The 1971 San Fernando Earthquake increased the level of seismic safety concern by local, state, and federal government agencies, and it catalyzed numerous policy changes by all three levels of government. New demands were placed on local planning and building departments to consider the seismic safety implications of their decisions. This article reports on research that describes how several southern California jurisdictions implemented these requirements and evaluates their success as tested by the 1994 North-ridge Earthquake. The research Found that seismic hazard information generally has not affected decisions on location, type, or intensity of land uses unless coupled with other concerns. Seismic safety policies, however, have created an environment in which it is easier to implement engineering initiatives, such as building codes or hazard abatement.

On February 9, 1971, a magnitude 6.4 earthquake struck the San Fernando Valley in southern California. This earthquake, the first to affect a major American urban area since the 1933 Long Beach Earthquake, killed 58 people, caused over 2,500 hospital-treated injuries, and caused over a half billion dollars in direct damages (Murphy et al., 1974). On January 17, 1994, the magnitude 6.7 Northridge Earthquake struck within a few miles of the 1971 epicenter. This earthquake killed 57 people, injured 9,000, and caused damage exceeding $20 billion (California Seismic Safety Commission, 1995).

These two earthquakes frame a period of rapid evolution of seismic safety policy. Although federal interest increased following the 1964 Alaska Earthquake, the 1971 San Fernando Earthquake was probably the one event that had the single greatest effect on seismic safety policy at local, state, and federal levels of government (Olson, 1996; Stallings, 1995; VSP Associates, Inc., 1988). During the 2 years following the 1971 earthquake several significant pieces of new legislation were adopted, addressing the safety of schools, hospitals, dams, development in active fault areas, and land use planning (Alquist, 1974; Joint Committee on Seismic Safety, 1974; Special Subcommittee to the Joint Committee on the San Fernando Earthquake Study, 1971; Special Subcommittee of the Joint Committee on Seismic Safety, 1972).

As a result of this legislation, the State of California placed new demands on local planning and building departments to consider the seismic safety implications of their development decisions.[sup1] Planning departments were required to include seismic safety elements as a part of their local land use plans[sup2], implement zoning and subdivision ordinances consistent with those plans, implement the State's prohibition of construction on active fault traces (Alquist-Priolo Special Studies Zones Act, 1972; Hart, 1994), and consider seismic hazards when reviewing the environmental impacts of new construction. Although it took many years, the 1971 San Fernando Earthquake was also the catalyst for several municipal efforts to reduce risks in the existing building stock (Alesch & Petak, 1986), which in turn led to statewide enactment of SB 547, the Unreinforced Masonry Building Law (1986). This law required local programs to identify and abate existing hazards (California Seismic Safety Commission, 2000). Planning and building departments have been involved in these programs, as well as in the development of earthquake recovery plans.

Several approaches are appropriate for reducing earthquake risks. To improve the safety of future development, adopting and enforcing seismic building codes is the most effective approach. But siting of new development is also important. This can be addressed by sensitive-area regulations, such as requirements for geological or geotechnical study, site design, infrastructure design, or land use restrictions. To improve the safety of existing development, cities may institute voluntary or mandatory building-strengthening programs, relocate uses, require hazard disclosure on sale of property, or provide public information to encourage individual actions (for a more complete list, see Olshansky & Kartez, 1998). And, of course, detailed hazard assessments form the necessary factual foundation for all these efforts (Deyle et al., 1998).

What is the role for planners in general, and land use planners in particular, in reducing earthquake risks? This is the fundamental question addressed by this research. Because earthquakes, unlike floods, have ill-de-fined source zones, have highly varying recurrence times, and affect widespread areas, some would argue that engineering design is the only way to reduce earthquake risk. Others would say that planning, too, has a significant role. French et al. (1996), for example, found in a post-Northridge Earthquake study that land use planning had a small but measurable effect in reducing damage in the 1994 earthquake. They found that although location and local geologic conditions were the most important predictors of damage, municipal plan quality also was statistically significant in predicting reduced damages. Burby et al. (1998) added to that study by showing that building code enforcement was also a significant factor. In contrast (and as a complement) to the descriptive, statistical approach of both of these studies, our research takes a case study approach to attempt to understand how local development management processes address seismic safety.

The similarity of the 1971 and 1994 earthquakes in size and location provides an unusual research opportunity. Twenty-three years is a long enough time span for policies to be adopted, implemented, and institutionalized. The 1994 earthquake provides an opportunity to see whether any of the lessons of 1971 had any effect in reducing damages that would otherwise have occurred in 1994. This research examines how several southern California jurisdictions implemented the post-1971 requirements and evaluates the success of these seismic safety planning tools as tested by the 1994 Northridge Earthquake. In particular, it focuses on evaluating the role of planning and planners in reducing earthquake risks.

The primary research questions were:

1. To what extent did the post-1971 policies affect development practice?
2. To what extent did they actually change the location, type, and intensity of land uses?
3. How effective were the policies in reducing damages in the 1994 earthquake?

And, finally, the normative question:

4. What is the role for planners in general, and land use planners in particular, in reducing earthquake risks?

The research used a case study method, consisting of review of planning documents and interviews with 39 planning and building officials in the City of Los Angeles, County of Los Angeles, City of Santa Clarita, City of Santa Monica, and City of Palmdale (see Figure 1). Several lines of evidence were sought:

• Content of seismic safety policies in plans and ordinances prior to 1994.
• Development review procedures prior to 1994 relevant to seismic safety.
• Extent to which land use development from 1971 to 1994 reflects seismic hazard maps.
• Opinions regarding relative effectiveness of seismic safety policies in reducing the effects of the 1994 earthquake.

