High-rise building

A high-rise building is, in fire protection engineering, a building that is nominally taller than about 25 m and from which occupants cannot be rescued from the outside during a fire. Because rescue from outside is not possible, occupants must leave by the stairs, and firefighters often have to climb the same stairs to reach the fire. Height changes several parts of the fire problem at once. The stack effect drives smoke up through stairs and lift shafts to floors far from the fire.

A full evacuation can take a long time, so many high-rise buildings use a staged or phased evacuation in which the floors most at risk leave first. Firefighters who climb with breathing apparatus need about 2 min per floor, which has led to calls for protected lifts for the fire service. The 2001 collapse of World Trade Center Building 7, a 47-storey office building, was the first known total collapse of a tall building primarily due to fires.

Definition and characteristics

The investigation of World Trade Center Building 7 explains two terms that are often mixed up. Architects and structural engineers speak of a tall building when it is taller than its surroundings, slender, or needs technologies such as wind bracing, and is nominally taller than 15 to 20 storeys. Fire protection engineers use high-rise building for buildings nominally taller than 25 m (75 ft) from which external rescue from fires is not possible. The fire definition therefore depends on whether rescue from outside is possible, not on the shape of the building.

The main consequence of height is that the risk to an occupant grows with the floor on which that person is. The National Institute of Standards and Technology (NIST) notes that fire rating requirements in the United States were originally developed from experience with buildings of less than 20 storeys. For tall buildings, the consequences of a given threat are more severe for an occupant on an upper floor than on the first floor. With lifts out of service, both full evacuation from upper floors and fire service access to those floors take much longer.

The fire load in a high-rise office floor is not unusual in itself. A table of fire load densities from Swiss studies of 1967 to 1969, reproduced in the International Fire Engineering Guidelines, gives an average of 800 MJ/m² for a high-rise office building. Some performance-based methods do not cover such buildings. In New Zealand, Verification Method C/VM2 (second edition, 2025) lists tall buildings of more than 60 m or 20 storeys as outside its scope, together with buildings that need a managed evacuation.

Smoke movement and the stack effect

Smoke spreads through a building under several driving forces: expansion of heated gases, the stack effect, wind, building services and pressure differences caused by temperature differences. In tall buildings, smoke movement can be dominated by the stack effect. When it is cold outside, the warmer and lighter air inside rises through shafts such as stairs, lift shafts and service ducts. This is called the normal stack effect. When the outside air is warmer than the inside air, the flow in the shafts reverses and runs downwards.

The size of the pressure difference depends on the difference in density between the cold and warm air columns and on the height considered. Klote gives a relation for the pressure difference between a shaft and the outside, at standard atmospheric pressure. It uses absolute temperatures and the height above the neutral plane, the level where the pressure inside equals the pressure outside:

$$\mathrm{\Delta }P=K_s(\frac{1}{T_0}-\frac{1}{T_I})h$$

Symbol	Description	Unit
$\mathrm{\Delta }P$	pressure difference between the shaft and the outside	Pa
$K_s$	coefficient, 3460 in SI units	Pa·K/m
$T_0$	absolute temperature of the outside air	K
$T_I$	absolute temperature of the air inside the shaft	K
$h$	distance above the neutral plane	m

Above the neutral plane the shaft pressure is higher than outside, and below it the shaft pressure is lower. If air leakage paths are fairly uniform over the height, the neutral plane lies near mid-height. Vented shafts and other uneven leakage can move it considerably. In unusually airtight buildings with exterior stairs, reverse stack effect has even been seen in cold weather. The relation describes the undisturbed building; the shape and position of openings, flow resistance, wind and mechanical ventilation all change the real pressure distribution.

Air movement caused by the normal stack effect (left, cold outside) and the reverse stack effect (right, warm outside). Based on Klote (2002), pp. 4-274–4-275.

Klote's worked example is a building 60 m tall with the neutral plane at mid-height, an outside temperature of −18 °C and an inside temperature of 21 °C. At the top, the shaft is 55 Pa above the outside pressure, and at the bottom it is 55 Pa below it. A recalculation in SI units gives 1.80 Pa per metre and 54 Pa at 30 m above the neutral plane; Klote's 55 Pa (0.22 in. of water) follows from the rounded imperial temperatures. On a mild winter day of 0 °C outside, the same building gives about 27 Pa. A summer day of 30 °C with 21 °C inside gives about −10 Pa, a reverse stack effect. These are of the same order as the 13 Pa that Klote calculates for the buoyancy of an 800 °C fire room, 1.52 m above its neutral plane.

