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.
Landslide Mapping and Monitoring T'' he identification and map portrayal of areas highly susceptible to damaging landslides are first and necessary steps toward loss-reduction" (Zeizel, 1988~. Because individuals or groups do not undertake mitigation actions when they do not understand what to do, lack training, or do not have access to appropriate and understand- able technical information, the communication and use of technical infor- mation is crucial for effective landslide hazard mitigation programs. There are four general categories of potential users of landslide hazard informa- tion (Word and Jochim, 1989~: 1. scientists and engineers who use the information directly; 2. planners and decision makers who consider landslide hazards among other land-use and development criteria; 3. developers, builders, and financial and insuring organizations; and 4. interested citizens, educators, and others with little or no technical experience. Members of these groups differ widely in the kinds of information they need and in their ability to use that information (Word and Tochim, 1989~. Most local governments do not have landslide hazard maps and do not have funding available for mapping activities, and such communities usually look to a higher level of government for mapping. The U.S. Geo- logical Survey (USGS) has provided maps in some areas (e.g., demonstra- tion mapping of San Mateo County, California; Brabb et al., 1972), but in general, landslide hazard mapping by the USGS has had limited geo- 31
32 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK graphic coverage. Although most local communities look to their state as the primary source of maps, few states have undertaken significant land- slide hazard mapping programs. However, there are important excep- tions. California and Oregon, for example, have undertaken landslide hazard mapping at standard USGS mapping scales. These maps provide an excellent starting point for local communities and, importantly, form the basis for state laws that require a certain level of compliance with the information they provide. The considerable variability among state geological agencies, particu- larly in terms of their existing mapping capabilities and projected funding environments, makes it difficult to provide detailed commentary and suggestions regarding the partnerships between the USGS and states for landslide hazard mapping and assessment. Historically, there have been strong ties between the USGS and state geological surveys in the realm of mapping (e.g., Ellen et al., 1993; Coe et al., 2000b) and, to a lesser extent, for the identification and mitigation of natural hazards. The suggestion in the national strategy proposal (Spiker and Gori, 2000) of mapping part- nerships, using a model based on competitive grants and matching funds (as with the existing National Geologic Cooperative Mapping Program), would undoubtedly provide resources for a considerable amount of much-needed mapping. However, such a model raises the possibility that hazard mapping assessed as having a high priority might not be possible if state matching funds are not available. It is important that the details of the cooperative mapping partnership be worked out carefully, in close consultation with state geologists, as the national strategy implementa- tion plan is being developed. The principles and scope of the landslide hazard mapping, assess- ment, and delineation task contained within the USGS National Land- slide Hazards Mitigation Strategy (Spiker and Gori, 2000) are defined in section 3.1. This section uses several important terms, such as "hazard," "susceptibility," "donation," and "vulnerability," that are defined in Box 3.1. Landslide hazard Donation is commonly portrayed on maps. Preparation of these maps requires a detailed knowledge of the landslide processes that are or have been active in an area and an understanding of the factors that may lead to an occurrence of potentially damaging land- slides. Accordingly, this is a task that should be undertaken by geo- scientists. In contrast, vulnerability analysis, which assesses the degree of loss (see Box 3.1), requires detailed knowledge of population density, infrastructure, economic activities, and ecological and water quality values and the effects that a specific landslide would have on these elements. Specialists in urban planning and social geography, economists, and engineers should perform these analyses.