The cities were selected to explore a range of issues. The City of Los Angeles is the largest municipality in the region and is a leader in seismic safety, in part because of its experience in 1971. It is also built out, which means that land use decisions concern urban in fill rather than large tracts of raw land. Santa Monica, though smaller, has similar land use characteristics. Despite not being affected by the 1971 earthquake, Santa Monica had been active in seismic safety activities and was badly affected by the 1994 earthquake. Santa Clarita is a new city with ongoing development on raw tracts of land. It is particularly significant because its location was strongly shaken in 1971, urban development has grown significantly since then, and the city was badly damaged in 1994. Palmdale was not directly affected by either earthquake, although its transportation link to Los Angeles was cut off both times. But Palmdale is of special interest because this rapidly expanding city is located upon the famous San Andreas Fault. We also chose to look at Los Angeles County, both because it provides a complementary framework to the cities and because it governed development in the Santa Clarita and Palmdale areas before the cities incorporated.

In addition to general planning materials, we also analyzed documents for approximately one dozen specific development projects within these cities, in order to determine the role of seismic safety issues in affecting local land use planning decisions. Finally, we created GIS maps for Los Angeles, Santa Clarita, and Palmdale to assess the effect of previously mapped geologic hazards on current land uses, planned land uses, and patterns of new urbanization growth over the 1971-1994 time period.

The seismic safety element of the general plan sets the policy framework for local government actions regarding seismic safety. As one of the mandated elements of general plans in California, all cities and counties must have one. Although its functions were later integrated into a more broadly defined "safety element" required under state law, these functions have not changed substantially since 1971, and it is still convenient to refer to it as the "seismic safety element." At minimum, it requires that seismic safety be addressed. The initial legislation asked that municipalities identify and appraise seismic hazards; it left it up to them how to deal with them (Mader, 1997). The City and County of Los Angeles and the City of Santa Monica all prepared their first seismic safety elements shortly after the requirement was enacted in 1971, and all three updated them in the 1990s. Santa Clarita and Palmdale were not incorporated as cities in 1971. When they incorporated they prepared general plans, including a safety (including seismic safety) element.

The seismic safety element is an information base and a policy document. It presents information on seismic hazards in the community: expected earthquake sources and their likelihood, expected levels of ground-shaking and its effects, areas with potential for amplification or ground failure, and potential secondary hazards such as dam failure. It then lays out ways to use the information: zoning, subdivision review, building codes and their enforcement, environmental review, response plans, recovery plans, and hazard abatement programs. In comparison to the older seismic safety elements, the newer ones focus more on process and institutional responsibilities and less on maps.

Other studies have looked at use of seismic safety information by local planners in California, focusing either on use of seismic safety elements or on use of mapped information. A 1991 review by Olshansky and others (1991) summarized the major findings of some of these studies (e.g., Wyner & Mann, 1986):

1. Very few jurisdictions have fully implemented even some of their seismic land use objectives.
2. Local governments have favored regulations that pass the costs on to developers on a project-by-project basis, rather than drawing on general funds.
3. The major effect of the seismic safety element has been to heighten the awareness of officials and has sometimes prompted the hiring of professional geologists.
4. In general, implementation of seismic safety land use policies does not stand in the way of efforts to develop land. Most local officials rely on building design or site-specific engineering to mitigate seismic hazards.
5. Seismic safety, in general, is not high on the priority list of local officials.

The 1991 study also included interviews with local planners, as well as a panel of technical experts in seismic hazard mapping. It concluded by describing a technically and politically optimal system of using hazard maps to improve seismic safety. Such a system would be based on generalized identification of areas of potential ground failure. These maps would then be used as input to requirements for special studies, seismic load factors in building codes, or general information to assist land use planning. A study conducted at the same time by the California Division of Mines and Geology (Holden & Real, 1990) came to similar conclusions, and it was the basis of the special-studies-zone approach developed under California's 1990 Seismic Hazards Mapping Act (Tobin, 1991).

In a recent review, Mader (1997) summarized several evaluations of the use of seismic safety elements. His review was mixed. He reported that, "while good demonstrations of the effective use of seismic safety policies in general plans can be found, they are not common" (p. 48). The elements vary in quality and tend to fade into disuse over time. On the other hand, he noted two "unexpected benefits" of seismic safety elements. First, the maps are sometimes used as a screening device to determine when site-specific geologic studies are needed. Second, they are often used to identify seismic hazards to be addressed in environmental impact reports. Finally, he observed that, thanks to seismic safety elements, planners in California are now much better informed about seismic hazards, and this knowledge "can find its way into day-to-day work in subtle ways" (p. 48).

As land use planners, we were particularly interested to see whether seismic safety information has affected land use type or intensity. We explored this question by several means. We asked planning and building officials directly and examined seismic safety elements. We also looked at individual project decisions. Another way to measure whether seismic safety information has affected land use type or intensity is to empirically examine the results, both in reality (current land uses and evolution of urbanized area since 1971) and in official policy (the most recent land use plans). We analyzed this data in graphic form and by statistical analysis.