Pressure difference between a shaft and the outside in a 60 m building with the neutral plane at mid-height and 21 °C inside, for three outside temperatures. Based on Klote (2002), pp. 4-274–4-275.

The stack effect can carry smoke far from the fire. With a fire below the neutral plane, smoke moves with the building air into the shafts and up them, and flows out onto upper floors above the neutral plane. If leakage between floors is small, the floors below the neutral plane apart from the fire floor stay relatively free of smoke, until the smoke production exceeds what the stack flows can carry. Smoke from a fire above the neutral plane is carried to the outside through openings in the facade. Hot smoke can also rise through buoyancy against a reverse stack effect.

Wind adds to these pressures. If a window of the fire room breaks on the windward side, the wind forces smoke through the fire floor and even to other floors, endangering occupants and hampering firefighting. A broken window on the leeward side tends to vent the smoke instead. Kandola points out that a pressurisation system in a high-rise building must account for wind-induced pressures, or it may create overpressure that prevents escape or underpressure that makes it ineffective. In cold weather, a broken window on a lower fire floor also increases the stack effect pressures.

Evacuation strategies

In multi-storey office buildings, Proulx distinguishes two types of evacuation. An uncontrolled total evacuation sends everybody to the stairs at the same time. A controlled selective evacuation clears floors in a planned order and needs a voice communication system to manage it. The International Fire Engineering Guidelines describe the same idea. In high-rise buildings with suitable warning and intercommunication systems, trained staff may manage the evacuation, with floors furthest from the fire first put on alert and evacuated progressively only if the fire keeps developing.

Canadian drills show how the two approaches compare. Consider a 15-storey office building with 70 people per floor and two stairs 1,120 mm wide. A total evacuation then gives a flow of about one person per second per exit and a descent speed of 3.6 floors per minute. In both strategies, about 9 min were needed to move everybody below the presumed fire floor. The selective evacuation moved 490 people to safety in that time, while the total evacuation brought all 980 occupants to the ground.

Schematic movement of the last person leaving each floor in an uncontrolled total evacuation (left) and a controlled selective evacuation with a fire on the 10th floor (right). Based on Proulx (2002), p. 3-355.

In the total evacuation, however, people from the most threatened floors would queue, and stair doors opened on the fire floor would let smoke into the stairs above it. On the other hand, a total evacuation is simple and needs no costly systems or training, whereas a controlled evacuation may delay the moment at which occupants who should leave are told to do so. In a real fire, delays are longer, smoke may block exits and occupants above the fire floor may be unable to leave, which can change the total evacuation time entirely.

A controlled evacuation depends on clear communication. In a 1971 drill in a 21-storey office building in Ottawa, the planned sequence cleared the fire floor first and then the floors above it from the top down. The first public address announcement came almost 3 min late and was ambiguous, and as many as 350 people left their floors before their turn. The World Trade Center towers used prepared announcements from a Fire Command Desk in each lobby, asking occupants to wait until their floor number was called. In Australia, the 2020 data sheets for the Fire Safety Verification Method note that alarm systems for phased evacuation commonly have an alert tone and an evacuation tone. Where alarms to some areas are delayed, the delay must be included in the analysis.

The scale of a full evacuation is shown by the NIST simulations of the World Trade Center towers. A phased evacuation of 600 occupants to three floors below the fire floor took 3.5 to 4 min without delay and 11 to 11.5 min with random delays of up to 10 min. A full evacuation of a fully occupied tower of 19,800 people took about 112 min in the buildingEXODUS model, with each extra thousand occupants adding 5.2 min. NIST estimated that such an evacuation on the day of the attacks would have taken over 3 hours. About 87 % of the estimated 17,400 occupants of the towers evacuated successfully, and 99 % of those below the impact floors. NIST recommended that tall buildings be designed for a timely full evacuation, with stairs wide enough for firefighters moving up against the flow.

