Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 Introduction The surface of the land is made by Nature to decay . . . Our fertile plains are formed from the ruins of the mountains. lames Hutton, 1785 The surface of the earth, both on land and beneath the oceans, is continually being modified by mass movements operating in response to gravitational forces. One effect of the mass movements termed landslides can be to reduce the gradient of hillslopes to stable angles. In this report, the term "landslide" will include all types of gravity- caused mass movements, ranging from rock falls, through a variety of slumps and slides, to debris flows. Both subaerial and submarine mass movements are included. Although precipitation, earthquakes, and vol- canic eruptions are the principal natural drivers of landslides, in many cases landslides result directly from disturbance of hillsides by road con- struction or other human activity. Landslides contribute to the erosion, transport, and deposition of earth materials. Over geologic time, they help produce stable land suit- able for agriculture and habitation and provide materials that form fertile plains and valleys, beaches, and barrier islands. Unfortunately, landslides are not completely benign to human beings, and because at the scale of the typical human life span the benefits accruing from landslides are over- shadowed by their destructive characteristics, they are viewed as hazards that should be understood and, if possible, mitigated. This report is focused on the identification, understanding, and mitigation of landslide hazards the destructive aspect of landslides. 6
INTRODUCTION 1.1 PROPOSAL FOR A NATIONAL STRATEGY The Disaster Relief Act of 1974 (now the Robert T. Stafford Disaster Relief and Emergency Assistance Act the Stafford Act) assigned respon- sibility for landslide hazard warning to the Director of the United States Geological Survey (USGS), providing a basis for the USGS to assume a prominent leadership role in national landslide hazard mitigation. The primary objective of the existing USGS Landslide Hazards Program is to reduce long-term losses from landslide hazards by improving scientific understanding of the causes of ground failure and suggesting mitigation strategies. The USGS Landslide Hazards Program has hitherto been funded at a modest level of $2 to $3 million each year. However, impetus for an increased emphasis on this program was provided by the House Report accompanying the Department of the Interior Appropriations Bill for FY 2000, which directed the USGS to develop a comprehensive strat- egy to address the hazards posed by landslides. During 1999-2000 the USGS convened a series of workshops and meetings to plan and develop a national strategy, resulting in the compilation of USGS Open-File Re- port 00-450, National Landslide Hazards Mitigation Strategy A Frame- work for Loss Reduction (Spiker and Gori, 2000~.~ This report proposed a national strategy based on partnerships between the USGS as the re- sponsible federal agency and an array of federal, state, local, commu- nity, and industry partners. This partnership strategy envisioned a sub- stantially increased federal investment for the USGS Landslide Hazards Program, requiring almost an order-of-magnitude increase from the present annual funding level of $2.6 million to at least $20 million. Of this total, $10 million would support increased USGS activities and $10 mil- lion would be provided to partners. The USGS strategy proposal (Spiker and Gori, 2000) presents an out- line of the elements required for a national approach to the landslide hazard problem, with the 10-year goal of reducing the risk of loss of life, injuries, economic costs, and destruction of natural and cultural resources caused by landslides. The report identifies nine elements of a national landslide hazard mitigation program: (1) research to develop a predictive understanding of landslide processes; (2) hazard mapping to delineate susceptible areas; (3) real-time monitoring of active landslides; (4) loss assessment to determine economic impacts of landslide hazards; (5) infor- mation collection, interpretation, and dissemination to provide an effective system for information transfer; (6) guidelines and training for scientists, engineers, and decision makers; (7) public awareness and edu- iA modified version of this report, with the same title, was recently published as USGS Circular 1244 (Spiker and Gori, 2003~.
