Fire load density

Fire load density is the fire load of a space divided by an area, usually the floor area. The fire load is the energy that would be released if all combustible materials in the space burned completely. It is normally given in megajoules per square metre of floor (MJ/m²), although some methods relate the fire load to the total area of the walls, floor and ceiling instead. It can also be expressed as an equivalent mass of wood per square metre. In New Zealand the same quantity is called the fire load energy density (FLED).

The fire load density tells engineers how much fuel a fire can consume, not how fast it burns. It therefore mainly governs how long a fully developed fire lasts and how severe it is for the structure and for fire-separating walls and floors. Values come from surveys of real buildings. Typical mean values range from about 230 MJ/m² for hospitals to 780 MJ/m² for dwellings, and much higher for storage and libraries. Because fire loads vary widely between rooms of the same kind, design codes use a high fractile or a fixed design value rather than the mean.

Definition and units

Introduction to Fire Safety Engineering defines the fire load of a building as the total energy released when all its materials burn completely. Related to the floor area, this becomes the fire load per square metre of floor. The glossary of the US Nuclear Regulatory Commission's Fire Dynamics Tools defines fire load density as the fire load per unit area, with the symbol Q″ and the unit J/m². It notes that the area is sometimes the floor area and sometimes the total surface area of the enclosure. The choice matters, because the same fuel gives a much lower number when divided by the larger surface area.

Regulations often separate two parts of the fire load. The variable fire load is everything that belongs to the contents of a building, such as furniture, stock and paper. The permanent fire load is formed by combustible parts of the construction itself. Published survey tables usually give only the variable part. The International Fire Engineering Guidelines state that combustible building materials should be added to these values to obtain the total fire load. The Australian characteristic values likewise ignore wall and ceiling linings and the structure, so an adjustment is needed where these are combustible.

The energy can be expressed in joules or as an equivalent mass of wood. Introduction to Fire Safety Engineering uses a net calorific value of 19 MJ/kg for spruce to convert energy into kilograms of wood equivalent. The Swedish design method converts a wood fire load at 18.8 MJ per kilogram of wood. Lie, for example, gives the normal range of building contents in kilograms per square metre of floor rather than in megajoules.

A further distinction is between net and effective values. A net value is based on the heat of combustion measured in a bomb calorimeter, where nearly all the fuel burns. An effective value allows for the incomplete combustion of a real fire. In real fires about 80% of the complete heat of combustion is usually released. Published tables do not always state which kind of value they contain, so users must check this before combining data from different sources.

Determining the fire load density

The International Fire Engineering Guidelines list three ways to establish the fire load of a room or compartment. These are direct measurement or estimate, a statistical survey of similar buildings, and the use of a published characteristic fire load density. Direct measurement is suitable where the fire load is unlikely to change much over the design life of the building. The mass of each combustible item is multiplied by its calorific value, and the energies are added up and divided by the floor area:

$$q=\frac{\sum _i{m}_i{H}_i}{A_f}$$

Symbol	Description	Unit
$q$	fire load density of the compartment, per unit floor area	MJ/m2
$m_i$	total mass of combustible material i in the compartment	kg
$H_i$	calorific value of combustible material i	MJ/kg
$A_f$	total internal floor area of the compartment	m2

The relation holds for any room or compartment, provided that the calorific values are net values for the material as found in the building. Wet or damp materials release less useful heat, because part of the energy is spent on evaporating the water. The guidelines give a correction for the moisture content. Introduction to Fire Safety Engineering notes that the calorific values in sources such as the Dutch standard NEN 6060 are already net values that allow for this effect.

As a worked example, consider a room of 4 m by 5 m, so 20 m² of floor. It is assumed to contain 400 kg of wooden furniture and one three-seat sofa. Introduction to Fire Safety Engineering gives 18 MJ/kg for beech wood and 1,884 MJ for a three-seat sofa. The total fire load is 400 × 18 + 1,884 = 9,084 MJ, and the fire load density is 9,084 / 20 = 454 MJ/m² of floor. Expressed as wood equivalent at 19 MJ/kg, this is about 24 kg of wood per square metre. That lies within the range of 10 to 60 kg/m² that Lie describes as normal for building contents.

