Fire and Arson Investigation: Cause, Origin and Burn Patterns

The fire triangle is the bedrock model. Combustion needs a fuel, an oxidiser (almost always atmospheric oxygen at about 21 percent), and enough heat to bring the fuel above its ignition temperature. Remove any one side and the reaction stops. The fire tetrahedron extends this with a fourth side, the uninhibited radical chain reaction inside the flame, which is the side that halon and dry-chemical extinguishers actually attack.

Three temperature thresholds are routinely confused and must be kept distinct.

1. Flash point. Lowest temperature at which the liquid gives off enough vapour for the vapour-air mixture above it to flash on contact with an ignition source. Combustion does not sustain. Petrol about minus 43 degrees Celsius, kerosene about 38 to 72 degrees Celsius, diesel about 52 to 96 degrees Celsius.
2. Fire point. Slightly higher than flash point, the temperature at which the vapour-air mixture, once ignited, keeps burning without further input.
3. Ignition temperature (auto-ignition temperature). Minimum temperature at which the substance ignites by itself without any spark or flame. Petrol auto-ignites around 280 degrees Celsius, paper around 230 degrees Celsius, wood around 300 degrees Celsius.

Combustion modes also matter.Flaming combustion is rapid gas-phase oxidation with a visible flame above the fuel surface, fed by pyrolysis vapours.Smouldering combustion is slower, lower-temperature, surface-phase oxidation without a flame (cigarette in upholstery, peat fire), which can run for hours before transitioning to flaming.

Heat moves through a compartment by three modes.Conduction through solids (a steel beam carrying heat from one room into the next).Convection through fluids, dominant in a building fire because hot combustion gases rise as a buoyant plume and entrain cool room air on the way up, then form a horizontal ceiling jet when the plume hits the ceiling and spreads outward.Radiation through electromagnetic waves, which carries heat across air without contact and is what ignites secondary fuels a metre or two away from the original burning object.

A compartment fire (a fire confined to a room) goes through four stages: incipient, growth, fully developed, and decay. The transition from growth to fully developed is flashover. As the ceiling jet and hot gas layer at the top of the room heat exposed combustible surfaces (curtains, polyurethane sofa, paper stacks) by radiation, those surfaces release pyrolysis vapours. When the gas-layer temperature reaches roughly 600 degrees Celsius and the radiative flux to the floor reaches about 20 kilowatts per square metre, every exposed combustible surface in the room ignites almost at once. After flashover, the room is fully involved and burn patterns become much harder to interpret because everything is burning at once.

Backdraft is a different beast. It happens in an under-ventilated compartment where the fire has consumed the available oxygen but combustion gases (carbon monoxide, unburnt hydrocarbons) have continued to accumulate. When a door or window is suddenly opened (often by a first responder), fresh oxygen is admitted to the rich gas mixture and the whole room can deflagrate or explode outward. The classic warning signs are blackened windows with yellow-brown stains, pulsing smoke through small gaps, and inward air rushes felt at the door. The key distinction is that flashover is a thermal threshold event, while backdraft is a ventilation event.

Burn-pattern interpretation is the heart of the fire and burn pattern interpretationworkflow. The patterns are read in combination, not in isolation, because no single pattern is pathognomonic for arson.

V-pattern. The classical origin indicator. On a vertical surface above a point source, the buoyant plume of hot gases spreads upward and outward, charring the wall in a V shape whose narrow apex is at or just above the point of origin. The angle of the V depends on heat-release rate and ventilation, not (as older textbooks claimed) on the use of an accelerant. A steep, narrow V means the fire grew fast; a wide, shallow V means slower growth or a ventilated room.

U-pattern. Similar to V-pattern but with a rounded base, often produced when the heat source is offset from the wall and radiant heat dominates over direct flame impingement.

Hourglass pattern. Forms when the base of the flame zone (cool, oxygen-starved core) intersects a vertical surface, leaving an inverted-triangle cool zone at the bottom and the upward-diverging V above.

Truncated-cone pattern. A three-dimensional version of the V-pattern. Visible on free-standing objects and in cross-section through char layers.

