RAYMOND J. BURBY

THIS BOOK IS ABOUT NATURAL disasters and sustainability—the capacity of the planet to provide a high quality of life for not only present but also future generations. Hardly a month passes without screaming headlines informing readers of the latest natural catastrophe that has befallen some part of the world. Newspaper subscribers might also notice indications, buried in the Op/Ed section, that scientists are becoming concerned about our planet's capacity to sustain a high quality of life. Disasters signal a serious breakdown in sustainability. The earthquakes, floods, hurricanes, and other natural events that cause so much concern have always been present. But now disasters wreak havoc that goes well beyond the ability of society to take them in stride. What is more, in trying to reduce vulnerability by taming nature, the situation is often made worse, not better. But there is a better way.

By planning for and managing land use to enhance sustainability, we can reduce our vulnerability to disasters, if not eliminate them. Land use plans enable local governments to gather and analyze information about the suitability of land for development,

so that the limitations of hazard-prone areas are understood by policymakers, potential investors, and community residents. In the process of preparing plans, local governments engage in a problem-solving process that works to ensure that all stakeholders understand the choices the community faces, and that they reach some degree of consensus about how these choices will be made. For example, should existing areas at risk of flooding be protected by levees or evacuated and maintained as open space? Should community growth be channeled away from hazardous areas or do economic benefits from their development justify the expense of protection? How can protection be provided most effectively? Plans require the systematic evaluation of alternative courses of action, so that the approach chosen to reduce vulnerability is an optimal solution given a community's present circumstances, future prospects, and the goals and aspirations of its residents.

While plans provide general guidance for managing development, land use regulations such as zoning set specific rules for the private sector about where development will be allowed and how development (and redevelopment) is to take place so that vulnerability from natural hazards is minimized. Examples of safe development practices include elevating buildings and transportation infrastructure above expected flood heights, and strengthening buildings to withstand ground shaking in an earthquake or high winds in a hurricane. Land use management also includes complementary governmental policies, such as campaigns to inform citizens about areas that face the greatest risks from hazards, and policies to locate infrastructure such as roads and sewer lines so that they steer development away from hazardous areas. In combination, plans and land use management programs enhance prospects for a sustainable future-one in which citizens and their elected officials make informed choices about using hazardous areas in ways that will not jeopardize the long-term viability of their communities.

The logic of land use planning and management, on its face, seems self-evident. But, for a variety of reasons few local governments, of their own accord, pay much attention to hazards before disaster strikes. The federal government and a few states realize the importance of proactive programs to reduce risk, but they have been reluctant to force local governments to employ land use planning and management techniques. As a result, unsustainable development practices (and large losses in natural disasters) have been the rule rather than the exception in urban areas across America.

In this book we highlight the promise of land use planning and man-

agement for achieving sustainability, explore reasons why this promise is not being realized uniformly by local governments, and propose ways to foster sound land use decision making. We will discuss why land use and sustainability have been ignored in devising policy to deal with natural hazards, and lay out planning procedures, a vision of sustainability, and concrete suggestions for policy reform.

Natural disasters exact a heavy toll, and it is increasing every year. Worldwide more than three million people have lost their lives in flood, earthquake, landslide, hurricane, volcanic, and other natural disasters over the past two decades, and almost one billion have been affected in other ways. The economic consequences have been devastating. Property damages have been estimated to range from $25 to $100 billion between 1965 and 1985 (Advisory Committee on the International Decade for Natural Hazard Reduction, 1987). But losses are obviously much higher than that since disasters cause extensive disruption to economic systems. In fact, one source put the total economic costs of disaster in the early 1990s at more than $100 billion per year (Clarke and Munasinghe, 1995). In any given year, however, they can be much greater. The Kobe earthquake of January 17, 1995, produced this toll of human suffering: 6,308 deaths, 43,177 injuries, 300,000 people homeless, and $100 billion in total property damages.

In the United States property losses also have been extraordinary in the 1990s: for example, $30 billion in Hurricane Andrew in 1992; $16 billion in flooding in the upper Mississippi and Missouri River basins in 1993; $20 billion in the Northridge earthquake in 1994; $5.5 billion in Hurricanes Luis and Opal in 1995; and more than $1 billion in Hurricane Fran in 1996. Most of these losses were absorbed by households and businesses, but the public costs are substantial, too. Congress appropriated more than $115 billion for disaster relief and mitigation between 1975 and 1995. State and local governments absorbed billions more in losses. Although accurate loss data for developing nations are difficult to secure, one estimate indicates that, as a percent of gross national product, losses are almost 20 times higher in developing countries than in developed countries (Burton et al., 1993).