The same room shows how much the reference area matters. With a ceiling height of 3 m, the walls, floor and ceiling together have an area of 94 m². The Australian Data Sheets convert the floor-based value to the total surface area by multiplying by the ratio of floor area to total area. The result is 454 × 20 / 94 = 97 MJ/m², less than a quarter of the floor-based value. Methods that use the total area, such as the Eurocode parametric curves, accept values from 50 to 1,000 MJ/m² on this basis.

The same 9,084 MJ of furniture in a room of 4 m by 5 m by 3 m gives 454 MJ/m² per unit floor area but 97 MJ/m² per unit total surface area. Based on National Research Council of Canada et al. (2005), p. 2.4-2 and Australian Building Codes Board (2020a), Data Sheet B3, p. 40.

Not every combustible item contributes fully. Materials stored in containers with some fire resistance, such as steel filing cabinets, are protected and will not be fully consumed. The reduction depends on the fire temperature and duration, the integrity of the container and the nature of the contents, and it is hard to quantify without tests. Where a statistical survey is used, the guidelines ask for enough buildings to reflect their variability. Those buildings should have a comparable use, size and contents, preferably in the same country or one with similar social and economic conditions.

Typical values

Survey data are collected per type of use. The International Fire Engineering Guidelines reproduce a CIB compilation of 1983 with the mean and the 80%, 90% and 95% fractiles for broad occupancy groups. A fractile is the value that is not exceeded in that percentage of the rooms surveyed. Introduction to Fire Safety Engineering gives mean values from the Dutch national annex to EN 1991-1-2, which agree with the CIB means for several occupancies. In New Zealand, C/VM2 (second edition, 2025) prescribes one design value per group of activities.

Mean, 80% fractile and 95% fractile of the variable fire load density for seven occupancies, from the CIB compilation. Based on National Research Council of Canada et al. (2005), p. 3.4-8.

Occupancy	CIB mean (MJ/m2)	CIB 80% fractile (MJ/m2)	CIB 95% fractile (MJ/m2)	Dutch national annex, mean (MJ/m2)	C/VM2 (2025) design FLED (MJ/m2)
Dwellings	780	870	970	780	400
Offices	420	570	760	420	800
Hotel bedrooms	310	400	510	310	400
Hospitals	230	350	520	230	400
Shops	600	900	1,300	600 (clothes shop)	800
Schools	285	360	450	285	400
Libraries	1,500	2,250	not given	not given	800
Manufacturing	300	470	720	300	400 or 1,200

Storage behaves differently, because the fuel is stacked. The Swiss survey data in the International Fire Engineering Guidelines give storage values per metre of stack height, which must be multiplied by the height of storage. C/VM2 (second edition, 2025) uses 800 MJ/m² per metre of height, with a minimum of 2,400 MJ/m², for storage over 3.0 m high and for bulk retail. Spaces of moderate fire hazard, such as workshops and storage up to 3.0 m high, receive 1,200 MJ/m², and car stacking systems 400 MJ/m² per tier. Retail data are also very spread out. A Canadian survey of 168 shops found a mean of 747 MJ/m², with values from 56 to 5,305 MJ/m².

Australia groups building classes into four fire load categories. The Fire Safety Verification Method Data Sheets give a low category with a mean of 300 MJ/m² and a medium category of 500 MJ/m², which includes residential buildings. Offices fall in a high category with a mean of 780 MJ/m². The very high category, which includes shops, has a mean of 1,000 MJ/m², with a standard deviation of 750 MJ/m² and limits of 300 and 2,500 MJ/m². The data sheets assumed normal distributions for these categories.

Characteristic fire load distributions of the four Australian fire load categories. Based on Australian Building Codes Board (2020a), Data Sheet B2, pp. 9–10.

Published values do not agree between sources. The International Fire Engineering Guidelines warn that the tables conflict because of differences between countries and survey methods. They add that fire loads have changed with new technologies and materials. Introduction to Fire Safety Engineering gives a single mean of 1,800 MJ/m² for libraries, compared with the CIB mean of 1,500 MJ/m². It also notes that the Dutch national annex differs on some points from NEN 6060.