Simulated evacuation times for a World Trade Center tower: phased evacuation of the floors at risk compared with full building evacuations (models with delays of up to 10 min). Based on Averill et al. (2005), pp. 160–162.

Each tower had three stairs, two 44 in. (1.12 m) wide and one 56 in. (1.42 m) wide, for up to 365 people per floor. Neither the 1968 New York City Building Code nor the more recent codes that NIST compared made the number and width of stairs depend on the number of storeys. People with reduced mobility need other solutions, and the use of refuges and lifts for their evacuation is a growing area of research. Evacuation models increasingly allow lifts to be used for escape.

Smoke control in stairs and shafts

Movement through smoke is common in high-rise fires. In the 1993 bombing of the World Trade Center, 94 % of the occupants of Tower 1 moved through smoke. In two high-rise residential fires, about 96 % of the occupants above the fire floor did so. Proulx explains that repeated opening of doors on the fire floor and on upper floors lets smoke into the stairs, even when they are pressurised. In that sense, the stairs of a high-rise building may be the most dangerous place in a fire after the room of origin and the fire floor.

Stair pressurisation aims to keep a stair free of smoke by supplying air so that it is at a higher pressure than the fire floor. Its main goal is a smoke-free escape route, and a secondary goal is a smoke-free staging area for firefighters. Air may be injected at one point, usually the top, or at several points over the height. In tall stairs, a single injection point can fail when a few doors near it are open, because all the air escapes there. Dividing the stair into compartments is an alternative, but it suits only buildings with a staged evacuation, since open doors between compartments cancel the effect. The number of doors that will be open at once also depends on the strategy: in a densely occupied building with a total evacuation, all doors may be open together.

Klote's design example for a 20-storey stair shows the role of temperature. With −10 °C outside and 21 °C in the stair, the temperature factor is 1.39 Pa per metre. A pressure difference of 20 Pa at the bottom of the stair grows to 106 Pa at the top. About 3.9 m³/s of supply air is needed, and 6.5 m³/s if the leakage area around the doors is larger than assumed. Because the leakage areas are difficult to estimate and control, Klote mentions fans with adjustable supply as one way to correct the airflow after installation.

Lift shafts are a second vertical route. Using lift shafts as smoke shafts prevents their use for evacuation and often spreads smoke to floors far from the fire. The MGM Grand Hotel fire in Clark County, Nevada, in 1980 stayed on the casino level. Its heat and smoke, however, spread through seismic joints, lift shafts and stairs to the residential floors above. Of the 85 fatalities, most were found between the 20th and 25th floors, including 9 people in stairs and 5 in lifts. The lifts had no special smoke protection and were one of the main paths of smoke to the upper floors.

Firefighting and water supply

Fighting fires on the upper floors of tall buildings is not the same as in buildings less than about 30 m (100 ft) high. In the World Trade Center, responders without breathing apparatus climbed at about 1.4 min per floor, and firefighters with protective clothing, breathing apparatus and equipment at about 2 min per floor. Evacuees coming down, including people with disabilities, slowed them further in the 44 in. stairs. At 2 min per floor, reaching the 60th floor from the ground would take about two hours. NIST found that climbing more than 10 to 12 floors makes it difficult to start aggressive firefighting and rescue at once.

For this reason NIST recommended fire-protected and structurally hardened lifts in tall buildings. These would give emergency responders timely access and allow the evacuation of occupants with reduced mobility, and they would be reserved for responders during emergencies. Lifts of this kind for fire service use are described under firefighting lift. Rescue by helicopter is not a general alternative. At the MGM Grand Hotel, about 300 guests were evacuated by helicopter and the fire department rescued about 900 others. In the World Trade Center, the New York police and fire departments did not consider roof rescue a viable strategy for a general evacuation.

Firefighting water must be brought to the upper floors through vertical pipes, called standpipes in the World Trade Center reports; such systems are described under dry and wet risers. Each World Trade Center tower had an alarm system, automatic sprinklers, a standpipe system and a smoke venting system. Each tower also had six storage tanks of 5,000 gal (about 19 m³), three on the 110th floor and one each on the 20th, 41st and 75th floors. Water could flow by gravity to the sprinklers, building staff could start pumps, and the fire department could add water through fire department connections. Although there were redundant vertical supply pipes, each floor had only one connection to its sprinklers. The standpipes stood in the stairs, so where stairs were close together the spacing between standpipes exceeded the 140 ft (43 m) maximum of the 1968 New York City code.

