Cover Image
close this bookCrucibles of Hazard: Mega-Cities and Disasters in Transition (United Nations University, 1999, 544 p.)
View the document(introduction...)
View the documentAcknowledgements
View the document1. Introduction - James K. Mitchell
View the document2. Natural disasters in the context of mega-cities - James K. Mitchell
View the document3. Urbanization and disaster mitigation in Tokyo - Yoshio Kumagai and Yoshiteru Nojima
View the document4. Flood hazard in Seoul: A preliminary assessment - Kwi-Gon Kim
View the document5. Environmental hazards in Dhaka - Saleemul Huq
View the document6. Natural and anthropogenic hazards in the Sydney sprawl: Is the city sustainable? - John Handmer
View the document7. Disaster response in London: A case of learning constrained by history and experience - Dennis J. Parker
View the document8. Lima, Peru: Underdevelopment and vulnerability to hazards in the city of the kings - Anthony Oliver-Smith
View the document9. Social vulnerability to disasters in Mexico City: An assessment method - Sergio Puente
View the document10. Natural hazards of the San Francisco Bay mega-city: Trial by earthquake, wind, and fire - Rutherford H. Platt
View the document11. There are worse things than earthquakes: Hazard vulnerability and mitigation capacity in Greater Los Angeles - Ben Wisner
View the document12. Environmental hazards and interest group coalitions: Metropolitan Miami after hurricane Andrew - William D. Solecki
View the document13. Findings and conclusions - James K. Mitchell
View the documentPostscript: The role of hazards in urban policy at the millennium - James K. Mitchell
View the documentAppendices
View the documentContributors
View the documentOther titles of interest

10. Natural hazards of the San Francisco Bay mega-city: Trial by earthquake, wind, and fire - Rutherford H. Platt

Editor's introduction

To outsiders and residents alike, the earthquake and fire of 1906 seem to be a permanent part of the image of San Francisco and the San Francisco Bay Area. Moreover, research has shown that the 1906 event is one of the most frequently employed yardsticks by which all earthquakes - and indeed all urban natural disasters - are evaluated throughout the United States. But it is not an entirely accurate symbol of metropolitan hazard. For one thing, geologists and historians point out that larger earthquakes and fires also affected an (admittedly smaller) San Francisco in the previous century. For another, the local record of natural disasters includes many different kinds of event from slope failures and flash-floods to urban wildfires, all of which have added contrasting strands to the region's hazard experience. Finally, preparedness and mitigation initiatives in the decades since 1906 have upgraded the Bay Area's hazard-resistance capabilities. Though there is still far to go before San Francisco - and more especially the East Bay communities - can be declared models of sound hazard management, there are also grounds for believing that they are not all teetering on the brink of an apocalypse.


The mega-city that enfolds San Francisco Bay holds a special status in North America. Its core community is considered by many to be the most beautiful and vibrant urban place on the continent. The city of San Francisco is justly renowned for spectacular views of ocean cliffs and sheltered anchorages, a temperate marine climate adorned with rolling fogs, the sublime Golden Gate Bridge, captivating public parks, striking architecture, cable cars, restaurants, theatres, opera, symphony, art museums, a flourishing Chinatown, and a peerless night-life. It is one of the world's truly cosmopolitan cities.

Several other parts of the Bay Area are celebrated in their own right. Marin County is the quintessence of American suburban affluence; the Sonoma and Napa valleys comprise California's premier wine-growing region; Berkeley and Palo Alto are seats of distinguished universities; and Silicon Valley contains the global nexus of post-industrial technology. These superlative characteristics notwithstanding, the Bay Area mega-city (fig. 10.1) is a place of great physical and human hazard. In 1906, San Francisco was substantially destroyed by an 8.2 Richter magnitude earthquake and subsequent fires that spread from ruptured chimneys and exploding gas mains. On 17 October 1989, the 7.1 magnitude Loma Prieta earthquake, centred 100 km south of San Francisco, reminded residents that another "Big One" may not be far in the future. In the words of a British Broadcasting Corporation film, this is "The City that Waits to Die." Meanwhile, the Bay Area's people have been dying in a different and more literal sense. San Francisco was where the AIDS epidemic first came to public attention in the United States as it ravaged the city's prominent homosexual community during the 1980s. Exuberance and foreboding are opposite sides of life in this mega-city.

The Bay Area urban region is officially designated as the "San Francisco - Oakland - San Jose CMSA" (Consolidated Metropolitan Statistical Area).1 With a population of 6.3 million, it is now the fourth-largest conurbation in the United States after New York, Los Angeles, and Chicago. The CMSA extends approximately 200 km from Santa Cruz in the south to Santa Rosa in the north, and about 65 km from the Pacific Ocean to the city of Livermore, east of San Francisco Bay. It contains nine counties and several hundred municipalities in a belt of territory that roughly encircles the Bay and its northerly extensions, San Pablo Bay and Suisun Bay. The city of San Francisco (724,000) contains just 11.6 per cent of the total population.

1 Since 1950, the federal Bureau of the Census has designated "Metropolitan Statistical Areas" (MSAs) that encompass central cities and suburban areas. Periodically, MSA boundaries are revised (normally by adding more county units) or new MSAs are designated to reflect the outcome of continued urbanization. Where two or more MSAs are directly contiguous or functionally overlapping, the Census Bureau also designates "Consolidated Metropolitan Statistical Areas" (CMSAs) and refers to the individual MSAs within the CMSA as "PMSAs" (Primary MSAs).

Like San Francisco itself, the Bay Area is a region of widespread physical and human-caused hazards. Foremost among these is the risk of earthquakes along a network of north - south-trending faults that underlies the mega-city and stretches far beyond (fig. 10.2). More localized physical risks include landslides, flooding, soil liquefaction, and wildfires. Compounding the effects of natural disasters is a significant potential for failures of lifelines, including transportation, communications, water and sewer pipes, dams, energy systems, food, and medical care delivery. And, like other mega-cities, this one also suffers daily human-created hazards of traffic congestion, air and water pollution, chemical and toxic wastes, and urban crime. But these latter conditions lie beyond the scope of this chapter. Although the term "Bay Area" is synonymous in American culture with natural calamity, the region is earning a reputation for leadership in the development of new approaches to disaster preparedness, response, and recovery. As recounted below, this mega-city can mobilize massive resources - fiscal, technical, and intellectual - in the ongoing process of fitting its built environment to the pervasive natural limitations of its site.

Fig. 10.1. The San Francisco Bay Area (Source: Association of Bay Area Governments)

Fig. 10.2. Faults of the Bay Area (Note: Percentage figures indicate chance of occurrence of an earthquake in the 30 years from 1988. Source: Plafker and Galloway, 1989)

Physical setting

Unlike the Atlantic coast, the Pacific littoral offers few safe natural harbours. Between Puget Sound to the north and San Diego Bay to the south, there is only one major opening through the treacherous sea cliffs and coastal barriers: the Golden Gate. This fabled strait leads from the open ocean into the vast, sheltered recesses of San Francisco Bay, which is - in effect - an inland sea extending about 100 km from north to south. The Sacramento River, which drains into the Bay, historically afforded access by riverboat to the state's interior central valley. San Francisco thus possessed three physical advantages that ensured its rapid growth: (1) a sheltered deepwater port; (2) easy water access to cross-bay satellite communities (e.g. Berkeley, Oakland, Alameda); and (3) river access to a vast agricultural and mining hinterland that extended eastward to the Sierra Nevada mountains.

The site of San Francisco has always appeared highly favourable to urban settlement. In the words of one of the city's early enthusiasts:

[It] seems topographically marked for greatness, rising on a series of hills, with a great harbor on one side, a great ocean on the other,... mighty waters ever passing by to the outlet of the widespread river system of the country. (Bancroft, 1888; quoted in Vance, 1964, pp. 4 - 5)

Lying at the northern tip of a broad peninsula, it is accessible from the sea or overland from a region of fertile valleys and low hills to the south. This site was originally bordered by bay-side marshes that were easily drained and converted to buildable land. Hills on the peninsula (e.g. Nob Hill, Telegraph Hill, Russian Hill) afforded fresh breezes and superb views for wealthy inhabitants. The imposition of a typical American grid street system on irregular and occasionally steep slopes fostered San Francisco's most famous means of transportation, the cable car, and contributed to the cardiovascular well-being of local residents!

In the midst of these favourable attributes lies the mega-city's nemesis, the San Andreas Fault. It marks a major boundary between the Pacific tectonic plate and the North American plate. Skirting the mainland north of the Bay Area, this fault passes San Francisco just offshore and crosses the coast a few kilometres south of the present city limits. It continues inland, bisecting the base of the peninsula and extending south-east towards southern California (fig. 10.2).

The eastern side of the bay ("East Bay") was originally fringed by a marshy shoreline and backed by steep topography (e.g. San Pablo Hills, San Leandro Hills). This shore has been extensively reclaimed by filling wetlands, and subsequent urban development has occurred on unstable soils that are subject to mobilization during earthquakes. Another fault (the Hayward Fault) marks the junction between bay-side flatlands and hills a few kilometres inland. This branch of the San Andreas system is "an offspring even more dangerous than its treacherous parent" (Yanev, 1991, p. 36). Most predictions of future Bay Area earthquakes focus on the Hayward Fault (see below).