Variability and design values

Fire loads vary from room to room and over time, even within one type of use. According to Lie, deviations of 50% or more from the most probable fire load are common. That alone can shift fire temperatures by hundreds of degrees and the fire duration by 50% or more. Hall notes that fire load density typically follows a log-normal distribution, in which the logarithm of the value is normally distributed. A log-normal distribution was also fitted to the Canadian retail survey mentioned above.

For well-defined occupancies such as dwellings, hotels, hospitals, offices and schools, the International Fire Engineering Guidelines suggest a coefficient of variation of 30% to 50% of the mean. The 80% fractile is then about 1.25 to 1.5 times the mean, and isolated peaks reach about twice the mean. For dissimilar occupancies such as shopping centres and industry, the coefficient of variation rises to 50% to 80%, and peaks reach 2.5 times the mean. In the CIB table for offices, the 80% fractile is 1.36 times the mean and the 95% fractile 1.81 times the mean.

Codes differ in the fractile they choose. The CIB compilation states that the mean cannot prudently be used for design, because it gives a negative safety margin for every room above the mean. It recommends at least the 95% fractile. Eurocode 1 Part 1-2 recommends the 80% fractile, assuming a Gumbel distribution, as cited by the Australian Data Sheets. With an 80% value there is a probability of 0.2 that the fire severity is underestimated. The Australian Data Sheets therefore describe such a design as one that limits the probability of failure to an acceptable level, rather than one that survives every burnout.

The Natural Fire Safety Concept, used in the OZone software, starts from the 80% fractile of surveyed fire loads as the characteristic value. It multiplies this by partial safety factors for the compartment area, for the danger of fire activation of the occupancy and for active measures such as sprinklers, detection and fire brigades. A combustion factor accounts for incomplete burning. The Australian Data Sheets advise against using such reductions for active measures to show compliance with the National Construction Code. In their view the effect of those measures should follow from the fire engineering analysis of the specific building.

New Zealand takes a simpler route. The 2013 commentary on C/VM2 explains that its design FLEDs resemble those of earlier compliance documents and are loosely based on the CIB study. Other values should not be used, because the actual fire load in a building is hard to control over its life. The second edition (2025) lets the design FLED be multiplied by a factor F_m. This factor is 0.5 for a sprinklered firecell and 1.0 for an unsprinklered one. For structural systems that cannot develop dependable deformation capacity after flashover it is 1.0 with sprinklers and 1.25 without.

Use in fire safety engineering

The fire load density is mainly an input for the later stages of a fire. In fire models it usually sets the fire duration, because the fire stops when the fuel is used up. For escape analysis it matters less, because the first phase of the fire is decisive there. It becomes important when the course and duration of the fire matter, such as for the thermal load on the structure and on fire-separating elements. The rate at which the energy is released is set separately, for example by the heat release rate per unit area and a growth curve.

A design fire combines these inputs. In C/VM2 (second edition, 2025), the FLED is one of the parameters of every design fire, next to the growth rate, the peak heat release rate and the species yields. After the growth phase the fire burns at its maximum heat release rate until all the fuel is exhausted, and the total fire load follows from the design FLED. In the OZone design fire, the decay phase begins when 70% of the design fire load has been consumed.

For the structure, the fire load density feeds the methods that describe a fully developed fire. In the time equivalence method of EN 1991-1-2 Annex F, the design fire load density per unit floor area is multiplied by factors for the linings and openings. The Australian Data Sheets add that the contribution of combustible construction should be included. A parametric fire curve according to Annex A uses the value related to the total surface area instead. The Swedish design curves in the SFPE Handbook are likewise drawn for fire load densities of 250 to 1,500 MJ per square metre of total internal surface.

Lie summarises the effect of the inputs simply. The fire load determines how long a fully developed fire lasts, whereas the openings determine both its duration and its intensity. Experiments with wood cribs in a 29 m² compartment illustrate this. The total heat released in a burnout was proportional to the fire load, but doubling the fire load did not double the burning rate when the window restricted the air supply. For such ventilation-controlled fires, a higher fuel load mainly lengthens the fire rather than raising the gas temperature.