8 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK cation; (8) implementation of loss reduction measures; and (9) emergency preparedness, response, and recovery to build resilient communities. The partnerships referred to in the USGS strategy document (Spiker and Gori, 2000) are described in broad outline: · Partnerships with state and local governments to assess and map landslide hazards, to be funded through competitive grants ($8 million annual allocation, requiring 30% matching funds) · Partnerships with other federal agencies (e.g., National Park Service, U.S. Forest Service [USES], Bureau of Land Management) to increase the capabilities of federal agencies to address landslide hazards ($2 million for USGS participation as requested by other agencies) · Partnerships with universities, local governments, and the private sector to support research and implementation efforts ($2 million annu- ally, distributed through competitive grants) The committee has reviewed the National Landslide Hazards Mitiga- tion Strategy (Spiker and Gori, 2000) and agrees that the nine major components identified in the proposed national strategy, ranging from basic research activities to improved public policy measures and enhanced mitigation, are the essential elements required to address the hazards arising from landslides at a national level. However, the treatment of these components in the strategic plan is brief and requires a more complete description of the comparative importance of each element. The com- mittee considers that in its analysis and discussion of each element of the proposed national strategy, it is essential that a sense of priorities be pre- sented. In framing its assessment and review of the National Landslide Hazards Mitigation Strategy, the committee has been particularly aware of the diversity of issues associated with the national landslide problem that arise from regional considerations and of the considerable variations in institutional capability and responsibility at the regional level. It is this range of capabilities, and the widespread demand at the local level for tools and information to address this national problem, that present such a clear argument for the coordination and assistance that would be pro- vided by a national program for landslide hazards mitigation. 1.2 COMMITTEE CHARGE AND SCOPE OF STUDY To be assured that the strategy advanced by the USGS was the most appropriate approach to this problem, the USGS requested that the National Research Council (NRC) conduct a review, with the charge presented in Box 1.1.
INTRODUCTION 9 The review committee established by the NRC to address this charge received input from a variety of interested parties during its information- gathering meetings from representatives of federal agencies, state agen- cies, local jurisdictions, private companies, and the academic community. Participants in these meetings uniformly supported a national approach to providing assistance for state and local agency landslide mitigation efforts. However, the wide variation in the nature and extent of existing state and local agency activities means that these are treated in a more generic sense when compared with federal agencies, where the committee was able to evaluate current activities on a nationwide basis and suggest specific roles in a future national partnership strategy. One issue that the committee grappled with was the extent to which other non-landslide ground failure hazard mitigation should or could be addressed in this assessment. The USGS national strategy proposal addresses only ". . . landslides, the most critical ground failure hazard facing most regions of the nation" (Spiker and Gori, 2000, p. 3) but asserts
10 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK that the strategy "provides a framework that can be applied to other ground failure hazards." In the absence of any detailed consideration of non-landslide ground failure in the national strategy proposal, the com- mittee considered that it was not possible to provide a review of non- landslide hazard mitigation and, accordingly, has restricted its assessment to landslides. However, the commentary on approaches to mitigation pre- sented in this report, particularly the emphasis on risk-based approaches, applies to ground failure hazard mitigation in the broadest sense. 