Grim as these statistics are, it is possible that society so far has seen only the tip of an enormous iceberg of risk. Forecasts of potential future losses lurking beneath the surface of the sea of public complacency re-

garding natural hazards are truly frightening. Consider these worst-case earthquake scenarios for Los Angeles, San Francisco, and Tokyo developed by Shah (1994). A magnitude 7.0 earthquake on the Newport-Inglewood Fault in Los Angeles could cause $170 to $220 billion in economic losses, $95 to $120 billion in insured losses, 3,000 to 8,000 deaths, and up to 20,000 serious injuries. A repeat of the 1906 earthquake in San Francisco (a magnitude 8.3 event) could result in $170 to $225 billion in economic losses, $80 to $105 billion in insured losses, 3,000 to 8,000 deaths, and up to 18,000 serious injuries. A repeat of the 1923 Tokyo earthquake would cause economic losses of $2.0 to $3.3 trillion, insured losses of $30 to $40 billion, 40,000 to 60,000 deaths, and serious injuries in the hundreds of thousands.

Governments traditionally try to cope with disasters using warnings before disaster strikes, emergency relief after a disaster occurs, and hazard reduction measures such as levees to reduce the likelihood of a future disaster. For a variety of reasons, none of these usual approaches is adequate to reduce losses in disasters to acceptable levels. In addition, each is enormously expensive.

Warning is the oldest way of coping with disaster, and it is still ubiquitous. Some warning systems try to reduce losses immediately preceding the onset of a hazardous event. Home phones that ring automatically to sound the alarm when flood waters begin to rise rapidly are an example. Others try to reduce losses well before a disaster occurs by letting people know that particular areas are affected by hazards. Examples include maps that delineate flood and seismic hazard zones, signs posted to warn of flood hazards, and laws that require real estate agents to inform their clients of hazards. More informative warnings are based on hazard assessments that not only identify the presence of natural hazards but also provide information on the amount of loss that may be incurred on average every year, or for a specified scenario such as a 100-year flood (a flood that has a 1 percent chance of occurring in any given year).

Warning is an essential ingredient in any strategy to cope with losses from disasters, since no method of coping—even the land use strategies

we argue for in this book-can completely eliminate risk. At the very minimum, therefore, people need to be informed of the risks they face so they can make well-considered decisions based on their own calculations of costs and benefits. But warning has weaknesses, so it cannot be relied upon as the sole means of coping.

A variety of psychological factors lead people to ignore immediate warnings of imminent disaster as well as cautions about using land that is hazardous. Some people are not able to accurately perceive probabilities of loss, even when they have been told a hazard exists. As a result, they tend to heavily discount any benefits from avoiding a hazard or taking action to reduce vulnerability. Even after they have been proven wrong and a disaster occurs, people tend to assume it will never recur (a conclusion likely to be refuted by subsequent events). Dissonance theory (i.e., beliefs follow behavior) predicts that, once people have exposed themselves to a hazard by locating in an area at risk, they will assume any hazards that might exist are trivial. Also, some people become fatalistic when thinking about hazards when they can neither prevent them nor estimate (very accurately) when and where losses are likely to occur. Thus, psychological difficulties in perceiving hazards lead most people to simply ignore or greatly discount the potential for disaster.

Difficulty in accurately estimating risk also limits the effectiveness of warnings. Hurricane Andrew offers an excellent example. In 1989 the South Florida Regional Planning Council published a vulnerability study that estimated maximum probable losses in south Florida at about $5 billion from a worst-case, Category 5 hurricane. Losses actually incurred three years later from Andrew, which was a lower intensity Category 4 hurricane, were five to six times larger! Geographers Ian Burton, Robert Kates, and Gilbert White (1993), in fact, have observed that experts frequently err on the low side when estimating vulnerability to losses from natural hazards. As a result, reliance on warnings may lead people to expose themselves and their property to a greater degree than they would have, had information about hazards been more accurate.

The limitations of warnings early on led governments to look to other means of coping with disasters. Relief and insurance, which take effect after catastrophic losses occur, reduce the adverse impacts of disaster and ease reconstruction and recovery. Research undertaken in the late 1970s provided strong evidence that relief and other emergency

measures work; studying communities some years after they had been hit by disasters, researchers could find no evidence of adverse effects (see Friesma et al., 1979; Wright et al., 1979).