Individual development projects. To what extent is seismic safety considered in individual project decisions? To see how seismic safety policies are implemented, we examined environmental impact reports for a number of large, controversial projects in the City and County of Los Angeles, Santa Clarita, and Palmdale. These reports are useful for several reasons: They document the project characteristics, analyze all possible impacts (including seismic hazards), recommend specific actions that the city can take before approving the project, and include citizen comments from the extensive public involvement required under state law. Projects included large hillside developments, large projects astride the San Andreas Fault, and high-rises within the San Fernando Valley area of Los Angeles (see Figure 2).

Despite significant public opposition to many of these projects, the public comment letters almost never stated seismic safety as one of their concerns. The actions recommended by the environmental impact reports consisted generally of standard site-design and engineering practice: conformance to state fault-zone procedures, avoidance or repair of unstable hillsides, and use of the Uniform Building Code. A typical conclusion: "The potential effect of groundshaking on structures can be satisfactorily mitigated by earthquake-resistant design in accordance with the latest Uniform Building Code and current state-of-the-art practice" (Planning and Design Solutions, 1990, pp. 4-6).

Citywide land use patterns. Despite the above findings, it is still possible that seismic safety information might affect land use patterns in more subtle ways. Perhaps developers might avoid hazardous lands in the first place, because of the additional design costs involved. To examine this question, we looked at land use at a macro scale, rather than the micro view of environmental impact reports.

Our independent variables were the mapped hazard zones in Los Angeles, Santa Clarita, and Palmdale. The question is whether these mapped zones affected subsequent land use location, type, and intensity. We measured land uses in three ways:

• Past: growth in urbanized land area over time, measured from a time sequence of aerial photos.
• Present: land use types and intensities existing in 1993.
• Future: planned future land uses, as expressed by the most recently adopted land use plans.

The City of Los Angeles has been built out to its city limits for many years. For this reason, land use locations are already well established, based on economic and infrastructure factors. Furthermore, because our analysis showed that 61.9% of the City's land area falls within at least one type of mapped hazard zone (fault zone, hillside, potential liquefaction area), avoidance is not a viable option (see Table 1). On the other hand, there is some evidence that land use intensity is sensitive to degree of hazard. For example, 61.8% of single-family residential acreage falls within mapped hazard areas, compared to only 50% of multi family residential acreage (see Table 2).

The land now occupied by the City of Santa Clarita urbanized very rapidly from 1971 to 1994 (see Table 3). We measured 9,600 urbanized acres in our earliest air photos in 1974, compared to 26,000 acres in 1994--a 270% increase. But this growth appears to have been remarkably insensitive to mapped hazards: Of these urbanized acres, 34% were in hazard areas in 1974 and 30% in 1994. With regard to 1993 land uses, and in contrast to Los Angeles, the denser residential land uses are more likely to be in hazard areas: Only 22% of single-family acreage versus 32% of multifamily acreage was in hazard areas (see Table 4). And 53% of commercial acreage was in a mapped hazard area (mostly potential liquefaction zones).

Palmdale has less acreage in hazard areas than the other cities (no potential liquefaction zones and more stable hillsides), but it does have the very large, complex, and hazardous San Andreas Fault Zone. In 1993, only 13% of single-family, 5% of multifamily, and 1% of commercial acreage was in a mapped hazard zone (see Table 5). Despite these relatively low percentages, the trends have not been good. Development boomed from 2,900 urbanized acres in 1974 to 21,000 acres in 1994--over 700% growth--and this growth increasingly encroached onto hazard areas, chiefly the San Andreas Fault Zone (see Table 6). Only 0.4% of Palmdale's urbanized land was in the San Andreas Fault Zone in 1974, compared to 7.5% in 1994.

In summary, all of the data and graphical analyses convincingly show virtually no relationship between published seismic hazard information and location of urban growth. In the case of Palmdale, post-1971 growth appears to have encroached onto hazardous lands.

This discussion raises the question of which land uses should be in hazardous areas. The conventional wisdom is that hazardous areas are most appropriate for less intense uses that reduce exposure. One could make a case, however, for high-value uses that can afford adequate site preparation and construction. This issue is particularly relevant to liquefaction hazard, which generally is not a threat to human life and is easily mitigable through foundation design. Such designs might be more affordable for higher-intensity, higher-value structures. Ideally, if a system of strict building codes exists and is well enforced, land uses will be sited where the benefits of location outweigh the costs of mitigation and expected damage, particularly where loss of life is not an issue. In the case of Santa Clarita's commercial center, which is sited on potentially liquefiable soils, the location apparently has economic advantages beyond the costs of site preparation and liquefaction-resistant foundations.

An important question is to what extent local planners should consult seismic hazard maps. How accurate are these maps in predicting relative damage? The Northridge Earthquake provided an opportunity to examine this question. I conducted a study to evaluate how well seismic hazard maps predicted damage in the Northridge Earthquake (Olshansky, 1997). In particular, the study evaluated a set of maps published in 1985 by the U.S. Geological Survey. I found that, after nor-malizing for residential density, structures built on certain geologic map units had significantly more damage than the areawide average. Buildings on young, soft soils, hillsides, and zones of at least moderate liquefaction susceptibility were 1.5 to 2.5 times more likely to be damaged than the area average. Looked at another way, being on good alluvial material 5 kilometers from the strongest shaking appears to be roughly equivalent to being on poor alluvial material at a distance off about 10 kilometers. These results were consistent with those of French et al. (1996), who found that at the scale of municipahties, liquefiable area was positively correlated with damage.