1.3 SOCIOECONOMIC IMPACTS OF LANDSLIDES Landslides within the United States constitute a major geologic haz- ard, occurring in all 50 states and causing on average some 25 to 50 fatalities and damage of approximately $1 billion to $3 billion each year (NRC, 1985; Schuster and Highland, 2001~. The socioeconomic effects from the thousands of landslides that occur each year impact people, their homes and possessions; industrial establishments; and transportation, energy, and communication lifelines (e.g., highways, railways, communications cables). Socioeconomic losses are increasing as the pressure of expanding populations causes the built environment to expand into more unstable hillside areas. Landslides are responsible for considerably greater economic losses and human casualties than is generally recognized although they repre- sent a significant element of many major disasters, the magnitude of their effects is often overlooked by the news media. The losses attributed to most individual landslides are relatively small, although they can be devastating to individual property owners. Because damage costs are borne mostly by individuals, with only some involvement of federal, state, and local government relief and rehabilitation programs, the nation has largely ignored the financial risk posed by landslides. Landslide costs include both direct and indirect losses affecting pri- vate and public properties. Direct costs can be defined as the costs of replacement, rebuilding, repair, or maintenance resulting from direct landslide-caused damage and destruction of property or installations (Schuster and Fleming, 1986; Schuster, 1996; Schuster and Highland, 2001~. All other costs of landslides are indirect, for example: · reduced real estate values in areas threatened by landslides; · loss of tax revenues on properties devalued as a result of land- slides; · loss of industrial, agricultural, and forest productivity, and of tourist revenues, as a result of damage to land or facilities or interruption of transportation systems;
INTRODUCTION 11 · loss of human or domestic animal productivity because of death, injury, or psychological trauma; and · costs of measures to prevent or mitigate potential landslide activity. Private costs to individuals or corporations are incurred mainly as landslide-caused damage to land and structures, including private prop- erty and corporate industrial facilities. A destructive landslide can result in financial ruin for property owners because landslide insurance or other means to offset damage costs usually are not available. Public costs are those borne by government agencies national, regional, or local. Probably the largest direct public costs resulting from landslides are for the repair or relocation of transportation facilities (e.g., Box 1.2~. Based on a survey of state transportation departments, Walkinshaw (1992) found that the average annual direct costs of maintenance and repairs to U.S. highways as a consequence of landslide damage from 1985 to 1990 were nearly $106 million. Although this survey reflected actions only for state and federal highways, which represent about 20% of the entire U.S. road system, these have the larger cut-and-fill structures and are likely to account for the majority of landslide costs. Nevertheless, the survey may have underestimated these costs because many state trans- portation departments do not maintain detailed records of their landslide- related highway maintenance costs. Indirect public costs are diverse and include such disparate elements as loss of tax revenues, reduced capacity or capability of lifelines, reduced productivity of government forests, and impacts on the quality of sport fisheries. An interesting comparison of indirect to direct costs of landslides was provided by the 1983 closure of heavily traveled U.S. Highway 50 in California as a result of landslide activity, which prevented tourist access to popular Lake Tahoe. Although highway repairs totaled $3.6 mil- lion, the estimated loss of tourist revenues was a staggering $70 million (San Francisco Chronicle, 1983~. Widespread and numerous landslide occurrences caused by storms- many of which were probably related to E1 Nino have plagued Califor- nia for the past 50 years. Exceptional landslide activity occurred in 1951- 1952, 1956, 1957-1958, 1961-1962, 1968-1969, 1977-1978, 1979-1980, 1982, 1995, and 1997-1998. As an example, in the six southern counties of California total losses due to landslides caused by heavy winter rainfall in 1979-1980 were estimated at $500 million (Slosson and Krohn, 1982~. In January 1982, an intense storm triggered 18,000 debris flows and land- slides in the San Francisco Bay area, damaging or destroying about 6,500 homes and 1,000 businesses. The direct costs of these landslides were in excess of $66 million, but in addition, 930 lawsuits and claims in excess $298 million were filed against city and county agencies (Smith, 1982~. In
2 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK January and March 1995, above-normal rainfall in southern California trig- gered damaging debris flows, deep-seated landslides, and flooding in Los Angeles and Ventura Counties (Harp et al., 1999~. The most notable land- slide that occurred at this time was the deep-seated La Conchita landslide (Figure 1.1), which, in combination with a local debris flow, destroyed or badly damaged 11-12 homes in the small town of La Conchita (O'Tousa, 1995~. In the late winter and early spring of 1998, heavy rainfall again caused major landslide activity and damage totaling approximately $156 million in the 10-county San Francisco Bay region (Godt and Savage, 1999~.