But relief, too, has its problems. Lack of adequate preparation, planning, and coordination of services are endemic, so that in some cases the relief effort (particularly by subnational governments) is inadequate, and short-term suffering can be acute, as was the case in south Florida after Hurricane Andrew. Because it subsidizes people and firms occupying hazardous areas, relief can produce complacency. If they believe someone else will pick up the tab, individuals and communities may not be willing to take the steps necessary to reduce their own vulnerability, even when such steps are feasible and proven to be cost effective. This complaint is frequently voiced by economists (see, for example, Lichtenberg, 1994). And, of course, the costs are astronomical, as the billions of dollars now appropriated every year for disaster relief in the United States demonstrate. Furthermore, relief does little to prevent recurrence of disasters, since people frequently rebuild in the same disaster-prone locations using the same unsafe building techniques.

Insurance compensates for losses in a way that provides greater assurance to property owners and avoids some of the pitfalls of relief. If insurance rates are set at levels that actually reflect the risk of loss, then insurance premiums can transmit accurate economic signals to policyholders. As risk (and premium rates) go up, some people and firms will decide that the benefits of locating in (or continuing to occupy) a hazardous area are not worth the added insurance costs they will have to bear, and therefore they will locate elsewhere. Even those who have already located in a hazardous area and cannot relocate may adopt loss reduction measures such as elevating structures in a floodplain or strengthening buildings susceptible to shaking in an earthquake, if in doing so they can reduce their insurance rates. However, insurance also has a number of limitations.

Because of the potential for catastrophic losses, insurance providers must have the ability to spread losses widely over a broad area or to draw on reinsurance markets or public financial assistance. Spreading losses over a wide area has proven virtually impossible, however, because the psychological barriers to risk perception noted earlier lead most people to view insurance against natural disasters as a poor value. Only those most at risk—and whose frequent losses would drive a company insuring their property into bankruptcy—willingly purchase such insurance. In the nine-state region affected by the 1993 floods in the Mid-

west, for example, only about 20 percent of the structures in the floodplain were insured, a rate that is typical of the United States as a whole (see Interagency Floodplain Management Review Committee, 1994). Similar findings have been reported for earthquake insurance in California, where only about one in five households is insured.

The enormous insurance losses in recent natural disasters make reinsurance very difficult for companies to find. In response to heavy outlays required by recent disasters, the insurance industry has tried to limit catastrophic losses of reserves by shifting risk to government. When insurance companies began canceling policies in Florida following Hurricane Andrew, for example, the State of Florida instituted a surcharge on property insurance policies to create the Hurricane Catastrophe Fund, a reinsurance fund for private insurers. Florida and Texas have created ''wind pools" to provide insurance coverage to property owners who otherwise would not be able to afford coverage. Finally, an effort has been made to shift more hazard insurance risk to the federal government. Bills introduced in Congress in 1994 and 1995 would have established a public corporation to provide all-hazards insurance, with the federal treasury providing reinsurance for catastrophic losses that exceed loss reserves. Congress, however, has not passed legislation that would have the federal government expand its the role from that of an insurer of risk from flooding to a re-insurer of risks from all natural hazards.

Areas susceptible to natural hazards often have attributes that make them attractive for economic use. Floodplains and volcanic soils, which are particularly fertile, are farmed intensively. Most major cities were founded and initially grew because of their proximity to rivers that provided a ready means to import raw materials and export goods to far-flung markets. Those same floodplains in more recent times have been favored routes for expressways, because little effort and expense is needed for clearing and grading rights-of-way. The economic advantages provided by river and highway access have attracted billions of dollars of investment in areas at risk of flooding. In coastal areas at risk from hurricanes and winter storms, a variety of economic advantages have attracted development. Coastal fisheries, transportation through ports and along intracoastal waterways, and outdoor recreation have each attracted millions of people and billions of dollars in investment. These large investments in areas at risk from natural hazards early on led engi-

neers and policymakers to look for ways to reduce risk while continuing to reap the rewards of vulnerable locations.

Throughout history a variety of engineering methods have been used to keep hazards such as riverine and coastal flooding away from people and damageable property. Dams and flood storage reservoirs, levees, dikes, pumps, channel improvements and diversions, sea walls, and groins—each of these measures provides protection for large areas, and thus can be both effective and efficient in reducing losses.

