NFPA Report: U.S. Multiple-Death Fires for 2007
NFPA Journal®, September/October 2008
By Stephen Badger
In 2007, firefighters in the United States responded to an estimated 1,557,500 fires. Of these, 414,000 were in residential structures, 116,500 were in nonresidential structures, and 1,027,000 did not involve structures. Tragically, these fires caused 3,430 deaths, with 2,895 in residential properties, 105 in nonresidential properties, and 430 outside of structures.1
Firefighters work at the scene of a beach house fire in North Carolina where seven people died and others injured.
Thirty-seven of the incidents are classified as catastrophic multiple-death fires, defined by the NFPA as residential fires that kill five or more people, or nonresidential or nonstructural fires that kill three or more people. Seventeen of the fires occurred in residential properties, eight in nonresidential properties, and 12 outside of structures. These fires represented 0.002 percent of the total U.S. fires in 2007.
There were 190 deaths in these fires, a figure that represents 5.5 percent of the total estimated U.S. fire death toll last year. Of the 190 victims, 41 (or 21.6 percent) were children under age six. Eleven of the victims were firefighters.
The three most serious fires each killed 10 people. Two of the fires occurred in residential structures, both resulting in the deaths of five children under the age of six; one was a wildland fire. These fires are discussed in more detail below.
Compared to 2006, the number of catastrophic multiple-death fires increased by one, but the death toll went down by 33 deaths. The under-age-six toll went up 13, from 28 in 2006. The deadliest fires in 2007 accounted for 10 deaths each, contrasted with 2006, when one aircraft crash and fire killed 24 people.
Catastrophic residential fires
Seventeen fires occurred in residential properties, including 13 in single-family homes (of which two were manufactured homes), two in apartment buildings (one was in a nine-unit, and the other was in a 64-unit building), one in a two-family duplex, and one in a 40-unit motel.
Residential fires accounted for almost half of the total multiple-death fires, and 109, or 57.4 percent, of the deaths in these fires. This is three more than in 2006, when there were 106 deaths. The death toll to children under six was 35 (32.1 percent of the deaths in these fires)—up 10 from 2006. There were 13 of these fires that started between 11 p.m. and 7 a.m. This was three quarters of the residential fires, resulting in 86 deaths. Of these, 27 were children under age six. See Table 1 for details of all the fires. The largest of the fires are discussed in more detail here.
As mentioned above, two of the most severe fires in 2007, in terms of loss of life, occurred in residential structures, with each claiming 10 lives. The first fire broke out just before 4:00 a.m. in the living room of a one-story, single-family home. The cause was not determined; but it started in or around a living room chair. A smoke alarm was found in a small hallway; it did not operate and the reason for that was not determined. This was one of two factors that impaired or delayed the residents’ escape. The other factor was the adults’ elevated blood alcohol levels. Two children and an adult were in a front bedroom, three children were in a rear bedroom, an adult and a child were located in a den near the kitchen, one adult was located in the kitchen, and one was found at the rear door. One person escaped this fire.
The second 10-fatality fire broke out in a four-story, four-unit home being used as a single-family residence when an electrical short in a first-story bedroom ignited nearby combustibles. A resident tried to extinguish the fire but was not successful. The fire spread horizontally throughout the room and through the open door to a stairwell and then to the second and third stories. Investigators located two smoke alarms within the residence, one in the fourth-story hallway and one in the basement. Neither alarm had a battery. Five of the victims were located on the fourth story, three on the third story, and two on the second story. Another six occupants escaped or were rescued by firefighters.
One fire resulted in nine deaths. This fire broke out at 11:00 p.m. in a second-story apartment in a five-story, 64-unit apartment building when smoking materials fell into a couch. The occupants thought the odor was from a bug bomb in a neighboring apartment and vacated their unit for one on the third story. Upon returning to air out their apartment, they found it on fire. They tried, unsuccessfully, to extinguish the fire with a handheld extinguisher. They then fled, leaving the door open. The fire spread rapidly out to the paneled hallway and throughout the building. There were single-station smoke alarms in the apartments and hallways. No information was reported on their operation. There was no automatic suppression equipment present. The victims were occupants of various units throughout the building, but no information was reported on where they lived or where they were found. Several additional people escaped or were rescued.