Originally cloaked in grasslands with pockets of oak, the East Bay Hills blocked expansion of settlement beyond the bay shore during the nineteenth century. With the advent of automobiles and improved building technologies, the steep west-facing slopes of the hills, which have spectacular views of the bay and distant San Francisco, became te residential districts beginning in the 1920s. There the proximity of structures, ornamental vegetation, and residual grasslands has always posed a chronic risk of wildfire at times of drought. Much of this area burned in the East Bay Hills "firestorm" of 20 October 1991.

Settlement and growth

European settlement first reached the Golden Gate in 1776 when the Mission of San Francisco de Asis was founded by Spanish Franciscan monks and military authorities. This mission and its associated presidio formed the nucleus of the future city. The United States gained sovereignty over California from Mexico in 1846. Until then, American contact with this area was largely via New England-based ships trading along the coast in competition with Russian vessels. An early (1835) American visitor, Richard Henry Dana, described the locality without making reference to its yet-unrecognized lurking hazards:

A few days after our arrival [in San Francisco Bay], the rainy season set in, and for three weeks it rained almost every hour, without cessation. This was bad for our trade.... The mission of San Francisco, near the anchorage, has no trade at all, but those of San Jose, Santa Clara, and others, situated on large creeks or rivers which run into the bay, and distant between fifteen and forty miles from the anchorage, do a greater business in hides than any in California. Large boats, manned by Indians, and capable of carrying nearly a thousand hides apiece, are attached to the missions, and sent down to the vessels with hides, to bring away goods in return. (Dana, 1964, pp. 221 - 222)

A few years later, San Francisco would be transformed, first by political annexation to the United States and second by the discovery of gold in the Sierra Nevada mountains. The promise of riches attracted thousands of "49ers" from Back East and abroad to pan for gold in the streams and to dig mines on newly established claims issued by federal land offices. San Francisco became the commercial centre for the region by serving the needs of would-be miners in transit and those who returned with sudden wealth. Banks, warehouses, retail stores, and hotels quickly appeared along wide, new streets leading from the wharves straight up the hills and across the peninsula. Agriculture and industry also contributed to the local urban economy. The city's West Coast economic primacy was assured when rail lines arrived from the East in the 1860s, terminating in Oakland.

By the 1850s, San Francisco was beginning to emerge as a city of wealth, philanthropy, gracious architecture, and culture. Separated from the East Coast by a week of rail travel, and from Europe by at least a month at sea, it took on the sobriquet "Paris of the West Coast." In 1866, the New York landscape architect Frederick Law Olmsted, Sr. was invited to propose a plan for what would eventually become Golden Gate Park, the San Francisco counterpart to his celebrated design for New York's Central Park. By 1875, the Palace Hotel, financed with Comstock Lode mining profits, opened as "the finest hotel in the world" (until its destruction in the 1906 fire). Thirty years later (1905), the Association for the Improvement and Adornment of San Francisco received a "City Beautiful" plan which it had commissioned from the great Chicago architect Daniel H. Burnham. This plan was adopted a month after the earthquake, but much of it was never accomplished. Although the Bay Area was tied commercially and culturally to the Atlantic world, it was also linked by sea to distant ports of the Pacific Rim. It was there that contemporary boosters envisaged further economic gains:

The great triangle of the Pacific is destined to have its lines drawn between Hong Kong, Sydney and San Francisco. Of these three ports, Hong Kong will have China behind it, Sydney, Europe, and San Francisco, America; and with America for a backing, San Francisco can challenge the world in the strife for commercial supremacy. (Keeler, 1903, p. 94)

San Francisco remained the financial and cultural capital of the region, while the East Bay communities of Oakland, Berkeley, and their neighbours flourished both as bedroom suburbs to San Francisco and as separate cities (table 10.1). Oakland later took on the role of major ocean port and military naval facility. Berkeley, home to the University of California's leading campus, became - and still remains - an icon of counter-culture and social protest.

Table 10.1 Populations of San Francisco and East Bay cities, selected years 1870-1990


San Francisco


































Sources: Vance (1964) and Bureau of the Census, Statistical Abstract of the United States, Washington, D.C., 1992.

The 1906 earthquake and its aftermath

By the turn of the century, San Francisco had a population of one-third of a million. Many other cities had experienced catastrophic fires, but no city in North America had been consumed by earthquake-related fires. Like the sinking of the ocean liner Titanic in 1912, the San Francisco earthquake and fire of 1906 shattered conventional assumptions about the ability of "modern technology" to overcome natural perils. And just as the loss of the Titanic led to tighter safety requirements for ocean vessels and an ice patrol in the North Atlantic, the San Francisco earthquake stimulated actions there and elsewhere to mitigate future disasters.

The earthquake struck at 5:12 a.m. on Sunday, 18 April, as a tremor that lasted 40 seconds. A major aftershock took place 13 minutes later, followed by many others of lesser magnitude. Because it occurred at an early hour, deaths were relatively few; falling masonry would have been lethal to daytime crowds, especially in the downtown business district of San Francisco. On the other hand, the duration of shaking proved too much for the city's hastily built housing stock and many victims were trapped under debris. Some structures collapsed, others leaned, and thousands of masonry chimneys fell onto roofs, in some cases starting fires. Fissures opened in pavements miles from the San Andreas Fault and hundreds of church bells pealed as the earthquake jolted terrified inhabitants awake (Bronson, 1986). Nearby cities such as San Jose and Palo Alto were badly damaged, but Oakland and Berkeley were less affected and were able to render vital assistance to San Francisco. The earthquake also struck many other communities along a 320 km corridor between Salinas and Fort Bragg.

Post-earthquake fires took some time to get under way. Meanwhile, residents fled to the streets, unaware that most of the damage was yet to come. Unlike other great urban fires (e.g. London in 1666, Chicago in 1871, Oakland in 1991), wind was not a major factor; nor was the region experiencing a drought. However, near total failure of the water supply contributed directly to the disaster. Before the quake it was already known that San Francisco's water system was inadequate for fighting a major fire (Bronson, 1986). The earthquake promptly disabled the system by rupturing water mains and cracking storage tanks. As gas lines exploded and cooking fires set homes ablaze, the fire department could not extinguish even modest fires. Water from the bay was pumped onto some nearby structures - an event that was to be repeated in the Mission District of San Francisco after the Loma Prieta quake 83 years later. Some use was made of dynamite to clear fire breaks. But, by and large, firemen looked on helplessly as the city burned.

The San Francisco earthquake and fire of 1906 are widely considered America's worst urban natural disaster and one of the world's great urban conflagrations. Flames burned over 1,146 hectares or 490 blocks - an area 50 per cent larger than that devastated by the Chicago fire of 1871 and six times bigger than the district that was burned by the "Great Fire" of London in 1666 (see Platt, 1991, and chap. 7 in this volume). Approximately 500 people died as a direct result. Three-fifths of the city's inhabitants lost their homes and the entire business district was destroyed (Bronson, 1986). The 1906 catastrophe illustrated many characteristics of a contemporary mega-city natural disaster. Multiple interrelated hazards were involved (i.e. earthquake and fire). Lifelines (water, communications, transportation) failed and caused secondary impacts. Widespread structural damage illuminated the inadequacy of existing building standards and the prevalent use of wood for smaller structures. Much of the working-class population lost homes and jobs. Finally, aid from external sources assisted the immediate response and to a lesser extent enabled the longer-term recovery.

Despite many similarities with modern disasters, there was one striking contrast. Missing from the San Francisco catastrophe was any significant economic or material assistance from the national (federal) government. The US Army helped to keep order and allocated supplies, but otherwise most of the aid came from private or quasi-public sources. The Red Cross was entrusted with overall coordination of the relief effort. Financial aid totalling US$ 9 million (1906 dollars) was provided by voluntary contributions from individuals and other cities (Bronson, 1986). International assistance totalling US$ 474,000 was also contributed from 14 foreign nations. But the response was largely a spontaneous, "grass-roots" reaction to a widely publicized disaster. Significant federal participation in disaster assistance and recovery did not begin in the United States until the 1950s.

The foremost public action in response to this disaster was a dam, reservoir, and aqueduct system that conveyed water from the Hetch Hetchy valley in the Sierra Nevada over a distance of 190 miles to the city. San Francisco's proposals for the Hetch Hetchy dam precipitated a precedent-setting 10-year-long controversy between advocates of wilderness preservation, headed by John Muir (founder of the Sierra Club), and proponents of the "wise use of natural resources" represented by Gifford Pinchot, Director of the US Forest Service and adviser to President Theodore Roosevelt. The dam was finally approved and construction began in 1913 during the administration of President Woodrow Wilson (Nash, 1982). The Hetch Hetchy dam and reservoir now provide San Francisco with high-quality drinking water, but the system's reliability in the event of another major earthquake remains in doubt. Pipes from the reservoir cross the Hayward Fault and could rupture in an earthquake, once again leaving San Francisco without its primary water supply.2

2 The East Bay Municipal Utility District faces a similar threat. Aqueducts from Pardee Reservoir just north of Hetch Hetchy also cross the Hayward Fault.