LANDSLIDE MAPPING AND MONITORING 33
34 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK Because landslides both leave a topographic signature when they occur and are driven largely by topographic effects, improved sources of high-resolution topographic information have the potential to greatly increase the accuracy of landslide hazard maps. The probable impacts of new remote-sensing tools on the creation of such maps are reviewed briefly in section 3.2. The fundamental importance of landslide hazard mapping, assess- ment, and delineation to the development of effective loss reduction strat- egies is discussed in section 3.3. It defines the role of landslide donation mapping in defining priorities for landslide investigations, monitoring activities, or mitigation programs within national, regional, or local land- slide hazard mitigation plans. In some cases, engineered mitigation may be undertaken immediately after the susceptibility of an area is identified, but frequently an engineered approach to mitigation is not cost-effective. In such cases, monitoring systems, discussed in section 3.4, may provide the most appropriate miti- gation option. Even in cases where engineered mitigation is eventually planned, interim monitoring capabilities are often installed to ensure that any landslide movement is identified as early as possible so that injury or economic costs can be avoided. Thus, monitoring of landslide-prone regions is an important adjunct to susceptibility and hazard mapping. 3.1 SUSCEPTIBILITY AND HAZARD MAPPING The USGS proposal for a national strategy (Spiker and Gori, 2000) identified three activities that will be required to provide the maps, assess- ments, and other information needed by officials and planners to reduce landslide risk and losses: 1. Develop and implement a plan for mapping and assessing land- slide and other ground failure hazards nationwide. 2. Develop an inventory of known landslide and other ground failure hazards nationwide. 3. Develop and encourage the use of standards and guidelines for landslide hazard maps and assessments. The USGS proposal states that "landslide inventory and landslide susceptibility maps are critically needed in landslide prone regions of the nation. These maps must be sufficiently detailed to support mitigation action at the local level. To cope with the many uncertainties involved in landslide hazards, probabilistic methods are being developed to map and assess landslide hazards" (Spiker and Gori, 2000, p.l3~. The proposal also notes that "these maps and data are not yet available in most areas of the
LANDSLIDE MAPPING AND MONITORING 35 United States." The committee concurs that the national strategy proposal appropriately identifies the landslide hazard Donation task as the primary responsibility of the USGS, to be undertaken in partnership with states; the following analysis and recommendations concern implementation of this task. Hazard Donations may be mapped at various scales; user require- ments and the intended applications determine the appropriate scale (Box 3.2~. Because a clear understanding of the different types of land- slide hazard maps is critical for successful implementation of a national strategy, the definitions from Spiker and Gori (2000) are reproduced in Box 3.3. In the absence of accepted national standards for landslide hazard maps, a variety of mapping styles have been employed for each type of map. This even applies to landslide inventory maps the most basic type of landslide map. These document the locations and outlines of landslides that have occurred in an area during a single event or multiple events. Small-scale landslide inventory maps may show only landslide locations and general outlines of larger landslides, whereas large-scale maps may distinguish landslide sources from landslide deposits, classify different kinds of landslides, and show other pertinent data (Figure 3.1~. The quantitative definition of hazard or vulnerability requires analysis of landslide-triggering factors, such as earthquakes or rainfall, or the application of complex models. Both tasks are extremely difficult when dealing with large areas. Consequently, the legends for most landslide hazard maps usually describe only the susceptibility of certain areas to landslides (Figure 3.2), or provide only relative indications of the degree of hazard, such as high, medium, and low. Over the past decades, geoscientists have developed several approaches to landslide hazard analysis, which can be broadly classified as inferen- tial, statistical and process-based (Hansen, 1984; Varnes, 1984~. All three approaches (Box 3.4) are currently applied to produce the different map types defined in Box 3.3, and there is no standard approach used in the United States. Not all methods of landslide Donation are equally applicable at each scale or for each type of analysis. Some require very detailed input data that can be collected only for small areas because of the required levels of effort and the high cost. Consequently, selection of an appropriate mapping technique depends on the type of landslide problems occurring within an area of interest and the availability of data and financial resources, as well as the duration of the investigation and the professional experience of the experts involved. When carefully applied by well-qualified experts, the inferential approach may describe the real causes of slope instability, based on scien- tific and professional criteria. However, due to the scale and complexity of slope instability factors, the basic inferential approach is unlikely to be
36 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK
LANDSLIDE MAPPING AND MONITORING definitive over large areas when mapping is conducted at small scales. For such applications, the combination of expert inference and qualita- tively weighted contributing parameters greatly improves the objectivity and reproducibility of the Donations. Combined statistical and process- based approaches may efficiently provide reliable regional landslide Donations over large areas, by classifying the terrain into susceptibility classes that reflect the presence and intensity of causative factors of slope- instability. For detailed studies of small areas, large amounts of data may become available, and in such cases, simple process-based models become increasingly practical for establishing landslide hazard Donations. They allow variations in the safety factor to be approximated and, thus, yield information useful to design engineers. In an environment where choices must be made from a number of possible mapping approaches, the pro- posed National Landslide Hazards Mitigation Program can play a vital role in evaluating methods, setting standards, and ratifying procedures.