The benefits to be gained from controlling the hazard, however, have to be weighed against the serious costs and various shortcomings of the structural approach. Structural protection against natural disasters comes at a high cost in dollars and does not provide complete protection, as massive flood damages to tens of thousands of homes and businesses behind breached levees in the upper Mississippi and Missouri River basins tragically demonstrated in 1993. The failure of structural protection in the 1993 flooding is not an isolated occurrence. A report prepared for the U.S. government indicates that fully two-thirds of national losses in flooding result from catastrophic events that exceed the design limitations of engineering works that are relied on to provide safety (Sheaffer et al., 1976).

In part, disasters such as the Midwest flooding in 1993 stem from a curious side effect of structural protection. People and businesses tend to view the structures as providing complete protection from loss, when in fact they provide only partial protection up to the limits of some storm event, such as a storm that happens, on average, once in 50 or 100-years. Even lower magnitude storms may overwhelm structural protection as a result of incorrect design, operational mistakes, or loss of capacity, such as the silting up of flood storage reservoirs. Because people do not understand that structural protection has limits, however, structures have been found to actually induce development in hazardous areas and to increase, not decrease, the likelihood that when a large flood or hurricane does occur, losses truly will be catastrophic (for evidence of this effect, see White et al., 1958, and Burby and French et al., 1985). Local governments have contributed to this loss scenario by waiving requirements for the elevation of buildings located behind levees, apparently because, like individuals, they overestimate the degree of protection structures actually provide.

In addition to catastrophic losses of life and property behind engineered structures, incalculable additional losses to nature have sometimes accompanied structural efforts to reduce vulnerability. For ex-

ample, levees along hundreds of miles of the Mississippi River erected by the U.S. Army Corps of Engineers following flooding in the late 1920s have, in turn, contributed to accelerated erosion in the state of Louisiana because the river's silt no longer replenishes wetlands. The conversion of wetlands to open water has ranged from a low of 25 square miles to a high of 60 square miles per year (see Kelley et al., 1984, and Louisiana Coastal Wetlands Conservation and Restoration Task Force, 1993). Elsewhere in the United States, thousands of free-flowing streams have been straightened, channelized, or converted to slack water behind flood control dams (L. R. Johnston and Associates, 1991, Chapter 12). Along the coast, hundreds of miles of shoreline have been "hardened" to protect adjacent property from hurricanes and coastal storms so that, in cases such as southern New Jersey and New Hampshire, solid walls of concrete and rip-rap now line the shore rather than dunes and dry-sand beach, which have all but disappeared (see Beatley et al., 1994).

The various shortcomings of warning and education, amelioration through relief and insurance, and hazard reduction through structural protection were well recognized by 1950. In that year, President Truman's Water Resources Policy Commission recommended that federal agencies work to foster a new approach—land use adjustments through zoning and other measures—as a way of reducing flood losses. The new land use approach the president's commission advocated had a simple premise: flood losses could be reduced significantly if, rather than trying to keep the flood out of people's way, government worked to keep people out of the flood's way by discouraging development of hazardous areas or, where development is warranted on economic grounds and little environmental harm results, by imposing special building standards that reduce vulnerability. These two types of land use measures, location and design, are still used today. (Land use approaches can mitigate the effects of an array of natural hazards, as described in Sidebar 1-1.)

The goal of the locational approach is to reduce losses in future disasters by limiting development in hazardous areas. This approach tends to be effective in reducing losses, preserving environmental values, and providing opportunities for outdoor recreation. But these gains come at the cost of giving up some of the economic benefits offered by hazard-prone land. The goal of the design approach, in contrast, is safe con-

struction in hazardous areas. This type of land use allows economic gains to be realized, but at the cost of greater loss of natural values and of susceptibility to greater damage when events overpower the design standards employed. A properly conducted planning process allows communities to find the right mix of these two land use approaches.

In managing the location of development, local governments work to shift existing development and steer new development to areas that are relatively hazard free. The tools communities can use for these purposes include land use regulations such as zoning, and a number of nonregulatory techniques, such as buying hazardous areas and using them for parks, buying repeatedly damaged structures and helping people relocate to a safer area, and locating development-inducing public facilities and services only in areas that are relatively free of hazards.

In managing the design of development, communities also can employ regulatory and nonregulatory techniques. Regulatory techniques include building codes and stand-alone ordinances that require actions such as elevating structures above estimated flood levels, bracing buildings to minimize damage from ground shaking during earthquakes, and

installing hurricane clips so that roofs stay attached to buildings in high winds. Nonregulatory techniques include public information and training programs to inform builders and building owners of damage-resistant design techniques, and low-cost loans and other types of subsidies that make the new designs affordable. In the almost half-century since the federal government first began espousing land use approaches to hazard reduction, literally dozens of manuals and other training tools have been prepared to show local governments and property owners how to apply locational and design techniques.