Pour patterns and irregular trailers. Liquid accelerant poured on a floor leaves an irregular, low-burn pattern that follows the puddle shape, often with sharp-edged margins where the liquid soaked into porous flooring (carpet, wood). Long, narrow trailers (a line of low burn linking two rooms) suggest a deliberately laid liquid trail. NFPA 921 warns that flashover and full-room involvement can mimic pour patterns by producing low-level burns on floors, so a pour pattern must be confirmed by laboratory analysis of the floor sample for ignitable liquid residue under ASTM E1618.

Multiple unconnected points of origin. Two or more separate fires in different rooms with no physical or radiative path between them is one of the strongest single indicators of arson. Investigators rule out drop-down (burning debris falling from above) and electrical fault propagation before classifying as incendiary.

Spalling on concrete. Loss of surface chips from a concrete floor or wall under rapid heating. Can be caused by an accelerant pool, but also by rapid water cooling during firefighting and by the natural pop-out of moisture in poorly cured concrete. Spalling is not pathognomonic for arson;, the correct framing is "consistent with but not diagnostic of accelerant use".

Char depth and alligator pattern. Char depth (millimetres of charred wood) was historically used to estimate burn duration at a point. Modern NFPA 921 treats it as a relative comparison tool within one fire, not an absolute timer. "Large shiny alligator" versus "small dull alligator" char interpretations (where large blistering was claimed to indicate accelerant) are now considered scientifically unreliable; the validator for an arson opinion is laboratory residue analysis, not blister size.

NFPA 921 (NFPA Guide for Fire and Explosion Investigations) is the methodology Indian fire investigators and SFSL fire divisions now follow as the international standard.

1. Recognise the need
   An incident has occurred and a fire-cause determination is required for insurance, prosecution or public-safety reasons.
2. Define the problem
   What exactly is being determined: origin only, cause only, classification (accidental, natural, incendiary, undetermined), or all of these.
3. Collect data
   Scene examination, witness interviews, photographs, samples, prior incident reports, weather data, building plans.
4. Analyse the data
   Reconstruct fire growth from burn patterns, electrical-system condition, fuel load, ventilation. Apply fire-science principles.
5. Develop and test hypotheses
   Generate every plausible hypothesis for origin and cause. Test each against the physical evidence by deductive reasoning. Reject hypotheses that contradict any datum.
6. Select the final hypothesis
   Only one hypothesis survives the testing step. If two or more survive equally, the cause is classified as undetermined.

In parallel, the scene work itself runs through a fixed origin-to-cause workflow. First an exterior survey of the building (where the fire vented, what neighbours saw, where the heaviest exterior damage is). Then an interior survey room by room, working from the area of least damage to the area of most damage, because the area of most damage is usually (not always) close to the area of origin. Inside the worst room, the investigator identifies the area of origin(a few square metres), then the point of origin(a defined spot or object), then the ignition source(electrical fault, lighter, candle, friction spark), then the first fuel ignited(the substance that took the initial heat and produced sustained combustion), and finally the cause as a combination of source and first fuel.

Sample collection at the scene follows photography under the forensic photographySOP, then debris sampling into clean unused metal cans or nylon bags (not glass jars, which can break in transit; not polythene, which is permeable to volatile residues). A separate comparison sample of unburned flooring is collected from outside the suspected pour area. Every can goes into the chain of custodyregister and on to the SFSL fire division for headspace GC-MS analysis under ASTM E1618.

NFPA 921 classifies every determined cause into one of four categories.

1. Accidental. The most common. Electrical faults (loose connection, overheated conductor, short circuit with arcing), cooking fires (unattended pan, overheated oil), smoking materials in upholstery, candles, children playing with matches or lighters, heater contact with fabric.
2. Natural. No human action. Lightning strikes, spontaneous combustion (linseed-oil-soaked rags self-heating to ignition, hay-stack fires, coal-storage fires), focused sunlight through a glass object.
3. Incendiary (arson). Fire set deliberately. Indian context includes dowry-related kerosene-douse cases, insurance fraud (shop and warehouse fires), revenge fires, and riot-related arson.
4. Undetermined. When more than one hypothesis survives the NFPA 921 test step, or when the scene is too disturbed by post-fire activity to support any one hypothesis to the exclusion of others.

