Sources
The fire in Fireline follows published fire-behaviour research, and the tactics follow wildland firefighting doctrine. Each entry lists what the game does, the source, and what was simplified.
Every cell of the map has a fuel type, a slope and the wind at that moment. From them the game computes how fast the fire advances, in metres per minute: a grass fire with a 6 m/s wind runs at about 20 m/min, timber litter at 2–3.
Simplified · We use Rothermel’s structure (a no-wind rate multiplied by 1 + wind factor + slope factor) but with three fuel classes tuned by hand instead of the full equations. The full model, with fuel moisture, arrives in phase 2.
The map has grass, shrub and timber litter. Grass burns fast and goes out fast; shrub burns hot; timber is slow but smoulders for hours.
Simplified · Three classes stand in for the 40 standard fuel models. Phase 2 swaps them for the real Scott & Burgan set.
Wind and uphill slope are added as arrows. A fire going uphill with the wind spreads fastest; downhill against the wind it spreads slowly.
Rothermel (1972), slope factor φs = 5.275 β^−0.3 tan²φ.
Albini, F. A., & Baughman, R. G. (1979). Estimating windspeeds for predicting wildland fire behavior. Res. Pap. INT-221. USDA Forest Service.
Simplified · The slope factor is fixed at 5 tan²φ (capped) and the wind at mid-flame height is a fixed fraction of the 10 m wind per fuel class.
A fire in open fuel grows as an ellipse stretched downwind. The head, moving with the wind, is fastest; the flanks and the back are slower. Stronger wind makes the ellipse longer and narrower.
Anderson, H. E. (1983). Predicting wind-driven wild land fire size and shape. Res. Pap. INT-305. USDA Forest Service.
Simplified · We use Anderson’s length-to-breadth formula from the effective wind speed, exactly.
The game finds, for every cell, the earliest time the fire can reach it along any path. That arrival time is also what draws the projection lines and warns crews about their escape routes.
Finney, M. A. (1998). FARSITE: Fire Area Simulator — model development and evaluation. Res. Pap. RMRS-RP-4. USDA Forest Service.
Simplified · A 30 m grid with 16 neighbours per cell, recomputed every 15 game minutes so wind shifts take hold.
Shrub and timber burning with flames over 2 m throw embers 150–650 m downwind. This is how fire crosses a control line.
Albini, F. A. (1979). Spot fire distance from burning trees — a predictive model. Gen. Tech. Rep. INT-56. USDA Forest Service.
Simplified · The distance and the chance are drawn at random within that range instead of from Albini’s lofting model.
Crews on foot can work flames up to 1.2 m, dozers and engines up to 2.4 m, aircraft up to 3.4 m. Above that, only indirect attack works. Selecting a unit colours the fire by these four bands.
Byram, G. M. (1959). Combustion of forest fuels. In K. P. Davis (Ed.), Forest Fire: Control and Use (pp. 61–89). McGraw-Hill. Fireline intensity I = H·w·R and flame length L = 0.0775·I^0.46.
Andrews, P. L., & Rothermel, R. C. (1982). Charts for interpreting wildland fire behavior characteristics. Gen. Tech. Rep. INT-131. USDA Forest Service. (The 4 / 8 / 11 ft thresholds of the “hauling chart”.)
Simplified · Flame length is taken from the spread rate perpendicular to the front, so heads are hot and flanks are workable, as they are in the field.
A hand crew digs about 15 m of line per minute in grass, less in shrub and timber. A dozer does about 60 in grass but can’t work slopes over ~45%.
Broyles, G. (2011). Fireline production rates. USDA Forest Service, San Dimas Technology and Development Center, 1151-1805-SDTDC.
Simplified · Single rates per resource with a multiplier per fuel class; the real tables vary with crew type and terrain.
The helicopter cycles between the fire and the nearest water, so where you send it matters. The tanker lays up to 600 m of retardant that slows the fire for hours, but not under very tall flames. Aircraft don’t fly at night.
NWCG Standards for Aviation (PMS 520) and the USDA Forest Service Aerial Firefighting Use and Effectiveness (AFUE) study (2020).
Simplified · Retardant cuts spread to 8% where flames are under 3.4 m and to 60% above that. Real effectiveness depends on coverage level and fuel.
A line must start on something that can’t burn (a road, water, rock, ground that already burned) or the fire hooks around its end. The game shows “anchored” or “no anchor” while you draw.
NWCG Glossary of Wildland Fire (PMS 205): anchor point, flanking, indirect attack.
Simplified · This isn’t scripted: the simulated fire spreads around the end of an unanchored line.
An edge only counts as held when it sits against a barrier and has gone cold. Heavy fuel stays hot for hours and can throw embers back across a line unless it is mopped up.
NWCG Glossary of Wildland Fire (PMS 205): contain, control, mop-up.
Simplified · Cooling times are 25 min for grass, 100 for shrub and 6 h for timber. The rekindle chance is a flat 0.4% every 10 minutes per hot cell touching a line.
Crews pull back on their own when nearby flames exceed their limit, to cold burned ground or to unburned ground well away from the fire, along a path the fire won’t cut. A lookout warns you when the fire is about to reach a unit. A crew that is burned over is out of action and has injuries.
Gleason, P. (1991). LACES: Lookouts, Communications, Escape routes, Safety zones.
NWCG Incident Response Pocket Guide (PMS 461).
Simplified · Communications are assumed perfect in Easy and Normal. Hard mode will make the picture lag behind reality.
Ground crews slow down after 12 hours on the line and go off shift at 16.
NWCG Standards for Interagency Incident Business Management (PMS 902): 2:1 work/rest and 16-hour shift guidance.
Simplified · Rest is a flat 8 hours at the station.
Fine fuels dry out through the day and pick up moisture at night, so the fire spreads fastest from late morning to evening and slows after dark.
Rothermel, R. C. (1983). How to predict the spread and intensity of forest and range fires. Gen. Tech. Rep. INT-143. USDA Forest Service.
Simplified · A daily multiplier on spread rate (1.0 in the afternoon, 0.45 at night) instead of computing fuel moisture hour by hour. Phase 2 replaces it.
The scenario’s front swings the wind from southwest to southeast in about 20 minutes. The east flank becomes the head.
Scenario written for the game, following the classic wind-shift pattern described in fire-weather training (NWCG S-290).
Simplified · Wind follows a fixed script with a forecast on the radio. Uncertainty in the forecast comes later.
Hourly costs, the $35k evacuation, the grading and the number of people hurt when fire reaches a town that wasn’t cleared are game numbers, not data.
To be calibrated in phase 3 against published USDA Forest Service and CAL FIRE cost data.
Simplified · They are set for gameplay and aren’t based on real data.
The valley, its towns and its roads are fictional, generated from noise. The 30 m cell matches the resolution of common US fire-behaviour landscape data.
LANDFIRE program (USGS/USDA), 30 m fuel and terrain layers, as the reference for grid size.
Simplified · No real place is modelled.
Fireline · sgura.app ↗