INTRODUCTION 13 Although landslides are common throughout the Appalachian region and New England, the greatest landslide hazard in the eastern United States, at least in terms of financial losses within a fairly restricted area, is from landslides affecting clay-rich soils in Pittsburgh, Pennsylvania, and Cincinnati, Ohio (see Box 1.3~. Landslides also occur across the Great Plains and into the mountain areas of the western United States, where weathered shales and other clay-rich rocks occur near the surface, and they are particularly common where there are steep slopes, periodic heavy rains, and vegetation loss following wildfires. Earthquakes and volcanoes
4 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK FIGURE 1.1 The 1995 La Conchita landslide, southern California. SOURCE: Schuster and Highland (2001).
16 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK also cause landslides: the 1994 Northridge Earthquake in the San Fernando Valley triggered thousands of landslides in the Santa Susana Mountains north of the epicenter. In terms of loss of life, by far the most disastrous landslides to occur within the United States and its territories have been caused by hurri- canes or tropical storms making landfall from the western Atlantic Ocean. Hurricane Camille in 1969 caused extensive debris flows in central Vir- ginia although the exact number cannot be ascertained; most of the 150 who died as a result of Hurricane Camille are thought to have been victims of debris flows triggered by heavy rains associated with the hurricane (Williams and Guy, 1973~. In October 1985, heavy rain in Puerto Rico from Tropical Storm Isabel caused a major rock slide that obliterated much of the Mameyes district of the city of Ponce. The slide destroyed about 120 houses and killed at least 129 people the greatest death toll in North American history from a single landslide Gibson, 1992~. 1.4 ENVIRONMENTAL CONSEQUENCES OF LANDSLIDES Although landslides routinely cause local changes to the land surface, occasionally much larger changes are produced that affect subaerial and submarine landscapes, natural forests and grasslands, the quality of streams and other bodies of water, and the habitats of native fauna, both
INTRODUCTION 17 on the earth's exposed surface and in its streams and oceans (Schuster, 2001; Schuster and Highland, 2001~. Giant prehistoric landslides have been identified on the slope of Sinking Creek Mountain in the Appalachians of Virginia. These are among the largest known landslides in eastern North America and among the largest in the world. They were probably trig- gered in colder and wetter conditions during the last Ice Age, and these areas appear stable today. However, they have produced many disrupted terrain features, including scarps, ponds, and wetlands, that form important local ecosystems within the National Forest. Although historic landslide events usually are not as large as these examples, many large landslides have had significant impact on topography. One example is the Thistle, Utah, landslide of 1983 that dammed the Spanish Fork River, causing the valley to be flooded and forcing the relocation of major transportation routes (see Box 1.2~. High volumes of landslide-derived sediment can be delivered to stream channels. Debris flows can follow stream channels for great dis- tances, causing substantial channel modification, and they also provide important sediment transport links between hillslopes and alluvial chan- nels. Two examples illustrate these points. Studies in northern Idaho show that rotational landslides produce about 40% of stream sediment, debris avalanches produce an additional 40%, and only 20% is derived from surface-flow erosion (Wilson et al., 1982~. In a similar study in Puerto Rico, Larsen and Torres Sanchez (1992) found that 81% of sediment transported within the Mameyes River basin was contributed by mass wasting. An equally important aspect is that sediment levels may remain high for decades following major landslide events, increasing flood risks for down- stream communities and threatening efforts to restore fisheries and aquatic ecosystems (e.g., Madej,1995~. Nearly 20 years after the eruption of Mount St. Helens, sediment eroded from landslide and debris flow deposits con- tinues to elevate sediment levels in the Toutle River by 10 to 100 times (Bernton, 2000~. Although most kinds of wildlife are able to retreat fast enough to pre- vent injury from all but the fastest-moving landslides, all wild creatures are subject to landslide-caused habitat damage and destruction. Birds and other animals that nest or live in underground burrows are at high risk. Schuster and Highland (2001) reported that springtime landslides in fluvial-lacustrine sediments along the Columbia River in south-central Washington probably kill very large numbers of nesting cliff swallows. Landslides can adversely impact fish habitats, especially those of anadromous fish (e.g., salmon) which live in the oceans but return to freshwater streams to spawn. Landslides cause many changes in aquatic habitat. Elevated sediment delivery due to landslides can lead to increased mobility and scour of spawning gravels, increased fine sediment in
