2
Research and Hazard Assessment

The ability of any organization or individual to translate observations of volcanoes into interpretations of present or future eruptive behavior requires grounding in the theoretical understanding of how volcanoes work. At the same time, the advance of theory depends on a steady infusion of new measurements. Thus, the interactions between the development of new concepts and volcanic data collection and analysis are inextricably intertwined.

Most of this report focuses on the daily activities of the VHP, what may be referred to as the “operational” component of its mission. In this chapter, the committee begins by briefly examining the changing contributions of VHP scientists to the development and promulgation of the theoretical framework of volcanology. Hazard assessment, the operational activity that most clearly connects with research, is then explored in some depth.

RESEARCH

It is difficult to separate the contributions to basic volcanological knowledge made by VHP scientists from those made by their colleagues in other parts of the USGS, other government agencies, universities, other countries, and the private sector. Nonetheless, throughout much of the second half of the twentieth century, members of the present day USGS Volcano Hazards Program were national if not global leaders in the formulation of ideas about how volcanoes work. Building upon a steady stream of fresh observations from the HVO, VHP personnel advanced the understanding of the roles and significance of earthquakes, deformation, explosivity, gases, and lava flow mechanics in the evolution of ocean island volcanoes. Other USGS scientists conducted long-term



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program 2 Research and Hazard Assessment The ability of any organization or individual to translate observations of volcanoes into interpretations of present or future eruptive behavior requires grounding in the theoretical understanding of how volcanoes work. At the same time, the advance of theory depends on a steady infusion of new measurements. Thus, the interactions between the development of new concepts and volcanic data collection and analysis are inextricably intertwined. Most of this report focuses on the daily activities of the VHP, what may be referred to as the “operational” component of its mission. In this chapter, the committee begins by briefly examining the changing contributions of VHP scientists to the development and promulgation of the theoretical framework of volcanology. Hazard assessment, the operational activity that most clearly connects with research, is then explored in some depth. RESEARCH It is difficult to separate the contributions to basic volcanological knowledge made by VHP scientists from those made by their colleagues in other parts of the USGS, other government agencies, universities, other countries, and the private sector. Nonetheless, throughout much of the second half of the twentieth century, members of the present day USGS Volcano Hazards Program were national if not global leaders in the formulation of ideas about how volcanoes work. Building upon a steady stream of fresh observations from the HVO, VHP personnel advanced the understanding of the roles and significance of earthquakes, deformation, explosivity, gases, and lava flow mechanics in the evolution of ocean island volcanoes. Other USGS scientists conducted long-term

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program studies of individual volcanoes in the western United States, especially the Cascades Range. Many of these studies were carried out for purposes other than hazard assessment, such as the identification of geothermal and mineral resources and the geologic mapping of wilderness areas and national parks, but many findings were applicable to VHP goals. The VHP has defined its scientific priorities for the next five years by the fundamental questions that must be answered to fulfill the program mission of effective mitigation and useful warnings (USGS, 1997): Where are potentially high-hazard volcanic areas? Where is volcanic unrest occurring and in what manner? Is a restless volcano going to erupt? When? How long and how dangerous will the eruption be? How will eruptive style change over time? How can the potential for short- and long-term volcano hazards potential best be communicated? In FY 1999, the VHP funded more than 60 science projects, most of which directly or indirectly addressed these questions. The committee did not review the individual VHP research projects nor did it conduct an in-depth assessment of the research component of the program. It started with an awareness of the outstanding reputation of much of the research carried out by the VHP. However, the committee heard several anecdotes about longstanding research projects that have questionable connections to the primary mission of the program. The committee feels strongly that USGS management must ensure that most, if not all, basic research projects are directed toward the above three priorities. Such assurance can come from stronger internal USGS programmatic oversight and from careful structuring and enforcement of the annual performance plans of individual research scientists. This oversight should include subjecting proposals for research projects to external peer review. Additionally, as emphasized elsewhere in this report, prompt publication of research findings is essential. The committee is aware of important research findings that have languished for years (or even decades) without being published. These problem situations must be addressed and solved. One of the most important long-range issues that the VHP must face is deciding how central in-house basic research will be to its mission in