Thus, such maps do have the potential to provide useful information regarding relative hazards. But what level of policy intervention is justified by the damage differentials observed? It is important to remember that by "enhanced damages" we mean only a slightly enhanced probability of damage over a broad area. One cannot see the hazard zones on the ground, even on the day after the 1994 earthquake. Thus, although these maps can improve the intelligence of planning, the hazard differential is not sufficient to justify regulating land use type or intensity. Such maps can be used as the basis for requiring further study and they can help governments set priorities in managing land use, enforcing building codes, conducting seismic-strengthening programs for existing buildings, and planning for emergency response and long-term recovery.[sup3]

A critical, but difficult, question is: How effective were the post-1971 policy changes in actually reducing damages in the 1994 earthquake? Many of the policies had other important mitigation objectives even if they did not change land use types and locations. We directly asked this question of all interviewees and also made our own inferences based on our understanding of how the policies were designed and implemented. We purposely defined "planning" broadly to include all local development management policies, generally covering all activities of planning and building departments. Summary evaluation of the most significant activities follows.

Seismic safety elements. Seismic safety elements do not generally restrict land use type or intensity, except in state-designated active fault zones and in hillside areas.[sup4] And restrictions in hillside areas are usually based more on hydrologic and aesthetic considerations than seismic ones. Furthermore, most of the planners and building officials stated that they rarely, if ever, refer to the seismic safety element.

Yet the officials are unanimous in being glad that the seismic safety element exists. First, it provides maps that are used on a daily basis by planners and building staff involved in environmental review and permitting. The maps act as triggers to initiate other actions, such as environmental impact report requirements, fault zone studies, further geologic investigation, and building permits.

Second, it provides information on hazards, risks, and possible actions to improve safety. This information improves planning decisions, and it is the seismic safety element that brings it into the process. The seismic safety element forms a permanent information and policy foundation for subsequent ordinances and site-specific development actions. In Santa Monica, for example, the 1978 seismic safety element initiated a limited building inventory and seismic retrofit process. The increase in awareness engendered by this program was directly responsible for enactment of a more comprehensive retrofit ordinance shortly after the 1994 earthquake.

Third, seismic safety planning is important in increasing awareness throughout municipal departments and helps to inform departments of each other's roles. All the cities pointed out that the process provided a focus for interdepartmental discussions of seismic safety issues. Santa Monica and Santa Clarita had recently completed their latest safety elements, and Los Angeles was in the process of updating theirs; all of them took these processes very seriously and provided the latest information on the seismic risks faced by their communities.

It is clear that the seismic safety element played a key role in setting a framework for local seismic safety programs (the advent of the California Seismic Safety Commission in 1975 played a similar role at the state level). The effects of these actions are intangible and difficult to measure, but they established policies to enable more specific actions to take place. They served as a catalyst for mapping and other risk assessment studies, increased awareness of the general public and of elected officials, and documented hazard and vulnerability at the local level. At minimum, they informed emergency response planning, which in turn provided further rationale for mitigation.

Fault zoning. One of the most significant post-1971 innovations, the Alquist-Priolo Act (1972), which regulates construction in active fault zones, was not relevant in the 1994 earthquake because of the absence of surface fault rupture. This is ironic, because it was the most significant land use act to emerge from the 1971 earthquake, in which several structures were split apart where the fault tore the surface of the earth. Until 1990 it was the only law in the U.S. that tied land use regulations to earthquake hazard maps. In that year, California enacted the Seismic Hazard Mapping Act (1990; California Division of Mines and Geology, 1997), which asked the State Geologist to undertake a statewide seismic hazard mapping program; in effect, it expanded the Alquist-Priolo Act requirements to areas prone to ground failure from liquefaction and landsliding (Tobin, 1991). At the time of the Northridge Earthquake only one of the maps had been completed, so evaluation of this act is not in the scope of our research.

Building codes. Probably the most significant policy activity has been the continued improvement in the building code and its enforcement, most notably the code changes instituted following the 1971 earthquake. The newer codes, based on lessons learned in 1971 and adopted into the 1976 edition of the Uniform Building Code, clearly reduced damages and saved lives in 1994. The three most serious building collapses were all structures built under the pre-1976 version of the Uniform Building Code (see Olshansky, 1998). Although some newer buildings were seriously damaged in areas of strong groundshaking, buildings constructed according to the latest seismic codes clearly suffered less damage than older buildings in general, as was also true in the 1995 earthquake in Kobe, Japan. The San Francisco Bay area had a similar experience in the 1989 Loma Prieta Earthquake. Because of the successful performance of buildings in the Northridge Earthquake, relatively few code changes were required. A notable exception was the surprising performance of steel-frame high-rise buildings, many of which had serious and unexpected cracks in their joints. These problems are being addressed by a major, multiuniversity research effort.

Hazard abatement programs. A number of abatement programs played a significant role in reducing damages, most notably the unreinforced masonry (URM) retrofit program adopted by the City of Los Angeles in 1981, as well as a mandatory retrofit program adopted by Santa Monica in 1992 (Santa Monica first initiated a voluntary program in 1978). The City of Los Angeles also had its own bridge retrofit program for city streets, similar to the State's program for highways. The URM retrofit program in Los Angeles was clearly successful in saving lives (K. Deppe, personal communication, October 20, 1994). None of the City's URM structures collapsed, which was the purpose of the program (see Figure 3). Still, about 200 of them had serious damage and another 300 had some damage. Although most were repairable, these damages were not acceptable to many building owners, and engineers have begun to develop methods for higher levels of safety. The Northridge Earthquake also made it clear that many other types of buildings require retrofit, and the City of Los Angeles has since initiated programs for some of these.

