider, particle density 1000 kg/m³) iterated against the Reynolds-number-dependent drag coefficient:
Re = ρ_air·v·d / μ
Cd = 24/Re (Re < 1, Stokes)
Cd = (24/Re)(1+0.15Re^0.687) (1 ≤ Re < 1000)
Cd = 0.44 (Re ≥ 1000, turbulent)
v = sqrt( 4·g·d·(ρ_p−ρ_air) / (3·ρ_air·Cd) )
Fine ash (<0.1 mm) settles at centimetres per second and travels hundreds of kilometres; coarse lapilli (>5 mm) fall at several metres per second and land within a few kilometres of the vent. The deposit-thickness panel accumulates where every particle lands, so its shape — a sharp near-vent peak thinning exponentially downwind — is a genuine output of the plume-height and settling-velocity physics above, matching the classic isopach thinning trend reported for real eruptions, not a hand-drawn curve.