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program the future. Such research is also being done at universities, government laboratories, and non-U.S. institutions. Thus, one might argue that the VHP could forgo its basic research activities without having a major impact on the state of knowledge of volcanic processes. On the other hand, eliminating this program element altogether would likely damage the intellectual vitality of the VHP and make it more difficult (if not impossible) for the program to hire top-flight young scientists. Furthermore, because most USGS scientists lack some of the other commitments of their academic counterparts, such as teaching and grant writing, they may be better able to pursue long-term research projects than university faculty and students. The committee believes that if the VHP is faced with continuing budget shortfalls, it could elect to reduce fundamental research activities and redirect scarce resources to monitoring and crisis response functions, which it is uniquely positioned to do (see Chapters 3 and 4). However, these savings would come at a high cost. The ability of the VHP to respond to volcanic crises would be compromised by a lack of expertise in hazard assessment or volcano process studies. One possible solution would be for VHP members to collaborate more on research projects with scientists outside the USGS, particularly those from universities and laboratories of other government agencies. More active collaborations, coupled with an extramural grant program for academic researchers overseen partly or completely by the VHP, would help ensure that more investigations that are directly relevant to the program’s mission would be carried out (see Chapter 5). In addition, a more proactive and vigorous approach to retraining of existing personnel could help maintain the breadth of expertise needed to understand and respond to volcanic behavior. HAZARD ASSESSMENT Most VHP scientists and technicians spend the majority of their time in operational activities that lie within continuum of assessment, monitoring, and crisis response. This section considers hazard assessment, which forms a foundation for the other two operational components. The committee first describes what volcano hazard assessment is and why it falls within the purview of the USGS. Next the different types of volcanic hazards are categorized. A status report on

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program hazard assessment within the VHP is then presented, focusing first on three important approaches (mapping and dating, theoretical modeling, probabilistic methods) and then on how the different observatories carry out these responsibilities. Finally the committee considers the future of assessment within the VHP and recommends that greater emphasis be placed on prioritization, collaboration, and consistent data archiving in order to help the program carry out its mandate more effectively and economically. What Is Volcano Hazard Assessment? Volcano hazard assessment aims to determine where and when future volcano hazards will occur and their potential severity. This kind of appraisal provides a long-term view of the locations and probabilities of large-scale eruptions and related phenomena such as volcanic debris avalanches and tsunamis. The boundaries between hazard assessment and basic volcanological research are indistinct. Maps of volcanic deposits can be used either to reconstruct the history of a particular volcano or to decipher eruption processes that occur at many volcanoes. Stratigraphic sequences of ash deposits can similarly reveal how one volcano has behaved, or they can provide a basis for comparison across an entire volcanic arc, leading to the discovery of fundamental principles. Volcano hazard assessment involves a combination of three methodologies. First, volcanic deposits are recognized and mapped, and the associated materials are dated to provide a chronology of the related eruptive events. Second, laboratory and numerical simulations of the physical and chemical processes that initiate volcanic hazards are used to better understand and constrain their underlying causes. Third, information from both field-based studies and simulations are combined to make statistical assessments of the probability of future events. This work relies on one of the fundamental axioms of geology: the past is the key to the present and future. In other words, volcanoes with the most eventful history of activity are those most likely to erupt again in the future, and the style and nature of past eruptions are the best guides to future behavior.