Geologic information used for project review. Both environmental impact reports and geologic review of subdivisions serve to reinforce the seismic safety element by reminding participants of earthquake hazards and of the need to carefully follow building and grading codes on sensitive sites. In one case, the environmental review process clearly facilitated the City's ability to require further detailed studies and additional project redesign. These processes have been demonstrably effective in reducing damage and loss of life from rainfall-induced landsliding (Erley & Kockelman, 1981; Olshansky & Rogers, 1987) and appear to have had similar effects in the 1994 earthquake, where landslides were numerous but damages and injuries caused by landslides were very few. These processes primarily operate by reinforcing building and grading codes or by rearranging or clustering land uses within a subdivision, but they rarely change land use type or intensity for hazard reduction reasons.

Recovery planning. Another notable planning success, although not contributing to 1994 damage reduction, was the City of Los Angeles' Recovery and Reconstruction Plan. Having this plan in place at the time of the earthquake greatly facilitated the City's institutional ability to initiate recovery programs and to integrate mitigation into these programs. It was the process of preparing the plan, rather than the document itself, that helped to clarify departmental roles following the disaster. This finding emphasizes the importance of periodic plan updates, even if the actual revisions are relatively minor. This plan is evaluated in much more detail by Spangle Associates (1997).

The California Seismic Hazards Mapping Act (SHMA; 1990), enacted shortly after the 1989 Loma Prieta Earthquake, extended the principles of the Alquist-Priolo Act to areas of ground failure and strong ground-shaking (California Division of Mines and Geology, 1997). The genesis of this Act actually goes back to the 1971 earthquake, when the legislature rejected a broad mapping program and decided instead to focus only on fault rupture zones (Tobin, 1991). The Seismic Safety Commission, however, persisted in asking for such a program, and in 1987 the legislature directed the Division of Mines and Geology to design a program that would provide improved information on seismic hazards to property owners (Real & Holden, 1991). This study was underway when the 1989 Loma Prieta Earthquake struck. The Speaker of the Assembly, reacting to damage due to soft soils in the Marina District of San Francisco, asked the Commission to help draft a bill to address this situation (Tobin, 1991). The Divison of Mines and Geology study was completed in 1990, with the principal finding that "preparation of maps delineating where earthquake ground shaking hazards are likely to occur, and the establishment of special studies zones would provide the most practical and effective seismic hazard information for reducing future earthquake losses" (Real & Holden, 1991, p. 182). This study, then, became the basis for the legislation.

In addition to directing the State Geologist to prepare appropriate maps, the SHMA (1990) requires several actions by local governments: posting of the official maps at the county offices, recording of the maps by the county recorder, requirement of a geotechnical report prior to the approval of any project in a seismic hazard zone, submittal of all approved geotechnical reports to the State Geologist, and consideration of the maps in preparing the safety element and in "adopting or revising land use planning and permitting ordinances" (section 2699). In contrast to the Alquist-Priolo Act (1972), which requires only avoidance of the fault trace, the SHMA permits any appropriate mitigation of the hazard. The program does not regulate land use type or intensity. Instead, local governments can use the information to ensure that new development is reasonably sited and designed.

Because of funding delays, as well as the need to establish mapping standards, startup of the program was slow, and none of the maps had been completed at the time of the 1994 earthquake. Therefore, this policy, which had its origins after the 1971 earthquake and was finally enacted after the 1989 earthquake, was not yet in effect to reduce damages in the 1994 earthquake. It was the 1994 earthquake, however, that finally gave the program the boost that it needed. Shortly after the earthquake, the Federal Emergency Management Agency (FEMA) provided funding for accelerated mapping of 16 quadrangles in southern California, and FEMA along with California's Office of Emergency Services have continued to provide additional funding. As of November 2000, 47 maps have been released, and an additional 47 are funded for completion in 2004 (California Division of Mines and Geology, 2000).

The SHMA reinforces many of the risk reduction strategies discussed in this article. It provides authoritative information to be used in safety elements and in project review. The maps can also be used to establish priority buildings for hazard abatement. Significantly, it also works in tandem with recent changes in the building codes. For example, the 1997 edition of FEMA's NEHRP Recommended Provisions for Seismic Regulations, which forms the basis of the nation's seismic building codes, specifically calls for an applicant to submit "a written report that includes an evaluation of potential site hazards such as slope instability, liquefaction, and surface rupture due to faulting or lateral spreading" (Building Seismic Safety Council, 1997, p. 107). In addition, the Provisions now classify sites into five soil categories to account for amplified groundshaking known to occur on softer soils. Because local planning and building officials may not know the locations of problematic soils in their communities, the publication of SHMA maps should dramatically improve their ability to implement these new code provisions.

All of these policy changes have been helpful, though not necessarily in the rational way that they were intended to work. Seismic hazard information and seismic safety policies have generally not affected decisions on location, type or intensity of land uses, unless coupled with other concerns, such as protection of hillsides or river corridors. Nor is seismic safety raised by the public as a significant concern even in the largest and most controversial development projects. And, based on my related study of the effectiveness of seismic hazard maps in Los Angeles County (Olshansky, 1997), these are, in fact, appropriate responses to the information.