One fire killed eight people, including a child under age six. This fire broke out about 7:21 a.m. in a two-story, single-family row house when mishandled smoking materials ignited the victims’ clothing or love seat in a first-story family room. The fire burned through the room and extended up the stairway to the second story. The home had no smoke alarms or suppression equipment. Six of the victims were located on the second story, four in a front bedroom and two in a rear bedroom. One victim was located on the bottom step of the stairway and the last victim was located on the walk in front of the house. In addition several people escaped or were rescued.
Two fires were responsible for seven deaths each, three killed six each and there were eight five-fatality fires. These fires are all described.
Catastrophic nonresidential fires
There were eight fires in nonresidential structures, including two in stores, two in office buildings, and one each in a tunnel, a chemical laboratory, a fuel storage facility, and a test site. All of these fires occurred during the daytime or early evening. There were 34 victims, including 11 firefighters and one child under age six. Compared to 2006, there was one more incident and eight fewer deaths. See Table 2 for complete details of these fires.
The largest loss-of-life incident in a nonresidential structure was responsible for the deaths of nine firefighters. This fire broke out at 7:08 p.m. in a one-story, 50,000-square-foot (4,645-square-meter) furniture store and warehouse. Arriving firefighters fought this rapidly spreading fire from inside. The roof of the showroom area suddenly collapsed, trapping and killing nine.2
One fire killed five workers at a hydroelectric plant, in a 4,000-foot (1,219-meter) long by 12-foot (3.7-meter) wide tunnel that connected two reservoirs with the plant. This fire broke out at 2:00 p.m., 1,000 feet (305 meters) below grade, as nine workers were applying epoxy. The workers were cleaning a sprayer with a flammable solution when vapor from the solution ignited. The source of ignition was not reported. Four workers were below the fire and were able to escape by way of the main entrance at the lower end of the tunnel. The five who died were above the fire. They ascended as far as they could, until they reached a bend in the tunnel and could not climb the steep incline. They were trapped in the smoke and died. The U.S. Chemical Safety Board (CSB) is investigating this incident.
Two explosions and fires killed four people each. The first one occurred at a convenience store/gas station. Two gas company employees were transferring propane from one 500-lb (227-kilogram) tank to another when a valve malfunctioned, allowing propane to escape. As the gas dispersed, it found an unknown ignition source and exploded. The explosion killed two firefighters, who had responded earlier to a report of a gas leak, and the two gas company employees. Four other civilians and a firefighter were seriously injured.2
The second incident was an explosion at a chemical laboratory during the production of a gasoline additive, when a runaway chemical reaction in an overpressurized steel vessel ignited in a tremendous explosion, and sent a fire ball and mushroom cloud over 2,000 feet (610 meters) high. Very little information was reported on this laboratory and explosion. The CSB is investigating this incident as well.
Four fires killed three people each. The first was deliberately set shortly after 5:00 p.m. in a fifth-story storage area in a six-story office building. There was a partial-coverage sprinkler system in the atrium and common areas of the first story. The system operated and kept the fire from spreading to the atrium. The only smoke detection equipment was in the elevator lobbies for elevator recall; it was not reported if it operated. The three fatalities were all located on the fifth story in two different areas. A firefighter became trapped during a search of the fifth story, but was rescued. An arrest has been made in this case.
The second fire was an incendiary fire in an insurance office. A person entered the office and poured gasoline on two women and set them on fire. The badly burned women, one of whom was pregnant, escaped but died as a result of burns. The pregnant woman’s baby was delivered by cesarean section. Sadly, the infant only survived for three days.
The third and fourth incidents were explosions involving tanks. The first, at a fuel storage facility, was caused when someone lit a match or lighter on or around a fuel storage tank. No other information was reported. The second was at a rocket testing facility when a nitrous oxide leak was ignited by an undetermined source.