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program Why Does the USGS Do Volcano Hazard Assessment? Assessment is integral to the mission of the VHP. Under the Stafford Act (Public Law 93–288), the USGS has the responsibility to issue timely warnings of potential geologic hazards to the affected populace and civil authorities. A narrow interpretation of this law might restrict the USGS to monitoring only those volcanoes showing outward signs of an imminent eruption, ignoring assessment altogether. This temptation could come from the twin desires to save money and to emphasize only the most obvious needs associated with volcanic hazards. However, assessment is an essential complement to monitoring: it provides an important means of prioritizing which volcanoes should be monitored and which types of data should be collected. It helps determine the potential magnitude of imminent volcanic hazards and allows the public to be educated about the likely consequences of volcanic activity. What Hazards Are Assessed? Volcanic hazards are highly varied in nature, frequency, size, area of impact, and complexity (Figure 2.1). There are two basic types of eruptions: (1) effusive, which generate lava flows and domes, and (2) explosive, which produce mixtures of ash, blocks, and gas, known as pyroclastic flows, capable of traveling large distances at great speeds. In the presence of water, from groundwater, precipitation, lakes, streams, or melting glaciers, ash and other loose deposits may become mobilized into highly destructive debris flows, known as lahars. Ash, consisting of small volcanic particles, can be thrown kilometers above the volcano during explosive eruptions and may be carried as far as hundreds of kilometers downwind. When ash settles out of the atmosphere it leaves thick deposits whose weight may cause the roofs of buildings to collapse. Ash deposits can also seriously alter drainage patterns and sediment loads, leading to widespread flooding. Volcanoes themselves tend to be unstable structures, occasionally collapsing to form landslides and, rarely, massive volcanic debris avalanches. Debris avalanches can bury extensive areas or, when they enter bodies of water, lead to large water waves or tsunamis. Gases escaping from subsurface magma can lead to respiratory distress or death in humans and other animals and may destroy forests and crops.

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program The potential impact of eruptions is not restricted to destruction on land (Figure 2.2). One of the greatest and least appreciated volcanic hazards is the threat that ash clouds pose to aircraft. Nearly all nonstop airline routes from North America to Asia pass over (or short distances downwind from) tens of potentially active volcanoes in Alaska and Russia. Ash may cause engine failure even when it is too fine grained to be visible or detectable by airborne radar. Not all volcano hazards are related directly to eruptions. Damaging mass movements can take place without any magmatic discharge, which occurred in 1888 when a large part of Mount Bandai, Japan, collapsed to form a debris avalanche. Giant submarine landslides have formed by partial collapse of nearly all of the Hawaiian volcanoes. One of these, originating from the flanks of Mauna Loa, generated a tsunami that washed more than 200 meters up of the slope of the neighboring island of Lanai. Damaging earthquakes may accompany the underground movement of magma, even if molten material does not erupt at the earth’s surface. The extensive range of hazards that must be evaluated requires the combined knowledge of a broad array of scientists, including geologists, geophysicists, hydrologists, geotechnical engineers, atmospheric physicists, and statisticians. Because assessment is inherently interdisciplinary, the VHP needs access to a diverse set of expertise, either within its own ranks or through collaborations with outside groups. What Is the Status of Assessment Within the VHP? Traditionally, volcano hazard assessment within the VHP focused on field-based and geochronological studies of individual volcanoes. Initially geochemists and petrologists from the GD undertook many of these investigations for purposes other than analysis of hazards. Since the eruption of Mount St. Helens in 1980 and the creation of the CVO, hydrologists, geomorphologists, and sedimentologists from the WRD have also taken part in these appraisals, incorporating laboratory and numerical simulations and field studies of the processes that lead to debris flows and steam explosions. Creation of the AVO in the early 1990s allowed glaciologists, atmospheric scientists, and remote sensing specialists to play a role as well. Because the VDAP team can be called on to assist with volcanic crises anywhere in the world, it generally has

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program Figure 2.1 Simplified sketch of a volcano typical of those found in the western United States showing a variety of hazards associated with volcanoes (USGS, 1998). Graphic designed by Sara Boore, Bobbie Myers, and Susan Mayfield.

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program Figure 2.2 Impacts of volcanic eruptions.

