NFPA Journal®, November/December 2008

By Stephen G. Badger

In mid-October, Southern California was in a drought, with only 5 inches (12.7 centimeters) of rain since January. Santa Ana winds were forecast for a wide area. The winds, along with temperatures in the nineties (low thirties Celsius), created the scenario that led to one of the costliest fire seasons in recent memory. At approximately 9:42 p.m. on October 20, the first fire erupted. Within a week, at least 23 named fires broke out. These fires would become known, collectively, as the Southern California Firestorm of 2007. The governor issued a state of emergency for several of the hardest-hit counties. The Fire Districts Association of California estimated that approximately a million people were evacuated or fled the approaching fires, heading to emergency shelters or to relatives and friends. Firefighters set up unified command systems at all the fires and directed operations. Several of the named fires burned together to form larger fires. Resources were strained for fire suppression crews.

Officials reported several different sources of ignition, including incendiary acts, power lines damaged by the Santa Ana winds, a vehicle crash, and a youngster playing with matches. Mother Nature played a large role in these fires, as well: a deficit in rainfall, temperatures well over 90°F (32°C), relative humidity as low as 10 to 15 percent, and hurricane-strength Santa Ana winds of more than 75 miles (121 kilometers) per hour, gusting to more than 100 miles (161 kilometers) per hour.

By the time the week ended, the fires had burned over 518,000 acres (210,000 hectares) and destroyed over 3,108 structures, including 2,180 residential properties. More than 500 other structures suffered damage, and 239 vehicles were destroyed. It has been estimated that damage to agriculture in the area was over $47 million.

At the time these major fires were burning, more than 250 additional fires broke out. These fires were contained to less than 10 acres (4 hectares) each. More than 1,100 fire departments throughout California and across the United States, as well as several countries around the world, sent aid. Police departments, specially trained fire crews of prison inmates, and county and state officials also responded.

Loss estimates have varied, but the most reliable reported figure is $1.8 billion in property damage. Tragically, the fires claimed at least 10 lives. Four additional deaths not involving fire effects occurred during evacuation.

NFPA reports each year on large-loss fires and explosions in the United States, defined as any event that results in property damage of at least $5 million. In 2007, fire departments responded to an estimated 1,557,500 fires that caused an estimated loss of $14.6 billion. To learn more, read “Fire Loss in the United States During 2007 ” by Michael J. Karter, Jr. in the September/October 2008 issue of NFPA Journal. Many of these fires were small or had little or no property damage reported. However, 71 fires resulted in losses of $5 million or more each. Together, these large-loss fires resulted in over $3.5 billion in direct property loss, killed 19 civilians, and injured 168 firefighters and 67 civilians. These 71 fires accounted for only 0.005 percent of the estimated number of fires in 2007, but accounted for 24.1 percent of the total estimated dollar loss.

There were 25 more large-loss fires in 2007 than there were in 2006, an increase of more than 54 percent in the number of large-loss fires. The total property loss in large-loss fires was up by almost $3 billion, which meant that the 2007 total was more than six times the total in 2006. Most of this increase and half the large-loss total for 2007 occurred in the Southern California Firestorm alone. In fact, large-loss wildland fires excluding the Southern California Firestorm accounted for more total loss than the large-loss total for all of 2006.

When adjusted for inflation to 1998 dollars, the number of large-loss fires in 2007 was the highest since 1999. Adjusted total loss in those fires was the highest since 2001, when the September 11 attacks on the World Trade Center occurred (see Table 1 and Figure 1 and Figure 2 ).

The Southern California Firestorm of 2007 was one of 20 fires that caused more than $20 million in property damage (see Table 2 ). Of these 20 fires, 13 were structure fires, 6 were wildland fires, and 1 was a vehicle fire. Together, they resulted in a combined loss of $3 billion, which represents 85.7 percent of the total large-loss fire losses and 20.5 percent of the total fire losses of 2007. The Southern California Firestorm was also one of five fires that resulted in a loss of more than $100 million. The combined loss for these fires was $2.5 billion, or 61.4 percent of the large-loss losses and 17.1 percent of the total fire losses in 2007.

From 1998 to 2006, there was a generally downward trend in the number of large-loss fires and explosions, but the 2007 total was a sharp departure from that trend. Total loss in large-loss fires is too sensitive to the specific loss in the costliest incidents each year to show any clear trend

Where the fires occurred
In 2007, 71 large-loss fires occurred in all major property categories, except health-care and correctional facilities (see Table 3 and Figure 3 ). Fifty-nine of the 71 fires occurred in structures, causing a total loss of $946.7 million. Ten wildland fires resulted in losses of $2.4 billion, and two vehicle fires resulted in losses of $107.5 million. One of these vehicle fires involved a gasoline tanker truck crash and fire that affected a series of roadway overpasses.

Fifteen of the 59 structure fires occurred in special structures, including 10 buildings under construction, 3 vacant or idle properties, a pier, and a boardwalk. These fires resulted in $211 million in losses.

Eleven fires occurred in manufacturing plants, resulting in losses of $269.8 million, and eight fires in residential structures resulted in losses of $78.5 million. Of these eight fires, five occurred in apartment buildings, two occurred in single-family homes, and one occurred in a lodging or rooming house.