18 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK spawning gravels, increased fine sediment in pools, bank destabilization, and diminished availability of food organisms (e.g. Swanston, 1991~. In the Pacific Northwest, the increased occurrence of debris flows due to timber harvest activities (Sidle et al., 1984; Schmidt et al., 2001) and the coincident rapid decline in aquatic habitat and salmon stocks have led to the enactment of numerous state and federal land-use regulations and to frequent lawsuits over their application and validity (e.g., see University of California Committee on Cumulative Watershed Effects, 2001~. The USES has conducted numerous studies of the relationship between destructive landslides, forest cover, and logging operations in this area (Swanston and Swanson, 1976; Megahan et al., 1978; Swanston, 1991; McClelland et al., 1999~. It has become standard practice in assessing timber management plans to conduct watershed-scale surveys to map landslides (and landslide potential) and their possible association with and sensitivity to land-use practices. There is debate as to the most effi- cient and accurate way to do this, and how to interpret such maps in terms of land-use decisions. No standards have been established in the United States, and maps vary widely in their detail and accuracy and, consequently, their usefulness. It is important that not only should land- slide occurrence or relative slope stability be established for an area, but also the sediment and wood production associated with landslides should be estimated, because sediment and woody debris strongly influence aquatic habitats. Federal leadership is needed to establish methods and standards of reporting and interpretation. Landslides have occasionally directly caused human health problems. Following the 1994 Northridge, California, earthquake, Ventura County experienced a major outbreak of coccidioidomycosis (valley fever) a respiratory disease contracted by inhaling airborne fungal spores (libson et al., 1998~. The earthquake and its aftershocks produced many highly disrupted, dust-generating landslides in canyons northeast of Simi Valley, and prevailing winds transported dust into the Simi Valley and to com- munities farther west (Figure 1.2~. In the following eight weeks, 203 coccidioidomycosis cases were reported, about an order of magnitude more than would otherwise be expected. The temporal and spatial distri- bution of coccidioidomycosis cases indicated that the outbreak resulted from inhalation of spore-contaminated dust generated by the earthquake- triggered landslides. 1.5 THE CONCEPT OF LANDSLIDE MITIGATION The preceding sections, and the suite of case studies presented in Appendix A, highlight the immense socioeconomic and environmental damage and losses that result from landslides. A considerable variety of
20 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK This emphasizes that mitigation strategies extend far beyond engi- neering or technical measures designed to stabilize a landslide. Achieving mitigation by enforcing strict building code, excavation, and fill ordi- nances, as applied in the Cincinnati area (see Box 1.3), depends upon a sound scientific knowledge of the distribution and properties of the earth materials and potential landslide processes. Local zoning and subdivision regulations in California have been used creatively by some communities to concentrate building on stable land while leaving unstable land as open space to serve the development. This model could be adapted in other states. Also in California, public districts have been formed to provide funds for potential landslide repairs. Although mitigation options could include this type of financial arrangement to pool the exposure of indi- viduals to landslide hazards and to compensate for losses, such mecha- nisms have not been widely applied in the United States and there seem to be considerable obstacles to providing a broadly based national land- slide insurance program (see section 5.2~. As a consequence, the mitigation of landslide hazards for many existing developments and transportation networks is accomplished by using a variety of monitoring and warning systems which protect lives and property but do not prevent land- slides or by resorting to expensive engineering stabilization solutions. Engineering solutions are often expensive and may not provide a per- manent solution. In some cases a "permanent" solution is unnecessary- the landslide hazard is high for only a relatively short time, such as dur- ing excavations for a new structure. One example of early and innovative temporary landslide hazard mitigation is provided by the construction of Grand Coulee Dam on the Columbia River in Washington State (U.S. Bureau of Reclamation, 1936; Hansen, 1989~. The north abutment of the dam was threatened when a huge mass of water-soaked silt began to creep into the excavation. Engineers placed numerous pipes into the silt mass and a refrigerant was pumped through these pipes, thereby freezing the water in the silt and temporarily stabilizing the landslide until the concrete for the abutment had been poured and cured. This temporary mitigation activity was accomplished in about six weeks, saving considerable time and expense. Occasionally, opportunities arise that allow for ingenious engineer- ing solutions of a more permanent nature. The construction of Inter- state 70 over Vail Pass in the 1970s encountered highly unstable conditions and numerous landslides along the route, especially on the western side of Vail Pass where several sections of almost continuous landslides are found along Black Gore Creek. Many innovative concepts and designs were utilized or developed on this project because slope stability and erosion control were paramount concerns. At one location on the western side of Vail Pass, two landslides on opposite sides of the valley were