San Francisco recovered rapidly from the 1906 catastrophe chiefly because the downtown business district was well insured. Total insurance payments amounted to US$ 5.44 billion (1992 dollars), by far the largest urban fire insurance loss in American history.3 Reconstruction of the central city began immediately. By 1910, 417,000 people were in residence, an increase of 22 per cent since 1900 (table 10.1). Housing was constructed rapidly both to replace what was lost and to accommodate newcomers. However, like London after 1666, leaders of San Francisco declined to alter the city's basic pattern of streets and land uses during the rebuilding process. They even ignored Daniel H. Burnham's "City Beautiful" plan for the redesign of the city:

Few cities ever found themselves demolished, with a ready-made plan for a new and grander city already drawn up, awaiting implementation, and with money pouring in to help realize the plan. San Francisco chose to ignore its Burnham Plan, and decided instead to build at a rate and manner which made the city not only less beautiful than was possible, but more dangerous. The rubble of the 1906 disaster was pushed into the Bay; buildings were built on it. Those buildings will be among the most vulnerable when the next earthquake comes. (Thomas and Witts, 1971, p. 274)

3 The Chicago fire of 1871 cost US$ 1.96 billion in insured losses, and the Oakland fire- storm of 1991 required US$ 1.75 billion in insurance payments (San Francisco Chronicle, 31 October 1993, p. B1).

This statement was prophetic: the city's Marina District, built on 1906 rubble, was badly damaged in 1989.

The Loma Prieta earthquake of 1989

The Bay Area was relatively unscathed by natural disasters for 83 years after 1906. During this period, the relative dominance of San Francisco diminished. Urban development flourished elsewhere in the mega-city, greatly facilitated by the building of the Golden Gate and Oakland Bay bridges during the 1930s. Decentralization was further encouraged after the Second World War by federal guarantees for new home construction in outlying areas, by new connecting freeways, and later by the Bay Area Rapid Transit (BART) system, which links East Bay cities together and includes San Francisco. The combined population of the East Bay (Oakland PMSA) now exceeds that of San Francisco and is growing faster (table 10.2).

Table 10.2 Population change in San Francisco Bay Area, 1970-1990 ('000)




% change 1980-90

San Francisco-Oakland-San Jose CMSA





San Francisco PMSA





Oakland PMSA





San Jose PMSA





Santa Cruz PMSA





Santa Rosa PMSA





Vallejo PMSA





City of San Francisco





City of Oakland





Source: Bureau of the Census, Statistical Abstract of the United States, Washington, D.C., 1991, table 36.

CMSA = Consolidated Metropolitan Statistical Area
PMSA = Primary Metropolitan Statistical Area (PMSAs are components of a CSMA)

Table 10.3 Probability of one or more large earthquakes on the San Andreas Fault System, as estimated in 1988

Expected magnitude


5 years

10 years

20 years

30 years

San Francisco Bay Area






Southern San Andreas






San Jacinto Fault






Source: US Geological Survey Working Group (1988).

During the long period of deceptive quiet that followed the 1906 earthquake, another major earthquake on the San Andreas Fault was awaited by scientists. In 1988, a US Geological Survey Working Group on California Earthquake Probabilities (1988) estimated that a magnitude 7.0 earthquake would occur on the northern San Andreas (Bay Area) with a probability of 10 per cent within five years and 50 per cent within 30 years (table 10.3).

Concurrently, the US Geological Survey and the California Office of Emergency Services (OES) were closely monitoring seismic activity in the region. Two small earthquakes in June 1988 and August 1989 persuaded OES to advise local governments about an impending larger earthquake. This represented "the most significant use of earthquake forecasting for public policy purposes since the 1975 Haicheng earthquake was predicted in the People's Republic of China" (FEMA, 1990, p. 15). These forecasts were soon fulfilled: the San Andreas Fault released a 7.1 magnitude earthquake at 5:04 p.m. (local time) on 17 October 1989. The epicentre was located about 96 km south-south-east of San Francisco, near the community of Loma Prieta. The fault ruptured over a distance of 40 km at a depth of 19 km below the surface; no surface faulting appeared (FEMA, 1991; Bolin, 1993). Seismic shaking lasted 15 seconds, followed by numerous aftershocks. The event was felt over an area of about 1 million km2 extending southward as far as Los Angeles and northward to the Oregon border (fig. 10.3). It caused 62 known deaths and 3,757 injuries, and left 12,000 people homeless. Property damage of over US$ 6 billion was sustained and there were severe disruptions of public transportation, utilities, and communications (Plafker and Galloway, 1989).

This earthquake coincided with the beginning of the third game of a baseball World Series that ironically matched two teams from the Bay Area: the San Francisco Giants and the Oakland Athletics. Sports commentators quickly shifted from reporting game preparations to the unfolding spectacle of disaster. It was an unparalleled opportunity to raise public awareness of seismic hazard. Warnings of a greater disaster to come were trumpeted by the news media around the world, as on the cover of Newsweek (30 October 1989): "Bracing for the Big One: The Lessons of the San Francisco Earthquake."

Infrastructure damage

French (1990) has identified four consequences of infrastructure damage associated with the Loma Prieta earthquake:

1. direct physical and economic damage to the systems;
2. diminished ability to carry out emergency response activities;
3. inconvenience due to temporary service interruption;
4. longer-term economic losses due to limits on recovery.

Although the quake's epicentre was in a rural upland well south of the mega-city, it caused dramatic and costly damage to infrastructure and older private buildings on both sides of San Francisco Bay. The most deadly outcome involved the collapse of a double-deck commuter freeway that was constructed on bay mud in Oakland. A 2.4 km section of this road failed, trapping hundreds of vehicles and accounting for 41 of the deaths attributed to the earthquake. For several days, national television broadcasts of rescue efforts transfixed audiences throughout the country. The last living victim was removed from his car after 90 hours (Newsweek, 30 October 1989, p. 32).

Fig. 10.3. Areas affected by the Loma Prieta earthquake, 17 October 1989 (Source: Plafker and Galloway, 1989)

The only direct road link between San Francisco and the East Bay was severed when a 15 m section of the Bay Bridge collapsed. Its closure for repairs forced commuters to make long detours around the bay, or to rely on the underground BART rapid transit system, which was not damaged. The cross-bay ferry service, which had previously been discontinued after new road and rail links had speeded up travel, was revived. In San Francisco, Interstate Highway 280 and the incomplete Embarcadero Freeway were also damaged. The latter was subsequently scheduled for demolition. Elsewhere, many local roads and highways were blocked by landslides that had been triggered by ground shaking. Damage to buildings was concentrated in communities close to the epicentre, particularly Santa Cruz. But many older unreinforced masonry buildings were damaged in San Francisco and Oakland (including the latter's City Hall). As predicted by Thomas and Witts (1971), damage in both cities was primarily associated with seismic mobilization of unconsolidated filled lands along the bay. Just as they had in 1906, local water mains ruptured, leaving firefighters without sufficient pressure to tackle large blazes. A fireboat helped to pump bay water and local citizens organized bucket brigades to save the Marina District.

Overall, much damage was related to the amplification of seismic energy in unconsolidated soils rather than to distance from the epicentre (FEMA, 1990). Critical facilities, including several hospitals, were damaged in locations where hazardous soil conditions could have been anticipated. Stanford University in Palo Alto, which was severely affected by the 1906 earthquake, suffered losses of US$ 160 million as a result of the Loma Prieta earthquake. Public schools were generally not severely damaged, probably as a result of statewide earthquake construction codes that were adopted after the 1933 Long Beach earthquake in southern California. The East Bay Municipal Utility District, which supplies water to 1.5 million people, recorded approximately 200 local water mains breaks, but its principal supplies were not affected (FEMA, 1990).

Emergency response

Preparation for a major earthquake in northern or southern California has been under way for many years. A large earthquake in either the Bay Area or Greater Los Angeles has the potential to become a national, or even an international, catastrophe (e.g. through its impact on the global financial and insurance industries). Planning involves not only local, county, and regional authorities but also the State of California, the Federal Emergency Management Agency (FEMA), and other federal agencies. Only three months before Loma Prieta, a training exercise ("Response '89") tested response capabilities in the event of a hypothetical 7.5 magnitude earthquake on the Hayward Fault. After the real earthquake, FEMA reported that emergency responders - both governmental and non-governmental - functioned smoothly but resources were stretched to the limit: "The entire system of emergency responders had reached a total saturation point, and if the disaster had been of any larger proportions, the total system would have broken down" (FEMA, 1991, p. 40).

Emergency response was hindered by overload of telephone circuits, blockage of highways, rupture of gas lines, loss of water pressure, and interruption of electrical power supplies. But several factors helped to reduce potential losses: the epicentre was relatively remote from most of the Bay Area; schools were not in session; it was a windless day. Commuter traffic was also less than normal for 5 p.m. because many people were indoors watching the World Series. Key personnel were still at their workplaces and available to staff emergency operations centres within minutes after the earthquake.

Property damage and individual mitigation

The earthquake damaged more than 22,000 residential structures, 1,567 commercial buildings, and 137 public buildings (FEMA, 1991, p. 21). Local building inspectors marked 3,957 structures with "red tags" to indicate that they were uninhabitable. Most of the damaged buildings had been constructed before the state's Uniform Building Code was updated in 1973 (and revised further in 1988) to reflect seismic risks. Modern office towers in downtown San Francisco swayed as they were designed to. The preventive removal of projecting masonry and ornamentation from most older buildings also helped to avoid showers of rubble on people in the streets below (Plafker and Galloway, 1989), although 12 people died from this cause. Unreinforced masonry construction had long been prohibited for new buildings, thus reducing the risk of total building collapse as occurred during recent earthquakes in Mexico City, Armenia, and elsewhere. Damage to neighbourhoods, homes, and personal property was widespread throughout southern and central parts of the Bay Area. A summary of sample damage incurred by residents of San Francisco and Santa Cruz counties is provided in table 10.4 (O'Brien and Mileti, 1993). Approximately two-thirds of San Francisco respondents and 83 per cent of those in Santa Cruz County reported that their homes suffered slight to moderate damage. Damage in "the neighbourhood" was reported by 85.3 per cent of those contacted in San Francisco and by 98 per cent in Santa Cruz.