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program had to rely on assessments carried out by organizations other than the VHP. Outside the USGS, assessment has tended to include a growing reliance on physical and probabilistic models. The following material describes some of the methods used in hazard assessment and their status within the VHP. Mapping and Dating Geologic mapping, stratigraphy, geochronology, and physical volcanology provide the backbone of volcanic hazard assessments by revealing trends in eruption timing, volume, and explosivity. Historically, the USGS has done an excellent job of incorporating these types of geologic data into its assessments. For example, insight gained from the geologic mapping of Mount Pinatubo, coupled with knowledge of magmatic and eruption processes, increased the accuracy of forecasts by the USGS and the Philippine Institute of Volcanology and Seismology (PHIVOLCS) and arguably reduced loss of life when this volcano erupted violently. Many USGS scientists first observed massive debris avalanches during the 1980 eruption of Mount St. Helens. Although not widely appreciated prior to this event, the potential for such hazards is now acknowledged in assessments for Mount Baker, Mount Rainier, and similar volcanoes. The committee commends VHP efforts to integrate findings of geologic studies into volcanic hazard assessments. An ongoing challenge is to quantify geologic data more effectively in ways that optimize their use in such assessments. The VHP has also benefited from having outstanding in-house instrumentation and the professional expertise needed for various kinds of age dating of geological materials. USGS management will have to evaluate whether it can continue to afford the associated high expense or whether it would be more appropriate to contract out such services to the private sector. Modeling Although mapping and dating of volcanic deposits can provide a good framework for hazard assessment, mechanical models of physical, chemical, and hydrologic processes help refine forecasts of the types and magnitudes of future eruptions. Both numerical models and laboratory

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program simulations can relate the boundary conditions on a volcano to the likely consequences of any incipient eruptive activity. To date, this modeling work has focused on landslides, debris flows, lava flows, and various types of explosive eruptions. For instance, results of large-scale flume experiments carried out by CVO scientists at a unique facility in Oregon (e.g., Iverson, 1997) now allow them to calculate how far downstream and how rapidly a lahar will travel if it is triggered by the melting of a glacier perched above a specific volcanic drainage. Another approach, the so-called energy-line model (e.g., Malin and Sheridan, 1982), can determine the area likely to be covered by the fallout from an explosive eruption column based on the observed height of the plume. Recent laboratory and field-based studies of lava domes relate the textural and structural patterns observed on their active surfaces to the likelihood that they will explode (e.g., Fink and Griffiths, 1998). Although there has been some VHP participation in the development of these models, especially those related to hydrologic and sedimentologic phenomena, most have been created by non-USGS scientists. VHP hazard assessments generally do not incorporate the latest of these methods. Besides potentially limiting the scope of such assessments, the lack of a strong theoretical focus within the VHP makes it more difficult to test these models thoroughly against real-time eruption data collected by the program’s scientists. The committee encourages the VHP to include more theoretical modeling of volcanic phenomena in its hazard assessments. Probabilistic Hazard Assessments Because it is impossible to predict eruptive behavior with certainty, particularly for dormant volcanoes, most hazard assessments are inherently probabilistic in nature. For example, analysis of stratigraphic and radiogenic data may suggest that a given volcano has a 30 percent probability of erupting explosively in the next 50 years. Such a statement does not provide all of the detailed information needed by preparedness officials, which is where conditional probabilities become useful. For instance, if an explosive eruption occurs, what is the probability that ash accumulation will exceed some threshold? This approach enables volcanologists to consider a complicated series of events discretely or to estimate hazards based on empirical or subjective information that in

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program practice may be incomplete or poorly understood. As another example, volcanic unrest often appears to build exponentially, and estimates of the probability of future eruptions may be made based on this kind of trend. However, these estimates must be conditioned by the likelihood that the exponential model is correct. Use of these three approaches to hazard assessment—mapping and dating, theoretical modeling and probability calculations—by the VHP reflects the training of its participants. Most of the GD scientists have backgrounds in petrology, geochemistry, and field mapping and thus are most familiar with the more traditional stratigraphic and geochronologic methods. Many of the WRD scientists are experienced with simulations and mechanical modeling, which explains why some of the finest theoretical explanations of debris flows and sediment transport have been carried out by members of the VHP. Probabilistic approaches are relatively recent additions to the VHP assessment repertoire, but they are receiving more attention lately because of their obvious utility in communicating with civil defense authorities and the general public. The committee strongly encourages the VHP to develop a balanced assessment program that takes advantage of the full range of techniques available to volcanologists today. The State of Volcano Hazard Assessment at USGS Observatories Assessment priorities vary from observatory to observatory, reflecting local differences in the nature of the volcanic hazard and the expertise of resident scientists and technicians. Here the committee offers a description of the state of volcanic hazard assessment in each of the regions overseen by the VHP’s observatories. Alaska Volcano Observatory In Hawaii and the Cascades, threats to nearby population centers are the focus of most volcanic assessment. By contrast, Alaska has small population clusters around its volcanoes. Thus, the most serious hazards associated with Alaskan eruptions are those that occur at a distance from the eruptive site. Tsunamis generated by volcanic landslides and earthquakes can potentially affect coastal areas throughout the Gulf of