Land use design measures are effective in reducing losses to new development but have less impact on losses to existing development in hazard zones. The reasons are economic and political. In the case of new construction, developers can take into account the higher costs imposed by more stringent design standards before they begin a project and can work these costs into their calculations, passing them backward to landowners through lower offering prices for development sites or forward to final consumers. Astute landowners will oppose land use regulations that affect new development, but developers and builders are

likely to be neutral or supportive if development opportunities are not wholly foreclosed. In the case of existing construction, however, the added costs of retrofitting to meet more stringent design standards are not likely to have been anticipated when homes and other buildings were acquired and are hard to shift to someone else. As a result, researchers have found that imposing new standards on existing buildings through regulation is politically contentious and usually not very effective (for evidence of this, see Alesch and Petak, 1986, and Wyner and Mann, 1986). For similar reasons, trying to persuade owners to voluntarily retrofit their buildings to reduce the chances of damage in a disaster in most cases also has proven futile (see Laska, 1991). However, if information and regulations are supplemented with economic incentives, they have some potential to bring about relocation and retrofitting, as shown by various successful community efforts following the 1993 Midwest floods.

With the right mix of regulatory and nonregulatory land use management measures, local governments can achieve striking reductions in losses from natural disasters while at the same time accomplishing envi-

ronmental and other community goals. Burby and his associates conducted field studies to see what a national sample of ten cities had accomplished through planning and managing flood hazard areas over a ten-year period. The ten cities included: Arvada, Colorado (a suburb of Denver); Cape Girardeau, Missouri; Fargo, North Dakota; Omaha, Nebraska; Palatine, Illinois (a suburb of Chicago); Savannah, Georgia; Scottsdale, Arizona (a suburb of Phoenix); Toledo, Ohio; Tulsa, Oklahoma; and Wayne Township in northern New Jersey. The data collected in these cities revealed that over the ten-year study period, floodplain development had been reduced by over 75 percent of what would have occurred without the local planning programs (see Burby et al., 1988). Comparison of the benefits and costs of managing development ($11 million per year in reduced property damages versus $1.3 million in administrative and private costs) showed substantial net benefits from the efforts of these cities. But this and other studies have uncovered a number of serious barriers that have limited wider realization of these potential gains.

As with the earlier approaches to hazard mitigation, land use management has had to contend with some serious problems that have limited its use and effectiveness. The most important of these problems are a lack of local political will to manage land use, deficiencies in management capacity, private sector evasion of development rules, and regional fragmentation that limits opportunities for areawide management solutions. Showing how such barriers can be overcome is an important goal of this book.

One of the most serious limitations of the land use approach is that without strong mandates from higher-level governments, few local governments are willing to protect against natural hazards by managing development. As noted earlier, in some cases land use measures are not appropriate. More often, however, local governments simply ignore the

Wave impact on houses in Scituate, Massachusetts, after a nor'easter in 1991 (Ralph Crossen, Building Commissioner, Barnstable, Mass.).

potential for losses. Like individuals, they tend to view natural hazards as a minor problem that can take a back seat to more pressing local concerns such as unemployment, crime, housing affordability, and education. One study, for example, found that on average local officials rank natural hazards thirteenth in importance, just behind pornographic literature, among the issues with which they are dealing (Rossi et al., 1982). Local officials become most interested in land use management when their communities suffer chronic losses. But by then planning is much less effective. Planning is, by definition, preventive in nature and not well suited to correcting problems with hazards once they have been allowed to develop.

Land use management was first advocated by the federal government in the 1950s, but by the end of that decade less than 100 of some 20,000 flood-prone communities were using zoning to limit more intensive use of flood hazard areas (Murphy, 1958). By the end of the 1960s, less than 500 had adopted such measures, and even after passage of the National Flood Insurance Act of 1968 (P.L. 90-448, Title XIII; U.S. Code, vol. 42, sec. 4201, as amended, 1968), which mandated building

regulation as a condition for community eligibility for federally subsidized flood insurance, only 3,000 took part in the federal flood insurance program. Widespread participation in the program and adoption of the required building regulations did not come until after passage of the Flood Disaster Protection Act in 1973 (P.L. 93-234), which threatened to withdraw federal insurance for mortgages and funding of infrastructure (such as sanitary sewers) in flood-prone areas if communities did not adopt the requisite regulations. Today, about 90 percent of local governments with identified flood hazards participate and have adopted the federally mandated design standards. But for other hazards where a federal mandate does not require local regulation, much less has been accomplished.