The geographic spread and scale of damage are all the more noteworthy in light of the much-documented failure of Californians to purchase earthquake insurance and their reluctance personally to engage in most other types of hazard mitigation. In 1990, the California Legislature established a mandatory earthquake insurance programme that provided up to US$ 15,000 in benefits; this was funded out of a surcharge on household insurance policies. The propensity to buy earthquake insurance appears to have been unaffected by the Loma Prieta experience, for Palm and Hodgson (1992) report that there was no substantial increase in earthquake insurance purchases thereafter.4 Moreover, those who bought insurance tended to be located further from Loma Prieta (table 10.5).

4 Before 1990, earthquake insurance in California was available from private insurance companies for premiums on the order of US$ 2 - 4 per thousand dollars of coverage, subject to a large deductible of 10 per cent of insured value. Thus home-owners had to pay a premium of several hundred dollars a year and were still not covered against moderate damage that did not exceed the 10 per cent deductible. Fewer than half the state's residents carried earthquake insurance. In 1990, after Loma Prieta, the state legislature established a mandatory programme of minimum earthquake coverage (Palm and Hodgson, 1992, chap. 2).

Table 10.4 Types and distribution of damage due to Loma Prieta earthquake reported by residents surveyed in San Francisco and Santa Cruz counties

San Francisco County

Santa Cruz County





Neighborhood damage





















Residence damage





















Household items





















Source: Modified from O'Brien and Mileti (1993), table 3.

Even in Santa Clara County, whose western boundary virtually coincides with the segment of the San Andreas Fault that ruptured in 1989, the prevalence of earthquake insurance reported by survey respondents was just 51 per cent in 1990. Only 29 per cent of respondents in Contra Costa County, which experienced widespread effects of Loma Prieta, carried earthquake insurance. On the other hand, most victims would not have benefited greatly from such insurance because only 3.8 per cent of residence damage reported for Santa Cruz County (adjoining Santa Clara County) was rated as "severe." Most of the costs of "slight" and "moderate" damage would probably not have exceeded the 10 per cent deductible and would thus have been borne by owners. Palm and Hodgson (1992) also found a comparable reluctance to undertake other personal mitigation activities involving any monetary cost (table 10.5).

About 12,000 people were made temporarily homeless by the earthquake, many of them poor. As in other disasters, the Red Cross moved quickly to set up shelters and emergency feeding stations. About one week after the event, 44 shelters were in operation. These housed approximately 2,500 persons per night, with the remaining 9,500 either sharing accommodation with friends and relatives or living in tents, vehicles, and other improvised quarters (Bolin and Stanford, 1993).

Table 10.5 Earthquake mitigation actions taken before (1989) and after (1990) Loma Prieta by residents surveyed in four California counties

Survey counties

% with earthquake insurance before Loma Prieta

% increase in earthquake insurance after Loma Prieta

% incurring expense for other earthquake mitigation actions



Santa Clara





Contra Costa





Los Angeles





San Bernardino





Source: Modified from Palm and Hodgson (1992), fig. 6 and table 13.

Federal emergency response and mitigation

Unlike in 1906, the US national government now regularly provides massive assistance in the wake of natural disasters. This is triggered by a "major disaster declaration" issued by the President and affecting specified counties. The lead federal agency - FEMA - had been widely accused of reacting slowly to hurricane Hugo, a major hurricane that spread devastation across a vast area, including the Virgin Islands, Puerto Rico, and North and South Carolina, two months before the Loma Prieta earthquake. In that case, 113 counties, most of them rural and isolated, were declared eligible to receive federal aid. By contrast, the Loma Prieta disaster prompted a Presidential declaration for only 12 counties in a compact zone within or adjacent to the Bay Area. In the glare of national media coverage, FEMA rushed to open 18 Disaster Assistance Centers, including five that were mobile. By the end of January 1990, 77,654 individuals, families, and businesses had registered at these centres for federal assistance, including about 40,000 applicants for temporary housing and 31,255 applicants for individual and family grants of US$ 11,000 per household. The Small Business Administration issued 34,976 home and personal property loans and 14,773 business loans. Federal assistance by then had amounted to about US$ 202 million (FEMA, 1990, p. 27). As of August 1993, total federal costs for the Loma Prieta disaster were projected to reach US$ 689 million (FEMA, unpublished data).

The victim population, like that of California in general, was highly diverse. Special problems in the administration of disaster assistance, particularly temporary housing, appeared in some localities. According to Bolin and Stanford (1993, p. B48), sheltering problems in Watsonville and elsewhere reflected longstanding ethnic and class antagonisms: "The earthquake and the subsequent housing crisis highlighted the preexisting deficiencies and inequities in Watsonville housing and provided the circumstance in which a new political agenda could be formulated, with housing as its focus."

One dispute involved demands by Mexican-Americans that mobile homes provided as emergency housing remain indefinitely because they were more habitable than the substandard housing that was damaged by the earthquake. It is an irony of US disaster policy that the federal government is empowered to address social needs in an emergency recovery context (albeit temporarily) that are often overlooked in the absence of a disaster. The post-disaster problems of California's poor foreign-born population may increase if the provisions of California Proposition 187 (passed by voters in November 1994) are observed; these may prohibit certain kinds of disaster relief to illegal aliens.

The Federal Disaster Relief and Emergency Assistance Act (Stafford Act) requires that:

the State or local government [receiving disaster assistance] shall agree that the natural hazards in the areas in which the proceeds of the grants or loans are to be used shall be evaluated and appropriate actions shall be taken to mitigate such hazards, including safe land-use and construction practices. (Sec. 409)

Accordingly, FEMA and the California Office of Emergency Services convened a State/Federal Hazard Mitigation Survey Team two weeks after the earthquake. The team involved 60 participants, including 32 federal, state, regional, and private utility agencies with earthquake-preparedness responsibilities. Their ensuing report (FEMA, 1990) offered 57 recommendations under six headings: (1) hazard identification and monitoring; (2) land-use planning and regulation; (3) repair and reconstruction; (4) response planning; (5) insurance; and (6) funding.

The next Bay Area earthquake

Once the dust settled, it was clear that the Loma Prieta earthquake was not the "Big One" that has been long expected in the Bay Area. A new working group was assembled by the US Geological Survey to assess the risk of further large earthquakes in the region. This sought to: (1) reassess earlier seismic data in light of Loma Prieta; (2) evaluate the effects of Loma Prieta on other faults in the Bay Area; and (3) incorporate new data on slip rate and recurrence interval on the Hayward and Rodgers Creek faults. The report of this group estimated a 67 per cent probability of one or more large earthquakes in the Bay Area within the next 30 years. This is a substantial increase over the 50 per cent probability estimated by the 1988 Working Group (US Geological Survey Working Group, 1990).

The Association of Bay Area Governments (ABAG, 1992) has compiled estimates of housing units likely to be rendered uninhabitable by a 7.0 magnitude quake (table 10.6). These figures reflect only initial damage inflicted by ground shaking and do not include possible further damage owing to fires, landslides, and other causes. Nevertheless, for almost all of the scenarios that were projected, the numbers of dwellings rendered uninhabitable are greater than those affected by the Loma Prieta quake. For one scenario (i.e. an earthquake that affects both north and south portions of the Hayward Fault) up to 29 times as many houses might be involved. The costs to insurers, government, and home-owners would all be vastly greater than those incurred in 1989. Post-disaster problems of emergency response, sheltering, and rehousing are likely to be overwhelming.

Table 10.6 Estimated total dwelling units rendered uninhabitable by selected earthquake scenarios

Fault involved

No. of uninhabitable units

San Andreas - Peninsula


Hayward - north and south


Hayward - north segment only


Hayward - south segment only


Healdsburg - Rodgers Creek


Loma Prieta (as per model)


Loma Prieta 1989 (actual)


Source: Adapted from ABAG (1992), table 1.

Ominously, the next major Bay Area earthquake is most likely to occur on the Hayward Fault in the heavily urbanized East Bay. The Working Group estimated a risk of nearly 30 per cent that an earthquake of at least magnitude 7.0 will occur on the southern or northern segments of the Hayward Fault within the next 30 years. The last earthquakes on those segments occurred in 1836 and 1868, respectively. The Hayward Fault is "probably the most built-on fault in the world" (BAREPP, n.d.). About 1.2 million people live within the epicentral region5 of a potential 7.0 tremor on this fault, 10 times the population within a comparable distance of Loma Prieta (US Geological Survey Working Group, 1990, p. 4). Ground shaking in the East Bay would be at least 12 times larger than during the Loma Prieta quake (USGS, n.d.). It is thus likely that the East Bay will be a vortex of disaster during the next major Bay Area earthquake. Loma Prieta, which stretched emergency response to the limits, may in retrospect be viewed as merely a dress rehearsal.

5 The epicentral region is the area of most intense ground motion on bedrock sites and was considered by the Working Group to extend for a radius of 10 km from the location of an earthquake fault rupture.

The East Bay Hills firestorm of 1991

Two years after the Loma Prieta quake, the Bay Area mega-city was again struck by a major natural disaster. This time the area affected was localized and the victim population relatively wealthy. The East Bay firestorm of 20 October 1991 reminded disaster planners worldwide that natural hazards may be vastly aggravated by unwise building patterns and neglect of commonsense mitigation.