The reliability of automatic sprinklers has been studied for this building type. In Australia, the 2020 data sheets summarise a fault tree study of wet-pipe sprinkler systems in Australian high-rise office buildings. The general sprinkler reliability of 90 % used in the data sheets lies within the range calculated for these buildings.

Structural fire resistance and collapse

High-rise buildings in the United States are designed to withstand a sizeable fire and remain standing. Their elements receive fire resistance ratings in hours from standard furnace tests. These ratings are relative: they do not mean that a structure will last that long in a real fire, which may be more or less severe than the test.

World Trade Center Building 7 shows what can happen when this protection is overwhelmed. Debris from the collapse of the north tower ignited fires on at least 10 floors, and the fires on six floors grew and burned until the collapse almost seven hours later. Their growth and spread were consistent with ordinary office contents fires. Had the sprinkler system had a water supply and operated as designed, the fires would likely have been controlled and the collapse prevented.

Local fire damage, made worse by the thermal expansion of long-span floors, led to a cascade of local floor failures, which left a critical interior column unsupported over nine storeys. Its buckling started a progressive collapse, in which local damage spreads from element to element until the whole structure or a large part of it falls. NIST concluded that structures must be designed for situations where sprinklers are absent, do not work or are overwhelmed.

Fire can also spread vertically outside the building. The Australian data sheets warn that a phased evacuation strategy may be sensitive to rapid fire spread to several floors over the facade, a facade fire, and that this scenario may need its own evaluation. The evacuation plan should then describe what happens when a fire spreads to several floors at once.

Limitations

Much of the detailed knowledge about high-rise fires comes from a few cases. The World Trade Center reports describe one event with unusual causes, and the evacuation models NIST used had not been validated for buildings as tall as 110 storeys. The stack effect relation assumes uniform leakage and a known neutral plane, and the real distribution of pressures depends on openings, wind and ventilation systems. Drill data, such as the Canadian studies, describe evacuations without smoke, which may differ greatly from a real fire. A common misconception is that a total evacuation is always the safest choice. The drill data show that it can expose people on the most threatened floors to queues and smoke in the stairs, while a phased approach depends on reliable communication and trained staff.

See also

• Fire service access
• Stay-put strategy
• Evacuation lift
• Mass timber

References

• Australian Building Codes Board. (2020a). Data sheets: Fire Safety Verification Method (Handbook annex, Print version 1.1).
• Averill, J. D., Mileti, D. S., Peacock, R. D., Kuligowski, E. D., Groner, N., Proulx, G., Reneke, P. A., & Nelson, H. E. (2005). Occupant behavior, egress, and emergency communications (NIST NCSTAR 1-7). National Institute of Standards and Technology.
• Bryan, J. L. (2002). Behavioral response to fire and smoke. 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. 3-315–3-341). National Fire Protection Association; Society of Fire Protection Engineers.
• Kandola, B. S. (2002). Introduction to mechanics of fluids. 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. 1-1–1-26). National Fire Protection Association; Society of Fire Protection Engineers.
• Klote, J. H. (2002). Smoke control. 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. 4-274–4-291). National Fire Protection Association; Society of Fire Protection Engineers.
• Lew, H. S., Bukowski, R. W., & Carino, N. J. (2005). Design, construction, and maintenance of structural and life safety systems (NIST NCSTAR 1-1). National Institute of Standards and Technology.
• Ministry of Business, Innovation and Employment. (2025b). Verification Method C/VM2: Framework for fire safety design (2nd ed.).
• National Institute of Standards and Technology. (2005). Final report on the collapse of the World Trade Center towers (NIST NCSTAR 1).
• National Institute of Standards and Technology. (2008). Final report on the collapse of World Trade Center Building 7 (NIST NCSTAR 1A).
• 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.
• Proulx, G. (2002). Movement of people: The evacuation timing. 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. 3-342–3-366). National Fire Protection Association; Society of Fire Protection Engineers.
• Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.