38 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK FIGURE 3.1 Detail of inventory map showing recently active and dormant land- slides near La Honda, Central Santa Cruz Mountains, California. Information shown on this map includes state of activity, dominant type of movement, scarp location, and depth and date of movement. SOURCE: Wieczorek (1982~. 3.2 NEW REMOTE-SENSING TECHNOLOGIES Remote-sensing is used here in its broadest sense to include aerial photography as well as imagery obtained from a variety of platforms, ranging from ground-based mobile units to airborne or satellite platforms. Because landslides directly affect the ground surface, remote-sensing tech- niques are well suited to slope instability studies. Remote-sensing images can provide diagnostic information concerning the overall terrain condi- tions that often are critical for determining susceptibility to slope instability. Landslide information extracted from remote-sensing images is related mainly to the morphology, vegetation, and drainage conditions of the slope. The interpretability of slope instability or movements on remote- sensing images is related to both the size of the landslide features and
LANDSLIDE MAPPING AND MONITORING 39 FIGURE 3.2 Map showing areas susceptible to landslides in the Green Mountain area of the Morrison Quadrangle, Colorado. SOURCE: Scott (1972). their contrast to "background" conditions in the vicinity. Results are greatly dependent on the skill and experience of the interpreter (Soeters and van Westen, 1996~. Interpretation of aerial photographs, especially stereographic images, for identifying slope instability has long been accepted as a valuable land- slide investigation technique. However, a number of new remote-sensing opportunities now exist for landslide investigators. Earlier satellite imagery with relatively low spatial resolution was of little use for landslide studies except for basic inventories of large regional extent. The comparatively recent advent of commercial satellites capable of providing images with 1-m, and even submeter resolution suggests that such satellite imagery will form an important component of future landslide studies.
40 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK
LANDSLIDE MAPPING AND MONITORING 41
42 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK Perhaps the most important of the new remote-sensing tools is the use of airborne mounted lasers (LIDAR [Light Detection and Ranging]) to produce remarkably fine-scale topographic maps. Landslides occur where the landscape steepens, and small differences in topography can produce large differences in the likelihood of ground failure. Computer-based analysis of topography will increasingly play a crucial role in identifica- tion of the location and analysis of the relative potential for landsliding. High-quality digital elevation data must be one of the foundations of a national program for landslide hazard mapping and mitigation. Currently, the best available digital elevation data derived from digitizing standard 1:24,000 USGS topographic maps (with 10-m spacing) are inadequate to perform this analysis. High-resolution topographic maps can now be made quickly over large areas using LIDAR methods, which allow a grid spacing of as little as 1 m. In forest terrain, LIDAR- based terrain maps can be more revealing than high-resolution aerial photographs (Figure 3.3~. The first landslide hazard maps based on LIDAR imagery are now being reported and analyzed by various groups, includ- ing the USGS. These early maps show that high-resolution airborne LIDAR surveys can reveal previously unrecognized deep-seated land- slides, can capture the sharp edges of small shallow landslide scarps and debris flow runout tracks, can show debris flow fans, and can greatly improve models for mapping shallow landslide potential. Much work lies ahead in learning how to exploit these data. At present, airborne laser mapping is carried out by one federal group, two universities, and a variety of private companies. The National Aero- nautics and Space Administration (NASA) Airborne Topographic Mapper (ATM) pioneered LIDAR surveying, with predecessors that go back to the 1970s. Two major contributions arising from the ATM flights have been the ALACE (Airborne LIDAR Assessment of Coastal Erosion) program and the mapping of glacier fields in Antarctica, the Arctic, and Greenland. Much of the coastline of the United States (excluding Alaska and Hawaii) has been covered by the ALACE program. The USGS and the National Oceanic and Atmospheric Administration (NOAA) have both established projects associated with the ALACE program. Unlike the coastal program, the USGS is using commercial sources to generate LIDAR for its work on fault and landslide hazards. The ATM has recently completed some flights to address inland problems such as floodplains development. The USGS Coastal and Marine Geology Program participated in coastal erosion studies along the Washington, Oregon, and California coasts to assess the damage caused by the E1 Nino winter storms of 1997- 1998 (USGS, 2003a). These studies used repeat LIDAR surveys to generate accurate topographic models of the coastal areas before and after the storms to strikingly illustrate topographic changes.