The cities of Oakland and Berkeley both contain prized residential sites on steep west-facing slopes of the East Bay Hills. The combination of bay views, accessibility to urban pursuits, and semi-wild surroundings has long attracted the wealthy. During the 1920s and 1930s, this district became interlaced with narrow winding roads lined with thousands of medium-sized homes on parcels whose average area was only 0.04 ha (5,000 sq. ft). Even by California standards it is an exceptionally hazardous district. The Hayward Fault marks the break of slope at the base of the hills and many public facilities lie directly over it, including the football stadium of the University of California at Berkeley. Landslides, flash-floods, and debris flows are additional geological hazards.

As in many other parts of California, the risk of fire along the boundary between wildlands and urban areas is a function of several related factors:

Vegetation: The native oak and grassland ecosystem has been transformed by the introduction of decorative but highly flammable landscape trees, especially eucalyptus and Monterey pine. Unless removed, dead biomass from eucalyptus trees accumulates as an excellent source of fuel.

Fire suppression: Fuel builds up because natural fires are controlled to protect local residences.

Drought: Drought desiccates vegetation (in 1991 California was just emerging from a five-year drought).

Building materials: Most homes are constructed of wood and are roofed with "shakes" (cedar shingles) in the characteristic style of California woodland architecture.

Narrow roads: Narrow, steep, winding service roads impede the mobility of incoming police, fire, or medical vehicles and obstruct evacuation by car. During fires, abandoned cars easily block such roads.

Wind: Santa Ana winds occur from time to time during the late summer and autumn in California. These hot dry winds are caused by high-pressure systems that form over the semi-arid interior Great Basin. They blow from the east and north-east towards the ocean at speeds that are amplified in narrow mountain passes and steep terrain.

Water supply: It is difficult to maintain adequate water pressure in hilly areas. Failures of high-elevation reservoirs are common.

Arson: It is widely believed that arson plays a role in igniting California wildland fires near urban areas.

Floods and debris flows: Once a fire occurs, the burned areas may experience subsequent flash-floods, mudslides, and debris flows triggered by heavy precipitation on devegetated slopes.

The Oakland - Berkeley hills and California as a whole have a long history of urban/wildland fire disasters, and these have increased markedly in the 1990s (table 10.7). Since 1930, 14 fires have occurred in the Oakland Hills, including seven that erupted in the same place (Wildcat Canyon) where the 1991 firestorm began (Topping, 1992, p. 5). Today, about 7 million people are estimated to live in regions of California that are susceptible to urban/wildland fires (FEMA, 1992).

Table 10.7 Major California urban/wildland fires, 1923-1993



Acres burned

Structures burned










Los Angeles








Santa Barbara








San Bernardino




Santa Barbara








San Bernardino




San Bernardino












San Diego












Santa Barbara




Alameda (Oakland)




Southern Calif.



Source: Adapted from California Governor's Office of Emergency Services (1992), Annex A.

Concern about these kinds of fires is both high and widespread in California. After a 1923 fire in Berkeley (Alameda County), a special commission offered more than 100 recommendations for mitigating future losses; most of these were ignored (Blakely, 1992). In 1980, following a series of southern California fires, a Governor's Task Force on Chaparral Fire and Flood Risk Management made recommendations affecting three areas of responsibility: (1) risk reduction, land use, and conservation standards; (2) vegetation management and fire-hazard reduction; and (3) fire command system improvements (Blue Ribbon Fire Prevention Committee, 1982, pp. 64 - 68).

Planning for natural disasters in mega-cities must be conducted at geographic scales that are commensurate with the scope of different hazards. Wildfires threaten many adjacent communities in the East Bay Hills but not the entire Bay Area. In 1982, a Blue Ribbon Fire Prevention Committee for the East Bay Hills was formed under the East Bay Regional Park District, which manages several large tracts of public open space in the Hills. The District served as coordinator for a comprehensive review of fire hazard in the entire Hills region. They issued a report (Blue Ribbon Fire Prevention Committee, 1982) that focused primarily on vegetation management and the need to extend existing "fuel breaks" along the entire 41.1 km length of the urban/wildland boundary. It was also proposed that a new special assessment district be created to manage vegetation throughout the area at risk. Unfortunately, actions in support of these recommendations were not accomplished before the 1991 firestorm.6

6 In view of the intensity of the 1991 firestorm, it has been questioned whether fuel breaks would have impeded the spread of the conflagration. The 1982 Blue Ribbon Committee perhaps overemphasized vegetation management to the exclusion of other mitigation approaches, such as coordination of fire-fighting amongst Bay Area cities (conversation with Tony Acosta, Oakland Park Department, 26 October 1993).

The firestorm

The firestorm was the third most costly urban fire disaster in US history after the San Francisco earthquake of 1906 and the Chicago fire of 1871. Although the area burned (about 647 ha) was modest compared with other California fires, it was by far the most destructive urban/wildland fire in terms of damage to buildings. Twenty-five people died and over 150 were injured. In the space of nine hours, the fire destroyed or damaged 2,621 homes and 758 apartment and condominium units. About 5,100 people, many of them elderly, were left homeless. Damage amounted to at least US$ 1.5 billion (National Fire Protection Association, n.d.).

The fire erupted under textbook conditions. Daytime temperatures hovered around 30°C, relative humidity was 17 per cent, and hillside vegetation was bone dry. Dead plant material littered the ground and trees overhung many homes, despite the long drought and warnings of fire danger. Hot, dry Santa Ana winds blew from the east on the morning of the conflagration. The State Department of Forestry had issued a "Red Flag" warning of potential fire hazard, but few residents took notice.

The blaze began on Saturday, 19 October as a small brush fire "of suspicious origin" high up Wildcat Canyon near the ridge line (fig. 10.4). It was apparently extinguished that day, but embers were re-ignited next day by Santa Ana winds blowing at medium strength. Flames rapidly spread downhill, leaping from house to house. Fire personnel who returned to the scene were overwhelmed. About 790 houses burned in the first hour. Turbulent winds generated by the fire formed a fire- storm that spewed burning material in all directions. Flames crossed an eight-lane highway (Route 24) and continued to consume homes and vegetation further downslope. Public orders to evacuate were difficult to communicate in the absence of sirens.

Fig. 10.4. Area affected by the Oakland firestorm, 20 October 1991 (Source: FEMA, 1992)

The presence of smoke was the primary warning to Hills residents. Many found it impossible to drive down obstructed roads. Cars by the hundreds were abandoned and destroyed as individuals literally ran for their lives. Few of the victims could do more than save themselves and whatever they could carry. Sixty years of building on the hills had created a hazard that no one could individually undo. It was a classic "tragedy of the commons"; all were swept up in the common peril and personal consequences depended upon the fluke of the winds, not individual actions. Emergency response was massive but ineffectual. The city fire departments of Oakland and Berkeley were depleted by funding cuts, and were poorly trained and equipped for fighting a fast-moving urban/wildland fire. Both cities issued urgent requests for assistance from other cities and from federal and state agencies, pursuant to longstanding mutual-aid agreements.7 Despite acts of great individual courage, fire-fighters were overwhelmed. Communication was hampered by overloaded radio and telephone circuits. Smoke and traffic congestion obstructed the mobility of the hundreds of emergency response vehicles and personnel. Even the Oakland hydrants were incompatible with the hose couplings of other municipal fire departments. That problem was however overshadowed by a loss of water pressure. Failure of the water distribution system stemmed from three causes: (1) all but 1 of 11 East Bay Municipal Utility District (EBMUD) reservoirs serving the area ran dry because of outflow through ruptured mains, melted pipes, and open faucets; (2) electrical power failed as transmission facilities burned thus disabling water pumping stations; and (3) EBMUD lacked emergency generators to operate the disabled pumps.8 As in the 1906 San Francisco fire, failure of water pressure left hundreds of fire-fighters as helpless observers.

7 In response, an armada of fire-fighting forces converged on the fire: 440 local engine companies, 6 air tankers, 16 helicopter units, 8 communications units, 2 management teams, and 88 fire chiefs (California Governor's Office of Emergency Services, 1992, p. 31).

8 The FEMA (1990) Loma Prieta report had urged installation of on-site emergency generators by water supply managers, but the district had not done so (Oakland Tribune, 1991, p. 56).

The conflagration was intensely hot and fast moving. Temperatures at the centre of the fire exceeded 1,000°C, hot enough to boil asphalt (National Fire Protection Association, n.d., p. 12). Firestorm conditions began to develop within 15 minutes of the first building ignition. A thermal inversion formed at approximately 1,000 m; this trapped heat from the fire and intensified the pre-ignition heating of vegetation and structures. Flames consumed hundreds of structures simultaneously. Some houses appear to have exploded from within, perhaps due to the intense heating of interior spaces by picture windows that faced the flames (a form of greenhouse effect).9 Propane and other flammable substances also contributed to the loss of many homes. Once a home was ignited, there was little possibility of saving it. All that remained next day were concrete foundations, chimneys, and charred tree trunks (photo 10.1). Fire victims lost personal mementos, pets, art and music collections, and the contents of computer disks - as well as their homes. Many such losses were not covered by fire insurance.

9 Personal communication, Tony Acosta, Oakland Park District, 26 October 1993. News reports on the firestorms in southern California in October 1993 mentioned a similar phenomenon.