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program Alaska, as well as elsewhere around the Pacific Rim. Additional threats result when eruptions melt glaciers that then generate debris flows capable of destroying various remote installations, such as logging and fish-processing facilities. Alaska has a rich and diverse fauna and flora. Species preservation plans, which now are partly the responsibility of the USGS (although not the VHP), may benefit from volcano hazard assessments. Volcanic ash interaction with jet aircraft poses the greatest danger from Alaskan volcanoes, because ingestion of ash can result in engine damage or failure. On average, approximately four volcanoes per year erupt ash clouds of sufficient height and volume to endanger aircraft in the heavily traveled North Pacific air corridor. Although responsibilities for monitoring and crisis response in Alaska are shared among the VHP, the National Weather Service (NWS), and the FAA, only AVO is capable of (1) establishing the historical context of future explosive eruptive activity, (2) providing advance warning of an impending eruption, and (3) conducting ground monitoring that can confirm an eruption is actually in progress. Because of the nature of these dangers, AVO has placed greater emphasis on monitoring and crisis response than on long-term hazard assessment. Only a few Alaskan volcanoes have even rudimentary hazard maps (Appendix C). The expense and logistical difficulties associated with access in Alaska preclude the kind of comprehensive mapping strategy carried out by CVO and HVO. There is ongoing debate within AVO and the VHP about the appropriate level of this sort of characterization. However, in order to fully understand and evaluate the risks from volcanoes to the flying public, better information about eruption frequencies and magnitudes is needed. Recent AVO-coordinated mapping campaigns at selected Alaskan volcanoes carried out by teams of USGS, other government, and university geoscientists have expanded the coverage of hazard assessment products. The committee concludes that basic yet rapid assessment of the eruptive histories of as many of the Aleutians volcanoes as possible is necessary to guide prioritization of the placement of instruments used to provide warnings to pilots and other nearby infrastructure. Mapping Aleutian volcanoes has potential benefits beyond hazards assessment and mitigation. The Aleutians constitute one of the most active volcanic arcs on earth. Joint geophysical, geological, and oceanographic campaigns have recently been proposed to improve

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program understanding of several different volcanic arcs. Although not the direct responsibility of AVO or the VHP, such studies could contribute significantly to the creation of a historical framework and better appreciation of eruptive activity in Alaska. Because the Aleutians are so active, they are important testing grounds for methods that will ultimately be applied in more populous areas. Hawaiian Volcano Observatory Although most visitors to Hawaii get the impression that its volcanoes erupt spectacularly but safely, HVO scientists have documented major volcanic hazards to both local and distant populations. The most common eruptive activity in Hawaii produces lava flows whose dangers are primarily to property (Figure 2.2(C)). Drifting clouds of noxious gas referred to as volcanic fog or “vog” represent an environmental health hazard in downwind areas. Less common, but more dangerous, phenomena are also well represented in the geologic record. Violent explosive eruptions in 1924 and 1790 generated pyroclastic surges and showered Kilauea’s summit area with large blocks. Recent studies have revealed many more such events than had previously been recognized. Sonar images of the seafloor adjacent to the Hawaiian Islands, collected and analyzed by VHP scientists in the 1980s, showed huge submarine landslide deposits that correlate with massive scars on the flanks of adjacent volcanoes. In at least some cases, these landslides must have been sudden and catastrophic, producing tsunamis that left marine deposits perched on the sides of nearby volcanoes, several hundred meters above sea level. These phenomena reflect a severe potential hazard for much of the Pacific Basin. Faced with this array of dangerous processes and products, HVO has to employ a complex assessment strategy consisting of mapping, dating, modeling, remote sensing, and estimating probabilities. Most but not all of this evaluation has been targeted at the island of Hawaii (Appendix C ), where volcanic activity is ongoing; and recent studies have been extended to Haleakala volcano on the island of Maui. Studies by HVO staff incorporating systematic mapping and radiometric age dating of lava flows and other deposits have created a comprehensive picture of the recent constructive history of the island. Based on this database,