Catastrophic nonstructural fires
There were 12 nonstructural fires, including five road crashes and fires, four road vehicle fires, two wildland fires, and one aircraft in flight. These fires accounted for 32 percent of the catastrophic multiple-death fires, one more than in 2006 when there were 11 such fires. There were 47 deaths in these fires; down from the 75 deaths in 2006, which included a major aircraft crash that killed 24. Five children under age six died in these fires in 2007. A medical examiner’s or coroner’s office verified that the deaths in the vehicle crashes and fires were due to inhalation of products of combustion or thermal injuries, and not impact injuries. See Table 3 for details on these fires.
The largest loss-of-life fire was the wildfire that killed 10. The headlines in most newspapers, television, and radio news stories in the United States and in many areas around the world told of the California firestorms of 2007. These fires broke out over several days in late October and burned over 518,000 acres or 809 square miles (210,000 hectares), destroyed over 3,100 structures, including 2,100 residences, and caused the evacuation of almost a million residents. Tragically, there were at least 10 deaths directly related to these fires. In addition, several more deaths were attributed indirectly to these fires, mostly due to natural causes at various stages during the evacuations and fires.
The 10 deaths directly resulting from the fires occurred in two of the larger named fires. Eight died in the Harris fire and two died in the Witch fire. Five of the 10 victims were caught in the open, in a ravine or a cave near the Mexican border, three were in their homes, and two were in a garage. The exact location of two victims was not reported.
Santa Ana winds of over 70 miles (113 kilometers) per hour with gusts of over 100 miles (161 kilometers) swept the fires very rapidly, catching most victims before they even had a chance to evacuate. Reports also indicated that at least 27 other civilians were injured, along with over 130 firefighters. It appears that the dedicated actions of over 11,000 firefighters, as well as many law enforcement and government agencies from California and throughout the United States and Mexico, kept the death toll to a minimum.
In another incident, a pilot reported smoke in the cockpit of his small plane and declared an emergency. Before he was able to make an emergency landing, the aircraft crashed into two homes in a residential neighborhood near the airport. On arrival, firefighters found a one-story home fully engulfed in fire and the plane and a second two-story home heavily involved in fire. An off-duty firefighter reported to firefighters that a child was trapped on the second story. Rescue attempts were made to no avail. The child perished, as did a child and an adult in the other home and two onboard the aircraft. In all, five people, including two children under the age of six, died in this fire. The National Transportation Safety Board is investigating.
Two motor vehicle crashes and fires each killed four. The first occurred when a vehicle slammed into the rear of a car stopped in traffic. The second involved a car and truck colliding, with the truck then striking a van. The truck caught fire and consumed the car and van.
There were eight incidents that killed three each. Three of these were vehicle crashes and fires. The first occurred when a car and a tractor trailer collided, then both collided with two other trucks as a fire broke out. In the second incident, a fast-moving car struck a parked construction vehicle and burst into flames. The third occurred when a car went off the highway, through a fence, into a ravine, and caught fire.
Four other three-fatality incidents were road vehicle fires—two in cars, one in a van, and one in a travel trailer. The first occurred in the parking lot of a warehouse when a father locked himself and his children, including one under the age of six, in a car, poured gasoline over the children and the interior of the car and lit it on fire. The second incident was very much the same. It occurred at 1:35 a.m. with a car parked behind a service station. Gasoline was poured over the occupants and set on fire. The report did not identify who set the fire. Two of the victims were under the age of six.
The third incident occurred overnight while three youngsters were having a sleepover in a travel trailer in the yard of a residence. The fire was discovered by a father who went to investigate the lack of activity at about 9:00 a.m. The cause of this self-extinguished fire was listed as an unattended candle igniting furniture in an eating area of the trailer.