The flames reached within a few hundred metres of the opulent and historic Claremont Hotel. If the winds had continued, the fire could have consumed much of Berkeley, including the University of California campus. However, at about 7:00 p.m. on Sunday evening, the wind shifted and declined, nudging flames back over areas already burned. This allowed fire-fighters to contain and ultimately control the fire.

The recovery process: Private sector

Much support was provided by relatives, friends, and neighbours who welcomed victims as temporary house guests and helped them deal with the emotional trauma by listening to their tales of horror and their fears about the future (Adler, 199210). Residents of neighbouring unburned hill areas in Oakland and Berkeley emerged with feelings of relief mixed with anxiety about future fires and perhaps guilt that they were spared.

10 This collection of individual accounts of the disaster through photographs, essays, and poetry was published privately. Proceeds from the book are being donated to the Alta Bates Burn Center.

Photo 10.1. Concrete foundations were all the remained of hundreds of homes burned in the Oakland firestorm.

It is usual for US public officials to vow to rebuild after disaster. The Mayor of Oakland followed this tradition. The day after the disaster he stated that the Hills community would be rebuilt. This was not simply a statement of solicitude for the victims but a clear-sighted recognition that the burned area provided some of Oakland's highest real estate tax revenues and that rebuilding would provide incomes to local merchants and contractors (Edward Blakely, personal communication, 12 August 1993). The alternative of leaving the burned area unbuilt and turning it over to an existing system of hillside regional parks was not seriously considered. A combination of high public acquisition costs and lost tax revenues provided powerful counter-arguments. Nevertheless, the failure to assess the economic and environmental impacts of not rebuilding represents a major shortcoming of the post-disaster recovery process.11

11 Acquisition might have been an economically feasible alternative to reconstruction. For example, the average price of lots sold in the burned area during October 1993 was approximately US$ 130,000 (East Bay Phoenix Journal, 18 October 1993, p. 17). If the approximately 3,000 vacant building lots could have been acquired at this price, the cost would have been around US$ 390 million less administrative costs and lost tax revenues. As compared with the public costs of reconstruction and restoring public streets and utilities to the area (in the order of US$ 100 million), this might not have been excessive. By this means the emotional costs of those who rebuilt under difficult conditions would have been eased. And the risk of future conflagrations would have been greatly reduced. However, Oakland relies on the Hills as its highest tax-generating district and the city would have required payments in lieu of taxes from an external source before it would be persuaded to adopt this approach.

Despite official reticence to explore this alternative, private owners and public officials alike shared a pervasive determination to understand the causes of the disaster and, to the extent feasible, to rebuild more safely in light of risks from fires and earthquakes. Both professional and journalistic post-disaster assessments were produced in abundance. Seldom has a natural disaster been so thoroughly and competently examined from a multitude of disciplinary and personal perspectives. One of the earliest of these was a special series by the Oakland Tribune newspaper entitled "Firestorm: Bitter Lessons." Another journalistic outgrowth of the disaster was the appearance of a bi-monthly regional newspaper, the East Bay Phoenix Journal, published in Oakland. This began as a voice on behalf of fire victims who had to cope with issues of governmental policy, finance, and insurance as well as mounting a search for personal support services.

Unlike its role after the Loma Prieta earthquake, private insurance was a dominant force in financing rebuilding of the East Bay Hills. One year after the disaster, 3,954 claims amounting to US$ 1.4 billion (c. US$ 350,000 per household) had been filed with 49 insurance companies. Many of these were disputed, leading to frustration and often to lawsuits initiated by claimants. However, as settlements were reached on most claims, the fire victims received infusions of money. About two-thirds of the owners had purchased guaranteed replacement-cost insurance coverage. They were also reimbursed for costs of temporary housing and they enjoyed several kinds of relief from federal and state taxes. The average net worth of families who lost homes was expected to rise from 10 to 25 per cent compared with pre-firestorm levels (San Francisco Examiner, 31 October 1993, pp. E-1 and E-5).

Reconstruction of the Hills produced jobs in the building industry at a time when California was mired in recession. During 1993 some 11,000 new construction jobs were created. By 12 October 1993, building permits had been issued for about half the homesites burned in Oakland (1,368 out of 2,777) and for over half those burned in Berkeley (39 out of 69). Homes completed in these communities numbered 434 and 20, respectively (East Bay Phoenix Journal, 18 October 1993, p. 16). This represented over US$ 1 billion already invested in rebuilding, with several years of construction still ahead.

The recovery process: Public sector

Governmental institutions - city, county, special district, state, and federal - assisted the recovery process in many ways. These included emergency response and rescue, infrastructure repair, grants and tax subsidies to the victims, administration of rebuilding activities, and research on causes of the disaster. Research was especially important as a basis for reducing future vulnerability.

Because the disaster occurred in the heart of a highly developed mega-city, there was no shortage of diverse expertise to address post-disaster issues. This included researchers and technical experts from the University of California at Berkeley, whose campus was situated directly in the path of the fire. Disaster response networks had already been primed by the Loma Prieta earthquake two years earlier and the City of Oakland had become particularly skilled in the art of disaster recovery. Post-disaster reports spanned many disciplines, including urban forestry, geology, urban planning and landscape design, fire-fighting, mental health, and water supply. Table 10.8 lists the most comprehensive of these.

Table 10.8 Selected post-disaster reports following East Bay Hills firestorm of October 1991

• Federal Emergency Management Agency, Hazard Mitigation Report for the East Bay Fire in the Oakland-Berkeley Hills (Federal Disaster Declaration 919-DR-CA, 1992).

• Governor's Office of Emergency Services, The East Bay Hills Fire: A Multi-Agency Review of the October 1991 Fire in the Oakland/Berkeley Hills. Report prepared by the East Bay Hills Fire Operations Review Group (27 February 1992).

• Task Force on Emergency Preparedness and Community Restoration, "Final Report" (3 February 1992). The Task Force was established jointly by the mayors of Berkeley and Oakland and included committees on (1) Emergency Preparedness; (2) Communications; (3) Forestry and Vegetation; (4) Infrastructure and Development; and (5) Planning, Zoning, and Design.

• Ad Hoc Council on Replanting Needs (ACORN) (a project of the University-Oakland Metropolitan Forum), "Guide to Landscaping for Fire Safety" -collection of materials issued March 1992; second edition issued October 1992.

• Kenneth Topping, AICP, "Report on Oakland Hills Fire Prevention and Suppression Benefit Assessment District." (15 October 1992). Prepared as consultant to city of Oakland.

• Hills Emergency Forum, 1993 Report. Ongoing intergovernmental forum consisting of two major components: (1) East Bay Fire Chiefs Consortium and (2) Vegetation Management Consortium.

• East Bay Municipal Utility District (EBMUD), Firescape: Landscaping to Reduce Fire Hazard. Public information document (1992).

• East Bay Municipal Utility District, Preliminary Report on EBMUD's Response to the East Bay Hills Firestorm (24 March 1993).

- Phase I Final Report (15 July 1992)

- Phase II Final Report (21 January 1993)

• National Fire Protection Association, The Oakland-Berkeley Hills Fire, October 20, 1991 (1992).

The federal government officially entered the firestorm recovery process under a "major disaster declaration" for Alameda County issued by President Bush on 22 October 1991. For the second time in two years, FEMA established a disaster centre in Oakland and activated the many forms of aid available under the Federal Disaster Relief and Emergency Assistance Act. But the issues that confronted FEMA this time were different. Proportionally less public infrastructure and more private property were involved in the East Bay Hills than in communities affected by the Loma Prieta earthquake. Most victims were affluent and well insured. The fiscal burden thus fell more heavily on private insurers and less on US taxpayers. The number of structures destroyed or badly damaged was roughly comparable in both disasters: 3,957 "red-tagged" after the earthquake and about 3,300 burned in the firestorm. But, by late 1994, the total cost of federal assistance after the firestorm was US$ 513 million; this is about 10 per cent of the funds so far committed to recovery from the earthquake (FEMA, unpublished data). About four-fifths of federal funding after the fire was spent on reimbursing state and local governments for their response and recovery costs

Part of the federal aid was used to fund operation of a Community Restoration Development Center (CRDC), which was set up as a field office of the Oakland city government. Situated in a former supermarket close to the fire scene, the CRDC provided a "one-stop" facility where fire victims could meet with representatives of city, state, and federal agencies, utilities, insurance companies, and medical and social services (photo 10.2). Planning staff at the CRDC reviewed proposed rebuilding plans. Permit approval was expedited by a computerized geographical database that provided information about ownership, site, vegetation, utilities, and other characteristics of every parcel in the burned area (FEMA, 1992, p. 9).

Photo 10.2. Community Restoration Development Center, funded by FEMA, provided convenient "one-stop" access to city officials, utility representatives, and other services to victims

Following established practice, FEMA convened an inter-agency hazard-mitigation team to identify ways that the risks of similar disasters might be reduced in the future, both in the East Bay Hills and in other areas subject to urban/wildland fires (e.g. southern California). The team made recommendations about vegetation management, roofing, road width, fire codes, flood hazards, and other matters (FEMA, 1992).12 The California Governor's Office of Emergency Services also conducted a review of emergency response, particularly communications and fire-fighting. This agency urged, among other matters, that communities adopt compatible hydrant connectors so that fire equipment from any community could function in any other community, and it proposed installing a permanent auxiliary pumping system that would be capable of supplying water to all hill areas (California Governor's Office of Emergency Services, 1992). In California after urban/wildland fires there is an immediate upsurge of public anxiety about the risk of landslides, debris flows, and erosion on steep slopes during the next winter rainy season (McPhee, 1989). The City of Oakland spent US$ 5 million of federal money on aerial reseeding of the burned area and placed hundreds of straw-bale check dams in gullies and channels. But researchers found that such measures may have been unnecessary and were possibly even counter-productive (Booker et al., 1993).