Six fires in stores and office properties resulted in losses of $100.1 million. Five of the six occurred in stores, and one occurred in an office building. There were also five fires each in industrial properties; educational properties, including two high schools, two elementary schools, and a college building; and storage properties, resulting in losses of $100 million, $75.3 million, and $71.5 million, respectively.

Four fires in public assembly properties—two churches, a library, and a country club clubhouse—resulted in losses of $30.5 million.

Information on operating status was reported for 49 of the 59 structure fires. Twenty-seven were operating to some extent. Twenty-two were at full operation or occupancy, three had construction workers on site, one was partially operating, and one had security present. Twenty-two properties were closed or were not known to have anyone on site when the fire broke out.

The cause was reported for 33 of the structure fires, 3 of the wildland fires, and 1 of the vehicle fires. Ten of the structure fires were confirmed to have been incendiary, and two others were suspected of having been incendiary. The 10 incendiary fires were responsible for 5 of the 19 deaths. There were no deaths in the two suspicious fires.

Twenty-four of the fires broke out between 11:00 p.m. and 7:00 a.m. Twenty-one of these were structure fires, one was a wildland fire, and two were vehicle fires. The cause was known in 12 of these fires, with 5 being confirmed incendiary fires. Of the five, four occurred in closed properties, and one occurred in a property that was operating.

Detection and suppression systems
Information about smoke detection equipment was reported for 41 of the 59 structure fires. Twenty-one of these 41 properties, or 51 percent, had some type of automatic detection equipment. Twenty fires occurred in properties with no automatic detection equipment. Of the 21 properties that had equipment, 5 had complete coverage systems­, 3 with smoke detection and 2 with automatic detection of an unreported type. Two were partially covered with smoke detection systems. Five other structures had smoke detection equipment, but the coverage of these systems was not reported. Eleven of the 21 systems operated effectively, while 4 systems failed to operate. Three were not operational yet, and one had been shut down before the fire because the building was vacant. The operation or effectiveness of six of the systems was not reported.

Information on automatic suppression equipment was reported for 48 of the 59 structure fires. Twenty-eight structures had no suppression equipment at all. For the 20 structures with systems, all were sprinkler systems. Of the 20, 10 were wet-pipe systems, 2 were dry-pipe systems, and 2 were combination wet- and dry-pipe systems. The other six were reported simply as sprinkler systems, with no type reported.

Five sprinkler systems were complete coverage. Of these, two were wet-pipe systems, two were combination systems, and one system was unspecified. Five systems had partial coverage. Of these, two were wet-pipe systems, one was a dry-pipe system, and two were unspecified. The coverage of 10 systems was not reported.

Ten of the 20 systems operated. Two of those 10 were partial-coverage systems that operated as designed, keeping the fire out of the protected area. Eight systems operated but were not effective, one because the sprinklers were not in the area of the fire, one due to poor maintenance, one due to poor head clearance, one due to a lack of enough agent available (no explanation was given), one because the main valve in the part of the system in the area of origin was padlocked shut, and three for unreported reasons. Eight systems did not operate. Four of these had been shut off prior to the fire for various reasons, and three were being installed and were not yet operable. No reason was given in the eighth case.

Complete information on both detection and suppression equipment was reported for 39 of the 59 structure fires. Eleven of the structures had only detection equipment, 7 had only suppression equipment, 8 had both detection and suppression, and 13 properties, or 22 percent of all structures, had no coverage.

What we can learn
Each year’s large-loss study focuses on the relatively small number of fires that account for a significant portion of the dollar loss that year. In 2007, 71 fires accounted for approximately a quarter of the year’s losses. Table 4 provides details on the fires resulting in losses of $20 million or more.

Details on all 71 large-loss fires in 2007 can be found in the full report, available through NFPA’s Fire Analysis and Research Division and on our website at www.nfpa.org under Research & Reports.

As stated earlier, the number of large-loss fires increased by over 50 percent, and the property loss increased by more than $3 billion, over half of which was the result of one fire. In seven of the past 10 years, at least one fire has caused a loss of over $100 million. In 2007, there were five such incidents, resulting in a total loss of $2.5 billion. In the past 10 years, there have also been eight wildland fires that each resulted in more than $100 million in losses.

Adherence to the fire protection principles reflected in NFPA’s codes and standards is essential if we are to succeed in reducing the occurrence of large-loss fires and explosions in the United States. A range of ignition causes and factors was reported among the large-loss fires in 2007, including incendiary, abandoned or discarded smoking materials, mechanical or part failures, short circuits, combustibles too close, and unattended candles. Proper design, maintenance, and operation of fire protection systems and features can keep a fire from becoming a large-loss fire. Proper construction, storage methods, and house cleaning will make fires less likely and help control or limit fire spread, if a fire occurs.

Financial and economic losses are one thing, but the emotional toll of losing your dreams and future hopes cannot be calculated in dollars. That is one thing all the fires in this report have in common.

Acknowledgments
NFPA thanks the U.S. fire service for its contributions of data, without which this report would not be possible. In many cases, the fire departments were unable to contribute complete details to NFPA because legal action was pending or ongoing, or they were unable to determine many pieces of information we need to make our study as complete as possible. The author wishes to thank Rita Fahy and Norma Candeloro for providing the support this study required.