The science of identifying hazards and designing to reduce their adverse impacts has far outrun the ability of local governments to put this new knowledge into practice. Land use approaches are founded on the accurate identification of areas affected by hazards, but hazard zone mapping is enormously expensive. In the case of flood hazards, for example, little progress was made in mapping areas subject to flooding until the 1950s, when the Tennessee Valley Authority, U.S. Army Corps of Engineers, and U.S. Geological Survey began helping with this effort. Even then, floodplain maps did not become widely available until the 1970s, when the federal government invested almost $1 billion in the production of maps needed to implement the National Flood Insurance Program. A similar federal effort has not been undertaken for other hazards, and local hazard identification often is sorely deficient.

Hazard zone mapping makes possible the adoption of both locational and design approaches to hazard mitigation. However, application of hazard-resistant methods requires knowledge and expertise often lacking in local government.

Few planning education programs provide detailed instruction in hazard mitigation and, as a result, few local planners have more than rudimentary knowledge of how to deal with hazards. Studies also have found that enforcement personnel may have insufficient knowledge of the hazard mitigation aspects of codes to enforce them effectively (see, for example, Southern Building Code Congress International, 1992). In addition, inadequate staffing may make it difficult for enforcement personnel, even if they are competent to interpret and apply code provi-

sions, to keep up with the demand for building inspection (see Building Performance Assessment Team, 1992).

A third critical barrier is lack of private sector compliance with land use and building regulations. Even if state and local governments adopt strict regulations to mitigate losses from natural disasters, the measures will not achieve their intended purpose if they are ignored, wholly or in part, by property owners, land developers, and builders. Massive breakdowns in enforcement and compliance with building codes have been implicated in the extraordinary losses experienced in recent disasters. In the case of Hurricane Andrew, for example, fully one fourth of insured losses were traced to shoddy design and construction practices. Analyses of building-damage distributions indicate that houses built after 1980 were 68 percent more likely to be uninhabitable after Andrew than older homes, even though building code standards applicable to newer houses were more stringent (Miami Herald, December 20, 1992, p. 3). Substandard construction practices were also reported as an important cause of heavy losses in Hurricane Hugo (All-Industry Research Advisory Council, 1989).

Yet another problem stems from the fact that hazards and the geophysical systems that engender them do not respect political boundaries. In some cases, land use management programs for hazard mitigation cannot be effective without intergovernmental coordination. Floodplain management experts Jon Kusler and Larry Larson (1993), for example, provide convincing evidence that the geographic scope of land use management needs to be expanded so that it embraces entire watersheds. When only a portion of a watershed is managed, serious consequences can result, such as when the hazard mitigation efforts of one jurisdiction lead to increased, not decreased, losses for neighboring communities. This sad result was documented for Jackson, Mississippi, which suffered millions of dollars in property damages over the Easter weekend of 1979 when flood waters of the Pearl River were pushed into the city by the levees of an adjacent city on the opposite bank of the river (see Platt, 1982). The multijurisdictional coordination needed to avert such disasters is very difficult to achieve. Even more difficult, however, is multi-

jurisdictional agreement to management programs that have enough substance to actually accomplish a marked decrease in vulnerability.

The various barriers to effective land use management strongly suggest that this approach is likely to be ineffective if used as the sole means of averting losses in natural disasters, just as warning, relief, insurance, and hazard reduction would each be inadequate if used alone. The appropriate mix of these approaches is likely to vary from one place to another depending upon local circumstances. Because of this variation, it is virtually impossible for federal or state agencies to prescribe from above what communities should do to lessen their vulnerability and enhance sustainability. Instead, we believe local planning has to be conducted if communities are to find the right combination of measures that at the same time are effective, efficient, equitable, and feasible.

Planning for hazard mitigation should ensure that

• Information on the nature of possible future hazard events is available to the public.

• Land subject to natural hazards is identified and managed in a manner compatible with the type, assessed frequency, and damage potential of the hazard.

• Land subject to hazards is managed with due regard for the social, economic, aesthetic, and ecological costs and benefits to individuals as well as the community, while taking into account the rights of private landowners.