The fourth fire occurred at 2:04 a.m. in a van parked in a driveway of a residence. A group had filled their van’s tank with gasoline then half-filled a 5-gallon (19-liter) plastic gas jug. During the ride home, the jug was on its side and gasoline seeped out and soaked the carpet. The group parked the van in a driveway while visiting a home. When they entered the van to leave, the gasoline vapors ignited from an undetermined source.
The last fire to kill three was a wildland fire. This fire broke out about 9:00 a.m. and burned in pine, spruce fir, and pinyon juniper as well as grass and sage. Three family members working in a field were caught by the fire’s spread.
Role of smoke alarms and sprinklers
Information on detection equipment was reported for 13 of the 17 residential fires. Seven properties had detection equipment present. Of these seven, only one had complete coverage and two had partial coverage. Details on coverage were not available for the other four structures with detection equipment. Alarms were known to have operated in just one incident, where seven deaths occurred. The reason for their inability to evacuate was not reported, but this fire began on an outside deck, which could have resulted in some extended growth before fire spread into the coverage area and activated the alarms. There were two fires where it is known that smoke alarms did not operate. In one of those fires, neither smoke alarm had a battery and in the other fire, the single alarm had been tested and found nonoperational prior to the fire, but the reason it was not operational was not reported. There were 20 deaths in these fires, 10 of the victims were under age six. The operation of the equipment in the other four fires was not known or not reported.
Six residential structures, or 46 percent, had no detection equipment. There were 37 deaths in these fires, including 13 children under the age of six.
Information on detection equipment was only reported for four of the eight nonresidential properties. Three of the structures had no detection equipment and the fourth only had smoke alarms for the elevator recall system. The operation of that system was not reported.
None of the 17 residential properties had suppression equipment. Only one nonresidential structure was known to have suppression equipment, which operated and kept the fire from spreading.
Smoke alarms have been proven effective in reducing the risk of death in home fires. The most effective arrangement is to use interconnected multiple-station smoke alarms that are supplied by hard-wired AC power with a battery backup. These should be located outside each sleeping area, on each level, and in each bedroom. Homeowners should routinely test smoke alarms according to manufacturers’ recommendations. NFPA recommends testing residential smoke alarms at least monthly. Batteries should also be replaced according to manufacturer’s recommendations but, in the case of conventional batteries, at least yearly.
Smoke alarms are only effective if occupants exit the building when they sound.
Children should be familiar with the sound of a properly operating smoke alarm. They should follow a practiced escape plan that emphasizes two exits from any location in the home with a designated meeting place.
Exit drills in the home are part of many school curricula. Practicing the plan helps families determine if children and others readily waken to the sound of a smoke alarm, and that, along with assistance for family members who require it, can be factored into the plan. Practicing fire prevention principles could have prevented many of the fires.
Where we get our data
NFPA obtains its data by reviewing national and local news media, including fire service publications. A news clipping service reads all daily U.S. newspapers and notifies the NFPA Fire Analysis and Research Division of catastrophic fires. Once an incident has been identified, we request information from the local fire department or the agency having jurisdiction. NFPA’s annual survey of U.S. fire experience and mailings to state fire marshals are additional data sources, although not principal ones. We also contact federal agencies that have participated in the investigation of such fires. The diversity and redundancy of these sources enable us to collect the most complete data available on catastrophic fires in the U.S. We understand that in many cases, due to ongoing litigation, a department cannot release information. Also in some cases departments have been unable to determine the information we request.
Michael J. Karter, Jr., “Fire Loss in the United States During 2007,” NFPA Journal, September/October 2008.
Rita F. Fahy, Paul R. LeBlanc, Joseph L. Molis, “Firefighter Fatalities in the United States—2007,” NFPA Journal, July/August 2008.
NFPA wishes to thank the U.S. fire service, and the medical examiners for their contributions of data, without which this report would not be possible. The author would like to give a special thanks to Rita F. Fahy, Norma Candeloro, and his coworkers for their guidance in the completion of this report.
Stephen G. Badger, a fire data assistant with NFPA’s Fire Analysis and Research Division, is retired from the Quincy, Massachusetts, Fire Department.