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program HVO scientists have produced hazard zone maps that take into account the types of danger, the magnitude of typical events, and their frequency (Figure 2.3). These maps provide vital information, but the sharp boundaries separating hazards zones pose problems for civil defense authorities and insurance companies charged with interpreting their implications. Actual hazard potentials are smoothly varying continua and do not exhibit step functions like those portrayed by such maps. During periods of sustained eruption, Kilauea emits about 2,000 tons of sulfur dioxide gas each day (Sutton, et al., 1997). This air pollution causes respiratory problems and contaminates rainwater-catchment systems that provide drinking water to many residents. HVO staff members closely monitor the amount and composition of gas emissions and collect and integrate information on volcanic air pollution from a variety of sources. They work closely with government officials and health professionals who inform residents and visitors about this hazard. HVO faces additional assessment complications for those volcanoes that remain capable of erupting but have not been active in the past century (Hualalai, Haleakala, Mauna Kea). Recent mapping of these dormant volcanoes has clarified their hazards (Appendix C). Part of HVO’s challenge is to convey this information to emergency managers and other public officials despite the widespread public perception that all volcanic activity is localized on Kilauea and Mauna Loa. Cascades Volcano Observatory The Cascades Volcano Observatory is responsible for assessing and monitoring the hazards of the volcanoes of the Cascades Range, which stretches from British Columbia to northern California (Figure 1.1). Two features of Cascades volcanoes most affect the assessment work of CVO: (1) many of them are located near major population centers; and (2) they typically lie dormant for decades, centuries, or millennia before returning to activity. These features mean that much of the work of CVO scientists involves public education about risks posed by seemingly benign mountains. Fortunately, widespread publicity about the Mount St. Helens eruption in 1980 raised general awareness about volcanic dangers. However, as time goes on with no nearby activity, this appreciation will fade. Because of the large number of volcanoes that could potentially erupt and the diversity of their eruption styles, CVO geologists have spent

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program Figure 2.3 Hazards map for lava flows at Kilauea, Hawaii. Relative hazards range from 1 (high) to 5 (low). Lava flows erupted since 1823, gray; principal subdivisions, dark gray; boundary between Kilauea and Mauna Loa flows, heavy black line (USGS, 1992). considerable effort mapping, dating, and collating information about the volcanic deposits of the region. Cascades Range volcanoes have erupted about 50 times in the last 4,000 years, leaving substantial deposits. Seven of these eruptions took place in the past 200 years, and four affected areas far beyond the margins of the volcano. These frequencies suggest that there is approximately a 30 percent chance of an eruption in the region every 10 years and an 18 percent chance of an eruption whose influence extends beyond the base of the volcano. Like other volcanic chains formed by the descent of one convergent tectonic plate beneath another, the Cascades have many different types of

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program eruptions. Lava flows of highly varied chemistry, explosive blasts of ash and rock, massive flank collapses, and voluminous debris flows are all possible consequences when a Cascades volcano awakens. Several of the Cascades volcanoes have permanent glaciers, increasing the likelihood of dangerous debris flows. Even slight increases in hydrothermal activity near a volcano’s summit may enhance glacial melting and debris flow formation. Thick ash deposits from explosive eruptions can also modify drainage patterns and choke streams and rivers, resulting in increased flooding even at great distances from the volcano. Although the 1980 eruption of Mount St. Helens had a serious economic and ecological impact on its surroundings, its scale is dwarfed by the magnitude of prehistoric eruptions in the region, such as those at Mount Mazama (Crater Lake) (Figure 1.1). The two basic techniques used by CVO staff for assessing past Cascades activity—mapping and age dating—are much the same as those used in Hawaii. However, with a much larger number of volcanoes to evaluate, CVO staff members face more difficult prioritization issues than their colleagues at HVO. Over the past three decades, successive groups of USGS scientists have been involved with volcano hazard assessment in the Cascades (Appendix C). Before the establishment of CVO, individual USGS geologists based mainly in Menlo Park took on long-term mapping projects of the major Cascades stratovolcanoes. These comprehensive studies were aimed as much at basic understanding of magmatic processes and regional geologic history as at determining the likelihood and distribution of future volcanic hazards. Many of these studies lasted for more than a decade, and they resulted in a large number of refereed scientific publications as well as geologic maps. A second group, based at USGS-Denver, rapidly generated more focused reports and hazards maps that outlined the dangers of individual volcanoes. In the 1990s, CVO scientists adopted a more collaborative approach, including sedimentologists and fluvial geomorphologists to identify a wider range of past activity. In addition, they placed greater emphasis on more rapid assessments, to ensure that all potentially active volcanoes have at least a basic hazards map that can be updated periodically when improved data and methodologies become available. CVO is also trying to evaluate the many smaller and less prominent volcanic centers between the major cones (Appendix C).