12 The report reiterated the vegetation-management recommendations of the 1982 Blue Ribbon Fire Prevention Committee, whose executive summary was appended to the FEMA report.

The most comprehensive policy review was conducted by a regional Task Force on Emergency Preparedness and Community Restoration, created by the mayors of Oakland and Berkeley a week after the fire. This group was given eight weeks to prepare a detailed report that would guide recovery of the burned area and would protect other parts of the East Bay Hills. The Task Force included city and county officials, university faculty, utilities, local businesses, and representatives of the fire victims. It was organized in five work groups that were charged with different responsibilities:

1. Emergency Preparedness - to evaluate the emergency response systems of the cities, including fire stations, communications, early warning systems, and the use of volunteers.

2. Communications - to improve media and public information during a disaster.

3. Forestry and Vegetation - to develop landscape standards and other guidelines that reduce the potential for fire to spread.

4. Infrastructure and Development - to increase the quality of street systems and utilities (water, electricity, and telephone) for the fire areas.

5. Planning, Zoning, and Design - to develop new zoning and design standards for the area to be rebuilt as well as for all potential fire areas.

Key policy proposals and outcomes are displayed in table 10.9.

Table 10.9 Task Force on Emergency Preparedness and Community Restoration: Key proposals and status, 1994



• Improve local emergency services through a major bond issue for new radio communications and fire protection equipment.


• Increase the training and use of community volunteers to identify fire hazards and fight small fires.


• Require all homes in designated fire hazard zones to remove wood roofs, use prescribed landscaping methods and materials, and incorporate fire control methods.

In progress

• Establish a fire protection district for the fire danger zone from funds allocated for fire prevention activities.


• Move public utilities underground in the fire area, and

In progress

improve water supply to the area.

In progress

• Revise building codes to require the use of sprinklers, and/or

Not adopted

more fire-resistant building materials.


• Improve street access for firefighting equipment, and

Not adopted

relocate some fire equipment to fire-prone areas.


• Limit the density of homes in fire-prone areas.

Not adopted

Source: Task Force on Emergency Preparedness (1992) as summarized in Blakely (1992). Status based on estimates by the author.

Collaboration between city and university was especially fruitful with respect to forestry and landscaping issues.13 A grant from the San Francisco Foundation funded the Ad Hoc Council on Replanting Needs (ACORN) - a city - university project that was designed to educate property owners about fire-resistant landscaping methods. The East Bay Municipal Utility District reinforced this effort with a public information document entitled Firescape: Landscaping to Reduce Fire Hazard, which was distributed as a public service to all of EBMUD's water customers in the Hills.

13 The work group on this topic, chaired by Edward J. Blakely, Professor of City and Regional Planning at the University of California, Berkeley, was assisted by the University - Oakland Metropolitan Forum, a non-profit organization also headed by Blakely. Professor Blakely further served as Policy Advisor to the Mayor of Oakland and played a key role in coordinating local responses to both the earthquake and the firestorm.

Oakland voters translated public education into political action during 1993 by approving an Oakland Hills Fire Prevention and Suppression Benefit Assessment District. The district encompasses all hill areas of Oakland, whether or not they were burned in the firestorm. All of the approximately 20,000 parcels of real estate in the district will be assessed at US$ 75 per year initially, which could rise to a maximum of US$ 300 per year. This revenue will be used primarily for reducing the amount of hazardous fuels on public and private lands; among other means, they will continue to employ herds of goats to graze public lands within the district. The programme will fund inspections of private lands to ascertain violations of the Uniform Fire Code and will generally strengthen the capability of Oakland to combat urban/wildland fires (Topping, 1992).14

14 The task of organizing regional responses to issues of vegetation management and fire response beyond the city of Oakland has been assumed by the Hills Emergency Forum, a coalition of Oakland, Berkeley, East Bay Municipal Utility District, the East Bay Regional Park District, and the Lawrence Berkeley Laboratory of the University of California. All of these groups own and manage open lands in the Hills.

The Task Force Committee on Planning, Zoning, and Design addressed the controversial issues of how the Hills should be rebuilt. It was firmly committed to rapid rebuilding and gave little public consideration to the alternative of buying burned areas for public open space (see endnote 11) or of limiting density by means of non-compensatory zoning regulations. Despite the very small size of house lots in the burned area, the Committee urged approval for reconstruction of homes within the same "footprint" (i.e. area covered by the previous building). It called for utilities to be placed underground and for the adoption of "innovative parking solutions" to avoid blockage of narrow roadways. (It did not, however, urge the widening of roads, presumably since that would further reduce the size of building lots and would be expensive.)

Fourteen months after the fire, the city of Oakland finally adopted a new zoning designation (S-14) for the burned area. It permitted enlargement of replacement structures by 10 per cent over previous building sizes and it exempted any plans already submitted, regardless of the size of structures proposed. Where it applies, the S-14 zone establishes a variety of rules concerning minimum lot setbacks and sideyards, heights of buildings, parking, and landscaping. These rules are specified in relation to specific slope and street frontage characteristics of different lots.

At the end of 1994 the physical results of this ordinance became visible on the hillsides of Oakland: hundreds of very large, free-standing homes of eclectic design, often sited within 3 - 5 m of each other on tiny lots (photo 10.3). Fire hazard has been reduced by using non-combustible roof materials, placing utilities underground, and limiting flammable vegetation. Off-street parking will be more abundant, but the roads are generally as narrow as before and larger homes may well produce more vehicles per household. Congestion of streets still remains as a public safety hazard. The proximity of homes, the lack of internal sprinklers capable of extinguishing common house fires, and the likelihood of street congestion all create a sense of unease about the potential consequences of an earthquake on the Hayward Fault during coming decades. Private home-owners may (incorrectly) assume that, if it were not safe to rebuild, the government would prevent them from doing so.


The Bay Area mega-city has been sorely tested by earthquake, wind, and fire since 1989, especially in the East Bay region. The Loma Prieta earthquake and the East Bay Hills firestorm each inflicted grievous - but not catastrophic - harm. Both disasters demonstrated the vulnerability of large mega-cities to lifeline failures - especially water, transportation, electrical, and medical facilities and their supporting systems. On the other hand, each displayed the advantages of possessing a well-organized "larger community," extending from the President of the United States to the professional staff of regional agencies, to whom private citizens and local city officials look for help in an emergency.

Photo 10.3 The rebuilding process has produced eclectic variety in house styles, built in extreme proximity to one another.

There is little cause for comfort in the fact that the Loma Prieta quake and the Oakland firestorm did not inflict heavier losses. As reported by FEMA, the emergency response capabilities of the entire federal, state, regional, and local governments were stretched to breaking point after the Loma Prieta quake. Differences in a few key variables could have produced a much worse outcome - if the epicentre had been closer to San Francisco, if the magnitude had been larger, if the wind had been blowing, if these events had occurred when workers were at their jobs and children were in school. Similarly, if the Santa Ana winds had continued to blow, the Oakland firestorm could have overwhelmed the Claremont Hotel, engulfed the University of California's Berkeley campus, and consumed many blocks of homes and businesses in Berkeley and Oakland.

The juxtaposition of two unrelated natural disasters in the same urban region within two years may be viewed as an unparalleled training exercise in disaster response and recovery. As suggested below, the implications for mega-cities elsewhere are salutory.

The importance of regional cooperation

Some American metropolitan areas, such as Denver, have established region-wide hazard-management agencies (Platt, 1986, 1987). Often, however, mega-cities are too large and individual cities are too small to serve as optimal geographic units for the purposes of disaster planning, disaster response, and disaster recovery. Mega-cities need to be divided into smaller sub-areas that are consistent with the functional requirements of particular services. Although the city of Oakland was a primary urban victim of both the 1989 earthquake and the 1991 firestorm, its leaders and citizens could not and did not face these trials in isolation. The "East Bay" provided a psychic and functional region that was drawn on to help Oakland recover.

Within mega-cities and their subregions, there is a high degree of interaction among institutions and individuals that have responsibilities for addressing common needs (e.g. government, corporations, non-profit organizations, and private citizens). Long before the Loma Prieta earthquake, California had evolved a strong tradition of mutual aid and intergovernmental cooperation. This is reflected in myriad intergovernmental agreements on public services, fire protection, and shared facilities. It also appears in strong and flexible institutions of regional governance. In the case of the East Bay, two key regional agencies are the East Bay Municipal Utility District and the East Bay Regional Park District. Other institutions crucial to the disaster recovery process have included the University of California at Berkeley, the Association of Bay Area Governments, Pacific Gas and Electric, and the Pacific Bell telephone company. Professional staff from these and other entities interact continuously on task forces, committees, and special projects such as the ACORN project of the University - Oakland Metropolitan Forum. Governmental, corporate, and philanthropic efforts are blended in an ongoing dialogue concerned with the welfare of the East Bay region.

Post-disaster hazard assessment

Both Bay Area disasters were followed by a deluge of post-disaster assessments drawn up under federal, state, regional, local, and private auspices. Some of these reports were prepared in accordance with Section 409 of the Federal Disaster Relief and Emergency Assistance Act. Section 409 evaluations are necessarily intergovernmental and interdisciplinary. Others focus on particular agency or corporate missions (e.g. urban forestry, soil stabilization, fire-fighting, medical care).