INTRODUCTION 21 stabilized by allowing them to buttress each other. Fill was added to the valley, the stream profile was raised and controlled to prevent further erosion of their toes, and the highway was constructed on the fill (Robinson and Cochran, 1983~. Where land values are high or the need for a lifeline is high, the cost of expensive engineering solutions can be justi- fied. In other instances, especially in larger developments, it may be pref- erable to build on stable terrain and use the potentially unstable terrain as valuable open space. One can argue that the most beautiful parts of the world are where the hazard of landslides is greatest and, because such areas are increas- ingly a focus for development, where the risks are most extreme. For this reason, the present report emphasizes the identification and mitigation of landslide hazards rather than either completely disallowing development or proposing to prevent landslides. By knowing the hazard and the risk, the informed citizen, developer, or public official can calculate the trade- offs among beauty, health, and wealth, and between fortune and fate. We must learn to live intelligently among the mountains and their ruins. 1.6 OVERVIEW OF NATIONAL STRATEGY PRIORITIES Landslides are widely distributed geographically and pose differing types of hazards depending on geologic setting and terrain type. The diversity of landslide problems, and the breadth of the needed elements of a national landslide hazard reduction program, can be illustrated by examples: · Debris flows triggered by extreme rainfall events have had devas- tating effects in mountainous regions of the United States, and there are indications that differences in climate, materials, vegetation, and topogra- phy may cause a variety of debris flow phenomena in different regions. Accordingly, debris flow hazard reduction will require improved under- standing of the initiation and propagation of these flow events and their region-specific characteristics. Once such basic science questions have been answered, the hazards posed by debris flows can be reduced by application of appropriate risk assessment and mitigation techniques. · Rock falls pose severe hazards, particularly along transportation corridors in many mountainous states. Although often involving only a small volume of material, the speed of rock falls and the hazard they pose to motorists have prompted several state departments of transportation to support rock fall research. The science of rock fall mechanisms is rela- tively well understood, and several computer simulation programs have been developed to aid in evaluating the hazard. However, improvements are needed in establishing standards for risk management and for certify-
22 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK ing the effectiveness of rock fall barrier systems. This can be achieved by encouraging more widespread adoption of established techniques through technology transfer. · Bedrock slides occur in many locations throughout the United States, displaying a range of movement types and resulting from a wide variety of triggering events. They can be local events involving small vol- umes or multiple events and large masses that involve a considerable area. Slides can move quite slowly or very rapidly, and the hazards they pose range from relatively minor to catastrophic. Once initial movement has occurred, bedrock slides can be mapped readily if the needed resources are available. Nevertheless, they continue to cause extensive economic losses due to ineffective regulatory controls on development in slide- prone areas. In general, improved risk assessment is needed for all types of land- slide hazards (see Chapter 4 below), as are advances in methods of cost- effective mitigation that might include hazard insurance and other financial instruments. Specifically, the establishment of landslide hazard mitiga- tion priorities should incorporate existing knowledge and the potential for cost-effective results. The matrix presented in Figure 1.3 evaluates the six broad landslide types2 against five activities that should be included in an effective national strategy to address the diversity of landslide hazard problems: 1. Improvement of the science base to provide an adequate understand- ing of landslide triggering and landslide movement mechanisms is an essential first step to fill gaps in current understanding and is a funda- mental requirement for other activities. 2. Technology integration and transfer is important for both the dissemi- nation of scientific understanding of the hazard and the identification of appropriate mitigation methods. 3. Mapping and monitoring provides the fundamental database for identification and delineation of landslide hazards. 4. Risk assessment integrates the many factors relating to slide occur- rence and consequence; it can be applied at various levels, ranging from qualitative to quantitative. 5. Mitigation takes many forms, with land-use regulation being the most important. Other mitigation activities include stabilization through engineering activities and construction of diversion works. 2A detailed classification and description of the numerous landslide types is presented in Cruden and Varnes (1996~; these have been simplified into six broad landslide categories for this report.