Seismic safety policies in general, however, have created an environment in which it is easier to implement engineering initiatives, such as building codes or hazard abatement. Seismic safety elements, though not used on a daily basis, are valued in various significant ways by planning and building officials. Seismic safety information in general, and seismic safety elements in particular, establish a policy framework within which engineering requirements and retrofit programs can operate more effectively. This, of course, is good planning, albeit not rational planning that would link land use types and intensities to specific mapped zones. And it is appropriately done by planners. It is often the role of planning departments to take this big-picture view. In the general plan process, the seismic safety element helps to inform the other municipal departments, as well as developers, of the relative hazards in the community. And the maps help planning departments to identify potential hazards early in the planning process, when they might be able to most effectively bring them to the attention of the building and engineering departments and the developer. The new maps being produced under the Seismic Hazards Mapping Act offer the promise of enhancing this function and broadening availability of hazard maps to a larger number of jurisdictions.

In sum, the post-1971 policies did significantly help to reduce damage in the 1994 earthquake. They did this primarily through building codes for new buildings and retrofit programs for older unreinforced masonry buildings. But there were more subtle effects as well, and planners played a key role in many of these. Building and design professionals became more aware of seismic hazards, of where the causative faults are located, of how the ground behaves, of which soils demand more careful analysis and design, of which buildings are most vulnerable, and of the odds of experiencing a damaging earthquake. Throughout the planning and development review process, all the players grew to understand the importance of taking seismic analysis and design seriously. Seismic safety is considered early in the planning process and is dealt with routinely by planners, building officials, engineers, and project applicants. Although difficult to measure, this is not an insignificant achievement.

Planners in seismically hazardous areas can draw some lessons from the Los Angeles County experience. First, seismic safety plans are an important first step. They establish the fact base for future planning and policy efforts, and they increase interdepartmental awareness of needed actions. Second, redundancy helps. Seismic safety is promoted by having the long-range planners, the environmental review specialists, and the building department all taking responsibility for identifying hazardous areas and requiring a higher level of design and construction in those areas. Third, in order to minimize the temptation to build in the most hazardous areas, seismic safety planning and hazard mapping need to be done early in the planning processes for newly developing areas. Although Santa Clarita and Palmdale now have established careful seismic safety development management processes, they perhaps missed opportunities by not instituting these processes sooner. Fourth, seismic safety planning need not, and should not, limit itself to new development. Planning has an important role to play in developing strategies for existing hazard-prone buildings and in establishing procedures, in advance, for post-earthquake reconstruction.

This work was funded by a grant from the National Science Foundation (Award number CMS-9416499). All the GIS data compilation and analyses were performed by Kimberley Knowles-Yanez, then a doctoral student in Urban and Regional Planning at the University of Illinois at Urbana-Champaign. I would also like to thank the numerous staffmembers of the Cities of Los Angeles, Santa Clarita, Palmdale, and Santa Monica and the County of Los Angeles, who generously allowed us the time to interview them and provided us with numerous planning documents.

[sup1]. Earthquakes can affect cities in several ways. An earthquake is caused by the sudden rupture of an earthquake fault (approximately a planar surface). This motion itself occurs within a limited area. But when the fault breaks, it sends out seismic waves that cause widespread ground-shaking. The most direct way an earthquake can cause damage is if the fault rupture breaks through to the earth's surface and directly damages overlying buildings. Surface fault rupture occurred in the 1971 earthquake, but not in 1994. The second, and most common, way an earthquake affects cities is by shaking buildings. This can damage structures and their contents, and damaged buildings can cause injuries and deaths. A third effect is ground failure triggered by the shaking. For example, groundshaking can trigger landslides, a type of ground movement called "lateral spreading," and soil "liquefaction." Liquefaction is a phenomenon whereby water-saturated sands lose all their strength to support overlying soils and buildings. Liquefaction effects range from ground settlement to total failure of building foundations. Because only certain types of soils will fail when subjected to seismic shaking, geologic maps can identify areas susceptible to ground failure. Finally, earthquakes can trigger secondary hazards, such as dam failure, hazardous waste release, or fire. Typically, geologic hazard maps, such as those incorporated into seismic safety elements, evaluate surface fault rupture, expected levels of groundshaking from future earthquakes, ground failure potential, and secondary hazards.

[sup2]. SB 351, enacted in 1971, required that all general plans contain a seismic safety element (Joint Committee on Seismic Safety, 1974). This was subsequently broadened to be a safety element, now codified as Cal. Gov't Code section section 65302(g) (West 1997).

[sup3]. In a few cases, if further investigation uncovers a major (and expensive) problem, such as a deep-seated landslide, then the de facto result may be a reduction in land use intensity, or even retention as open space.

[sup4]. Sometimes the coincidence of other natural features may result in land use restriction in a seismic hazard zone. For example, in Santa Clarita, a sensitive ecological area (riparian woodland) along the Santa Clara River coincides with potentially liquefiable soils. According to the staff planners, it is the riparian woodland issue that is more likely to result in restrictions on land use location or intensity, but the restriction is reinforced by the seismic safety issue.

Zone type           Percent[sup*] city area

No hazard              38.1
Hillside               44.1
Fault zone              3.3
Liquefaction zone      25.5

Source: City of Los Angeles digital maps.
[sup*] Total >100% because some areas have more than one hazard.