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program Long Valley Observatory As a “virtual observatory” located in Menlo Park, California, and receiving monitoring data telemetered from Long Valley, the LVO has few resources available to devote to hazard assessment. However, Long Valley caldera and vicinity have been the subjects of several comprehensive mapping projects over the past 25 years. A priority for LVO staff has been to update these earlier reports and make them more consistent with assessments prepared by other parts of the VHP. Despite its relatively restricted geography, the area monitored by LVO contains evidence of a diverse set of volcanic hazards. The greatest of these would be a caldera-scale explosive eruption like the one that formed the present physiography about 730,000 years ago. Neither scientists nor modern society has ever witnessed such a process, so recognizing its precursors is especially difficult. Seismic unrest and rapid uplift of the caldera floor in the 1980s caused widespread concern that a major magmatic explosion might be imminent. This uplift coincided with similar events at Yellowstone National Park; outside the city of Naples, Italy; and at Rabaul Volcano in Papua New Guinea. When the activity at the first two of these three sites subsided uneventfully, it highlighted the limitations of our knowledge of these most violent of eruptive phenomena. Assessment of the potential for this type of activity must rely exclusively on interpretations of mapped relationships caused by prehistoric events. The most recent activity in the vicinity of Long Valley took place along the Inyo-Mono Crater chain within the past few hundred years. These events included the formation of several lava domes and explosive products. Geologic mapping suggests that the activity, fed by one or more dike-like intrusions, stretched for 11 and perhaps 25 km. Recognition in the late 1980s that potentially simultaneous eruptions could extend such great distances led to the construction of an escape road out of the town of Mammoth Lakes. Earlier mapping also revealed numerous young basaltic lava flows in the area. Other Areas The VHP may create other “virtual” observatories in the future in response to renewed activity at other volcanic centers. One such example

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program is at Yellowstone National Park where, in the past several years, enhanced monitoring has been combined with earlier geologic and geophysical studies carried out by USGS and academic scientists. Bringing the state of assessment at this and other potential eruption sites around the western United States to a high standard requires access to as much compiled geological and geophysical information as possible. Thus, the scientist in charge of each observatory must strongly encourage all scientists to publish their results in a timely fashion. Future of Hazard Assessment The VHP’s four volcano observatories have developed different approaches to hazard assessment that derive chiefly from differences in their histories and hazards. HVO is gaining excellent knowledge of the geologic history of those Hawaiian volcanoes most likely to endanger significant population centers. CVO is responsible for some volcanoes, such as Mount St. Helens and Mount Lassen, whose hazards are relatively well understood and others, such as Mount Baker and Mount Hood, for which documentation is much less complete (Appendix C). The volcanoes that AVO monitors are, in general, the least well known. These differences in baseline knowledge influence the types of data collection carried out by each observatory. HVO and LVO have fairly complete frameworks in which to insert newly mapped relationships. CVO oversees a mixture of both well-known and obscure volcanic centers. AVO is still in a mode of basic mapping and data collection. If the VHP continues to be faced with a flat budget, it must find ways to carry out its mission more efficiently. The committee recommends that the VHP initiate a form of collaborative prioritization with respect to hazard assessment (Sidebar 2.1). This might include a broader application of the team approach now being used at AVO and CVO. It would require strong leadership from both VHP administrators and the scientist in charge of each observatory to ensure that program priorities are set and maintained and that the desires of individual scientists do not drive the program. In other disciplines and branches of the federal government, a demonstrated ability to prioritize is often rewarded with increased funding. Because volcanology is inherently multidisciplinary, volcano hazard assessment could provide the VHP and USGS with a flagship example of collaborative science.