Characteristic temperature curves of ventilation-controlled fires for fire loads of 5 to 40 kg of wood per m² of bounding surface, opening factor 0.05 m^1/2, heavy linings. Based on Lie (2002), pp. 4-203 to 4-205.

The amount of fuel also influences whether a fire is limited by ventilation at all. Lie notes that contents of 40 to 100 kg/m² occur with a considerable probability. In that range the fire is probably mainly ventilation controlled, even with large windows, as experiments have confirmed. In large open floors, fire may not involve all fuel at once but spread as a travelling fire. The Australian Data Sheets describe this as a fire that progressively spreads through a large enclosure rather than involving all combustible surfaces simultaneously.

Regulations also use the total fire load directly. The New Zealand Building Code clause C3.8 sets radiation limits for unsprinklered firecells within 15 m of a relevant boundary that hold more than 20 TJ of fire load or exceed 5,000 m². The 2013 commentary explains that the area limit was derived from a FLED of 800 MJ/m² per metre of storage height and a storage height of 5.0 m. Indeed, 5,000 m² × 800 MJ/m² per metre × 5.0 m gives 20 million MJ, which equals 20 TJ.

Limits and common misconceptions

A common misconception is that a high fire load density means a fast or intense fire. The quantity says how much energy is available, not how quickly it is released. The rate of burning depends on the fuel type, its geometry and the conditions in the room, including the air supply. Two rooms with the same fire load density can therefore burn very differently.

A second misconception is that a value from a table describes the building being designed. Survey values are averages or fractiles over many rooms, and the spread is large. Introduction to Fire Safety Engineering calls the tabulated values examples only and states that the reliability of each value must be assessed for the scenario. A design based on an 80% fractile still leaves a 20% chance that the actual fire load is higher.

The reference area is a frequent source of error. The same fuel gives very different numbers per unit floor area and per unit total surface area, as the worked example shows. Mixing net and effective values, or kilograms of wood and megajoules, causes similar mistakes.

Finally, the fire load density is a property of the contents at one moment. The amount and surface area of combustibles change with time and vary from compartment to compartment. The 2013 commentary on C/VM2 likewise notes that the actual fire load is hard to control over the life of a building. The Australian Data Sheets also stress that tabulated values exclude combustible linings and structure, which must be added where present.

See also

• Warehouse fire
• Heat release rate
• Burning rate

References

• Australian Building Codes Board. (2020a). Data sheets: Fire Safety Verification Method (Handbook annex, Print version 1.1).
• Cadorin, J.-F., & Franssen, J.-M. (2003). A tool to design steel elements submitted to compartment fires – OZone V2. Part 1: Pre- and post-flashover compartment fire model. Fire Safety Journal, 38, 395–427. https://doi.org/10.1016/S0379-7112(03)00014-6
• Cadorin, J.-F., Pintea, D., Dotreppe, J.-C., & Franssen, J.-M. (2003). A tool to design steel elements submitted to compartment fires – OZone V2. Part 2: Methodology and application. Fire Safety Journal, 38, 429–451. https://doi.org/10.1016/S0379-7112(03)00015-8
• Hall, J. R., Jr. (2002c). Statistics. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 1-193–1-202). National Fire Protection Association; Society of Fire Protection Engineers.
• Iqbal, N., & Salley, M. H. (2004). Fire Dynamics Tools (FDTs): Quantitative fire hazard analysis methods for the U.S. Nuclear Regulatory Commission Fire Protection Inspection Program (NUREG-1805). U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation.
• Lie, T. T. (2002). Fire temperature-time relations. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 4-201–4-208). National Fire Protection Association; Society of Fire Protection Engineers.
• Ministry of Business, Innovation and Employment. (2013). Commentary for Building Code clauses C1–C6 and Verification Method C/VM2.
• Ministry of Business, Innovation and Employment. (2025b). Verification Method C/VM2: Framework for fire safety design (2nd ed.).
• National Research Council of Canada, International Code Council, Department of Building and Housing, & Australian Building Codes Board. (2005). International fire engineering guidelines (2005 ed.). Australian Building Codes Board.
• Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.
• Walton, W. D., & Thomas, P. H. (2002). Estimating temperatures in compartment fires. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-171–3-188). National Fire Protection Association; Society of Fire Protection Engineers.