This companion sim traces the fire perimeter itself instead of a grid of cells. A single ignition point grows outward every frame as a Huygens-wavelet fire ellipse — the analytic front-tracking method (Richards 1990) used by real spread-prediction tools such as FARSITE — rather than a cellular automaton. In uniform wind, an unobstructed ignition point always grows into an ellipse with the ignition point sitting at one focus, not the centre, so the head and the back of the fire race away from the origin at different rates:
R(θ) = R_head · (1 − e) / (1 − e·cos θ) θ measured from the downwind bearing
R_back = R(π) = R_head · (1 − e) / (1 + e)
The eccentricity e comes from the fire's length-to-breadth ratio, itself a real published empirical fit — Anderson's (1983) formula against effective wind speed U (mph), also the ratio BehavePlus and FARSITE use internally:
L∶B = 0.936·e^(0.2566U) + 0.461·e^(−0.1548U) − 0.397
e = √(L∶B² − 1) / L∶B
Rhead itself is scaled by the same Rothermel moisture damping coefficient ηM as the 3D companion sim — sampled locally at each point on the growing perimeter, not just at the ignition cell:
η_M = 1 − 2.59·r + 5.11·r² − 3.52·r³, r = M_f(x,y) / M_x
η_M = 0 once r ≥ 1 (that stretch of perimeter cannot advance)
Because moisture is sampled independently along every bearing from the ignition point, the perimeter genuinely stalls and dents inward wherever it meets a damp pocket while the rest of the front keeps racing ahead — a distinctly different failure signature from the 3D grid's blocky cell-by-cell extinguishing, even though both share the same underlying physics.
- Mf slider — raises moisture everywhere; push the head-fire direction past Mx≈30% and Rhead collapses to zero.
- Wind / Slope — combined into one effective wind vector that sets both the L∶B ellipse shape and the head-rate multiplier — a genuinely different combination rule from the 3D sim's separate φw·φs multipliers.
- Ignite — starts a single point fire at the upwind edge; Reset Fuel Bed reseeds the moisture terrain and clears the burn.
Simplified teaching adaptation — real analysts still need fuel load, surface-area-to-volume ratio and heat content to get quantitative rates; the elliptical growth method itself, however, is exactly the one professional tools use to propagate a fire perimeter through time.