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program The Decade Volcano Program, set up by the International Association of Volcanology and Chemistry of the Earth’s Interior (IAVCEI) as part of the United Nation’s International Decade of Natural Disaster Reduction, provided an instructive illustration of how collaborative prioritization can work in volcano hazard assessment. Fifteen volcanoes around the world were selected for comprehensive study on the basis of the size of the population at risk and the style of volcanic activity. Although the efficacy of the program was hampered by a lack of funding, in the most successful cases international teams of government and academic researchers carried out a coordinated regime of mapping, monitoring, and public education. Particularly noteworthy were the successes at Mt. Rainier, one of the “Decade Volcanoes,” upon which scientists from the USGS; other federal, state, and county agencies; and a number of universities focused attention. Significant new insights were gained about the processes and dangers associated with these volcanoes. Importantly, the study of other volcanoes was postponed while these concerted campaigns were carried out. In some SIDEBAR 2.1 Collaborative Prioritization Collaborative prioritization occurs to varying degrees in many scientific disciplines. Astronomers, particle physicists, marine geoscientists, and planetary geologists have been forced into this mode by limited access to crucial facilities (telescopes, accelerators, research vessels, and spacecraft). Because of the large scale of these projects, overall programmatic funding levels tend to be higher, but grants to individual investigators may be smaller. In recent years, branches of science that have traditionally been less “high tech” have moved toward the collaborative approach. Field-based ecologists, geographers, and chemists have played key roles in the National Science Foundation (NSF) sponsored Long Term Ecological Research (LTER) program. Well-coordinated teams of life, earth, and physical scientists, in some cases supplemented by social scientists, study 22 LTER sites. Frequent meetings and widespread use of electronic communications ensure that individual projects are integrated and directed toward some common research goals. Another large NSF-supported program oversees the Science and Technology Centers, which are typically funded at $4 million to 5 million per year for up to 11 years. These cover a wide range of disciplines including earth science (e.g., Southern California Earthquake Center; Center for High Pressure Research).

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program cases, committees of scientists and public officials collectively decided what lines of research should be conducted and by whom. In addition to prioritization, volcano hazard assessment within the VHP would be improved by greater consistency of data collection, storage, presentation, and interpretation. For instance, hazard maps are not prepared in a single easily understood format. Terms such as “high risk” and “low risk” mean different things to different scientists within the VHP and to members of the public. This lack of consistency also is found in VHP hazard maps. In general, communities accept hazards that could result in devastation of property and loss of life with annual probabilities of less than one in one million, but require mitigation strategies in cases where the probability is greater than one in one thousand. Between these values, decision making is more complicated and may involve questions about how much effort should be devoted to the mitigation. Some individual users may tolerate a completely different range of hazards. For example, a higher hazard level is acceptable for many structures and roads, such as those found in national parks, because such areas are easily evacuated during times of volcanic unrest. On the other hand, critical facilities, such as nuclear power plants, dams, and other large, expensive structures, generally require lower annual probabilities because damage to these facilities cannot be mitigated by evacuation and may greatly compound the disaster. SUMMARY Basic research in the VHP, although reasonably well integrated, is being threatened by budgetary and personnel constraints, that may diminish the program’s ability to meet appropriate scientific goals. If these problems are not solved, the program will likely be forced to reduce levels of in-house basic research and/or to increase collaboration with non-USGS scientists. Hazard assessments, although traditionally strong in geologic mapping, radiometric age dating, and related activities, must be strengthened in modeling and probabilistic approaches, if the program is to continue to meet appropriate scientific goals. Existing hazard assessment activities at individual volcano observatories are effectively integrated and applied to hazard mitigation issues. The one-volcano, one-scientist projects currently under way, although scientifically

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program appropriate, may not be effectively integrated with other studies or with the VHP as a whole.

OCR for page 29
Review of the U.S. Geological Survey’s Volcano Hazards Program This page intentionally left blank.