                                 Percent          Percent: not
Land use                    in a hazard zone   in a hazard zone

Commercial/service               50.5                49.5
Government/education             58.0                42.0
Industrial                       34.1                65.9
Transportation/open
   space                         73.5                26.5
Single family                    61.8                38.2
Multifamily                      50.0                50.0

Sources: City of Los Angeles digital maps; land use map from
Southern California Association of Governments.

Legend for chart:

A1=Air photo date
A2=Developed area (acres)
A3=Fault zones
A4=Liquefaction zones
A5=Landslide areas
A6=Totals[sup*]
            Percent developed area in

A1       A2     A3     A4    A5     A6

1974    9,580   0.7   28.9   8.9   34.2
1975   12,064   0.8   28.2   8.4   32.9
1978   13,278   0.8   27.3   8.7   32.1
1979   13,605   0.7   28.1   8.5   32.6
1983   16,012   0.7   26.8   8.2   31.2
1989   20,234   0.9   27.3   8.6   31.1
1994   25,700   1.0   26.8   8.9   30.0

[sup*] Totals <sum of the three columns because some hazards
overlap.

Legend for chart:

A1=Land use
A2=Fault zone
A3=Liquefaction zone
A4=Landslide area
A5=Stable hillsides
A6=No hazard
                                 Percent[sup*] land use in

A1                          A2     A3     A4     A5     A6

Commercial/service          4.7   46.0    6.6   12.6   34.8
Government/education        0.0   23.7    8.2   24.0   44.1
Industrial                  1.1   17.3   16.4   51.2   14.7
Transportation/open space   0.4    5.8   13.3   73.2    7.6
Single family               0.2   14.2    7.3   34.9   43.6
Multifamily                 0.0   22.5    9.2   20.2   48.1

Sources: 1991 Santa Clarita General Plan; land use map from
Southern California Association of Governments.
[sup*] Totals >100% because some areas have more than one hazard.

                                 Percent[sup*] land use in

Land use                 Fault zone   Hillside zone   No hazard

Commercial/service          0.4             0.6         99.0
Government/education        0.3             0.0         99.7
Industrial                  3.5             7.2         90.4
Transportation/open space   7.8            32.0         62.7
Single Family               8.3             5.3         87.3
Multifamily                 5.2             0.0         94.8

Sources: 1993 Palmdale General Plan; land use map From Southern
California Association of Governments.
[sup*] Totals >100% because some areas have more than one hazard.

Legend for chart:

A1=Air photo dates
A2=1978-1981
A3=Developed area (acres)
A4=11,317
A5=17,291
A6=21,062
                      Percent developed area in

A1      A3      Fault zone   Hillside zone     Totals[sup*]

1974   2,933       0.4            0.0            0.4
A2     8,386       3.2            4.3            6.5
1983   9,058       6.0            2.7[supa]      8.6
1985   A4          6.8            0.5[supa]      6.8
1989   A5          7.8            3.4           10.3
1994   A6          7.5            2.9            9.7

[sup*] Totals <sum of the two columns because some hazards overlap.
[supa]. Apparent decrease due to incomplete aerial photo
coverage for these years.

MAP: FIGURE 1. Location of Cities of Los Angeles, Palmdale, Santa Clarita, and Santa Monica within Los Angeles County, California.

PHOTO (BLACK & WHITE): FIGURE 2. View of San Fernando Valley, looking north. The high-rise buildings were built between 1971 and 1994, most under the post-1976 building code. (October 1994)

PHOTO (BLACK & WHITE): FIGURE 3. This unreinforced masonry building in Hollywood, strengthened according to the 1981 Los Angeles retrofit ordinance, was damaged in the 1994 earthquake, but did not collapse. (October 1994)

Alesch, D. J., & Petak, W. J. (1986). The politics and economics of earthquake hazard mitigation (Monograph #43). Boulder: University of Colorado, Institute of Behavioral Science.

Alquist, A. A. (1974). The state role in seismic safety. In U.S. Department of Commerce, National Oceanic and Atmospheric Administration, The San Fernando, California, Earthquake of February 9, 1971 (Vol. I, part A, pp. 5-17). Washington, DC: U.S. Government Printing Office.

Alquist-Priolo Special Studies Zones Act, Cal. Stat. ch 1354 section 4 (1972), amended by Cal. Pub. Res. Code section section 2621-2630 (West 1984 & Supp. 2000).

Building Seismic Safety Council. (1997). NEHRP recommended provisions for seismic regulations for new buildings and other structures (FEMA-302). Washington, DC: Federal Emergency Management Agency.

Burby, R.J., French, S. P., & Nelson, A. C. (1998). Plans, code enforcement, and damage reduction: Evidence from the Northridge Earthquake. Earthquake Spectra, 14(1), 59-74.

California Division of Mines and Geology. (1997). Guidelines for evaluating and mitigating seismic hazards in California (Special Publication 117). Sacramento: Author.

California Division of Mines and Geology. (2000,November 21). Currently available seismic hazard zone maps and planned release dates [On-line]. Available: <http://www.consrv.ca.gov/dmg/shezp/schedule.htm>.

California Seismic Safety Commission. (1995). Turning loss to gain. Report to Governor Pete Wilson in response to Governor's Executive Order W-78-94 (SSC Report No. 95-01). Sacramento: Author.