• All reasonable measures are taken to avoid hazards and potential damage to existing properties at risk.

• All reasonable measures are taken to alleviate the hazard and damage potential resulting from development in hazardous areas.

An example of a land use planning process is summarized in Sidebar 1-2. The detailed planning illustrated in the sidebar can be undertaken as the first step in a stand-alone hazard mitigation effort, or it can be incorporated as one element of a local comprehensive planning program. Less intensive planning for hazard mitigation also can be effective, if communities systematically consider hazards in the course of ongoing comprehensive and neighborhood planning.

Recent studies in Australia, New Zealand, and the United States have documented a number of benefits that follow when governments plan be-

fore they act (Burby and May et al., 1997; May et al., 1996). First, through the planning process, individuals and community policymakers can acquire information about the location and nature of various hazards, and about the liabilities of building in hazardous areas. Second, planning can yield information about the most appropriate uses of land in a community, showing that in many cases hazardous areas do not have to be used more intensively for communities to realize economic and other development objectives, and making it possible for communities to consider restrictions on building in hazardous areas, and to adopt them where economically efficient. Third, linking natural hazards to other, higher-priority issues confronting decision makers can lead to increased priority for hazard mitigation. Finally, the planning process, by involving all stakeholders in finding mutually beneficial ways of dealing with hazards, can help build a political constituency that will demand action to deal with hazards and will support appropriate hazard mitigation measures. Because little can be done if policymakers are unwilling to devote public resources to hazard reduction, the last benefit is particularly noteworthy and, by itself, probably justifies the time and effort required to plan.

Progress toward development and implementation of land use plans for hazard mitigation requires commitment to and capacity for hazard mitigation at all levels of government. Developing the necessary commitment and capacity requires that partnerships be formed horizontally at the community level and vertically with higher-levels of government. The needed horizontal partnerships can be forged by a hazard mitigation committee and plan as described in Sidebar 1-2. The key vertical linkages that are essential to land use planning for hazard mitigation require attention by the states and the federal government.

The states and the federal government must play a lead role in building local capacity for and commitment to land use planning for hazard mitigation. The capacity of local governments can be enhanced by developing and providing information about natural hazards and by conducting research to formulate model hazard mitigation codes and standards. Training is equally important. Hazard mitigation is a new concept to many personnel in local government. Hazard mitigation training programs could include instruction in each of the components of mitigation planning, together with case studies showing how mitigation measures have been used in different types of communities.

But capability, once developed, is of no use without the willingness to use it to deal with natural hazards. Here higher levels of government have two important roles to play. The first is to lead by example. Governments at all levels have to take care that new facilities they fund are located, designed, and built in ways that minimize exposure to loss from natural hazards. The president issued executive orders in 1977 (floods) and 1990 (earthquakes) to direct federal agencies to avoid hazardous areas when at all practical and to construct all new buildings to be hazard resistant. The states need to take similar actions. The second and by far most important role of national and state governments, however, is to require, not just encourage, lower-level governments to prepare and implement land use plans for hazard mitigation.

If local governments are to prepare and follow through with land use planning for hazard mitigation, experience suggests that nothing short of strong mandates backed by the commitment and effort of the

state and federal governments are required. Without them, most local governments will continue with business as usual. A few innovative jurisdictions—those with extraordinary local leadership and those that have suffered severe losses from hazards in the past—will plan for and manage land use in hazardous areas. Most, however, will not, either because they lack adequate information about hazards and planning, or, more importantly, because there is no local constituency pushing in this direction. Thus, hazard mitigation requires a partnership. Impetus for land use planning and management must come from above, but the actual planning and conduct of programs must occur at the local level.

Hazard mitigation—actions taken to prevent and reduce risks to life, property, and social and economic activities—is a key focus of the Inter-

national Decade for Natural Disaster Reduction. This introductory chapter has described past responses to the threat of natural disasters and has outlined an approach to hazard mitigation based on local planning and land use management. In the remaining chapters of this book, a team of planning scholars expand on the ideas introduced here and illustrate how they can be realized.