LANDSLIDE MAPPING AND MONITORING 45 by state geological surveys in partnership with local governments or other state emergency organizations. The preparation of such maps was largely the result of mandates within the Stafford Act. A hazard mitigation clause is incorporated into FEMA-state agreements for disaster assistance, stipu- lating the identification of hazards and the evaluation of hazard mitiga- tion opportunities as a requirement for federal assistance. In the case of state-declared disasters, some states also require the development of local hazard mitigation plans, and thus landslide zoning maps, as a prerequisite for receiving state emergency relief funds. Some areas have been subject to extensive landslide mapping and inventorying programs, in collabora- tion with federal agencies (e.g., Cincinnati, see Box 1.3; the San Francisco Bay Area, Box 3.5; and Colorado, Box 3.6~.
46 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK A national landslide inventory would form an important first step toward an appreciation of the true scope and distribution of landslide hazards. An accurate inventory would provide metrics for national poli- cies and would greatly reduce the present uncertainty concerning the magnitude of economic loss and environmental damage caused by land-
LANDSLIDE MAPPING AND MONITORING 47 slides. Individual components of a national inventory should be compiled by relevant state and local agencies in partnership with federal agencies. In an environment where choices must be made from a number of pos- sible approaches, the proposed National Landslide Hazards Mitigation Program can play a vital role in evaluating methods, setting standards, and ratifying procedures. The USGS National Cooperative Geologic Map- ping Program's STATEMAP component provides an excellent model for the provision of federal assistance to states, on a matching-funds basis, to carry out the mapping that ultimately will populate the national inven- tory. The combination of resources from both federal and state spheres, with input from local agencies where possible, offers the potential for effective identification and increased understanding of landslide hazards. The long-term management and maintenance of a national inventory would require commitments and resources that could best be provided by the federal government, through the USGS. 3.4 LANDSLIDE MONITORING TECHNIQUES Monitoring existing landslides and sites of potential landslides plays an important role in landslide loss mitigation and landslide research. Monitoring serves several important purposes: · Identification of initiation of sliding or increased rates of sliding to provide a basis for alarms and warnings that can reduce landslide hazards · Determination of the depths and shapes of landslide masses as an adjunct to susceptibility and hazard mapping · Development of improved understanding of landslide processes and triggering mechanisms · Development of improved understanding of causative factors, such as earthquakes or high rainfall events · Evaluation of the effectiveness of control measures. Geophones, tape extensometers, piezometers, and rain gages have been used to monitor landslide movements for many years (USGS, 1999~. However, monitoring technologies have improved rapidly in recent years, reducing the cost and expanding the range of monitoring possibilities. This has resulted in more widespread deployment of landslide warning systems that will also provide detailed data to enhance understanding of landslide mechanisms. Programmed, automated Electronic Distance Measurement systems have been employed in many locations where slopes must be monitored continuously. Global positioning system (GPS) and differential GPS (dGPS) technologies have been used to monitor land- slide movements since the mid-199Os. The USGS conducted field tests at
48 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK the Slumgullion landslide in Colorado (Box 3.7) and demonstrated that the observed rate of change of high precision GPS coordinates could be used to determine slide velocities within 10% on a time scale of a few days Jackson et al., 1996~. More recently, dGPS techniques have been used at the Panama Canal to measure landslide movements with extremely high accuracy. Satellite-based interferometric synthetic aperture radar (InSAR) technology is capable of measuring vertical ground displacements with ~_GP~5 ~GPO] : .. ~ 'S ~ ; If- aria cruddy