For post-disaster assessments to be useful, they must be: timely; scientifically accurate; prescriptive; and practical. But they must not be overly constrained by political considerations. Members of official post-disaster assessment teams should be charged to "tell it like it is" and make whatever recommendations seem desirable without fear of criticism for suggesting politically unpopular approaches.

In the case of the East Bay firestorm, the post-disaster assessments were professionally competent but politically cautious. The most prominent of the non-federal assessments, the Report of the Task Force on Emergency Preparedness and Community Restoration, articulated dozens of sound hazard-mitigation measures. But the ultimate mitigation measure was not mentioned; that is, the alternative of not rebuilding the Hills, or at least of reducing density and widening streets. Perhaps there was only a small chance that there would be significant public support for those alternatives but they should have received serious consideration (see endnote 11). Other communities have chosen to support public buy-out of disaster-affected communities (e.g. Rapid City, South Dakota; Times Beach, Missouri). The Mississippi Valley floods of 1993 prompted the federal government's purchase of several thousand damaged homes (Interagency Floodplain Management Review Committee, 1994; Myers and White, 1993).

Modest efforts to change land-use and building patterns in the Hills after the fire were ineffectual. Streets remain narrow and homes have become larger, albeit roofed with tile rather than shingle. The post-disaster assessments have led to improvements in public sector services, but little or nothing has been done by private owners. The possibility of an earthquake on the Hayward Fault was not a significant factor in deliberations by the Task Force on Emergency Preparedness and Community Restoration.


This chapter has emphasized lifeline vulnerability because it has been a central problem of past disasters in the Bay Area. It cannot be restated too often that there is no more significant issue for natural-hazard planning in mega-cities than lifeline reliability. In California, the seismic survivability of individual structures has been markedly improved as a result of changes made to building codes. Public education programmes have also provided individuals and households with information that is essential to ensuring better protection during a disaster. But, as clearly demonstrated by the Northridge earthquake of 17 January 1994, improvements in individual safety are in effect nullified by the increased vulnerability of public and private lifelines.

Twice found lacking, the East Bay Municipal Utility District is now vigorously seeking to improve the resilience of its water supply system against earthquake and fires. More difficult to upgrade, the Bay Area's ageing transportation arteries remain vulnerable to the kinds of failures that in 1989 led to the collapse of an interstate highway in Oakland and closed the San Francisco - Oakland Bay bridge. Where failure is likely, it must be anticipated and alternative means of coping must be devised. Before a disaster occurs, the public must be informed about appropriate actions to take if major highways and bridges are disabled.

The unexpected

Perhaps the most daunting challenge for disaster managers in mega-cities is to plan for the unexpected. The interconnectedness of modern urban regions triggers a cascading series of failures, not necessarily catastrophic individually, but cumulatively producing chaos and misery as well as economic hardship. Dealing with the unexpected requires trained professionals equipped with reliable technology for damage assessment, decision-making, and communication. Disaster response is both a professional skill and an art. Perhaps a disaster's only silver lining is that it can enhance the ability to prepare for the next one. Short of first-hand experience, the need to learn from disaster experience elsewhere is obvious. In that respect, the San Francisco Bay Area has much to teach its sister mega-cities.


ABAG (Association of Bay Area Governments). 1992. Estimates of uninhabitable dwelling units in future earthquakes affecting the San Francisco Bay region. Oakland, CA: ABAG.

Adler, P., ed. 1992. Fire in the Hills: A collective remembrance. Berkeley: Patricia Adler.

Bancroft, H. H. 1888. History of California. San Francisco: The History Company.

BAREPP (Bay Area Regional Earthquake Preparedness Project). n.d. Living on the Fault. Public information brochure. Oakland, CA: BAREPP.

Blakely, E. J. 1992. "This city is not for burning." Natural Hazards Observer 16(6): 1 - 3.

Blue Ribbon Fire Prevention Committee (for the East Bay Hills Urban-Wildland Interface Zone). 1982. "Report." Oakland, CA: East Bay Regional Park District, mimeo.

Bolin, R. 1993. "The Loma Prieta earthquake: An overview." In R. Bolin (ed.), The Loma Prieta earthquake: Studies of short-term impact. Program on Environment and Behavior Monograph No. 50. Boulder: University of Colorado Institute of Behavioral Science.

Bolin, R. and L. M. Stanford. 1993. "Emergency sheltering and housing of earthquake victims: The case of Santa Cruz County." In P. A. Bolton (ed.), The Loma Prieta, California, earthquake of October 17, 1989 - Public response. US Geological Survey Professional Paper 1553-B. Washington, D.C.: US Government Printing Office.

Booker, F. A., W. E. Dietrich, and L. M. Collins. 1993. "Runoff and erosion after the Oakland firestorm." California Geology, November - December: 159 - 173.

Bronson, W. 1986. The earth shook, the sky burned. San Francisco: Chronicle Books.

California Governor's Office of Emergency Services, East Bay Hills Fire Operations Review Group. 1992. The East Bay Hills fire: A multi-agency review of the October 1991 fire in the Oakland-Berkeley Hills. Sacramento: OES.

Dana, R. H. 1964. Two years before the mast, 2 vols. Boston: Little, Brown.

FEMA (Federal Emergency Management Agency). 1990. Hazard mitigation opportunities for California. State - Federal Hazard Mitigation Survey Team Report for Disaster Declaration FEMA 845-DR-CA. San Francisco: FEMA.

FEMA (Federal Emergency Management Agency). 1991. The Loma Prieta earthquake: Emergency response and stabilization study. Contract report by the National Fire Protection Association. Washington, D.C.: US Government Printing Office.

FEMA (Federal Emergency Management Agency). 1992. "Hazard mitigation report for the East Bay fire in the Oakland - Berkeley Hills." Prepared pursuant to Disaster Declaration FEMA 919-DR-CA. San Francisco: FEMA, mimeo.

French, S. 1990. "A preliminary assessment of damage to urban infrastructure." In R. Bolin (ed.), The Loma Prieta earthquake: Studies of short-term impact. Program on Environment and Behavior Monograph No. 50. Boulder: University of Colorado Institute of Behavioral Science.

Interagency Floodplain Management Review Committee. 1994. Sharing the challenge: Floodplain management in the 21st century. Washington, D.C.: US Government Printing Office.

Keeler, C. A. 1903. San Francisco and thereabout. San Francisco: The California Promotion Committee.

McPhee, J. 1989. "Los Angeles against the mountains." In The Control of Nature. New York: Farrar, Straus, Giroux.

Myers, M. F. and G. F. White. 1993. "The challenge of the Mississippi River flood." Environment 35(10): 6 - 9, 25 - 35.

Nash, R. 1982. Wilderness and the American mind. New Haven, CT: Yale University Press.

National Fire Protection Association. n.d. The Oakland/Berkeley Hills fire - October 20, 1991. Quincy, MA: NFPA.

Oakland Tribune. 1991. "Blocks could have been saved with more water." Bitter Lessons (special series by the Oakland Tribune published the week after the firestorm and reprinted as a special section).

O'Brien, P. W. and D. S. Mileti. 1993. "Citizen participation in emergency response." In P. A. Bolton (ed.), The Loma Prieta, California, earthquake of October 17, 1989 - Public response. US Geological Survey Professional Paper 1553-B. Washington, D.C.: US Government Printing Office.

Palm, R. and M. E. Hodgson. 1992. After a California earthquake: Attitude and behavior change. Geography Research Paper 233. Chicago: University of Chicago Press.

Plafker, G. and J. P. Galloway, eds. 1989. Lessons learned from the Loma Prieta, California, earthquake of October 17, 1989. US Geological Survey Circular 1045. Washington, D.C.: US Government Printing Office.

Platt, R. H. 1986. "Metropolitan flood loss reduction through regional special districts." Journal of the American Planning Association 52(3): 467 - 479.

Platt, R. H. ed. 1987. Regional management of metropolitan floodplains. Program on Environment and Behavior Monograph No. 45. Boulder: University of Colorado Institute of Behavioral Science.

Platt, R. H. 1991. Land use control: Geography, law, and public policy. Englewood Cliffs, NJ: Prentice-Hall.

Task Force on Emergency Preparedness and Community Restoration. 1992. "Final report," mimeo, 3 February.

Thomas, G. and M. M. Witts. 1971. The San Francisco earthquake. New York: Stein & Day.

Topping, K. C. 1992. "Oakland Hills Fire Prevention and Suppression Benefit Assessment District Report." Pasadena, CA: mimeo, 15 October.

USGS (US Geological Survey). n.d. The next big earthquake in the Bay Area may come sooner than you think. Menlo Park, CA: USGS.

US Geological Survey Working Group on California Earthquake Probabilities. 1988. Probability of large earthquakes occurring on the San Andreas Fault. Open File Rpt. 88 - 398. Washington, D.C.: US Government Printing Office.

US Geological Survey Working Group on California Earthquake Probabilities. 1990. Probabilities of large earthquakes in the San Francisco Bay region, California. USGS Circular 1053. Washington, D.C.: US Government Printing Office.

Vance, J. E., Jr. 1964. Geography and urban evolution in the San Francisco Bay Area. Berkeley: Institute of Governmental Studies, University of California.

Yanev, P. I. 1991. Peace of mind in earthquake country. San Francisco: Chronicle Books.