California Seismic Safety Commission. (2000). Status of the Unreinforced Masonry Building Law (SSC Report No. 2000-02). Sacramento: Author.

Deyle, R. E., French, S. P., Olshansky, R. B., & Paterson, R. G. (1998). Hazard assessment: The factual basis for planning and mitigation. In R.J. Burby (Ed.), Cooperating with nature (pp. 119-166). Washington, DC:Joseph Henry Press.

Erley, D., & Kockelman, W.J. (1981). Reducing landslide hazards: Aguide for planners (Planning Advisory Service Report No. 359). Chicago: American Planning Association.

French, S. P., Nelson, A. C., Muthukumar, S., & Holland, M. M. (1996). The Northridge Earthquake: Land use planning for hazard mitigation (Final Report to the National Science Foundation). Georgia Institute of Technology, City Planning Program.

Hart, E. W. (1994). Fault-rupture hazard zones in California (Special Publication 42). Sacramento: California Division of Mines and Geology.

Holden, R. J., & Real, C. R. (1990). Seismic hazard information needs of the insurance industry, local government, and property owners in California (Special Publication 108). Sacramento: California Division of Mines and Geology.

Joint Committee on Seismic Safety. (1974,January). Meeting the earthquake challenge (Final Report to the Legislature, State of California).

Mader, G. G. (1997). Enduring land use planning lessons from the 1971 San Fernando Earthquake. Earthquake Spectra, 13(1), 45-54.

Murphy, L. M., Steinbrugge, K. V., & Duke, C. M. (1974). The San Fernando Earthquake. In U.S. Department of Commerce, National Oceanic and Atmospheric Administration, The San Fernando, California, Earthquake of February 9, 1971 (Vol. I, part A, pp. 3-4). Washington, DC: U.S. Government Printing Office.

Olshansky, R. B. (1997). The role of earthquake hazard maps in loss estimation: A study of the Northridge Earthquake. Earthquake Spectra, 13(4), 721-737.

Olshansky, R. B. (1998). Promoting the adoption and enforcement of seismic building codes: A guidebook for state earthquake and mitigation managers (FEMA-313). Washington, DC: Federal Emergency Management Agency.

Olshansky, R. B., Foster, F., & Cook, S. (1991). Microzonation: A planner's perspective. In Earthquake Engineering Research Institute, Proceedings, Fourth International Conference on Seismic Zonation (Vol. III, pp. 195-202).

Olshansky, R. B., & Kartez, J. D. (1998). Managing land use to build resilience. In R.J. Burby (Ed.), Cooperating with nature (pp. 167 201). Washington, DC: Joseph Henry Press.

Olshansky, R. B., & Rogers, J. D. (1987). Unstable ground: Landslide policy in the United States. Ecology Law Quarterly, 13(4), 939-1006.

Olson, R. A. (1996,February). Twelve seconds and thirty years: The policy and planning legacies of the San Fernando Earthquake (Presentation at 48th Earthquake Engineering Research Institute Annual Meeting, Los Angeles, California).

Planning and Design Solutions (1990,November). Final environmental impact report, Oak Springs Estates (Report for City of Santa Clarita). Newport Beach, CA: Author.

Real, C. R., & Holden, R.J. (1991). Special study zones: A viable approach to zonation for ground shaking hazards. In Earthquake Engineering Research Institute, Proceedings, Fourth International Conference on Seismic Zonation (Vol. III, pp. 181-187).

Seismic Hazards Mapping Act, Cal. Stat. ch 1168 section 3 (AB 3897) (1990), Cal. Pub. Res. Code section section 2690 2699 (Deering 2000).

Spangle Associates & Robert Olson Associates. (1997). The recovery and reconstruction plan of the City of Los Angeles: Evaluation of its use after the Northridge Earthquake (Report to National Science Foundation). Portola Valley, CA: Spangle Associates.

Special Subcommittee to the Joint Committee on the San Fernando Earthquake Study. (1971,October 31). Earthquake risk conference proceedings (Sponsored by the Joint Committee on Seismic Safety to the California Legislature).

Special Subcommittee of the Joint Committee on Seismic Safety. (1972). The San Fernando Earthquake of February 9, 1971 and public policy. Sacramento: California Legislature.

Stallings, R. A. (1995). Promoting risk; Constructing the earthquake threat. New York: Aldine de Gruyter.

Tobin, L. T. (1991). California's urban hazards mapping program: A bold experiment in earth science and public policy. In Earthquake Engineering Research Institute, Proceedings, Fourth International Conference on Seismic Zonation (Vol. III, pp. 103-110).

Unreinforced Masonry Building Law, Cal. Stat. ch 250 section 2 (1986), amended, Cal. Gov't Code section section 8875-8876 (Deering 1999 & Supp. 2000).

VSP Associates, Inc. (1988). To save lives and protect property: A policy assessment of federal earthquake activities, 1964-1987 (FEMA-181). Washington, DC: Federal Emergency Management Agency.

Wyner, A.J., & Mann, D.E. (1986). Preparing for California's earthquakes: Local government and seismic safety. Berkeley: University of California, Institute of Governmental Studies.

~~~~~~~~

By Robert B. Olshansky, Olshansky, AICP, is an associate professor of urban and regional planning at the University of Illinois at Urbana-Champaign, where he teaches land use and environmental planning. His current research interests include planning for natural hazards, watersheds, and local growth management.