LANDSLIDE MAPPING AND MONITORING 49 millimeter-level accuracy. Lower-cost, aircraft-deployed InSAR is being developed through NASA-sponsored research at Brigham Young Univer- sity and is being used to monitor movement of the Slumgullion landslide. Five landslides along U.S. Highway 50 in California have been moni- tored in real time using 58 instruments since heavy rains in lanuary 1997 caused slope failures that destroyed three homes and blocked the highway (USGS,l999~. This network, operated in cooperation with the California Department of Transportation, provides engineers and geologists with early notification of landslide activity and with information useful in the design of remedial measures to halt these slides. Real-time data from one of these landslides are available to the public on the Internet (USGS, 2003b). The USGS operates other remote real-time landslide monitoring sites. Near Seattle, Washington, real-time systems monitor the Woodway land- slide and two other unstable bluffs that threaten a major railway (USGS, 2003c). Remote monitoring by the USGS in Fremont, California, monitors movement of a landslide that threatens homes, and in New Mexico, remote monitoring has recorded the effects of wildfire on slope stability (USGS,l999~. In Colorado, remote monitoring provides notification of ground movement caused by the very large landslide in DeBeque Canyon that threatens Interstate 70 (see section 7.3~. In Rio Nido, California, the county government has assumed responsibility for operating a former USGS system that monitors a large landslide threatening more than 140 homes. The national strategy proposal suggests that all partners (federal, state, local, private, and academic) be involved in (1) improving capabili- ties for monitoring, (2) monitoring landslides, and (3) establishing land- slide warning systems. One part of the strategy proposes that real-time monitoring be integrated with NEXRAD (next-generation radar) data to improve warning capability. An existing partnership between USGS and NOAA's National Weather Service seeks to understand the relationship between rainfall intensity and duration with the thresholds for landslide initiation and the geologic determination of areas susceptible to land- slides, so that real-time rainfall monitoring can be used for landslide hazard warning (e.g., Wieczorek et al., 2001, 2003~. There is potential for broad application of such efforts to critical areas, with the likelihood that the involvement of FEMA in the partnership would assist emergency management. The USGS has taken the lead in applying the latest monitor- ing technologies in its volcano hazard program, and in its use of such technology for landslide monitoring at the Highway 50, Rio Nido, and Mission Peak landslides in California, it has worked with state agencies with the intention of ultimately transferring these capabilities to the state. It is desirable that the USGS maintain a research program that stays abreast of rapidly advancing monitoring technologies, so that it is able to
50 PARTNERSHIPS FOR REDUCING LANDSLIDE RISK assist state and local government agencies to acquire the capability to deploy the latest systems where they can be used to enhance public safety. The USGS also has a logical role in using these technologies to develop detailed field data that will improve understanding of landslide mecha- nisms and in supporting university-based research toward this end. Private firms in the electronic and telecommunication fields will play an essential role in developing new technologies and bringing them to market. Private firms in the geotechnical consulting field will make use of these technologies to improve the state of practice in landslide monitoring for their private and public sector clients. The USGS, together with NSF and NASA, should also play a role in developing monitoring capabilities by supporting studies to explore new technologies and reduce their costs.