Electrical-fault signatures are a recurring examiners question. When a copper conductor melts in a fire, the appearance of the bead distinguishes cause from indicator.Arcing beads(formed by an energised short circuit at the time of the fire) are smooth, spherical, and have a sharp boundary with the unmelted conductor; they indicate the conductor was live and the arc happened.Fire-melt beads or spatter(formed by external heat without electrical involvement) are irregular, often elongated, with a gradual transition zone. The Carlson and Hagimoto criteria (microstructural grain analysis under metallography) refine this further. Crucially, the presence of an arc bead is an indicator that current was flowing and an arc occurred at that point; it is not by itself proof that the arc caused the fire, because the fire could have damaged a previously sound circuit and produced the arc as a secondary event.

Spontaneous combustion is the textbook example of a natural cause. Linseed-oil-soaked rags piled in a bin oxidise exothermically; if the pile is large enough that the centre cannot lose heat to the surroundings, the temperature climbs to the auto-ignition temperature of the oil and the pile bursts into flame. Hay-stack fires and wet coal storage follow the same physics.

The Indian fire-investigation casebook has several recurring themes relevant to practicing investigators.

Stove-burst pattern in dowry-death cases. A long-standing diagnostic problem in Indian forensic pathology and fire investigation. The scene narrative offered by in-laws is that a kerosene stove burst while the woman was cooking and her saree caught fire. Genuine kerosene-stove bursts do happen, but the burn-pattern signature differs from a kerosene-douse arson: a true stove burst produces a tight low-level burn radius around the stove, splash patterns on adjacent surfaces from pressurised fuel, and intact peripheral floor; a douse-and-ignite produces a wider pour pattern, soaked-in flooring, and (often) absence of fuel inside the stove tank. The legal frame for death of a woman within seven years of marriage in connection with cruelty or harassment over dowry is BNS 2023 Section 80(corresponding to the older IPC Section 304B). Bihar, UP and Rajasthan have produced the bulk of these cases over the past three decades.

Major Indian fire incidents in the syllabus area.

• Uphaar Cinema fire, Delhi, 1997. Transformer fire in the basement; 59 deaths from asphyxiation (carbon monoxide and toxic smoke) across the cinema, with 34 of those deaths occurring in the upper balcony. Triggered the modern Indian focus on cinema-hall fire safety and the criminal-negligence prosecution that ran for two decades.
• AMRI Kolkata fire, 2011.Basement-stored combustibles in a hospital; 89 deaths, most from smoke inhalation. Led to revisions of hospital fire-safety norms.
• Surat coaching-centre fire, 2019. Takshashila Arcade building, illegal additional storey, single staircase, no fire NOC; 22 student deaths. Reinforced the NBC 2016 enforcement push.

Statutory framework. Property offences involving fire or explosives are charged under BNS 2023 Sections 324 to 326(the modern equivalents of IPC Sections 425 to 438 dealing with mischief by fire or explosive substance). Scene investigation in serious cases follows BNSS Section 176(3)which makes forensic-team attendance mandatory for offences punishable with seven years or more. Building-side compliance is governed by the National Building Code (NBC) 2016 with fire-detection-system standards under IS 2189 fire-extinguisher standards under IS 15683 and water-based suppression systems under various IS 12469 / IS 13039 sub-parts.

Court evidence. The fire investigator gives expert opinion under BSA 2023 Section 39(the modern equivalent of the Indian Evidence Act Section 45), and the analyst from the SFSL fire division corroborates the residue analysis under the same section. Defence cross-examination typically attacks origin determination, alternative hypotheses (electrical fault as the real cause), and chain-of-custody for the debris samples.

Indian institutional landscape. The Central Forensic Science Laboratory at Hyderabad runs a fire-investigation division that handles serious or high-profile cases. State SFSLs run their own fire divisions of varying capacity. The Bureau of Police Research and Development (BPRD) publishes investigation manuals. The National Forensic Sciences University (NFSU) at Gandhinagar runs postgraduate and certificate programmes in fire investigation.