The chapters of Part One, ''The Choices of the Past," describe the evolution of land use adjustments to natural hazards and diagnose causes of the current failed intergovernmental system. In Chapter 2, "Planning and Land Use Adjustments in Historical Perspective," geographer and lawyer Rutherford Platt chronicles the long struggle of societies around the world to come to grips with the forces of nature. He shows that many of the key ingredients of a land use approach to hazard mitigation have been known and experimented with for literally hundreds of years, but a variety of obstacles have limited their use—the most recent being

the growing property rights movement. Nevertheless, in the 1988 Stafford Act and the 1994-1995 effort to "reinvent" the Federal Emergency Management Agency, mitigation has been identified as the single most important means of reversing escalating disaster losses (Robert T. Stafford Disaster Relief and Emergency Act, P.L. 100-707, 42 U.S. Code sec. 5121 et seq.). Thus, after repeatedly coming to the well of land use planning for mitigation but not actually drinking, federal leadership now may be ready to help adopt the approach we advocate.

The job of Chapter 3, "Governing Land Use in Hazardous Areas with a Patchwork System" by political scientist Peter May and planning educator Robert Deyle, is to explain why governance systems have failed to prevent the breakdown in sustainability that disasters signal. Likening the nation's approach to hazard mitigation to an old house that has undergone repeated remodeling and cosmetic "fixes" to meet the needs of new owners, they conclude that the resulting patchwork "system" is

awkward, inconsistent in design, and not functioning well. By examining each of the system's component levels of government, however, they are able to isolate critical weak points and point toward needed reforms.

The chapters of Part Two, "The Land Use Planning Alternative," present four key components of successful planning for hazards—plans, information, development management, and partnerships with the private sector. In Chapter 4, "Integrating Hazard Mitigation and Local Land Use Planning," land use and environmental planning experts David Godschalk, Edward Kaiser, and Philip Berke share their experience in crafting land use plans for hazard mitigation. They describe the planning functions governments must engage in and illustrate what the resulting plans should look like. Because planning is not a static activity, they also discuss how plans can be evaluated so that planners learn from their experience and improve future planning products and processes.

In Chapter 5, "Hazard Assessment: The Factual Basis for Planning

and Management," planning educators Robert Deyle, Steven French, Robert Olshansky, and Robert Paterson explain why natural hazard information and hazard assessment are an indispensable part of a land use planning approach. They describe how hazard assessment has been and can be applied to land use planning and management. Their analysis highlights advances and limitations of these methods and the choices that must be made if hazard assessment is to be effective in supporting mitigation.

Chapter 6, "Managing Land Use to Build Resilience," by planning educators Robert Olshansky and Jack Kartez, examines the various tools governments can use to shape land development and redevelopment decisions. The authors also review what has been learned over the past 20 years about the willingness of local governments to actually employ various measures to reduce risk. Obviously, accurate information and good plans mean little if they are ignored by policymakers. The authors show

that policy innovation and adoption follow a fairly regular course in localities that are pursuing aggressive land use planning programs. The analysis highlights the various elements that ease adoption, separating those that are external to government from those that are internal and can serve as policy targets.

Planning, hazard assessment, and land use management can take communities a long way toward increasing sustainability by reducing risk from natural hazards. In addition, however, governments need the capacity to actually undertake the new tasks the land use approach demands. Chapter 7, "The Third Sector: Evolving Partnerships in Hazard Mitigation," by planning educator Robert Paterson discusses creating this capacity through partnerships between nongovernmental groups. Paterson argues that nongovernmental organizations and private voluntary organizations can play a number of roles to make mitigation work successfully. In arguing this case, he provides examples from all parts of the nation of new voluntary efforts that address a variety of hazards.

The two chapters of Part Three, "Looking to the Future," lay out our conception of sustainability as a goal for land use planning and management, and then examine what the federal government can do differently to move society toward more sustainable urban development by mitigating hazards. In Chapter 8, "The Vision of Sustainable Communities," planning educator Timothy Beatley explains why sustainability is critical to the welfare of the country and shows how it can guide decisions about mitigation. Importantly, Beatley demonstrates that sustainability does not mean non-use of areas at risk from hazards, but balance of the important economic, environmental, and social interests at play there. We conclude our exploration of the land use approach to disaster reduction in Chapter 9, "Policy Directions for the States and Nation" by planning educator Raymond Burby, with contributions from each of the other chapter authors. Here we present five principles to guide the nation toward our vision of sustainable communities. Using these principles as a standard, we discuss the accomplishments and shortfalls in federal and state policy from four perspectives (objectives, approaches, governance, and scope) and then look at each principle and identify reforms in federal and state policies and programs that could foster the sustainable future we envision. These include policy changes involving incremental alteration of current programs, presidential and gubernatorial executive orders, and major new legislation. We conclude with suggestions for research to improve the performance of existing hazard mitigation programs and to facilitate policy reform.