24 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK · Rock Fall: Rock fall processes are relatively simple and reasonably well understood. The Federal Highway Administration and some state highway departments have made substantial progress in developing rock fall hazard rating systems and in technology integration and transfer. It appears that widespread dissemination of this information would encour- age implementation and have a high payoff potential. At the same time, improved mitigation methods, such as improved criteria for testing, certi- fying, selecting, designing, and installing rock fall barriers, are needed. Establishment and broad acceptance of appropriate risk assessment tech- niques are also required. · Bedrock Slides: There is reasonable understanding of the mechanics of bedrock slide initiation, although additional case histories would add significantly to the body of knowledge. Although rare, very large, fast- moving bedrock slides are potentially life-threatening, and their move- ment dynamics are poorly understood and require further study. Post- failure movement dynamics and deformations of many bedrock slide types are poorly understood, and further expansion of the science base in this area is desirable. Once initial movement has occurred, bedrock slides can be identified with current technology and there is high payoff poten- tial associated with mapping them in areas of high risk in order to assist regulation. Improved mitigation methods and the establishment of appro- priate risk assessment techniques are needed. · Liquefaction Flow: Liquefaction flows are often caused by earth- quakes, but they can also occur in some types of glaciomarine clays. The basic science of liquefaction flow has received considerable attention in recent years, and liquefaction susceptibility criteria have been established and tested in the field. A high payoff potential can be expected from map- ping this hazard. As with other landslide types, improved mitigation methods and the establishment of appropriate risk assessment techniques are needed. · Soft Clay Slides: Geotechnical engineers have devoted substantial effort to understanding the mechanics of soft clays; as a result, the initia- tion and movement of landslides in these deposits are the best understood of all landslide types. Mapping is straightforward. Improved mitigation methods and the establishment of appropriate risk assessment techniques are needed. · Submarine Landslides: Because of the likelihood that submarine landslides will cause highly destructive tsunamis (e.g., the 1998 Papua New Guinea tsunami; Bardet et al., 2003; Liam Finn, 2003; Wright and Rathje, 2003), there is an urgent need to better understand the mechanics of submarine slide movement, particularly the role of gas hydrates in caus- ing shelf edge instability. At present there is a vast amount of geotechnical data scattered among hydrocarbon exploration and development compa-
INTRODUCTION 25 nies, offshore geotechnical companies, and academic institutes, and these data have to be collated and the gaps identified. Areas of potential sub- marine failure will have to be mapped and procedures determined for submarine landslide risk assessment. The elements of a national landslides hazards mitigation strategy are dealt with in more detail in the following chapters; Chapter 2 describes requirements and priorities for research into landslide processes; Chap- ter 3 describes the status of mapping and monitoring techniques and their application; Chapter 4 describes the importance of loss and risk assess- ment; Chapters 5 and 6 describe the technology transfer and integration components of a national mitigation strategy; and Chapters 7 and 8 describe the partnerships and funding that will be required for implemen- tation of an effective national strategy. Chapter 9 contains the committee's conclusions and recommendations.